MARKETING BOARD TAXATION. ..~ 1 ;  
OFNIGERIANAGRICULTURE;_,  . ;

SOME IMPLICATIONS ;

Thesis for 'the Degree of «Ph .. D
MICHIGAN STATE UNiVERSlTY

FRANCIS SULEMANU LDACHABA .

1972

 

 

 
  

LID RA R Y
Michigan State
University

    
   

This is to certify that the

thesis entitled

Marketing Board Taxation
of Nigerian Agriculture:
Some Implications

presented by

Francis Sulemanu Idachaba

has been accepted towards fulfillment
of the requirements for

Ph.D. fiegreein Agricultural Economics

y/ZZWILfil/WW I
Majorp/Cesso [

Date ”(/é/7Z

0-7839

 

  
 
 

 
  
  

"'6‘"; " —.—_’ 1
HOAG'lrSUI'S'"
TECC‘K BZN'J‘P‘.’ INC
. LIBRARY BlNDKRS

I 3 1293 10447 8023

We»?
twig/aw

 

 

r T a
T C» “It MHUTI‘HMl M a”; 3 Turn.» u. ..
IF {11,13‘ . y T .1» I .1.» 1H-
l T mllnMHU?» I: \t-l T.‘ €241
-. :3
t .
K _
7“..-
Tu¥_a..‘
- . _ ..2
.io .

 

 

1 O

     

I
l O
. u

.1 Mb stud

‘fbr alvw

ind \‘r

'.O

   
   
    
 
  
 

'h;‘1euune "
V . .i‘éunta f
3 MI pr . 2.9" ,,

V_ 2” exam.“ , ~: --,A ;

Item‘s. arm
m 11".»11129 .fa-

3" - '.. ..
,7‘ ,. a v6: (homop- v.7

* 5$§39 '_-{’ fir‘rrdh'r 1" i . _‘»,‘-':4-..- \a

4 ‘ .

‘eitre: 1" Mann! 1:? sacrpcty- , p.
.. * extends: r, accaunn by *he mfrmc-nt'ai
_’t|...

'Mi‘mtion and fiu}\l.qu~ I'lZ‘IQ-s, the.

 

" Went reunites as; changes in (are;

.w,
scum-2y eaaea‘ Transmit predicted
cub:- '. . c- a (C: 1513.; 5,6531;

‘. -.~r..

 

 

' r
,4. gr" ’

ABSTRACT

 
  
   
  
      
 

‘; W MARKETING BOARD TAXATION OF NIGERIAN
Fr" ' AGRICULTURE: SOME IMPLICATIONS

BY

Francis Sulemanu Idachaba

\

r:.: 'A' ‘H -
r». at” This study had three main aims: first to prov1de

s? I V;_ .
3: “model was developed to explain the effects of

"'5 6n the scale of farming in the Northern States

i’fias extended to account for the differential
fissduisition and salvessjfélyssr the
ignitestment responses to Changes in taxes
fiftifrfered,‘ in many cases, from those predicted

v ,

3“
.__
1:;me , -
.J‘

 

"“3"_ ‘L

 

r»; ra-zclassical mO‘

. . F
“. +‘ A:
'.e:;.acqu151..cn a

1
. - 1. u
3;. anneal e v e

Ilsa-5.5: by r 3;:

1: |.IVF.‘~9"S arr:

‘
It. .ll'v: u-‘bbov A.“

.';:;a. quantities
I A.' . p ' .

~ . . -
OIO‘H'R‘F ‘“‘~"
5"...Ei bU ELLOOLOKLLI

related to ”e

5‘.

.
"n n
“.5 '90.... 4°

.m-Nus a re
3’35 ‘0 “Pette

Sales 53:

b: ;
I... 95:.2Ated b.
': Qt.‘

u “.613 CE grO‘J

5.33:.“
-"‘ tG the r

-:
-.1
it J:.es
to tuE

*«21
Q‘ “a?
‘3‘5 ‘}b\
U‘ ‘vu
i“ _C.
‘
P.
“:33. 5‘
V‘t CB‘
' Q
v-
I
h
v.
u“?~
.
§
‘o “s
‘33 cr‘
n.“ “_
H r

 

Francis Sulemanu Idachaba

by the neoclassical model that fails to differentiate
between acquisition and salvage values for durable inputs.
At the empirical level, these taxes induced cotton and
groundnut farmers in the Northern States (1950—66) to use
less labor and land than would have been the case in the
absence of these taxes. These adverse effects operate at
two levels: by reducing the rate of entry of new resources
and investments and by increasing the rate of exit of
initial quantities on hand. It is shown, within a Cobb-
Douglas framework, that compensating subsidies on inputs
required to minimize the allocative distortions of taxes
are related to the output elasticities of these inputs.
This provides a rational basis for relating subsidies on
inputs to expected tax rates on marketing board crops.
Sales supply equations for cotton and groundnuts
were estimated by ordinary least squares. The elasticities
of sales of groundnuts and cotton (to the board) with
respect to the marketing board's producer price were
estimated to be 1.3139 and .9028 respectively. Predictions
of sales to the board were made assuming the board paid
farmers higher prices than they actually received. These
predicted sales exceeded actual sales whenever the hypo-
thetical prices were higher than actual prices. These
larger volumes of sales together with the implied lower
taxes on cotton and groundnuts were used to estimate the
(crop tax) revenue impact of paying farmers higher prices.

It was found that in many years in the 1950's and 1960's,

"‘ .I'fl' I.

 

.. We: prices W0!

r'
cu: ‘ .u u
a" c

mu; ‘i‘hOV‘t SUbSt a”

at, 11150133 Years:

101‘

2.575.216 coupled w:

I"...
I
«'LOIT‘I iFACCT-le. fiat.

ffect on s:
'12! :rzces has n
M511 sales t
2::reated as H.212
3.3::- ransom v :ia':
:iarces and ccva:

2525 Market mg 5:

fin.’ ‘ l“‘\fi

'w-:¢Uu-&Cus from

in
'3?- ! venues are

u. “.3 SPA!“

u.“ .that t

\‘ O;
I "A"

'-~€ ‘qvierfi‘e e‘t ]

o;

:39”; .
““‘zatlon w

-‘
l "
.. “’1

.
‘ ‘
Q 530'n'le'im
i‘
‘

L41; the s ale S

2-.15is Sign nals

 

Francis Sulemanu Idachaba

these higher prices would have increased farmers' monetary
income without substantially effecting tax revenue from the
crops. In some years, there were substantial reductions in
tax revenue coupled with substantial increases in farmers'
monetary income. This Offset to tax revenue loss from the
positive effect on sales to the board of paying farmers
higher prices has not been analyzed by previous writers.
When sales to the board of marketing board crops
are treated as random variables, tax revenues from them
become random variables. It is shOWn with evidence on
variances and covariances of sales Of crops to the Northern
States Marketing Board that grounds for expecting revenue
contributions from these boards to stabilize State Govern-
ment revenues are very tenuous indeed. Using budget data,
it is shown that this reliance did in fact destabilize
State Government revenues (1956-66) and that the degree of
destabilization was consistent with our expectations from
apriori knowledge of the supply and demand factors influ-
encing the sales of these crops to the boards. The
analysis signals the danger of relying on revenue contri-
butions from these boards particularly for planned capital

expenditures of the States.

 

mmmc BOP-.1

AA

AGRICULT'F.

Franc1;

 

 

in h . Mic
Hartlal 4

Part:

MARKETING BOARD TAXATION OF NIGERIAN

AGRICULTURE: SOME IMPLICATIONS

BY

Francis Sulemanu Idachaba

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Agriculture Economics

1972

 

Dedicated to

Aya

who made it all possible.

ii

 

I an very grate
f;:=.::;a‘.1y S‘JPP‘Orting
32:95. I "‘ particul.
webeer. very COOPera'

Hy ind btemes
’2;::?rcfessor and Th
22%;. His intellectu
:erszrel encouragement
3222‘s; infl

. uenced my

me Ynivers ity . I t

 

: gratitude to him.

I
l”“ R

. A
-""55V:

5 Lester V.

W
“1’ greatest

5..) .
unit‘zers ‘

«811 the

v .ukiy fruSQ’rat
3:14,“ ‘
-~.e girls -

  

 

  

ACKNOWLEDGMENTS

I am very grateful to the Rockefeller Foundation for
financially supporting my graduate studies in the United
States. I am particularly grateful to its officials who
have been very cooperative and nice in every way.

My indebtedness to Professor Glenn L. Johnson, my
Major: Professor and Thesis Supervisor, extends beyond this
StUdY. His intellectual stimulation, incisive criticism,

Personal encouragement and his confidence in me have
greatly influenced my entire graduate program at Michigan
State University. It is hard for me to adequately express
my gratitude to him. I want to thank the following:
Professors Lester V. Manderscheid and George K. Dike who
read earlier drafts of the thesis; Professor Robert L.
Gustafson who was very helpful in the early stages of this
study and to Professor Carl K. Eicher who was generally
helpful.

My greatest debt is to my wife, Aye, on whose
shoulders fell the responsibility of looking after a
freq"lently frustrated man and also of looking after our
two little girls, Achenyo and Qmada. I want to thank them

fOr
their patience and understanding.

 

 

 

.... n- no V" "
i ' n- " - .
.... I“ casual- O .

'Izfilfl'

‘ q”! rfiy'flmf \!
n $."up-JJUG‘LO.‘ o

The Role of A:
Economy. .
Farketing Boa
State Cover
Official Atti
of Marketir
Objectives of
The Significa
Plan of This
Footnotes.

 

i-iaE'fl‘ OF FRET-1':

Xarketing g.
50! State
Petemial p

u. TFE Erh'V‘I-h
t:t,'\"s C
Er“? m
JnA\D LA‘XNLI
." nu .'
$.‘PYV‘IS 0‘; '
V1,- ‘
{mime 1“
Ti
Grit . ‘
“ Pats
POOd Crr
es:

 

TABLE OF CONTENTS

L2,“.

“IX" 1‘ JIZT OF TABIIES - s o s o o o o o o c o

 
 
  
 
   
  
 
 
   
   
 
   
  
  
   
 
 
 
 
  

INTRODUCTION . . . . . . . . . .

The Role of Agriculture in the Nigerian
Economy. . .
Marketing Boards, the Export Sector and
State Government Revenues. . . . .
Official Attitudes on the Revenue Role

of Marketing Boards. . . . . . .
Objectives of This Study. . .
The Significance of These Problems . .
Plan of This Study. . . . . . .
.frootnotes. . . . . . . . . . .

REVIEW OF PREVIOUS STUDIES. . . .

Marketing Boards as Revenue Collectors

for State Governments . . .
Potential Producer Income and Potential
State Government Revenue . . . . .
'Footnotes. . . . . . . . . . .

THEEPEECTS OF GOVERNMENT AND MARKETING

' BOARD TAXATION OF CROPS AND SUBSIDIES ON
INPUTS ON THE COMMITMENT OF RESOURCES TO
- EARNING IN THE NORTHERN STATES. . . .

V..;The Effects of Taxes on the Relative
‘ ,.0utputs of Groundnuts, Cotton and
Peed Crops. . . . . . . . . .

EfféCts of Taxes and Subsidies on
:yzquilibrium Amounts of Resources Used.

iv

 

23

28
32

34

35

42

 

 

 

 

 

 

Elasticities a:
Comparative St‘
Estimate of
Earketinq B:
3 Resource
The Bias in E:
Got-'errs'ent ;.
of Crops a:
Resource 115v
Production 2
Aggregation P
fie Implicit
Eq‘ail ibr in:
as Our Poin
Government a:
of Crops a:
'Second-Bes
Governent a:
0f Crops E
Resources

The Data.
S‘LTflafy and
Footnotes

 

3'. sexes or ceop

GO‘FD“VZT\‘T

b.““ Luc‘

mu Taxes

Interp‘cti‘
The Deters.
cFrom Tax
”ales to t

”vernflfi

From the
ReSource 1
GOVErnr

  

 

 

 

Chapter

IV.

\I

 

Elasticities and Their Limitations . .
Comparative Statics and Biases in Our
Estimate of Effects of Government and

Marketing Board Taxation and Subsidies

on Resource Use. . .
The Bias in Estimates of Effects of

Government and Marketing Board Taxation

of Crops and Subsidies on Inputs on
Resource Use Using a Cobb-Douglas
Production Function . . . . . .
Aggregation Problems. . .
The Implicit Assumption of Allocative
Equilibrium: Schultz's Hypothesis
as Our Point of Departure .
Government and Marketing Board Taxation
of Crops and Subsidies on Inputs: A
"Second-Best" Problem. . . . .
Government and Marketing Board Taxation
of Crops Expressed as Taxes on

Resources Used in Farming . . . .
The Data. . . . . . . .
Summary and Conclusions. . . . . .
Footnotes . . . . . . . . . .

SALES OF CROPS TO THE MARKETING BOARD AND

GOVERNMENT AND MARKETING BOARD REVENUE

FROM TAXES ON CROPS . . . . . . .

Introduction . . .

The Determinants of Government Revenue
From Taxation of Crops . . . .

Sales to the Marketing Board and

Government and Marketing Board Revenue

From the Taxation of Crops . .
Resource Fixity, Supply Response and

Government Revenue. . . . . . .
Summary and Conclusions. . . . . .
Footnotes 0 I I O C O O I I O

ESTIMATION OF IMPACT ON GOVERNMENT AND
MARKETING BOARD TAX REVENUE FROM GROUNDNUTS
AND COTTON OF CHANGES IN THE BOARD'S NET

PRODUCER PRICES FOR THESE CROPS . . .

Introduction . . . . . . . . .
The Model . . . . . . . . . .

Page

62

63

64
69

70

100
100

101

105
113

118
121

126

126
128

 

5:?
. wM-‘ied 34cc.
fl ntv‘é‘ A. .
amt Run “c..- !'
“e Long Ron 3“.
Er:.rical Pes-
Preiiction .
n C.
Errary ..d
Footnotes .
v ~ 'n" :v‘ :-
71. mama on. _..
-.-— NY'::
3? SmuE GU.-.

 

 

Introduction
Marketing Boa
'3‘. State Go
Total State G
Without Re‘:
Marketing E
Errary and C
Footnotes .

7.1. 521113.11 REID CZ

t7~§lications
Footnotes

1"... .Pfls‘ fi.’
nay-.JJ“_.:I:

 

 

  
 
 
  
 
 
 
 
 
 
 
 
 
  
  
 
 
 
   
 
 
 
 
   

M “fi-Mbdified Model . . . .
";.Short Run Modified Model .
4. The Long Run Modified Model
‘ Bupirical Results. . .
Prediction . . .

r. »Summary and Conclusions.
' - Footnotes . . . . .

_OE STATE GOVERNMENT REVENUES .

Introduction . . .

_ of State Government Revenue.
‘ Total State Government Revenues
’ Without Revenue Contributions
Marketing Board. . . . .
”Summary and Conclusions. . .
5:560tnotes . . . . . . .

WSUMMARY AND CONCLUSIONS . . .

” Implications of Findings . .
Fmtnote s O O O I I O '

With
From

:MARXETING BOARDS AS POTENTIAL STABILIZERS

‘Marketing Boards as Potential Stabilizers

and
the

Page

135
135
138
140
158
181
185

191
191
192
204
210
215
218

232
235

239

 

 

 

1.3

0‘.

vi".

' o’
Connection 0.

' 5% rr;
:ores..y, a..-
7‘ ‘.. 5 V“ _
:ro:uc.' 5“C(-

Total Value, at
Exports to t:
fizgeria, 195

Share of Agric
Forestry in '

1930-1968 .

The Contributrl
CICPS to St;
1956-1966 .

T0151 RQVEn‘Je
Rehmd to .
and Statute
NiSeria, 19

Ratios Of COf

‘0 COtton :

for lefen

1963/64-19

 

 

TC
Taxes on ;
GIOJDONW

68“»

ul'_

 

 

     

Table

1.1.

LIST OF TABLES

Page

Contribution of Agriculture, Livestock,
Forestry, and Fishing to Gross Domestic
Product, Nigeria, 1950-1968 . . . . . . 2

Total Value, and Contribution of Agricultural
Exports to the Agricultural Economy of
Nigeria, 1950-1968 . . . . . . . . . 4

Share of Agriculture, Livestock, and
Forestry in Total Exports, Nigeria,
1950—1968 I I I I I I I I I I I I 6

The Contribution of Taxes on Marketing Board
Crops to State Revenues, Nigeria,
1956-1966 0 O O I I I I O C I. I O 10

Total Revenue and Investment Expenditure
Related to Revenue From Marketing Boards
and Statutory Corporations, by States,
Nigeria, 1970-74. . . . . . . . . . 12

Ratios of Cotton Prices in the Western State
to Cotton Prices in the Northern States
for Different Grades of Cotton, Nigeria,
1963/64-1971/72 . . . . . . . . . . 26

Effects of Government and Marketing Board
Taxes on Relative Prices of Cotton and
Groundnuts Produced on Northern States
Farms, Nigeria, 1950-1965. . . . . . . 40

Effects of Taxes on Groundnuts on Real
(Relative) Prices of Land and Labor
Employed on Northern States Farms, Nigeria,
1950-1965 . . . . . . . . . . . . 58

Effects of Taxes on Cotton on Real (Relative)
Prices of Land and Labor Employed on
Northern States Farms, Nigeria, 1950-1966 . 59

 

 

I I
'I
nil

I
4".

w; y
. Fartinzer Cons ..

"Transating (Fe:

for a One Per
of a Marketing
the Quantity c
State Farms C:

States, 1930-1

 

Fertilizer Cons
Sigeria, 1966

Unit Tax Elasti
Producer Pric
Producer Pris

Cotton: Tax 9.
Respect to t'r
CCttOl‘. for [:3
Elasticities
iiorthern Sta

chun'inuts: '1
RESpect to
Price for D
1 , _ .

 

o ..
.reolcted Gr:
Board priCE

TC 1
1967 ' ‘ r

'11

r1
— ~ :7 't" (D
(1" C! '1 LL

 

 

 

 

3.4.

 

 

Compensating (Percentage) Subsidies on Labor
for a One Per Cent Decrease in Net Price
of a Marketing Board Crop so as to Keep
the Quantity of Labor Employed on Northern
State Farms Constant . . . . . . . . 78

Fertilizer Consumption in the Northern
States, 1950-1967 . . . . . . . . . 86

Fertilizer Consumption in Kano State,
Nigeria, 1960/61-1969/70 (Tons) . . . . 87

Unit Tax Elasticities for Cotton Using
Producer Prices, Taxes and Potential
Producer Price, Nigeria, 1950-1966 . . . 109

Cotton: Tax Revenue Elasticities With
Respect to the Board's Producer Price of
Cotton for Different Assumed Price
Elasticities of Sales to the Board,
Northern States, Nigeria, 1950-1966 . . . 110

Unit Tax Elasticities for Groundnuts Using
Producer Prices, Taxes and Potential
Producer Price, Nigeria, 1952-1964 . . . 111

Groundnuts: Tax Revenue Elasticities With
Respect to the Marketing Board's Producer
Price for Different Assumed Price
Elasticities of Sales to the Board,
Northern States, Nigeria, 1952-1964 . . . 112

Predicted Groundnut Sales Assuming Marketing
Board Prices at 55 Per Cent of World
Level, Northern States, Nigeria, 1950-
1967 . . . . . . . . . . . . . 160

Predicted Groundnut Sales Assuming Board
Prices at 55 Per Cent of World Level,
Northern States, Nigeria, 1950/51—1967/68 . 161

Predicted Groundnut Sales Assuming Board
Prices at 60 Per Cent of World Level,
Northern States, Nigeria, 1950/51-1967/58
(Based on Equation 6). . . . . . . . 162

Predicted Groundnut Sales Assuming Board
Prices at 60 Per Cent of World Level,
Northern States, Nigeria, 1950/51-1967/68
(Based on Equation 7). . . . . . . . 163

viii

 

 

”fps-v;p~vb
J.Jdu.44enb

vs ow. _
*ucaa..y a

'I III R \\ .
.3X 16'! The.

‘
Prawnnnwfi
JQVIAA‘IOAIU

u.’ .. ‘1 .
b.dualAY 6

LL .oX Revenu

in,

'1.)
'4
(D
‘34
’J
(I
(f
(D

3~~ Predicted

oard P:
Rorthery

I ‘1 _. “
a“. «3X “Eve“!
From a;

Viduall

 

 

$5.

5.11.

5.12.

Predicted Groundnut Sales Assuming Board
Price at PN =max (PN 6Pw) where PNj is
er Eric

the Actual roduc e and Pw is the
World Price, Northern States, Nigeria,
1950/51-1967/68 . . . . . . . . .

Tax Revenue: Estimated Changes From Higher
Groundnut Prices, Larger Volume, Indi-
vidually and in Total, Nigeria, 1950—1965*

Tax Revenue: Estimated Changes From Higher
Groundnut Prices, Larger Volume, Indi-

vidually and in Total, Nigeria, 1950—1965**.

Tax Revenue From Groundnuts: Estimated
Changes From Higher Prices, Larger Volume,
Individually and in Total, Nigeria,
1950-1965 . . . . . . . . . . .

Predicted Cotton Sales Assuming Marketing
Board Prices at 85 Per Cent of Potential
Level, Northern States, Nigeria, 1950/51-
1967/68. . . . . . . . . . . .

Predicted Cotton Sales Assuming Marketing
Board Prices at PN-=max (PNj, .85P ),
Northern States, NIgeria, 1950—196 . .

Tax Revenue From Cotton: Estimated Changes
From Higher Prices, Larger Volume, Indi-
vidually and in Total, Nigeria, 1950-1967

Tax Revenue From Cotton: Estimated Changes
From Higher Prices, Larger Volume, Indi—
vidually and in Total, Nigeria, 1950-1967

Variances and Covariances of State Revenues
With and Without Revenue Contributions
From Marketing Boards and Allied Corpo—
rations, Nigeria, 1956-1966 . . . . .

ix

164

166

170

172

174

176

177

178

211

 

 

v ' ' “
71:5 1:: ocac:

 

     
   
   
   
  

up; a;

- a
.. agricu.ture

a

‘.
:-'. ~.. .‘1'

.,........al exports

:fiexgected future I
Lzzllilrg the past ar,
Brazing boards and
are open ment rev".
3:;ecti'2es of this 5‘

2.1:“ AC - ‘
sue U‘ thls St‘dc

 

CHAPTER I

INTRODUCTION

   
  
  
  
  
  
  
  
  

“This introductory chapter briefly examines the

i
H of agriculture in the Nigerian economy, the role of

government revenues. With this background,

 

‘ Eves Of this study will be stated. Finally, an

The Role of A riculture in

the Nigerian Economy
Agriculture is still the dominant sector in the
.:.79 ebonomy. In 1950, agriculture alone constituted
. "1 cent of the Gross Domestic Product (GDP); in 1967,
{atituted 41.8 per cent. In 1950, agriculture,
‘VJVIiShing, and forestry constituted 67.5 per cent‘
{in-1967 they constituted 55.9 per cent (see

From Table 1.1, even when agriculture alone

 

 

 

CC-

 

 

 

 

 

 

 

 

 

 

 

..o a... «.5. a4.» r.--
oc— 2: . . réo n can t uq-
00— cc“ 0 c.-
Nuq! 2‘
00‘ I 000‘ «CID MOE
3—squula.‘ 1:. .
.hholi.it .1L?.Il>ad Quid-flBuHO‘ ICU MS CdethChk 32.:Iufl 5:.
.‘h-uoucduuq .5.‘ Chan: HRH Gang than-‘nuh-n I D. ‘u:_:1m—..E -.P-\ .‘NQJUH-JDVLINV< \th's'uA11i \‘rdadu..>flq h..*
:II. t I ' D- Iliad-an.u— h§v< o>->I.I-..Lu ‘1....-.qi>d\~ Esau~=ylv‘.h\u‘
5.: fun-d and... I 9...... Inc [gotten-— In... vnfld... ‘3‘
.II‘O». awful- ulaa I". o ..- in... .- Dillon-s..- '- u on p
z s v n u u . I. in o 5. .a..l .l F...‘ a) I ol' newt .lha.'-II|>-u- a'II-tnnavh! Isi‘ Nun. :qu‘ can-id n u-s...vi¢ . u - ~ “JP-IF

 

 

a.0HA¢H«¢>s nos. 23!:
...a.:.. .mou :H ousuasoauma «0 0.35m on... c.“ .23 we 6:0“... ecu choose 0... us!» as 00:: Dana anon on» an on: :onoau 25: H
€09.65 0:» now 5:55..” .93...» uofiumu no anon... Eouu houuflu van» avenues scauguno use nausea. Scams—u one.” Bean on mac uou
mouse-Hume £053.30." 5 Juan on: 093960 uouuwb Eouu mcduasnou nuouuoa use» u0u can: :owuguuo uo avonuos cu noucououuup
ou mnazo .hmnommd nucaooua Hacoauuz auauumflz. m.onmaxu .O .z .: .ua a: anon» ou canauumaou aduuouav uo: smuomaa you cannons
.unou nouOSu ufiuncou Nona us Ono: nousmau Sauna." “nouaum 5mg as one: 032505" union: now nonsuuu he "louoz

.Hsaa .saan .Hauma .ou .Ho>
.uomun ..m.o.s .uuaunauaum uo uno an .asadnmoaa I .aauou z .nouuu an» o Donna-n: Hanna: .soaaummaa .voaa .uomuq ...m.o.s.
nuauuauaum uo uoauuo aqueous .<soH .ufiuomaz .muauufiuuum uo uoauuunt Hanan: .o>«u=Hu:a smuomma uou nausea» "nonunom

 

 
 

 

 

 

 

3.: 6.: 6.: 5.: oévm «.02 no:
mm 2. 9.: m.nm m.ovm n.5n So."
3 on Dav Evm rims Toma we:
3 2. Q: «Gm n63 «65 no:
cm 3 m6. «.3 ~63 To: :3.
an no Imv fits Evan 9.95m mom.”
.3 cm v.3 «:3 wévw «.3» «woe
«a an 7: TS uéno a6: Hum.”
ma um I: o.vo En; a6: 83
mm mm 0.3 Tn» oémm m4: an:
em .3 We... :23 «Jam «4.3 an:
no we Tum .13 we: «.mum 5a.—
; mm v.3 «.3 a6: oémm one."
mm no o.mm m.nm Emm‘ n.uhm mmoa
no on 0.3 0.3 m6: mémm en:
so co.— o.um «.3 ~43 wéum mm:
mm mm «.mm v.3 s6: «:33 «m3
co." «3 .15. IS «.2. 5.3m 3a..”
2: o3 o.um .515 adon o.mov on:
2: I omaa u:uu mom :3. 2m
5:23..» use
.Euuouom 3.03.3.3 ousuasuauz .50 no scant—omens ands-uh use
35,330.74: mac «0 cannon—cum a as 952..."..— uss china—6:3 .huu-goh £09553 us.»
a no mucus—6E0: .buuouom .xoouno>.3 ensues—Owns
no :oausfluucou uo Mensa oneumaoauu<

 

.336an dun-3: Jon—nous own-088 :05 ca 8.3-: v:- Iran-ouch £03.25.— .ousuusoquud «0 83:33:81.”; 3

 

 

::::s‘.ie:ed (excmdir‘

22.321 of the econ/3:7;

 

 

22:-2:32: position t'

_ mar-acted indice s

acti‘g'it iES' the‘

r:::: LPEOE agricUlt"

22::orce ard it is‘
22:22: pin period, a
.‘s‘: a; will still
;::;e::ed GDP.1
The export cor-

u:.

. 3.15.3:1cal 1y 1:9:

 

.:... exports origina

......

o
IIIII

fiance. those 6
€1.36?) +L t

1 we} had
.3: .iLQ 1_2) . Ty.
2L1}

.erne .
15. yalm c

 

a: “Gig 2
“ 0
‘ 8 PE:
1": ,a
' 3-: Ft.
“"111 1
'9. '1

 

 

  

 

is considered (excluding livestock, forestry, and fishing),

its dominant position becomes evident. As indicated by
the constructed indices of the share of agriculture and
allied activities, there is a declining trend in the
importance of agriculture. Yet, agriculture is still the
mainstay of the economy, employs about 70 per cent of the
labor force and it is projected that at the end of the
current plan period, agriculture, livestock, forestry,
and fishing will still account for 44.2 per cent of the
projected GDP.1

The export component of the agriculture sector has
been historically important and is still important. In
1950, exports originating in agriculture, livestock, and
forestry sector totaled LN80.345 million; in 1960, at
Independence, those exports totaled LNl4l.478 million and
by 1965, they had reached an all-time high of LN163,663,000
(see Table 1.2). The items underlying Table 1.2 are cocoa,
palm kernels, palm oil, beniseed, goat skins, cattle
hides, cotton (raw), rubber, fresh bananas (mainly up to
1960), timber logs, sawn timber, plywood and veneers,
groundnuts and groundnut oil and groundnut cake where the
last two items are included for completeness. These
exports as a percentage of total value of agriculture in
GDP were 20.8 per cent in 1950, 27.1 per cent in 1954,
21.6 per cent in 1966, and 24.5 per cent in 1967 (GDP

valued at 1962 factor cost), maintaining a proportion of

 

 

GOA I

 

 

 

 

 

 

 

 

.26... Lo..-
Omo~
A. ~v ANV A.~\-v Fh‘ ’=‘-. -L AIM-W. \ ‘.> \N
_. .

U:«ZI«L VII .>vaILC§ LCD Cu .M..II~CL .1..C.I..w>u.~ finalih3.‘ 32‘ 14:.- .‘ L-I‘>
slcnvuInvur—‘u s--:.~:...qhnr‘ -anU-qafi.‘ .~v< L: -5~'> r . u : \'.~=u~:.eq-\a< h.» Is,~:.~IE

66-.IL- Idh...nXu_- 54v h... ','~.~-...-Iv.- o-:d-—.u<- -< h - -< N: 0-3—‘3

'5A') h.~ 1-91:! | In: I.-...¢Xv- -=-|> 5.. ‘Ny‘..::-.~h m

' III‘ 'o n... It
.l": n1". o'c CCU-1.2 so.- >‘.|-1-.p...|l -II no: o-oou-uiivl‘ ‘a. I .t. ' Olav-III 'Il‘lu- 0-...I‘i,‘ .hnu .I‘.‘ can... I coho: - .d-u' I‘Qi‘-)

.6 .9}? . I 1‘ . u ‘Q‘c’?

 

 
 

 

5011.325 uo:.. 2.30:. -.¢.:.. "ouoz

500.3“. «m
ucuuu:oo uu unoo uouoau can» nonuau TS «Em TC How umoo uououu ”£05.50 an no: «wrong nan n:o«ua«oo«ao now anon no «cow—«s
moo :« ass»: «:3 .huunouOu £00302: .ouou«=o«umu mo os«u> 0:» van» :« >«o>«uoumn0u «3 «Eu .5 Scan uouflé «.«v and «.5
.«.« o«nd.«. oon $09.50» 2.0:» 50.3 utgxo o cum u o o :o um :0 so now owns a: no; one :« 95:03 «:3 ~aha-0.3M .xoou-obfin
.ausu«ou«um0 mo 02«u> now .33 .33 .wu«um«unum mo nonhunnd «3:34 :85 counofioo noun—36 .933 no on«c> “oounom

 

.m:..£u«h can .huunou0h .xOOuno>«.« .0u=v«50«nm< 2.9505" one: .«0H9u«ao«um4-u

 

 

 

 

 

.3 .1: 6.: .1: 6.: 2342 83
«3 m.«« m.v« «.3 «.o« on«.««« 53
«o« «.«« o.«« v.3 o.o« ¢««.ov« no:
vo« m.v« «.«N «.3 «.3 «ow.«w« no:
a: m.«« «.«« «.o« «.5 oo«.«m« =3
«3 v.«« m.«« «.3 «.3 o««.on« «om.—
hm« n.«« «.3 «.3 «.3 «o«.w«« «o««
«5 9: v.«« o.a« «.5 ovm.o«« «mg
2." «.m« o.«« o.o« 5.: «Sin: 33
a: «.3 «.m« «.«« «.o« o««.«v« «m3
4 a: m.«« w.«« «.3 «.5 «:6: «mo«
v«« v.«« v.«« o.«« o.«« u««.ho« 53
n: u..« w.v« w.o« w.o« «mo.m«« om:
««« m.«« m.«« v.«« v.«« m««.o«« mm:
«o« «.5 «.5 «.«« «.«« «w«.o«« on:
««« m.«« w.«« «.o« «.o« «34.3 «m««
o: w.m« o.m« o.«« o.«« m««.««« «mm«
o«« «.c« «.o« m.o« m.o« «3.2: «mg
oo« «.3 «.o« «.5 «.5 «on...» on:
oo« .. omm« ucoo you «.3: an
«.2 3V «.3 .«v
mag-«.— 65 §uuu0uo~ «no :« moo :« 9.3-: c:-
ioouogflu .0Mou«=o«uu< 9.335033 «0 031' tau-ouch .xuounofifi aka-ouch is. £03.25.—
loun nukes no we omcu:00uom Gauguin: «0 355.33.»: «0 canon—nu no.»
. 2 3«o> no sconu c an 3.396 oa«c> no 0o¢ucoou0m «.2 no 31>
c an nun—OE

 

.3973: 3.3032 «0 «lacuna «ounu«=u«um¢ on» 3 school” dauguauuni no co«unfl...uu:ou u:- .Qs.nc> «003......«4 a

 
 

 

51:: one-fifth for
=;:::::tion of tot
rzszzck, and fish
3.3;:2 cent in 19

u- uto~r~ but

Ann.
.:0.:vvauh on UUUA

gears to be a 5.6
12:21 exports.
Tazle 1.3
figure. livestcc

25.11; YE'EXport

“2.33314 5 ac

 

 

  

5

about one-fifth for the period of nearly two decades; as
a proportion of total value of agriculture, forestry,
livestock, and fishing in GDP, these exports amounted to
17.3 per cent in 1950, 23.3 per cent in 1954 (partly
reflecting the boom periods for primary products in world
markets around this time), 16.8 per cent in 1966 and

16.2 per cent in 1967. In this latter series, there
appears to be a declining trend in the share of agri-
cultural exports.

Table 1.3 shows the share of exports from agri—
culture, livestock, and forestry in total exports (in-
cluding re-exports) for the years 1950-68. Up until 1960,
these exports accounted for well over 80 per cent of all
Nigerian exports annually, with an all time high of
89.1 per cent. As from 1960, there is evidence of a
declining trend, reaching only 51.5 per cent of all
exports in 1966. We expect this decline to continue with
increasing exports of petroleum2 and increasing domestic
use of some agricultural exports.

Nigeria displays great diversity in agricultural
resource endowments.3 The agricultural economy of the
Northern States can be divided into cropping subregions
according to climate and ecology: (1) land where ground-
nuts and food crops like guinea corn, millet, beans, etc.
are grown; (2) land where cotton and food crops are grown;

(3) land where cotton, groundnuts and food are produced,

OCH

 

“EMUvU -Nnxonm fififi: 2‘
\

@«2

Hi

OOH .I Unrnwd

72!. OUBOQXE Cnuuh
d4< lune :terCnqux-rbrn I. If:
:1“. w athvdz Rise)

ma U -u 33X:

.fihuvnurflhnuau
~oJ-.;u~:,deth<

said-«mispri 7:1.
uni 53.0 Haiti» - hug/K

. 13?. . I€> «A
.R 0....UFHvu5

a InuAUU-eunv) ¢.H
--0.-uu Ind arugu‘rn

 

 

 

on. .lu 'Ii:-I.... .

A ---I‘ \V-‘u..-l“>-.h Ins€ivhcl4huu II-I‘ ill

. Ill.ll.\l~... ‘

. I u N CI-III\-

 

..nn...nafi..- -v\

 

 

 

 

 

 

 

 

 

 

.N.H manna 0mm .>uummuom can .xooumo>aa .musuasofiumm mo wuuomxm mo wsHm> mom .moma .vwma
.mmmH .mowmq ..w.o.m .moaumfiumum mo uomuumnd Hudccm .mwuswfim uuomxw cmHummflz Hmuoe "mousOm
oh m.Nm mmo.HHN mwma
Hm n.vm mam.avn nwma
mm m.am mmo.vmm coma
mm o.Hm www.mom mood
mu m.on omw.vHN vwma
vm v.mm www.mwa mood
vm o.m> mmm.moa Nwma
om m.mn www.mna Homa
em v.mm vah.moa coma
mm m.nm hmv.moa mmma
mm m.>w 0mm.mma mmma
mm m.¢m VMm.hNH bmma
,6 mm m.mw mnm.qma wmma
vm m.mm vam.mma mmma
mm o.nm Nmm.mva «mma
hm m.mm NMN.¢NH mmma
hm n.0m omm.mNH mmmH
mm h.mm voo.o~a Hmma
ooa H.mm mum.om omma
ooH u omma ucoo you Hafiz zm
huumoNOh was muuomxm sawummwz
.xuoumo>wq .musuasoflumd Had no wmmucoouwm m mm muuomxm
Scum muuomxm zuumwuom can .xUOumo>Hq cwwuomfiz snow
no oumnm mo xwvcH .wusuasufiumm scum muuomxm Hmuoa

 

.mwmalomma .m«uomfiz .muuomxm dance :a auumUHOh can .xooumo>flq rousUHaoHde mo oudnmll.n.a Mandy

 

 

 

1,; Lad in the v.

:12: at some grains
its Sachem State
:25: {1) land uhe
55:5 Si. 93‘... a.
:.L_:a'.:, rather, a:
care 3.51.113: oil pa

. I
"“" ":r'
-a-¢u.'..

n.' '0"

"‘5 orig'r‘al

I..'~‘n

no: Quit

. I
“"4.

u H ‘

I .
6".u' .
* “ 11110 the

'1. ‘
I’Hg .
Nata: frcpl ‘d"‘ ‘ ‘
.r

A“ I

-,. .

’CVI‘
I‘l' I" ‘k
I ‘ ‘

I sat: L..e 3‘

13 .

u "U rr

Cw
..' q
‘1: F ‘
‘ .-
tb h.‘at r
J
‘v
‘A
:u'e-S
I tanc']
-
4
“A
i" v.
§..‘ “
“‘ tfc.
‘
.
:3"!-
I' uv .
"a‘azl

 

and (4) land in the Middle Belt where mainly root food
crops and some grains are grown. The agricultural economy
of the Southern States can be divided into four production
zones: (1) land where cocoa and food are grown; (2) land
where oil palm and food crops are produced; (3) land where
oil palm, rubber, and foods are produced; and (4) land
where mainly oil palm, rubber, cocoa, or foods are
produced.

Marketing Boards, the Expggt

Sector and State Government
Revenues

 

It might be argued that it is now irrelevant to go
into the original intent for setting up the marketing
boards. Not quite. In June, 1971, the report of the
Study Group on Oil Palms4 felt a need to go "a little
deeper" into the original laws establishing the marketing
boards from which it concluded that "unless the law
governing the marketing board is changed . . . there can
be no improvement in the industry."

Arguments about the origins of the marketing
board system have become very extensive.5 Both in the
Cocoa White Papers6 and subsequent legislation, it is
evident that some form of stabilization of producer
prices, money or real income was intended as a primary
goal within the general framework of securing "the most

favourable arrangements for the purchase, grading, export

 

“,

 

 

 

'3"!

View he
51% academicians
1: acre cases

:5 setting up of

'-‘- "as stated:

the first cha:
working capit.
' - . Price 3
Prmary aim 0
resources am~
CEPital have

By the ti

:"thaes had Sw‘

‘:
“a

'n. in .
w contrlbld.

:9}
. u,
~94: “R

-¢3;3 Board

31"‘9‘13 Sign“ a
5“
l‘:-
. .
"“NEC to tells
“21:: fig
‘ Fred

-. ,ie a
.{j .
pV" CL
t
»
~.:Pq
s'a‘ A
't
y “‘
""5: .‘fl
5v S‘
La‘
5
, .
f
‘J'K u'h
“‘brfi'}- ‘
¥‘°‘
«1’

 

and selling of . . . cocoa, and to assist in the develop-
ment by all possible means of the cocoa industry . . . for
the benefit and prosperity of the producers."7

Official Attitudes on the Revenue
Role of Marketing Boards

 

 

The view has now evolved, among policy makers and
some academiciams alike, that price stabilization was not,
and in some cases should not, be the main objective behind
the setting up of the marketing boards. In the EEEEE
Annual Report of thgggigerian Groundnut Marketing Board,
it was stated:

The first charge on the board's funds must be for
working capital to finance its purchases of groundnut.
. . . Price stabilization must always constitute the
primary aim on the board's actual and prospective
resources after the critical requirements of working
capital have been met.

By the time of the First Development Plan,
attitudes had swung almost full circle. After spelling
out the contribution of the (then) Northern Nigeria
Marketing Board to its capital development program, the
Northern Nigerian Government 1962-68 Development Plan
planned to relieve "the board of its liability for subsi—
dizing the producer price of crops in lean years, and
fixing producer prices annually at such a level in relation
to world prices as to anticipate a surplus on the year's
operations to cover operating costs." The plan went on

further to state that "the marketing board will inevitably

look increasingly to Nigerian sources to provide the

 

 

 

 

:.;::-terr. credit re
"72;:;,'9 W0 thi
5:5:;r~g boards ha
525‘. sup. uses f
axis: tce Stat
25 regianalizatio
:::::c:. to see t
1'13“; agital ds
“f revenue
32:13: to a prod
titre of Specif
937539 F“ ocat l(
1"?” GEE-t of e

06 credited

5““? States

.w
.11 Per Ce‘.t as
if." “'P
"J Parts;
n: :_a L ’A
-S a

“2‘4
an” +:
'“b‘Or S .
Erhva
ads; *‘
bEYGS
\H '
M. I'.‘
I
.a“~E
Q
s"“
0'.-
'L.“IC.v-_‘
x. -S r
x
it;
."Er :3 Fr
y‘zr
l
‘h.
”affine.
-h‘r‘: 'P
a uv

 

 

short-term credit required to finance crop purchases and

movement."9

Two things have occurred: first, the
marketing boards have come to treat the so-called surpluses
as real surpluses freely to be used by the marketing

boards and the State Governments at will.10

Second, with
the regionalization of marketing boards, State Governments
have come to see them as sources of revenue for their
various capital development programs. In addition to this
source of revenue from crops, export crops have also been
subject to a produce sales tax and an export tax that is a
mixture of specific and ad valorem rates. Under the New
Revenue Allocation Decree retrospective April, 1969,
40 per cent of export duties on commodities other than oil
will be credited to the distributable pool while the
producing states will receive 60 per cent, instead of
100 per cent as previously. The pool will then be divided
into two parts: one-half to be divided equally between
the states and the other half to be divided on the basis of
population.11 Olakanpo and Teriba imply a strong positive
correlation between the existence of marketing board
institutions and the imposition of produce sales taxes and
exPort taxes on export crops.12

Table 1.4 shows the relative dependence of State
Governments on marketing board crops (we shall henceforth
refer to crops subject to marketing board control as

marketing board crops because for some of these crops

 

 

 

I3CC>§3 I; «C a: I . .1... ~32

.55.).z IO-I one. 3-.Iullfl
C. l._....u CHIC: I3C0>SK C. ILC~L Quark. I3EI>IK {a 1;, :5 :
’C—H.X.II :3 I‘XIH. iu‘oaehu ’Cv «tax-E! u... Iii-the ~CbCL. Dn-‘usxh‘z :3 .QX'H‘ ~¢6r\l.
he.» c..w&=-«~.d:.o.v

has C..u barandh canary LC Cguv bah-u .n puff...

 

 

 

 

"d'aa nuh'JI.“I I'o‘umu u.~.l!—dh..2 flaw—D“ :h‘ul".

 

"n-suvaO- .‘an'nvcz .I--.:nl>" -u'tfl- .ut I.-a..:hrb ‘h'c‘ bruub-luu-! :c .‘X'Lr has {AU-.Uanuuulflcco -P-L. o'.~ I.H:<..F

 

n.0Hnnflucbu uoc. cases 3.1.:. "ouoz

.aunum shay-ox cad uuaum mom-A
pouaumsoo ua .nmma nouns louuum wound was .0uuum uucslpw: .ouuum cuouwoz accuoum any as: noma Have: a: :oHuuu cuouuot 6H0 0:9
.ouduauconxo can. nos—.95.." Hasuoa Ban—«acumen banana 03"qu ninauocou aqua—5000‘ on» Gown :9?» dado .m.o.h 05 0.3.5» nouns—«yum
uomunn can” hon» oasouon 3 was... managnuum .93» on» «0 =05 who...» Eouu unuuwu non—:25..— ouuua noun;
a) csomuouuoo uo: 0v macauanauucoo omaucoouom .ha .m ..UaAH .nouuon aauoxud: 0:» mo iceduloua

     

 

  

95.3qu: no uncausfluucou eunucoouom 6;." .33». 30.35036 no an. an
$me .33 .nomuq ..m.o.m .auzuaunum no young-n: H5552 £35500» aquamuu v.3 acouusov 02:33— uuaum .309. Mon "cannon

 

 

 

 

 

 

 

 

 

 

 

.l.: «co.~n .u.: nmv.v~ win on“. ma moan

min nvv.ou .n.: th.an néu hum. 5N moo."

h.oN mnn.NN N.NN nhn.on «Una 0.34.3 voma

at: oGw.ON v.v~ «Nh.wN H.Nu hhm.vu mood

0 a.Nn FOH.NN o.mN ooméu N4. ovaéu «and
1 dig nnméfi N.@H mum.wu m.ow one...“ HuoH
«.mn wmm.o.n mén aoanN mév vroéu can."

mi...” 6.3.1" h.mN mpm.mu o.on ovnfia «mm.—

w.m.n oun.n.n h.vN nnh.w.n m.mN no.2” 0mg”

0.0N vnH.NH «Eva avatnn n.vn muo.vv hmad

a.vu wood otnu avr.n.n h.on Hon.v.n omod

ufluu Hon Add: 2.“ undo now Add: a 080 “Oh HHH! a
ouco>ox sou-um Emu-um 09:25.» nouuum 52.3.32 35:3— Hanna—won
ou nmouo canon 09:25: 3 30.6 933 3:25: 0» nuouu Econ 332,3 has»
9303.5 :0 nouns Hanan. weaves—nu: so .939 Hanan. 93387.3: no can. A-uoa
no couunauusou «o gausnuuucoo no soda—afluucoo
nouuum Econ-a nouavw cuonuuoz sown!" Eon-OB

 

63.732 .3332 33:26.. 3-3.. on :66 canon 2.38.: no non-u. no coaunfluugoo 3.7-...” 33

 

 

 

n‘ a ‘
VT: :vetw‘l o ‘* LY
A
' "' 7 Fur
2:55 :;~J&e)|
‘ 11 nrr
:ii'tso-ca..ed 5a.,

teting board I

crud-D it. :Me {:65

asserts" :.-; 1
1.2211961-62.
Eéiiince reac‘: ed
air: of 14.1 per

“1:3: ‘Ia

-~= :ms relat
21:1‘1958‘59 aj

:::;::ihg to t‘ Le

For the 5

fig" ti...
...:.e....g boards

D...
d“. [A]

”“9 (Table

.5“
‘A

v-IiTru'e Far-YE

k“

.u Eastern C
412 .. 5
U'as. ' F
e 5 P15
15 :
«qafis
3b deper‘r
‘3‘».
"Mt"?

 

  

ll

 

like cotton, only a very small proportion will be exported
in the future). For the purposes of this study, marketing
board so-called surpluses are seen as taxes on the producers
of marketing board crops. By 1960—61, revenues from taxes
on crops in the Western State and Lagos State accounted

for 60.5 per cent of total financial resources of the

State Government; this dependence reached a low of 7.2 per
cent in 1961-62. In the Eastern States, this relative
dependence reached a high of 27.8 per cent in 1961-62 and
a low of 14.1 per cent in 1962-63. In the six Northern
States, this relative dependence reached a high of 29.9 per
cent in 1958-59 and 15.6 per cent in 1960-61 (we are here
referring to the old regions in the terminology of the new
’ states).

For the Second Development Plan period, the
marketing boards are expected to play an even more impor-
tant role (Table 1.5). In the 1970-74 Development
Programme, marketing boards and statutory corporations
(mainly financed by marketing boards) provide more than
one-third of the total internal government revenue re—
sources in Western, South-Eastern, and Kwara States; in
South Eastern State these sources account for one—half of
the State's planned investment expenditure. Rivers State
is least dependent on marketing boards and statutory
corporations as sources of revenue. (Kano and South-

Bastern States would have deficits rather than surpluses

 

A a at In.
III-‘I .l
0-15.)-..

‘3‘ a. 2.1. :39
>n..fi3 a. a“.
Finn. Fv b'nv‘

.3. uIlth lip-d!-

..-'n- 'CI’P‘|'I

 

.'l .vh-l I'CIII,IVA

uCI . .2..-

 

1...!"
.uflduw‘CILII
U:"-I'>C—

.025. ‘1 ditch lit—I
'3 . .' U-h-‘a'-\oc—
1.. 12:...
F.-‘." uni! 'LI-anh.‘

.‘o unlit-..-

   

u lel‘lII-v- >1II. 0.40-... Dunn- "I'tvll

in. n'.-..

 

. I1- . 4.1.5.1....
6"...- oII>Cn
fl. inn-h.

,!-1.I-I I! III.‘

 

UC.I.-.>pv3
Cola: L:
SEIDQK ~‘uGF
&.. ‘\.|u-...-.Ia-
I I. '::‘>'I
I‘A4-U‘-.lnn....
*I..bo—U'O.m “you.
Ian'Avql IrnquI-II!

-Iog')! I'd IIIIU‘IIIIQ

' In) .0 i'ln-n.lu

Ida‘d'?‘
fl'Uu-".J
i'£ 1. fps.
.CA.~ t'hAIn-
¥.-..u-.~'~I
b.'.-'ao:
III- --l~'!
ivl.‘ -anb-\r“

0C!~C)S
.- .1 T... ~1h
'u-").‘
«'0.‘

0.5!. -:8

OIJ
‘0' at.

Inh""'\l-II Ivafi‘ ‘l4-'\a" I "Ifbull .l . I .III‘.’

 

12

.HO« .m .orod .005 .aofluafiuounn uo aha-.15: Hugh .vhlohad 5am a

 

 

 

v.5”

n.«.n

«ION

tvu

 

 

 

 

>6 o.n a... quota—
ni. or: n.9— m.w« wé «.n «Kn Hula-id!
o. «.3. «.« m.m.n 0.9 $4 1nd 53:.“ g
'6 «.ma v6 odd n6." o.n o.o« Ice-A
w.on min 0.3: «.3. m6." 0.0 n h 1.5on scan
w.n.. 9.3.. v.« adv H.0H 0.0 h.«v call
méu Sn.— n.nu o.vu 0.: n n via 533 fine:
«.0: 1m." 05. 1.3 0.3” « n v..n« g €03
0.3: 0:: rd: Ton m6." « n v.3 Hug Auto.
Ema h; $.17 «.5 min a o n.« g
in at?“ n6 «:3 nén 0.: m an glen—6m
o.«n «.3 OJ: «.«h v2: 0.0« o no goal
«.3: In 0600 won «.3: In 0600 now and. a
3325— 55.38.98 9305 nun—«80¢
83-838 salvo->3 3.8 no .7396
hug-um 38. no H32. 33: cog->0: 108. so can veal-g
can canon glaucoma.— - .- nsects“— !!an uo coca—loud.— gluonuoo 33a 88 E
nae-bound: 35: on: . «duos Sign a u- g 33.53% all:- 330
a Cong... 9:0: “as Gui 309 3305.30 .3 39.—
063013: #3033» on. g
9.33 8.33.3: Sun 3'»!

 

(
s ~
.252...— 3103.‘ ecu-nu «a 3.3393930 Ens-an 1:- ion angul— Ivuh ill o» go- 053 0"; 1a- .n.n .

 

 

 

255135;:er ccrpc
salable to these
3:: Central, Ncr1
32322522111 have I:
mating board ar.
IE'.'E.‘.‘.29.) Table 1
733:5 state gov

The Keste:

rezeigts for cap:
E3‘.;e:ia Marketin:
mm Prices,
1231:2111 adequat.
g... |
"1.6 S developr‘
0n the a
3:33.21" 5%,. u
‘ utvuid I.
NC! f”Tia-tic”
331:)..."
«ufille Evid
teyk
:PliCat:

30v .
1.25 ls their

-
5“

 

 

  

13

On total current and capital account if marketing board
and statutory corporation sources of revenue were no more
available to these state governments; East Central, Kwara,
North Central, North Eastern, North Western, and Western
States will have much larger deficits in the absence of
marketing board and statutory corporation sources of
revenue.) Table 1.5 shows the projected dependence of
various state governments on marketing board crops.

The Western State Development Plan 1970-74,

statesl 3

that "as at present, the main source of internal
receipts for capital development remains the Western
Nigeria Marketing Board" and that in determining net
producer prices, "emphasis would be given to the need to
marshall adequate revenue from this vital area to aid the
state's development effort."

On the academic side, Helleiner states that
"growth should have carried greater weight in Nigerian
policy formation than stability, and there exists no
conclusive evidence that the two are correlated."14
Teriba and Olakanpo have stated that "one important fiscal
implication of the regionalization of the marketing
boards :is their conversion from potential stabilizers of
Producer prices into potential stabilizers of Regional
G°VernInent Revenue."ls

Policy makers being so concerned about the revenue

I”tential of marketing boards and their allied statutory

 

("it 1’

 

.rs n-Jv -

 

«craticns have

~::::: and sales

m‘nt
-

”.v- .er‘fi $.5‘
wine vow-1.950

:W‘i ens of z:

.::.f-OV
- i : .‘Vf fig.
5533 0'. ma. me u .L

N." F... v r- ~r‘. a
La! ioub‘ r‘U.‘

s

.-:" F'f"

'10.! 3"“ .
o

.w:' 9%. “-..Hre
"" uni 5"“.

.'-:”: fly”, 17

I!‘.‘ .‘Ur .

h: “h"! F‘
no hv‘.-_le‘.ba
"H35;

“"39 Price

:‘a‘nal . .
....;.is 13.911:

:3.”
'fl‘: A u “
“nua LY
o
‘F
1.53:9
:‘W .
“‘ can.
“LJS‘ S
‘ I
“Q I
U fi’fl
in.‘ c‘ s
U“ (
K
5";
'«b :r:f:whn
I “A
v. s
u
2"“.
.‘ '-
-- R
§ A..." mpg“
d av‘bu.
d
. V
‘
a“,
‘ a
N : e5, ‘ P
C. ‘ ‘
{4&4 'PE
‘
.“
”-v

 

 

 

 
 
 
 

14

corporations have not bothered themselves with the negative
output and sales effects of the heavy taxes on marketing
board crops. Their policies are based on the implicit
assumptions of zero price elasticities of supply and/or
sales of marketing board crops. Recently, state governors
and other public officials have expressed grave concern
over what they term the "falling output" of marketing

board crops.l6 This, it seems to me, is really a concern
over the future of the revenue potential of marketing

17

board crops. On the academic side, though Helleiner in

his monumental study18

admits of the possibility of
positive price supply elasticities, the core of his
analysis implicitly assumes zero price elasticities of

19 Nowhere in the liter-

supply of marketing board crops.
ature is the possibility of drastically cutting down taxes
on crops, raising producer prices and still leaving state
government and marketing board revenue from agriculture
substantially unaffected thoroughly discussed.

There is an absence in the academic literature of
any analysis of the effects of taxation of marketing
board crops (represented by the marketing board surplus,
the produce sales tax and the export tax) on the scale of
farming measured either in terms of acreages or size of
lab“? employed. But this problem has become increasingly
imp°rtant in recent years as farmers have reallocated

res(“tees to different crops in response to differential

 

.z. vates on these

.4-

a
u-

governcrs a:

5:219 increasmg

-.¢~:!;I“"s if‘ ‘

hunt-tnlb “b L
.

94.0?pr 9" fr .uVR
“vow“: kll' r.OCub

‘
-!-':n- It «G u:
'I!I.lloe uh at u

, 23

fuovv=o

-.
,-
U

The firs

“'4'"; g

......e a frame"

at" fin; ,_ e
s - u.‘ h.c:.f‘e‘
»
“ l'ru‘
II. c n V
.“r“"‘ 0-6 .I h

bu.

.fi‘ '
w awllCdtir
\
"' "t ’P“n ‘
. g‘Vu;:I‘t‘

l“
"h.
' JCHL
a Er“
~a
.‘r: ‘
- in
ad‘bn fi'f.“
3~ Junr
N
‘I

I
6
a": ‘e.
' “‘91 w
p
p
‘ we '.
autc“
Q».
\.~
filters
q
v
Q“: ‘
I: ‘
v~ §,.
\l‘e ?;v‘
“I“
V.
“i: 3‘ ‘
‘s

 

  

15

 

tax rates on these crops. In the Northern States,
state governors and other public officials are worried
at the increasing diversion of land acreage away from
groundnuts into food production resulting from substi—
tutions in production in response to the effects of
government and marketing board taxes on relative crop

prices.20

Objectives of This Study

The first objective of this study then is to
provide a framework for examining the effects of govern-
ment and marketing board taxation of crops and subsidies
on inputs on the commitment of resources to production,
with application to the use of land and labor resources

. on the groundnut and cotton farms of the Northern States.

More specifically, the framework will be addressed
to the following questions. What have been the effects of
these taxes on the supply, "retirement," or salvaging of
used resources from and disinvestment in these resources
(on, Northern States farms? Have these taxes tended to
induce groundnut and cotton farmers in the Northern States
to use exactly the same, more, or less of land and labor
thatifliey would have used in the production of these crops
in the absence of these taxes and subsidies? If policy
flakers (are willing to live with the institutional reality
°f the Inarketing board system but are anxious to minimize

the a1locative distortions of these taxes with respect to

 

‘ ‘-
':E:.::e use, “3a \-
a ,c:A"'f‘e5? A I

.¢UU“'

{52:15 of differf
:32 relative 0‘

Tne seccr.

{2:22th zga

 

qr :t'pn in“ d
m-‘n.u “‘3 “Ha: t

, .
----- g . A 1
.45.”; Jitters h.

 

' n
mg... m‘n '
Juvgfiec| tells ‘

.; a"!

.. «u the (ere

-v‘~‘
.23... .amers *

5.

22:52:, ‘1; ..

6 Ac '

op .‘h

~ ...e beard.

Given p

p
..,_
3:”..P'u

.hmfints on

‘1" .
‘(c“.ln

W3 boar:

'b.;
N»: '

Varies
I

2:5 3b.. ‘
be:

‘1 the
53be

‘H .
*I«Stec

..A‘
v... LN»
‘ awed wi

tad?
.

‘43 r v .
t~3blen

 

 

  

16

  

resource use, what are the required compensating subsidies
on resources? A related question is: what have been the

effects of differential tax rates on cotton and groundnuts
on the relative outputs of these crops?

The second aim of this study is to examine the
quantitative impact on Northern States Government and
marketing board tax revenue from cotton and groundnuts of
paying farmers higher prices than the actual prices they
received. This will enable us to examine to what extent,
if any, the (crop tax) revenue loss that results from
paying farmers higher prices can be offset by the positive
effect, if any, of higher prices on sales of these crops
to the board.

Given past and planned dependence of many State
Governments on contributions (loans and grants) from
marketing boards and allied statutory corporations for
their various State Development Plans the third aim of
this study then is to examine whether these contributions
can be trusted to stabilize state government revenues over
projected plan periods as has been recently suggested.
This problem is important for the stability of State
Government operations and even political stability in the

future.

 

...e 81'

__....——
Auseful (
agricai measare
effects of tnese
aside epolicy d
:z‘lcrtnen Stat

”' '7’” Hf: * S u

u... «gunman,» ‘

u- :a~* ‘;at a”

‘1'. Let .‘

u! o... \ F‘P‘_ v d
not .uE JG: 5“ -nL

2" ' v‘ A '
ctgoc {late-e“ 1..

r I.
V: :(p‘o‘l

.I
\

".over
'5. urban areas.
to isvest
;:::easir.gly ch
353515 also c
3:11;? gal CI

oVu--‘ n E
. 1’ he“ U

v
-Q
“as:

.
I
a: 3;.Q
“"S F‘
v‘ C,
'9
:v'
ne‘flet‘ll F
. .g an,
N’.
~.:C.>.“ ‘
.‘ '- ‘
Sbahcl‘h
..j
a
5.15 P
. ‘ F
v

 

  

17

The Significance of These Problems

A useful conceptual framework provides a basis for

 

empirical measurements. Empirical measurements of the
effects of these taxes on resource use are necessary for
sensible policy discussions. There is mounting concern in
the Northern States over the diversion of acreages away
from groundnuts and cotton to other crops and also over
the fact that in some states such as the North—East State
and the North-West State, suitable additional land has not
been planted in these crops. .

Unemployment is currently a serious problem. There
is concern over the drift of farmers and youth from rural
to urban areas. Rural people deciding (at the margin)
where to invest their future labor time are known to have
increasingly chosen the non-farm, urban labor market.
There is also concern over the slow transformation of
traditional agriculture in Nigeria. Why are few farmers
adopting new recommendations, or buying new inputs, etc.?
These are all issues a good conceptual framework should
help throw light on.

We have no estimates of the price responsiveness
of sales of cotton and groundnuts to the Northern States
Marketing Board. Yet this knowledge is crucial for an
understanding of the responsiveness of tax revenue from

these crops to changes in the board's price.

 

 

hm regar
:erecen: experie
giggle. POI the
Ezaze, 'cagital r
23:: of estimate

"we" ,1, gr‘v
1 .95

I
”tn-avu. o s
.

h‘,
:n‘
\vudr -er

:::;:;s as they

.
" 3’7!

“"mz’ listed

I
:apte

be.

:-:‘-. .‘ .
“Mme *' e

2153 or. cr

OPE
55.6 of

-”o'..
_I
4. Intern Q

N

“‘n ‘\

s F

.n; ‘lie Ch‘r Ir
V~~

“mm

Hm" Fr

L‘s...
‘~ GIL
‘ ~ ‘I‘ P
R.-
vflwgri ;
t 5 ‘PI
u
W.
U..:
S to t":
s.‘
\_
'3
o. A
- n
‘Vus ..
are
.-
A‘s
‘5 :Ci'

:‘I p:
I
.L‘
b\th
a Pv-
U.
‘I
1.".
I V.
"kes
V‘Ev-s
‘Q

 

  

18

With regards to the third objective of the study,

 

the recent experience of the Western State is a good
example. For the 1971-72 budget year for the Western
State, “capital receipts of some EN 7 mill fell EN 5 mil]
short of estimates because the state of world markets,
particularly for cocoa, made it impossible for the

marketing board to make a contribution. . ."21

Plan of This Study

Chapter II will review previous studies. Short-
comings as they relate to the themes of this study will be
briefly listed.

Chapter III will present the framework for
analyzing the effects of government and marketing board
taxes on crops and subsidies on inputs on resource use and
scale of farming in the cotton and groundnut farms of the
Northern States. A production function approach is used
With the Cobb-Douglas serving as an example. Chapter IV
develops the theme that in examining problems of State
Government Revenue from marketing board crops, the relevant
concept for most of the crops is the price elasticity of
sales to the marketing board. Some illustrative calcu-
lations are carried out to show the dependence of govern—
ment revenue from marketing board crops on this parameter
and the level of taxes. Chapter V presents some empirical
eVidence on the sales elasticity parameter and other

panmeters and other effects of government pricing policy.

 

‘11 examines t':

a
“gin-a
.

galizers of state
"fir.“ngs, draws

.
n... I a

2:221:25 for agric;

 

 

 

19

Chapter VI examines the potential of marketing boards as

stabilizers of state revenues. Chapter VII summarizes the

main findings, draws some conclusions and offers recom-
mendations for agricultural development in Nigeria in the

1970's and 1980's.

 

 

2..._
...e count I;

   
    
   
   
  
     
  

11570, b}: 270 mil‘
ezzzazge earnings a"
.09, 58 per cent l.’
i3. 1:

. .
'1 n
< 7‘
'l 3-H»; man

R..." n . Ridden;
"oi-Jada“ Spinning a
:rtgof bales of cc

fate: use was 115.2
treasury-over sto
3113? Green on Catt

finest of Agric

 
 

‘ 3Glenn L.
-'~~..ach t ‘

 

  

FOOTNOTES

1Second National Development Plan, 1970-74 (Lagos:
Federal Ministry of Information, 1970), p. 53.

2The country's revenue from oil was EN 80 million
in 1970, EN 270 million in 1971; the share of foreign
exchange earnings accounted for by oil was 41 per cent in
1969, 58 per cent in 1970, and 71 per cent in 1971. See
New Nigerian (Kaduna, February 26, 1972), p. 1. Whereas
Nigerian spinning accounted for only 8.6 per cent of total
crop of bales of cotton in 1960-61, by 1970-71, the esti-
mated use was 115.2 per cent, with the shortage to be met
from carry-over stocks from 1969-70. See Report of the

Study Group on Cotton and Other Fibres (Lagos: Federal
Department of Agriculture, 1971), p. 14

3Glenn L. Johnson, et al., A Generalized Simulation

Approach to Agricultural Sector Analysis With Special
A i

cation to Ni eria (East Lansing: Michigan State
UniverSity, 19715, p. 12

 

4Report of the Study Group on Oil Palms (Lagos:
Federal Department of Agriculture, June, 197ITT p. 33.

5See P. T. Bauer, West African Trade (Cambridge:
Cambridge University Press), pp. 263-75; P. T. Bauer and
B. 5. Yamey, Markets, Market Control and Marketing Reform
(weidenfeid and—Nicolson, 1968)? pp. 133-55.

6P. T. Bauer, West African Trade, p. 268.

71bid., p. 276.

Ma 8First Annual Re ort of the Nigerian Groundnut
-—£§§ting Board, p. 16.

 

 

20

 

9 Develo ment
”.4" Elgerla. all".
243.

 

12 .'
”The Cocoa v
establis‘ment of the
.3211" 3*. world *9
:2: will be les;
2:: eassa es. “he
a 'sxp‘..s.' There ‘
22.22. the average we
Erziazers. On these
‘ which will t
'11-;33 Of high we-
‘2rzfits‘ will be 2.:
3%“- 8953539018 st-<
grzixer. Thus, or.
ingested t)“ at tt.
5‘3 Sws antial
31.12911 0:} “0:1 6 YT
-:i1‘ri2tra Battier.
.. LS Significant t

7
'suulas" adtd “lbs

 

. H
V‘ “ES 9v
us. {Ly ‘V‘
‘U ‘ \
I- . “1 Cf “h
2" 3% L~Cr
.

n o‘ \

-A

“‘s'j" a ls E

. 4“»er1 -

~25- “ 1‘
‘-de;“ L *‘ I

\.
n~

 

21

9Developmgpt Plan 1962-68 (Kaduna: Government of
Northern Nigeria, Ministry<f Economic Planning, 1962),
p. 45.

10The Cocoa White Papers which preceeded the

establishment of the Cocoa Marketing Board state "In some
seasons when world prices are high, the price paid to the
producer will be less than the average realization on
overseas sales. The boards will, on such occasions, show
a 'surplus.I There will, however, be other seasons in
which the average world price is below the price paid to
producers. On these occasions the boards will make a
'loss,' which will be financed from the 'surpluses' accrued
in years of high world prices. The intention is that
'profits' will be utilized primarily to maintain the
maximum possible stability in the price paid to the
producer. Thus, on the average of a period of years, it

is expected that the average net price paid in West Africa
will be substantially equal to the average net price
realized on world markets and that the boards' buying and
selling transactions will therefore approximately balance."
It is significant that the authors used the words "profits,"
”surplus," and "loss" in inverted commas realizing the true
nature of these balances. See P. T. Bauer, West African
Trade (Cambridge: Cambridge University Press, 1954),

p. 269.

11See Annual Economic Review (Nigeria: Standard

Bank Ltd., London, June, 19705, p. 16.

12O. Teriba and O. Olakanpo, Fiscal Monetar and
Investment Implications of the Marketing Boards (NISER:
International Conference on the Marketing Board System,
1971), p. 14.

13Western State Development Plan, 1970-74 (Ibadan:

Ministry of Economic Planning and Reconstruction, 1971),
p' 58.

14Helleiner does not indicate which type of corre-

latéon he is referring to. He would presumably be most
dellghted if there was perfect negative correlation
belWeen growth and stability! We might add also that the
?V1dence of zero correlation between growth and stability
13 ?Ot conclusive. More basic is the fact that the corre—
latlon coefficient itself proves nothing.

 

i

 

nai'w-_

 

15C). Teriba a
saitapter VI below

,
6The Federa
':.=;a:i, .as suggest:
'-r"l’OGuCtl\lty c
3.2:“: narratihg boa
skating system. S

l7,L . .

“VHS a cc:
5' tern State G023;
i:-:;rv- ard Vorth E:
33:5'190‘3rcit by I
“”5 ' a ‘35-" Se-
;.7. “9 "igx-t add‘
163:“:an board
-:.E 5C3! ‘ed a'd 3"
-5 and haVe 586;

 

 

 

a
GKHe‘.

Belleine

"5352139 baards e
4.3 yctential ‘
2:3,“? 1‘35 oi s
:‘a'. av} Side: tar a~
{3&3 Ci gove
: :fe‘ C'ops ‘ "2),:

h‘ust‘ ‘
1:172 5.1?! Oi S

“.1 a “his
.‘n;:‘ 3a;n it
,‘U‘é'iie £0:
3'3" :28 Pre
1‘ ‘ C‘OCS
‘uer; ‘ ' a
5- ‘a 38m.
2:37 . .

22

15O. Teriba and O. Olakanpo, o . cit., p. 10.

(See Chapter VI below for an examination of this potential.)

16The Federal Finance Commissioner, Alhaji Shehu

Shagari, has suggested that “failure by farmers to increase
their productivity could be traced to the poor prices paid
by the marketing boards" and he has called for a new

marketing system., See West Africa, #2858 (March 24, 1972).
p. 374.

17Thus a correspondent writes in the Daily Times,
"Northern State Governments, especially Kano, North
Western and North Eastern States have had their revenue
grossly undercut by having to lose sales tax which is LN
2:15: = a ton." See Daily Times, Lagos, February 3, 1972,
p. 7. We might add that these State Governments also lose
the marketing board's so—called trading surplus that could
have accrued and also a refund of some of the export tax
that could have been imposed.

186. K. Helleiner, Peasant Agriculture, Government
and Economic Growth in Nigeria (Richard Irwin, Inc., 1966).

19Helleiner realizes the revenue potential of

marketing boards and he even advocates maximization of

‘ this potential. He not only plays down on the price
elasticities of supply of marketing board crops, he does

not consider that this may not be the relevant concept for

purposes of government and marketing board revenue from

these crops. The price elasticity of supply (output) of

a marketing board crop may be zero and yet the price

elasticity of sales to the marketing board may be positive

and very large.

20Thus a correspondent for West Africa writes,

TOnce again it appears that there has not been sufficient
Incentive for farmers to plant out a higher acreage and
while the present high prices are obtainable for other
f90d crops, a pattern of rather smaller groundnut crops in
N1geria seems likely to continue." See West Africa,

#2857 (London: March 17, 1972). ‘

c‘ »- . .

 

21See West Africa, #2865 (Apapa: Times Press, 2
May 12, 1972). p."T_5 7. '-

\~_

 

   

 

I: .
~u

‘!
‘g‘fiv

 

REV I

This chapte:

    

3::iies as these re

Mar

Our brief
32:1 to those partsl

tun-g 5*tqd“, .
l. AGE}:

:2; x
5l “e arg‘des £0

.
"Ha:

 

CIOpS I for t

~
I‘: ‘V‘ R
“q ‘

I19 profit

.Mse effects 0

antes that the?

O
(1

the I
‘ use of ex~

a thrv,
‘ '91 of c
s. “-

...e 5'3ch Ea H

0C0

“‘sv‘rtion Cc

L

 

“ taX‘es

 

 

CHAPTER II

REVIEW OF PREVIOUS STUDIES

This chapter contains a review of some previous

studies as these relate to the objectives of this study.

  
  
 
 
  
   
   
 
 
 
  

 

Marketing Boards as Revenue

Collectors for State
Governments

Our brief review of past studies will be limited
only to those parts that are relevant to the objectives of
this study.

Helleinerl argues for economic growth. To finance
this, he argues for the maximization of taxes on marketing
board crops, for the marketing boards "to maximize their
Own trading profits." After listing some of the possible
adverse effects of taxes on marketing board crops, he
states that they are not in themselves "sufficient objection

to the use of export taxes." He cites some evidence from

a SuIVey of cocoa farmers (1951-52) and oil palm producers
in the South East state, and concludes that the largest
Pr°POrtion of any increase in income to farmers through

higher taxes would be consumed and “that since a much

23

 

 

' :a: proportion of

a.
000’

 

been consumed t'r
:ch marketing boa‘
:13: on peasant in
greater than those (2
fire funds in the
1::siiered superior.

:azefrre marbiguor

 

as been aided thr0‘|
:1" its export earni
:agroixmer to OU

'vie too giV'
tree seriously q

{ac-h“
M! of economic

.
2‘-
..N S

353.9: 4 .
. al‘ZEd le'fl

3.7.5:”

‘ .al, A Prec

Stitle
(Emu-ment:

" 3°56 taxi

I
.‘

‘M

“9

ErlOr f5:

V
p
i 3",“

“V
a

:2 new .
£111:

 

24

larger proportion of the increase in peasant income would
have.been consumed than that which was actually consumed
out of marketing board trading surpluses, the rates of
return on peasant investments would have had to be much
greater than those on marketing board ones if peasant uses
of the funds in the aggregate were really to have been
considered superior." He goes on further to state "it can
therefore unambiguously be stated that Nigerian development
has been aided through the device of channeling a portion
of its export earnings via the marketing boards away from
the producer to other (governmental) decision makers."2
We too give primary importance to economic growth.

But we seriously question his implied prescriptions. Any
theory of economic growth derived from investments in new
and superior sources of future income streams must take a
generalized view of capital accumulation,3 human and
physical. A preoccupation with the revenue needs of
State Governments should not lead us to neglect the effects
of those taxing policies on the generation of savings, new
and superior forms of farm-produced capital and investment
in new capital forms on Nigerian farms. Glenn Johnson's
warning in his Encyclopedia article is appropriate at this
point:

Public policy is an important determinant of the rate

of capital formation in agriculture. Some policies

and programs exploit farmers heavily, leaving little

surplus to accumulate in agriculture. If such taxes
are levied on commodities in whose production the

 

 

of capital accum
so; be reduced s
are reinvested l‘.

sentry has a co]I

mot substantial

221‘. exp cted to co

 

137a. overall econo:

Government a
:‘Izrops have disto:
rats: on investmen'
mired capital wt.

'iEital are considel

   
  

':r~g 4,.

-.. ...trodoced all
:rrast:er.ts in dur‘

l...
“and
l

ago shown , by

EdCECUate emp‘
went agenClr
Ia:

 

 

 
 

25

country has a comparative advantage, the over-all rate
of capital accumulation in the country's agriculture
may be reduced substantially even though tax moneys
are reinvested in agriculture.
Without substantial capital formation in a sector that is
still expected to contribute 44.2 per cent of the GDP in
1974, overall economic growth will be impeded.

Government and marketing board selective taxation
of crops have distorted the pattern of relative rates of
return on investments in both farm produced and non—farm
produced capital when both human and physical forms of
capital are considered. These taxes are most likely to
have introduced allocative distortions in the patterns of
investments in durable resources. Galletti, etc. have
long ago shown, by implication, the investment responsive-
ness of Nigerian farmers to crop price changes. A more
liberal view of capital may turn a substantial part of
what Helleiner terms "consumption" into investments in
human capital. The author feels that Helleiner does not
give adequate emphasis to the consumption propensity of
government agencies. The evidence is strong that there

(has been a lot of waste, that a lot of the farmers' money
Went into "political capital formation" for various
Political parties and individuals.5

A frequently neglected source of agricultural
growth is increased specialization, the "reallocations of
agricultural production from farm to farm within regions,

fflbm region to region and between the farm and non-farm

 

211101136 In the l;

;:‘.;:ies are likely

 

f12‘n‘estern State ha
111:: operating ac:
, ‘ |
1212 toe Western Sta'
11:11.1 prices in th‘
1512: than those i'
file 2.1.
~-~‘-:2.1.--Ratios
Cotton
Differs
l97l/7’
\‘i

 

 

economy."6 In the Nigerian case, government taxation

26

policies are likely to distort specialization patterns.

The Western State has for years sustained losses on its
cotton operating account because of its declared policy to
make the Western State self-sufficient in cotton. Thus,
cotton prices in the Western State have been consistently
higher than those in the Northern States as shown in

Table 2.1.

TABLE 2.l.--Ratios of Cotton Prices in the Western State to

Cotton Prices in the Northern States for
Different Grades of Cotton, Nigeria, 1963/64—

 

 

 

1971/72.
Grades
Year
NAl NA2 NA3
1963/64 1.00 1.00
1964/65 1.20 1.02 .90
1965/66 1.37 1.00 .88
1966/67 1.43 1.05 .92
1967/68 1.52 1.13 1.00
1968/69 1.17 .90 .78
1969/70 1.50 1.20 . .
1971/71 1.50 1.20 . .
1971/72 1.38 1.09 . .

 

Source: Ratios computed from Western State and
Northern States cotton prices in Re ort of the Stud Grou
on Cotton and Other Fibres, Federal Department of Agri-
culture, Lagos, I971, p. 64 and p. 7 respectively.

From this table it is clear that apart from

1963/64, NAl cotton farmers in Western State were

 

   

raistently paid hi

:2 Scrthern States;

uuv‘
”a

 

1.... in the West t
r; for NAB cotton
rite West. The e:-
Mast in the pro.
grasance of consist
grimlarly evider.

.r...ibution of NP.‘

A

”estern State was 2

    
  
   
 

36363.7 These a:
tritited the form
tr-se crops in whi
airartage. The re

p
n
-"Vr
’5‘“
‘6

"dth in the ;

 

we: statema w

.‘H _
‘“ 5911c

ultural P

G‘-
‘ C'r‘ntrit
?;E:‘1 A
US‘Jn is
tn:
"eria

27

consistently paid higher prices than their counterparts in

the Northern States; NA2 cotton prices were similarly
higher in the West than in the North most of the time;
only for NA3 cotton were Northern prices higher than those
in the West. The extreme lack of comparative advantage of
the West in the production of NAl cotton even in the
presence of consistently higher relative prices becomes
particularly evident from the fact that the percentage
contribution of NAl cotton to all sales of cotton in the
Western State was zero in all years except 1963/64 and
1967/68.7 These artificial incentives no doubt must have
inhibited the formation, use and allocation of capital in
those crops in which the Western State has a comparative

advantage. The reallocation of capital has been a source

of growth in the agriculture of other countries, e.g., the
United States;8 we should expect unimpeded specialization
in agricultural production within Nigeria to be a source of
growth. To the extent that these government and marketing
board taxes distorted patterns of specialization they have
inhibited economic growth of Nigerian agriculture. And the
issue is not whether marketing boards have contributed or
have not contributed to economic development--the real
question is the contribution of investment in new and
superior sources of income streams that would have occurred
in the absence of taxation of marketing board crops. To

the extent that investments in new and superior biological

 

 

grains, new cultiva
iiamers were in‘n;
axes have inhibite
:Efset by farm sett
afar'niversity so
ray that were no:

‘
cosine-n

_......ing areas (r:

 

:15 Eric and Ibadar

Helleiner1

.1. r
4-3

-the farmer w
:CWEXE given t

stance: income <

1.8 PCSlti'Je Sup

“5,
l“

4.1. ‘
“*9- Various

ital

Percentag.

28

strains, new cultivation practices, etc., by the millions
of farmers were inhibited by the heavy taxes, then these
taxes have inhibited economic growth that could not be
offset by farm settlements, expensive corporation buildings,
a few university scholarships and roads built with this
money that were not, in many cases, even in the major crop
producing areas (road construction in Ijebu Province vis-a-

vis Ondo and Ibadan Provinces are classic examples).9

Potential Producer Income
and Potential State
Government Revenue

 

Helleiner10 calculates "potential producer income,"
what the farmer would have received if all taxes on the
crop were given to the farmer. His estimates of potential
producer income contain a downward bias insofar as there
are positive supply price and/or sales price elasticities
for the various crops, and therefore his calculation of
total percentages of potential producer income that are
taken away in taxes contain an upward bias, given the
elastic nature of the demand curves facing aggregates of
producers of most marketing board crops. For short-run
detenmination of revenue to the marketing board it is the
Price elasticity of sales to the board of any crop that is
more relevant. The supply (output) price elasticity could
be zero and estimates of potential producer income would
still be biased downward if the price elasticity of sales

to the marketing board is positive. For a given increase

 

~ "cheer prices .

a biased downward 1

 

skating board is t
I
:::i::er prices, po'

its price elastir

szggly (output) pri

 

32;;1'1 (output) or
;::ce elasticity of

The one s t iv.

-.:.

:.._o gQ-Jernment a

gazes that are hi

O

seated by 01 ay i d

a.‘

t A

' 0f exPorts

I u
I.“

iait'da ‘-

itae _. ,
at. The I: E
iS:-~

SSE a Zero

- \
~32 (a; ;

“mg bOar

3‘:er Q .
Tn, . P
’ I
v“ ls
~.,. ,
"Ravi?

29

in producer prices, potential producer income would still
be biased downward if the price elasticity of sales to the
marketing board is positive. For a given increase in
producer prices, potential producer income would be larger
if the price elasticity of sales to the board exceeds the
supply (output) price elasticity. The price elasticity of
supply (output) of a crop could be positive and yet the
price elasticity of sales to the board is zero.

The question of revenue surplus accruing to the
state government and the marketing board at moving average
prices that are higher than actual producer prices was
treated by Olayide and Olatunbosun11--where they compute
this surplus revenue at the moving average prices as the
value of exports at world market prices minus the value on
moving average pricing basis. Their estimates contain a
downward bias for all those crops which have positive
price elasticity of sales (or supply) to the marketing
board. Their surplus government revenue figures implicitly
assmme a zero price elasticity of sales of the crop to the
nmrketing board.

Let us formalize their argument. Suppose

S = (Pw - P M (l)

M)

Where S is their "'producers surplus' in the hands of the

CMB," PW is the world market price, P is their calculated

M

moving average price and M is the quantity of the crop sold

 

.. “re board. Then
“recount; average 1

I." fly! to

112185, we find tha

 

zine hands of the

   
  

::;:e (is. ) is tht
‘M

:ariating board wit

3,3) and the ela;

;:;:es and the mov

sting average pri

53 1013 as the in;

“9‘98 Price do
From (23
It?! A
1’ 341C F
ers surr
is 9‘. .
“re moving a.

in
es 011 these

6

he vs ..,-
“5'1“? aver

5:11-11 ' -
S l“ the

C
d‘ "pvh ‘
Lqr‘
‘u’n
‘ ‘ w
E p L'
t” the
§“' ,
‘ 'IL'N

30

to the board. Then differentiating (1) with respect to

the moving average price PM and converting into elasti—
cities, we find that the elasticity of "'producers surplus'
in the hands of the CMB" with respect to the moving average
price (ESPN) is the sum of the elasticity of sales to the
marketing board with respect to the moving average price
(6MPM) and the elasticity of the gap between world market
prices and the moving average price with respect to the

moving average price where the latter is always negative

so long as the implied tax (Pw - PM) is positive, i.e.,

 

MP P -P (2)

where we have assumed that small changes in the moving
average price do not affect world price.12
From (2), if 6MP = 0, then the elasticity of

M
producers surplus' in the hands of the CMB" with respect

III
to the moving average price is always negative so long as
taxes on these crops are positive. In this case, raising
the moving average price tends to lower the "'producers'

surplus' in the hands of the CMB," and conversely for the

lowering of the moving average price (e.g., compare

Tables 19 and 27 in their article). However, if 8MP is
M

sufficiently positive, it is possible for the elasticity
0f "'producers' surplus' in the hands of the CMB" with
respect to the moving average price to become positive,

implying that it is then possible for increases in the

 

gnrg average Fri“
5::2rs' surplus' i“
.:..1 ' ( 0

a..." if Ey‘P I

M
raised, ' 'producers

 

€111 fall more than

rather Nigerian f 3

  
  
  
  
  
   
 
  

ziicatec'. by a 099i
rabcara as the-Y 2
Also, in their cal
Zlay'zde nd Olatur.
elasticity of sale
to treat. Their
aural sales at t':
aio'inward bias f

E‘astiCitY 0f sa‘

2'3:

\\\\\\

31

moving average price to lead to increases in the "'pro-
ducers' surplus' in the hands of the CMB." On the other

hand, if 8MP < 0, then, as moving average prices are

M
raised, "'producers' surplus' in the hands of the CMB"

will fall more than in the case where 6MP = 0. We doubt
M

whether Nigerian farmers behave irrationally in the way

indicated by a negative 8MP --that they will sell less to

the board as they are paid fiigher moving average prices!
Also, in their calculation of potential producer incomes,l3
Olayide and Olatunbosun implicitly assume a zero price
elasticity of sales to the marketing board for the crops
they treat. Their estimates--arrived at by evaluating
actual sales at the higher moving average prices--contain
a downward bias for those cr0ps with positive price
elasticity of sales to the marketing boards.

In the CSNRD studies perhaps not enough emphasis

was given to the possibility of raising producer prices

and still leaving government revenues from agriculture

 

substantially unaffected, without any resort to petroleum
revenues,14 or new production functions. The recent

Simulation study15

which takes a comprehensive view of the
Nigerian agricultural sector is a powerful application of
the simulation approach to agricultural sector analysis.
The basic shortcoming of the above (arbitrary)
Sample of previous studies is that the effects of govern-

ment pricing policies on resource use in Nigerian farms

were not adequately treated.

 

1 _
G. K. Hel
3.22113 in Nigerian
2.1.. Eicher and C
”an Boone-1r: '

A...

    

33-‘..a.ion hpbrr
5%: Bkal‘lgl Eda '

 

it‘ll“ f0! the "
:moration {an
:ECEr OP tfle “Av-
vrftne Corn“;
:5:st 0f +he \:
figment Pr“
1'.

FOOTNOTES

" 1G. K. Helleiner, "The Fiscal Role of the Marketing

Boards in Nigerian Economic Development, 1947- 61, in
C. K. Eicher and C. Liedholm, Growth and Development of the

_ Nigerian Econom (East Lansing: Michigan State University
WkPress, 1970).

. h...“

Helleiner, op. cit., p. 141.

3H. G. Johnson, "Towards a Generalized Capital
Accumulation Approach to Economic Development," reprinted

in M. Blaug, ed., Economics of Education 1 (Penguin Books,
1968), p. 34.

4See Glenn L. Johnson, "Capital,” International

Encyclopedia of the Social Sciences (The Macmillan Company
andithe Free Press, 1968), p. 231.

 

5See A White Paper on the fiMilitapinovernment

Policy for the Reorganization of the Northern Development
Corporation YKaduna: Government Printer, 1966); A White
Paper on the Northern Nigerian Military_Government's Polipy
for thefi Comprehensive Review of the Past Operations and
Methods of the Northern Nigeria Marketing Board (Kaduna:
Government Printer, 1967); Coker Commission of—Inquirylnto
Operations of Statutory Corporations in Western Nigeria

6See Glenn L. Johnson, op. cit., p. 231.

7See Report of the Study Group on Cotton and Other

Fibres, o . cit., p. 64.' A more subtle argument'for
alfferentiaI pricing would be differential transports

Costs--but this is not the basis for the present policy in
the Western State.

8Glenn L. Johnson, op. cit., p. 231.

32

El .4 .I
I". '4'- '

 

 

9Though th-
Les significant to
xeecrops of the S
:zzt‘teht relative I
serge of these ta

0 .
lJ'r'ellexre:

. I ' F
:5 fiaxezmg :5
I l ‘q n f"
::.:.:ers a

I
:e Mamethg Boar

12 .

This ass
Sigeria in competi
mines, the pm
matters depends
Local world marke‘J
:0; and the e1as°
azthrzigerian expi
greens of other 1
P3, i.e.,

EN+ BO

.. follows that

[*1
Z
"G
In!

There '1 -
A'tLers of the

Ely .' -
‘at‘clty 0f de

:IaSticity 0f 5 L1

 

 

33

9Though the effects of these taxes are probably
less significant for groundnuts and cotton relative to the
tree crops of the South, they are still nonetheless very
important relative to what would have obtained in the
absence of these taxes.

10Helleiner, op. cit., Table 2, p. 123.

11D. Olatunbosun and S. O. Olayide, "The Effect of
the Marketing Boards on the Output and Income of Primary
Producers" (NISER, Ibadan: International Conference on
the Marketing Board System, 1971), p. 64.

12This assumption need not hold in reality. With
Nigeria in competition with other groundnut exporting
countries, the price elasticity of demand facing Nigerian
exporters depends on their share in total world exports,
total world market elasticity of demand for the export
crop and the elasticity of supply of all others competing
with Nigerian exporters. With Nigerian exports (EN) and
exports of other competitors (E0) a function of world price
(P), i.e.,

 

EN + E0 = f(P)
it follows that
S
l 1- N
D,= n” E
ENP§§M sNo

where nEN,p is the elasticity of demand facing Nigerian
exporters of the crop, “M is the total world market
elasticity of demand for this eXport crop, so is the
elasticity of supply of all other exporters and SN is
Nigeria's share of the market.

13Olatunbosun and Olayide, op. cit., Tables 22-25.

14Glenn L. Johnson, 0. J. Scoville, G. K. Dike,
and C. K. Eicher, Strategies and Recommendations for

Nigerian Rural Development, 1969-1985, CSNRD 33 (East
Lan31ng: Mic 1gan State Un1ver51ty, 1969).

15Glenn L. Johnson, etc., A Generalized Simulation
A roach to A ricultural Sector AnalySis With_§pecial

A lication to Nigeria (East Lansing: Michigan State
University, 1971).

 

 

THATION OF

 

Our first
:alyzing the eff
ressurce use in N
Will help us to c'

I": S‘

wm‘.
wt] of us!

CHAPTER III

THE EFFECTS OF GOVERNMENT AND MARKETING BOARD
TAXATION OF CROPS AND SUBSIDIES ON INPUTS ON
THE COMMITMENT OF RESOURCES TO FARMING

IN THE NORTHERN STATES

Introduction

Our first objective is to provide a framework for
analyzing the effects of taxes on marketing board crops on
resource use in Northern States farming. This framework
will help us to determine the effects of these taxes on
the supply of used resources from Northern States farms
and the long run disinvestment in productive resources.
It will help us to determine the effects of these taxes on
the levels of resource use on Northern States farms: have
these taxes tended to induce cotton and groundnut farmers
to use exactly the same, more or less of the resources
(particularly land and labor) that they would have used in
the absence of these taxes? Looking at the problem from
a "second best" point of view, given that we have these
taxes on marketing board crops, what compensating subsidies

on inputs used would minimize the allocative distortions

34

 

#31:: respect to Ii

:5 induced effect:

:::;225 of cotton

 

 

 

his c..apt
bet at the inducel
:ttor. and ground:
:rtps: second we

sizsidies on inpu‘

:esc'arces from fa

“I: . Y!- F
dead for re

 

Anthem States F
C. 5

..a.e Go‘v'emrtxentc
lY“ I 1"
coupensating'

3?ny n
4 “tent taXat

The F
“‘
C

S:
~ J
1“

d

35

(with reapect to resource use) of these taxes? What are
the induced effects of differential taxation on the relative
outputs of cotton and groundnuts in the Northern States?
This chapter has three main sections: first we
look at the induced effects of differential taxation of
cotton and groundnuts on the relative outputs of these
crops; second we look at the effects of taxes on crops and
subsidies on inputs on the supply or retirement of used
resources from farms and the disinvestment process, and
the demand for resources and the investment process on
Northern States farms; and third, the question of whether
State Governments and their marketing boards should engage
in "compensating" subsidies on inputs in the presence of
government taxation is viewed as a "second best" problem.
The Effects of Taxes on the iRelative

Outputs of Groundnuts, Cotton
and: Food Crops

 

The government imposes a produce sales tax and an
export tax that is sometimes partially on a specific tax
basis and partially on ad valorem basis. The marketing
board imposes a tax on producers approximated by the so-
called "trading surplus" of the marketing boards.1 These
three elements comprise taxation by the government. At
the same time the government subsidizes many new inputs
like fertilizers, chemicals, sprays, new seed varieties,
information supplied by the extension services of the

government, etc. Let the farmer's production function

 

he; both traditior
regresented by (i)
:.;;‘.s that are us.
asine services of
"z; durable inputs
firiod; and (c) 56
:erte of use of

1-6 or simply ‘rr

L-o

p
no

.. the durable res

WNW

R:
h. .&.:o-.“,S are to ‘4

 

311: being.

We can w:

~~~~~10n in inplj

“Ems frc

36

when both traditional and new inputs are considered be
represented by (1) below, where the xi's include (a) current
inputs that are used up in a single production period such
as the services of hired labor for a given crop season;
(b) durable inputs that last more than one production
period; and (c) service flows that come from increasing
the rate of use of a unit of a durable resource per unit of
time or simply from more service flows per unit of stock
of the durable resource. Both the input prices and units
of inputs are to be understood in general terms for the
time being.

We can write the representative farm's production

function in implicit form as:

F(Y1' Y2, OOOOO’Y' X1, .0000, Xn) =0 (1)

m

where Yj (j = l, ..., m) is output of the jth crop and X1
(i = l, ...., n) is quantity of the ith resource where at
this level of abstraction some of the Xi's represent service
flows per unit of stock or per unit of stock per unit of
time coming from changes in the rates of utilization of
durable resources.

Let input Xi be subsidized at rate ti per unit so

that the price paid per unit of the input by the farmer =

(l-ti) Px. = Aipx. where Px. is the market acquisition
1a 1a 1a
price of X1 exclusive of government subsidy. Though some

components of government and marketing board taxation are

 

127126 on a specii
surgius and some t
;a:tiy determined
:5 crops in the a‘:
:ftaxation of c:

' !
groportzon of tne

 

taper ton of g:

(4 a. -, . _

:15 tax rate 1;
J

the crop = (1-1 .
3

 

|
emit PIOduce r i:

u“.

"t tathlQn' '1

his
‘ ‘ ECEH
DSQIH
J
a:
:1. =

37

levied on a specific tax basis (e.g., the marketing board
surplus and some of the export tax), these taxes are

partly determined each year on the basis of market prices

of crops in the absence of taxes. Let the three components
of taxation of crops be converted to a tax rate as a
proportion of the potential market price so that a EN 10

tax per ton of groundnuts with a potential market value
(inclusive of the tax, i.e., the potential producer price)
of SN 50 amounts to a tax rate of 20 per cent.2 Let us call
this tax rate Tj so that the price received per unit of

the crop = (l-Tj) PY- = uij. where Py is the expected

3 3 3'
market producer price of the jth crop inclusive of govern-
ment taxation, i.e., the potential producer price.

The profit equation for a multi—crop farm in the

face of these government policies is:

D

m
H = E p. P Y. - A. P X. (2)

Then the farmer maximizes his profit function subject to

the production function in implicit form. Form the

Lagrangian:
m n

L= 2: u.P Y.- 2 LP X.+¢F(Y, ...,X) (3)
j=l 3 yj 3 i=1 1 xia 1 l n

The necessary conditions for equilibrium are:

3L _ 8F _ ._

fi- — 1.1ij + 4) W - 0, 3-1, ...., m (4a)

3 j J

 

3:2: {4a) above:

P.|
E‘i L- J

I: __
I“??? ‘H P
I J

r 1

 

 

itch says that i:
ratsfornation be
;:':ern:ent and me
quantities of ot'
ital the ratio

Eighted by the

"11318 farmer a.

"'I m.

From (4

38

aL__ 3? _ -_

Fii - Ai Pxia + ¢ 5§h+i — 0' 1-1, ..., n (4b)
3L=F(Y X)=O (4)
EE' 1' "" n C

From (4a) above:

BY

F.

_ _ r -
Fl“ :57?- p . ‘4‘”
r 3

yr

1: 1:
HL
"U
Le
U
H
II
|'-’
B

which says that in equilibrium the marginal rate of
transformation between two crops that are subject to
government and marketing board taxation, holding the
quantities of other outputs and inputs constant, must
equal the ratio of their potential (producer) prices, each
weighted by the proportion of the unit crop price received

by the farmer after deduction of all taxes, u., u ; j, r=l,

j r
00., m.
From (4d), it follows that:

P P
u. y. y- u.
J—lg—las—lgl. (48)
u P < P u <
r Yr Yr r
In the special case in which u. = u = 1 (i.e., there is

j r
no taxation of any of the two cr0ps), a point will be

chosen on the transformation curve between, say, millets
and guinea corn that reveals no distortion in the relative

amounts of crops produced by the farmer:

 

P
y.
4‘
P

he
2! yr

.J‘LJ‘

{2:11 leads to “‘0’:

32311326 case in t

In Southe:

:2' Earth East and

.9

:grow either cc
resources at the
atrial taxation 'w
‘rgroduction whi

tents of cotton

 

would have been

snows empirical
Ln the relative
in“. '

...e Northern

uta West, and
A!

" 3'3 model ,

All or}

V

I
r"

d.:‘

3 taxation

§~Uiur~y
. V .1
on of ,
556.“.

the Case

1: .
' 19a

rs and Q
hid A

nave bee
3: Coo

4..th had

39

P P
H Y- Y
.1 _.l
11] > “I“? U r1 > P
r yr Yr

which leads to more of Yj being produced than would have
been the case in the absence of taxation, and vice versa
for “j < ur.

In Southern Katsina and Northern Zaria and parts
of North East and North West States where it is possible
to grow either cotton or groundnuts with the same fixed
resources at the beginning of each growing season, differ-
ential taxation which renders “j ¢ ur induces substitution
in production which leads to the production of relative
amounts of cotton and groundnuts different from what they
would have been in the absence of these taxes. Table 3.1
shows empirical evidence of induced allocative distortions
in the relative amounts of cotton and groundnuts produced
in the Northern States, particularly in North Central,
North West, and North East States, given the specifications
of our model.

All other things equal, government and marketing
board taxation of groundnuts had the effect of inducing
production of groundnuts at a level lower than would have
been the case in the absence of these taxes in 10 out of
16 years and of inducing production on a larger scale than

would have been the case in 6 out of 16 years. Taxation

of cotton had the effect of inducing production of cotton

HUI!)

‘illlilulll‘
XEF

 

 

kahuna—«CHE

 

...,!
CA..~\\2£
.Uth
.0 w ICQEOU
u :2

     

 

H EOE... ‘EL. Ibuhk
H: his—av... rm .0 « IEEOU BIDS—Icha-
HIuUCOoH.“ 092

fi: EDD-.13.:-
nIu .u...au3p~

 
  

   

IUICCCDOEO

1

 

 

 

Ii

leaaou
o‘.‘..b?.2 ~|Ehlh I‘u‘gl.
.1... EIQL.

.66su~ 12mm. n
Ibrvikkufill

 

Phul'PnJLl P- . . .
:1- 3 Fu-r-Jfiu.eia- Ibno.-Fsru3.uh° “a..- :3d.«n.»V Lav I'rvu.hnn -51d'd-‘ in» I"§.F Fw.h|n..n- DC‘UEXH'! menu- UKE-hkpu>°fiv Nae

 

40

.muooa uauauo onu who: moon» on unommam H50 now ouoa escape»
a no: ma onuhwma name no >uw~anuaam>ocoz .cou won xuu ouamomaoo «can oowum mono uoc aw oowum uoosvoua Huwucouom «ouoz
.neuuu «madam cuonuuoz co moosooum auscocdoum can
couuoo no mowufiucusw o>fiucaou on» c« ncofluuoumao 0>aumooHHo on on peso: upon» can A n u:\ a + H I u: I : .omouo moon» no cuon
co moxou oc nu“: “uncanny kn oo>wooou maausuou um3 unsu couuou uou ouaua uouscoum Hoaucouom awed mo scauuomoum on» ma una

.nuceuam ha vo>wmomu haacsuoc on: uonu muscpcdoum you oowum umoacoua Hoaucmuom awed mo cowuuomoum on» mg and

Hugo .mo £8338

 

.m .o .HmnommH couuooimco «muomma nuacocsoum uou mmmmmmw xaa ..ewnu .Go-~mma coupon uou nouamau gearuhh..m ..omma. .a nmzmo
O

.uaumwfiz cfl mcfiuwxuoz couuo
now mousmwm xoe

 

~

U .Hmowfiux .z .wolmmmH ..uwu .

 

 

.uocaoawmm .x .o .~mnommH mouaua couuou “.oanu .mounmoa nunnuasouo
nacho uo mcauoxuox .Homuauz .m .owammaa .‘unuHH nanny .mmma .nazuu

 

 

 

 

 

 

.uwuomwz ca £u3ouu oHEocoom can accecuo>00 oucuasofiu 4 acumoom .uoconaum .& .U .leomma noowum vacdcdouo uom «oouaom
Aha. Hem. mm.vm a.» m~.hv mmh. m.hn m.m m.o~ mead
ooa. com. ~o.wm «.HH ~a.vv as». m.mm m.a ..an voaa
sea. we». mn.am a.w as... Hma. ~.vn a.» m.o~ noaa
mom. eds. uq.vm n.v mp.a. use. n.5n ..G a.~m «was
«mm. ama. «uh.mm woo. oe.mm Haw. ¢.oe o.v o.mm Home
nan. oqo.a sm.nm o.~u ha.mm vac. «.vc «.a o.on coma
Ham. omo.~ pv.~m G.~- so.mm mam. m.sn o.~ p.vn amma
mo.H ooo.a mm.vm .o.- Hm.vm ova.“ m.mm F.H- c.6n amma
omo. was. om.mw H..H oo.vm Ham. m.a. o.ma n.nn smog
mo.” no». m~.~h H.AH ma.mm com. ¢.nv o.o «.mm umaa
em.” mew. mo.am o.p~ om.vm Hm». G.mn a.m p.mn mama
ham. vow. Hm.ah n..~ oa.mm «mo. o.mm n.oH “.mn emu”
GH.H mmm. HH.mm m.mv Hm..m sac. a.om ~.o~ a.on «mad
mac.a Ame. em.cw m.o~ so.mm mas. n.om n..a o.on «man
mo~.a mmm. HH.~HH qo.mm po.mm mew. n.mm n.p~ o.on Hmaa
as». New. hp.~s so.mm sn.on Ame. o.~v o.m« «.Hu cam"
:oa\zm nos\za
a: u sedan ads ocuum n ocuum Ida oowum
H1. : Monsooum ova-oaaoo nooavoum a : noosooum ouaoomaoo noosvoum uoo»
: n adducouom uoz Monacouom uoz
couuoo nuscccsouu

 

.moma-omoa .aaummaz .nauau nouaum
cuonuuoz :0 couscoum nuacvanouu can couuou no nonuum o>duuaom co «ouch canon ucwuoxun! can ucoaduo>oo no uuoouunll.a.n uqnda

 

 

g; a smaller scale

  
    

absence of these t
reduction on a 1"

o

i: the absence of

ritpctential pr
sanity. The at

M

..tton-groundnuts

 

grcrdnuts induc
focd crops than \
these taxes for '
are affected 5;
reSpect to food
Some ii“
by He) should

at
a. other crops

ind the quanti

remain Constan

3
n

.
‘ v

(n . .

“‘0 Crop mixt

t

at ‘ 4
a~es' wit}.

re: : -
”ins th.

V.

41

on a smaller scale than would have been the case in the
absence of these taxes in 6 out of 16 years and of inducing
production on a larger scale than would have been the case
in the absence of these taxes in 10 out of 16 years.

Food crops are not taxed. Thus the proportion of
unit potential producer price received by food crop growers
is unity. The above model therefore implies that in the
cotton-groundnuts-food crops zone, taxation of cotton and
groundnuts induced farmers to produce relatively more of
food crops than would have been the case in the absence of
these taxes for most of the period concerned. This must
have affected specialization patterns in the country with
respect to food production.

Some limitations of the special.case represented
by (4e) should be noted. The assumption that the quantities
of other crops other than the two crops under consideration
and the quantities of inputs used in a given crop season
remain constant may be particularly restricting where in
a given crop season farmers grow more than two crops whose
quantities change during a given crop season. However,
two crop mixtures are common throughout the Northern
States,4 with quantities in the input vector more or less
remaining the same in a given crop year. In such instances,
the model does approximate reality.

The above analysis, therefore, provides strong

reasons to expect allocative distortions induced by

 

   
   
   
  
 

serpent and mar
is relative amour
minced in the Ni
Implicit
glared relative
matted values c
grices. It is e:
influenced not 0.
treats of these
if these prices.
AI .
" a given Cr0p
we mOdel rep
{Mr
av ‘, .
9.5101]. the S
ha.“
. u CIOPS. In

need not hold .

E1
hp
t xa
& ~11.
i? ‘ F
Y

42

government and marketing board taxation with respect to
the relative amounts of cotton, groundnuts, and food crops
produced in the Northern States (1950-65).

Implicit in the model is the assumption that the
planned relative outputs of crops are functions of the
expected values of the probability distributions of crop
prices. It is eminently plausible that farmers are
influenced not only by mean prices, but also by higher
moments of these distributions--particularly the variances
of these prices.5 Insofar as farmers are risk averse,6
we would expect that the elasticity of the planned output
of a given crop with respect to the standard deviation of
the price of this crop is negative. In this sense, the
above model represents the special case when this is zero
for each crop or one in which,in the presence of risk
aversion, the standard deviations are the same for the
two crops. In the more general case, these conditions
need not hold. In addition, other variables like weather,
amount of information possessed by farmers, etc., that
also influence farmers' decisions have not been treated.

Effects of Taxes and Subsidies on

Equilibrium Amounts of
Resources Used

 

From (4d), for the ith input and the jth crop:
A.p
1 x a -F . BY.
1 _ m+i _ .
3;;T——--. Fj — Sii.' from which

 

   

eating that an i.
Librim be demand:
value product of
verzlcsive of qov
1:35 the subs idyl
ti one minus the

v
Oh :‘
-Ibcd'

y

r:- _ 1
"fix. ‘ — T—-. P‘
1'13 U- .

LJ

:
“ We
‘ e

2?-
a‘J‘Pgh

“;‘Hts I
,"A ’
U
v‘lerp"

43

3Y0 A. i = 1' 0.0, n
I 1
Yj xi "j “la j = 1, ..., m

meaning that an input will, in profit maximizing equi-
librium be demanded up to the point where the marginal
value product of the input equals its acquisition price

(exclusive of government subsidy) times the ratio of one

th

minus the subsidy rate per unit of the i input divided

by one minus the tax rate on the jth crop. In equilibrium

then,
1i i = l, ..., n
xi'Yj “j xia j = 1, ..., m
from which it follows that:
A. A. .
1 > 1) 1:1 n
MVP =-—P =P as—=l ’ "" (Sb)
xin “j xia < xia ”j < j = 1, ..., m

where A. is the proportion of market unit acquisition

1
price of the ith input paid by the farmer and “j is the
proportion of the unit potential producer price of the

jth crop received by the farmer. This means that for

given subsidy rates on other inputs, quantities of other
inputs and product price, if the proportion of unit price
of, say, fertilizers in groundnuts production paid by
farmers exceeds the proportion of market unit price of
groundnuts received by the farmer, then as a result of

government policy, the farmer is being induced to use less

2a: the equil ibri

2e obtained in t

 

:‘32 tax rate per 1
equivalent form o:
"wits taken
subsidy rate per

Earner would be i

 

fiend the same

 

braid have demant
arlreting board .
355ertilizers e
the farmer WOUh
more than the e
will have beEr

fixation .

Before

:0 EXtEnd the

‘n"’95tment in

be neither in
"mile thei

taut .'
Lon arid

44

than the equilibrium amounts of fertilizers that would
have obtained in the absence of government policies. If
the tax rate per unit of groundnuts (expressed in the
equivalent form of the proportion of the unit price of
groundnuts taken away by the government) equals the
subsidy rate per unit of fertilizers,then the groundnut
farmer would be induced under our stated conditions to
demand the same equilibrium amount of fertilizers that he
would have demanded in the absence of government and
marketing board policies. If the subsidy rate per unit

of fertilizers exceeds the tax rate per unit of groundnuts,
the farmer would be induced by government policy to employ
more than the equilibrium amounts of fertilizers that
would have been employed in the absence of government
taxation.

Before presenting the empirical evidence, we need
to extend the model to take account of salvaging and dis-
investment in resources and the possiblity that there may
be neither investment nor disinvestment as farmers re-
organize their farms in response to marketing board
taxation and subsidy policies.

In response to changes in the rates of taxes on
marketing board crops and subsidies on inputs, farmers are
induced to reorganize their farms. This reorganization
nay lead to the salvaging of and disinvestment in certain

inputs; it may lead to purchases and investments in new

 

132325 or it may

t‘r-
..g crops on .
51:2.) opportunit‘J
reorganizing the

or marketing boa;

{Ere uj is the

,::;ce of the ma.
is the Polite:
‘3

up,“ 0

.Uk” Y). and Y-
in 'th '
‘3 CrOp' re
L." i #-

.J S the amOu:

~:‘
:‘3e

7
“1' 0n )- ..
. ‘atgd. Le

”he i L '
“dch Says
:13 ‘

is '
lts mar‘KEt

 

 

45

inputs or it may lead to a reallocation of a fixed quantity
among crops on the basis of an internal (within the farm-
firm) opportunity cost principle. The gain (G) from
reorganizing the farm-firm in response to changes in taxes
on marketing board crops and subsidies on inputs is:7

c = ? u.p (y. — ij) - r ai(Xi. - x10) (6)

j=1 3 Yj 3 J 3

where “j is the proportion of the unit potential producer
price of the marketing board crop received by the farmer,
Py. is the potential producer price of the marketing board
crgp, on and Yj are the initial and reorganized output of

h

the jt crop, respectively, ai is the unit value of Xi'

Xij is the amount of X1 used in producing the jth crop
after reorganization and XiO is the initial quantity of

Xi on hand. Let us define three price equations for Xi:
one which says if the reorganized quantity exceeds the
initial quantity on hand, then the relevant price for X1

is its market acquisition price times one minus the subsidy
rate per unit of Xi; if the reorganized quantity of Xi is
less than the initial quantity on hand, then the relevant
price for X1 is the market salvage price; and, if after
profit maximizing reorganization it does not pay to vary

the quantity of Xi, then Xi is fixed and the relevent

price for Xi is its on-farm opportunity cost.

 

ThUSI
m >
if 2 ij
1‘1
,.
g: ._ ( X.“
A. l] l

- if Zx-- = X

here? is th
x.s

if ancunt of X

ire farm-firm w:

these restrictit

 

 

46

Thus,
“‘ o

if 2 xi. > Xi , then ai = AlPX a ,
j=1 3 i

if 2X.. < X.0 then a. = P
. 13 i ' i x.s '
j 1

- _ O .=

and if inj — Xi , then Pxis < ai < Aipxia' i 1,...,n.

where Px s is the salvage value of the 1th input. Let
i

Wi = amount of Xi purchased and Vi = amount of Xi sold.

The farm-firm wishes to maximize (G) above subject to

these restrictions:

2 x.. = x1 + wi - vi (6a)

x. - V. 3'0 (6b)

where (6a) says the reorganized quantity of inputs equals
the initial quantity on hand plus new purchases less the
quantity of inputs sold, while (6b) says sales of a given

input cannot exceed initial quantity on hand. Form the

Lagrangian:
m 0 n n n O
L a X u.P (Y. - Y. ) + X Px 1 — 2 lipx i Z 611[Xi - v1
m n O
+ wi - z xij] + z GZiIXi - Vi] (7)

i=1

"*e Karin-Tucker c

‘ . and W. ar

J

are:

. 3r.

:- -13 —l—- -
g'Y_W"‘y 8X

Jx. J ‘- ..

'3 a l)
—: . - XP
I ll 1 )1
l

LII-:F - I _
i its 11

' 0

‘F‘=Xi - V. +
.1 l

' O

‘gx - v >

“‘e of the
151635 tt |

 

 

47

The Kuhn-Tucker conditions for maximizing with respect to

x.., Vi and Wi and minimizing with respect to 61i and 621

 

 

 

1)

are:

BY
3L ' 3L
Bxij - uij. 3X. - Gli E-O' 8X.. Xij - 0' Xij 2.0 3 (7a)
3 l 1)

3L 3L

aw. ‘ 51. ' MP)... 5. 0' aw "i - 0: W1 1 0: (7b)
J. 1. 1.

3L 3L

3V. 3 Px.s - éli - 621 5-0' 3v, Vi ' 0' Vi 3.03 (7C)
1 1

%_ ‘3 X.O " V. + W. - X X. . = 0 (7d)
11 i i 1 j 1]

3L a x O - v > 0 3L 6 = o 6 > 0 (7e)

8321 i i - ' 3321 21 ' 21 — '

(7a) states that in optimal reorganization of the farm-
firm in response to taxes on marketing board crops and
subsidies on inputs, the marginal value product of the
input times the proportion of the unit potential producer
price of the marketing board crop received by the farmer
is less than or equal to its on-farm opportunity cost,
611. (7b) states that in equilibrium reorganization, the
optimum purchases of Xi involve a net acquisition price
(i.e., acquisition price inclusive of subsidies) equal to
the on-farm opportunity cost of Xi' (7c) states that in
equilibrium, the optimum sale of xi involves a salvage

Value equal to the opportunity cost for Xi when less than

 

 

2;“! is sold.

selling Xi 1
eguiiibriu'f.

"" ‘i‘.
13., acqurs

12;? ) eq

‘.

'I

‘ = 3_

2X w-P
“94 J 3

h
“r
4.

a
J

f’A-I V" ' v
nun-a finlcn

«a
V.
“E n
Prop“,
“4.
I.
Q3“.
H

48

xi0 is sold. In (7e), 621 turns out to be the gain from

selling Xi rather than using it, while (7d) says that in
equilibrium, the constraint in (6a) is satisfied. From
(7bh acquisition of Xi involves an acquisition price

(AiPx a) equal to its on-farm opportunity cost in equi-
i

librium. Substituting this optimum equivalent of the

opportunity cost in (7a), we get:

BY.

31. _ _ 3L _
fit—7 ‘ “ij. nil.- AiPx.a 5- 0' B‘X‘T Xij ’ 0' Xij 10' (7a)
13 J 13 1 13
or
BY. 1.
3L 1 8L
____ = p §_l_‘:_—— P , 5——— x.. =cL x..3_o (8)
axij yj Xij uj xia Xij 1] 13
av. Xi
from which P '5—1_ > P as -— > 1, where P is, in
j xi- Xia ”j xia

equilibrium, the on-farm opportunity cost of Xi' The on-
farm opportunity cost is the relevant price for allocating
a fixed amount of resources among competing crops.

If there are neither sales nor acquisition of a
given input after reorganization, from (7b) and (7c), the
on-farm opportunity cost of the input will be between its
acquisition price inclusive of subsidies and its salvage
price. This on-farm opportunity cost of the input is its
MVP evaluated at the crOp's potential producer price times
the proportion of the unit potential producer price of the

crop paid by the board to the farmers.

 

 

Also

2151035 is

'5 real price
:2 past: a r
1:. taxes on I?
1;:reased in‘.
25.5 ch age J'
tersely for a
resources hat
garticularly
sztsidies on
me wi .11 ma
Ef'nornal“

fiePastures f
hzmrs' inv

exp-e

St a re:

Of IEHr ‘
9 888 '
~\l'

response 5+1

49

Also important for investment-disinvestment
decisions is the most recent history of relative prices.
If real prices of resources have been persistently high in
the past, a reduction in real prices through a reduction
in taxes on marketing board crops may not result in
increased investment in resources if farmers do not expect
this change in marketing board policy to last, and con-
versely for a sudden increase in taxes if real prices of
resources had been persistently low in the past. This is
particularly so for small changes in taxes on crops and
subsidies on inputs. Farmers on the basis of past experi—
ence with marketing board pricing policies form concepts
of "normal" real prices for resources. Any sudden
departures from these "normal" prices may not affect
farmers' investment-disinvestment decisions if farmers
expect a return to these "normal" prices. The importance
of regressive expectations is thus emphasized in our
response studies.

Let there be a production function relating output
of marketing board crop Y to inputs X1 and X2. Assume it
does not pay to vary the quantity of X2 on hand, i.e.,

Px25 < uyPy gé; < A2px a
for all possible changes in taxes on marketing board crops
and all subsidies on the inputs (where ”y is the proportion

of the unit potential producer price of Y received by the

Ir
—.{l"’

 

farm's and x2
31:8 Of X2 pai
In Figu

iii absence of

ss-calied tradi
tax, where all
Talsrem equival
(i.e., 55’ = r
l
:arrer to Chang
;;'.'en level of
value, P b‘
x s' ‘
1
Suppose
a: .
Jt x1 is EE".
Elastity of X
3C"? "in
iaeere is
5a"

“aging of i:

.:.e land and l

 

 

50

farmers and A2 is the proportion of the unit acquisition
price of X2 paid by the farmer).

In Figure 3.1, let EE by the MVP curve of X1 in
the absence of taxes on Y. If the initial quantity of X1
on hand happens to be Kl,8 X1 is fixed in an economic
sense. With the imposition of taxes on Y (marketing board
so-called trading surpluses, produce sales tax and export
tax, where all these taxes are converted into their ad
valorem equivalents), the adjusted MVP is given by EE’
(i.e., EE’ = uyEE). With EE’, it still does not pay the
farmer to change the level of X1 employed because at the
given level of X1, its MVP is still higher than its salvage
value, les’ but lower than its acquisition price, lea'

Suppose taxes are raised so that the adjusted MVP
of X1 is EE”. It pays to salvage some of the given
quantity of X1 since its salvage value now exceeds its
MVP. There is disinvestment by the amount KZKl' Such
salvaging of inputs seems to have occurred withrespect to
the land and labor resources employed on Northern States
farms in recent years. If the initial quantity of X1
happens to be Ki, K3 is optimum quantity in the absence
of taxes. With taxes leading to EE’, optimum amount of
X1 is K3 implying that taxes have reduced acquisition in
the optimum by K3K3. Taxes have thus reduced acquisitions
of X

1 below what they would have been in the absence of

taxes. With the heavier taxes leading to EE”, not only

51

 

 

/
V

 

 

 

TN

 

K1 K1 “3 K3 K\ E “I"
Figure 3.1

is the additional acquisition in the absence of taxes
represented by KiK3 not forthcoming but also some of the
initial quantity on hand (KZKI) is salvaged. The closer

to the origin is the initial quantity on hand of X1, the
more a given tax leads to less acquisitions of X1 than

would have been the case in the absence of taxes. The
movement of farm youth and school leavers from rural areas
and non-use of available suitable land for cotton and
groundnuts in the North-East, North-West and North-Central
States are evidence of acquisitions that are less than would

have been optimal in the absence of taxes.

1m

 

 

In Figurt’
ratings given t‘.
that the farmer

where ll is the

:5 X1 paid by t}.-
facing the farm

Whether
or K3, with sub
line, there wil
Kim K; as the

.: , - .
.. SQbSIClles as

in

E

/ /

 

[

 

 

 

'___

52

In Figure 3.2, let EE, lea and les have the
meanings given them above. Let the subsidy rate be t1 so

that the farmer pays (1 - tl)lea = lllea
where A1 is the proportion of the unit acquisition price

per unit of X1

of X1 paid by the farmer. The new supply curve of X

1
fac1ng the farmer is Alpx a'

1
Whether the initial quantity on hand is Kl'9 Ki
or K3, with subsidies leading to AlPx a as the new supply
1
line, there will be additional acquisition of X1 up to K4.

With Ki’ as the initial quantity on hand, the same amount

of subsidies do not lead to new acquisitions of XI.

in

E‘\

 

(ha
r\\\\ AJ§A

 

 

 

 

 

 

 

l
I
l
i
I
l
l
i
K3

 

i

£5.
p<

K" :K‘KqK? E

Figure 3.2

 

The case
1‘; will be brief
2 ’lea' le):

ratings. If t”

 

aftaxes on Y 1i
stillead to he
czher hand, wit;
sate amount of
evacguisitior.
cse,a resourdI
Variable as a J
Effie initial

3

_--—-.——

"E

14

 

 

 

l

53

The case of simultaneously taxing Y and subsidizing
x1 will be briefly presented. In Figure 3.3, EE, EE’,
EE”, lea' lllea and les all have their previous
meanings. If the initial quantity on hand is K3, imposition
of taxes on Y leading to EE” and the granting of subsidies
on Xl lead to neither acquisition nor sale of X1. On the
other hand, with initial quantity of XI of K1 and with the
same amount of subsidies and taxes leading to EE’ there is
new acquisition up to the new equilibrium point Ki; in this
case, a resource that was initially fixed now becomes
variable as a result of taxes on Y and subsidies on X1.

”

If the initial quantity on hand is Kl , then taxes

\

\ \I
;\l\\ \ ~ 1.3m

["3

 

/
/

 

 

I 1
K5 “3 KI K:

 

 

 

 

 

2»:

Figure 3.3

 

ref-resented by E:

 

lead to acquiSit'
represented W E
sitions of X1 in
, .I

From F19
fcllcwing propor'
szbsidies, if t:

given

w...mic sense ,

 

seed to be subs
to occur. For

the differentia

 

:ze larger the
I.“ n
.. occur and c{
|
niti
.al guanti
‘iill‘ure 3.1)
I
“I99 taxes Is

a 91Ven dif fe]

54

represented by EE’ and subsidies represented by Alpxla
lead to acquisitions by the amount of Ki’Ki; with taxes
represented by EE” and with the same subsidies, acqui-
sitions of XI increase by the amount Ki’Ks.

From Figure 3.1 and the accompanying analysis, the
following propositions can be derived. In the absence of
subsidies, if the resource was initially fixed in an
economic sense, taxes on the marketing board crop would
need to be substantial for salvaging (and disinvestment)
to occur. For this initially fixed resource, the larger
the differential between salvage and acquisition values,
the larger the required taxes on the crop for disinvestment
to occur and conversely for a small differential. If the
initial quantity on hand were the equilibrium amount K3
(Figure 3.1), small taxes leave the resource fixed while
large taxes lead to some salvaging and disinvestment, for
a given differential between salvage and acquisition
values. If the initial quantity on hand in the absence of
taxes is less than the equilibrium amount K3, small taxes
lead to acquisitions that are less than would have been
the case in the absence of taxes while large taxes lead to
pppp_reduced acquisitions as well as some salvaging of the
initial quantity on hand. Apart from the basic result of
resource fixity, other results are obtained that are

different from those of the neoclassical theory of the

firm when the assumption of equality between salvage and

 

 

3""'Sltl0n value

‘.
not
-‘ r

catand of K3 (Fl

:rearcts a reduc.

:axes, the exter.
resource enployr
salvaging up to
eg;als MVP (e.g
'33 in Figure
35319 large th6
tie is the d;
between lea a:
an initial qua.
that adjust BE
vertiCal line
Elf-Ended me
“ilisitions l

taxes. For 1-
‘4

Figure 3.1).

‘
'0»...
““u'

Simi]
mi 3.3 bUt V

saith prOduCJ
|

 

:::#‘lp
.u.& have
Pht. I
-:
er . i
EigCeS ‘

55

acquisition values is dropped. With an initial quantity

on hand of K3 (Figure 3.1), whereas the neoclassical theory
predicts a reduced use of the resource as a result of
taxes, the extended theory predicts that small taxes leave
resource employment unaffected while large taxes lead to
salvaging up to the equilibrium point where salvage value
equals MVP (e.g., K2K3 is salvaged with taxes leading to
EE” in Figure 3.1). The corresponding salvaging for the
same large taxes ('EE”') using the usual neoclassical
model is the distance between K3 and the intersection

between P 10 With

an initial quantity on hand of Ki (Figure 3.1), for taxes

and EE”, which is many times K2K3.

that adjust EE down up to the intersection of lea and the
vertical line at Ki, both the neoclassical theory and its
extended form produce the same result--taxes reduce
acquisitions of x1 below optimal levels in the absence of
taxes. For larger taxes (as represented by EE” in

Figure 3.1), the neoclassical theory predicts disposals or
reductions in quantity on hand that are much larger than
those predicted with the extended theory.

Similar propositions can be drawn from Figures 3.2
and 3.3 but we leave these out because labor and land
which produce the bulk of value added in Nigerian agri-
culture have not been subsidized significantly in the
past.

Some clarifications are necessary for drawing

inferences from available empirical evidence. From (5b),

 

marginal va
exceeds the acq
ratio of the p!
the farmer to 1
price of the C1
:rthis case, t
induce farmers
been the case
onverse resul
Equilibrium us
1'? equals its
fixed resource
2;;ortunity cc
356).

When t

Ilfferential

   
  

P~'
a?

.0 of the

input . -
Held by

56

the marginal value product of a resource from a given crop

11 of the resource if the

exceeds the acquisition price
ratio of the proportion of unit acquisition price paid by
the farmer to the proportion of the unit potential producer
price of the crop received by the farmer exceeds unity.

In this case, government and marketing board policies will
induce farmers to use less of the resource than would have
been the case in the absence of taxes and subsidies. The
converse result holds when this ratio is less than unity.
Equilibrium use of a resource here is defined as when its
MVP equals its acquisition price which in the case of a
fixed resource allocated to two crops is the foregone
opportunity cost (i.e., its foregone MVP in alternative
use).

When the model is extended to take account of the
differential between salvage and acquisition values, a
ratio of the proportion of unit acquisition price of the
input paid by the farmer to the proportion of unit
potential producer price of the crop paid by the board
to the farmer greater thanunity also implies that less
of the resource is being used than would have been the

12 In this case,

case in the absence of these policies.
however, "less use" includes both the acquisitions that
are less than would have otherwise been the case in the

absence of these policies as well as the induced salvaging

and disinvestment in the quantity on hand as the salvage

 

 

value of the res
result of these
relevant acquis:
is the on-farm <
ressurces that
erztezc'ed model,
:‘isinvestment.
this resource fl
1:.tersection p5

Tables

.
-
’i?

"“59th boar
aid and labor
£111

‘1‘ C051”- of l“.
::ޤ1
““‘Y labor (
aerxet aCQUis:

“f resOurces)

 

ngrnmeht ar.
P'l
& C65; C)f Cyt}:

n, .
53ch

Ct priCel

sEQfi .

u: t111:e(i :1-
"4-48 in the _|
SCEE

as salvfl
‘V ltiOns wt
:52“

' palm
.533: (

 

57

value of the resource relative to its MVP rises as a
result of these taxes. For resources that are fixed, the
relevant acquisition price for allocating between crops
is the on-farm opportunity cost of the resource. For
resources that were initially fixed (economically) in the
extended model, large taxes would lead to salvaging and
disinvestment. Even small taxes lead to disinvestment in
this resource the nearer the initial quantity is to the
intersection point between the BE and les curves.

Tables 3.2 and 3.3 show the effects of taxes on
marketing board crops on the real (relative) prices of
land and labor employed on cotton and groundnuts farms in
the Northern States, 1950-66. So long as farmers pay the
full cost of hired labor or bear the full real costs of
family labor (where these costs may be represented by the
market acquisition prices or the on-farm opportunity costs
of resources), i.e., xi = 1, then for any crop subject to
government and marketing board taxation and for given
prices of other inputs, quantities of other inputs, and
product price, farmers would, in equilibrium, be demanding
less hired labor and/or family labor than they would have
done in the absence of government taxation, or salvaging
some as salvage values exceed MVPS. Under our stated
conditions, this means that for cotton, groundnuts, soy-
beans, palm oil and beniseed in the Northern States less

labor (family and hired) is being used than would have

I l “I. D . \Ft\l ' v.5

 

 

 

 

U _ ...L mr.3

«G .wb «23 ~ SCH.
CU « ML LEUSfiOHfl

Cf..r>‘ ‘1
Cu WJHQMJW EQ*L:C ~f¢€$
U “549%va

xcE OUDUOha ozu

a 32a
1 utfirb . _ . 0
Ca Lbbfimaolufi afldUCWUCL ..u
‘ .-N
K ~r.« UCSuCL CCdU..QLChl m.
MC CCthOQ XEL. ACUrvL.
lryh; u fl:
Inglis I
.....a..—|A...s.a~ ‘su— ~r..:J‘z \u.::.-...~ rise-.03. .~,~...~.u.~..z
.~go ltL-.-.~..-:.-v .hn. ~mro‘...~. s... n~.o.4.-s ~.~ l...lc. ~oh.~a\~

 

.' ll I II. I

Nu-nnt

In..—l.u.-

.111». . . (3.3..

F‘s-l... 1‘... :li.4¢ I‘-

 

 

A-u). au.

5.0:» «u...»-

58

.uoan cam Gama nuon mom wuwcd mamswo .mHmEhmm mn cwmm mUHSOmmu may no mowum coflufimwswom
used map mo cofluuomoum on» .H& umcu 0m vmuflpflmndm no: mums Honda can bend umnu omesmmm ma uHm

.waanHm>m mousmwm xmu Hmuou waco .NmIOmmH
Mom .uau .mo .uocflwaamm .M .w mom .NmuommH wousmwm xmu can mdUaHQ mom .mm .m .ma QmZmU .mHHmMaz
cfl muscocsouu molmwflumxumz one .Homwflux .m mom .omnmmma mwusmflm xmu cam mmoflum mom “mouaom

 

 

 

 

hm.H mmh. mHN. m.H H.H ¢.m Hmm.¢ moma
hm.H own. «Hm. m.a ~.H m.m mom.v coma
o~.H 0mm. ova. m.H mo. b.v omb.m mama
mH.H mum. FNH. m.a v.m| v.m mmn.m moma
vH.H mum. mud. o.H ~.~u H.o mmo.m Homa
~m.a 0mm. oma. o.H N.H m.m omo.~ coma
wo.H mmm. «no. o.H m.m| m.m NH¢.m mmma
mm. mmo.a mm0.I o.H m.n| m.m 5mm.~ mmma
mv.a mow. 5mm. o.H w.h N.h mmo.~ hmma
>~.H can. cam. o.H ¢.H N.m NmH.m mmma
5H.H 5mm. mva. o.a «.mn .m.c mwm.~ mmma
m¢.a mmo. mam. o.H ~.m .m.c om¢.m vmma
>m.a mmm. mom. .m.c Ha .m.c mom.m mmma
ov.H man. mam. .m.c .m.c .m.c .m.c «mma
mv.H who. mmm. .m.c .m.c .m.c .m.c Hmma
m¢.~ Hmv. mom. .m.c .m.c .m.: .m.c omma
zm ooo. 2m
muoeumm ou xme commcousm oommzousm cos you couscousm mofiuso Mum»
n oflmm mmz mo o>flwsH0cH cos you oumom mcaumxumz cos uom uuomxm
.1 page ooflum mowum noosooum xma moanoum map 0» msHQHSm moauso Hmuoe
dMl noosooum Hmwucmuom mo uuomxm
K Hmfiucouom cowuuomoum mm
mo c0wuuom xma Hmuoe
loam an:

 

 

.mwmauomma .mauomwz .meumm mmumum cumsuuoz ll:
:0 coonmEm uonmq can pang mo mooflum Ao>flumammv Hmmm no muscocsouu co mmxme mo muommwman.~.m ma
m5.

l

n
1

d

 

 

. cm... a 110th. a

It’s-ch 34
with ll) «I5
09‘s.; uOUSHXF-B

.I. 2.156.. u:

C°<.ou.:—:.-E .«1

~': hI-Iql sIln'K .Ionubn-

Ififlh. ..

In- 1“

Inc—.95.

Ha gulq.~.J=H
'2tu h..:.ul»nfi
~de—.S.~..L LC
Cayudhanruu~53 I‘
IIL. u'GCL.

...l.‘ thus! ’2. P..>-~.~Er—

uni-..-

 

 

CCF ..OA It...
[a — I.!nI-C.»

pup-C Fun-Id h an

..U— I..-

 

32. :3. :t

I: a EU
Isfahan»

A.)1d'~.~hw

 

‘1'! s Sun ..JL

nhAu

«Que:

IIII... M: Iablhulaa4... I452.

.uonua can undo auoa no oases gangs.
.Holudu on... an coda coupon...— 23 no nova gwuouqsvoa you: no gouuonoua 05 .«x 0.5 on conga-n5. you on: hon-H can and." unnu vol»:- 3 and

22.005” «0 0-93 on 3 can: on.) «7.33 uou 3:303 33) .00 .m _. . . gum 05

no aha-8 26‘- 05 ha 3:32. coca Iouu 2.333 393 ucououuad guano uou .05.?! .n .m 72.3 .305 Juana—.633 «0 9338 Haven.— 303.3380

uaoaoodo>oa onusuaaoouoz aaaoouaz no. .Houoooo aoauoo no nouauo uauuoouav no noooum no» .oonoo .oo ..Hsoa .aauaoo canon oaauoxuux any so oocouooaoo

30.33535” “g. MHHZ .6958 8- 356505 «0 «037.3: and How was can: 0.30080 33 5 2.0.39.3 on. gun. on» no m ¢ Jim

i .t and {0.393. .0 .x o: .oonomoa couuoo uo non-am 05333 new as» can: ousnoum no 90309000 usuua «83» was won .mm .m 23 330th .3 oo-

Jbloooa can: 03.3 23 you 3:32. uom .3230. «.3033. 05 no 2:803 2830.3 >5: 5.“: uouon no None.“ 0383!. 1 930259300 $2530: .308

oauuoao co nouns counoog: a can: ans.» oau “on you nooouo couuoo H ovouu once can: Aos-vsv .oo. «.muda conga ..uoaaoadom .uoauoa ousu-ouu nus: noouu-u

Iguana-«u h! luau wanna-cu noon: 305.3300 56 a! and sauna Bacon—noun «duufiuom on» you 3 6023 and scuba» ova-0900 on» noun: 3 gloom." coon.-
ouuum SE. .2 .m .3 9:80 .3 run-ooh. .3 .3qu- Houoqux vows and moo—vacuum ”anions uou noon: Ova—um 39.58

 
 

 

549

 

 

oa.~ poo. no“. o.~u p.o~oa- v.0oa o.ooo wood
as.” omo. ova. o.» o.«- H.~oa o.¢oo mood
ou.~ woo. «on. «.HA m.oov H.~o~ o.n~o «cod
oa.a poo. no”. o.o v.oou H.oon v.o~o noon
o~.~ moo. moo. o.v o.ovon o.o- v.uno «no»
co.” ooo.~ coo. voo.ou n.oo~au H.~au o.oooa Hood
mo. "no.” sno.- 0.“- v.ooou o.vo H.nno oooa
no. one.” coo.- 0.“- «.voouc n.ao o.ono oooa
co.” ooo.~ Hoo.u oo.o. o.o~o«- o.oaa ~.Hoo anon
on.» ooh. van. fl.oH v.ooa a.oo o coo soon
~o.- nor. an". a.p~ o.oov v.oo a «do wood
ov.a ooo. Hon. o.o~ o.nooa H.«o o coo mood
vo.a coo con. o.vu o.ooo~ .a.n a mop vooa
on.a ooo. How. o.nv o.~oo~ .a.a « «do moo”
on.“ How. ova. m.o« n.ooaa .¢.a o our «mod
on.” ooo. How. oo.on o.ooo« .a a non Hood
ow.“ «no. non. vo.oo v.~o- a a u.a ooou
2a coo. so

nun-nan 0» nouns acoanouso ounce nos ..«uan an.»
on: uoao up: can» no o>ausaoau :99 non nae oaauoxunz ousuouo vacuum
.I: ooouo nauseouo guano nauseouo cannon-cu on» on «quay

ox unauauuoo no unannouoa no undouso
guoaoum . n: cofltoaoum on
any annoy

 

.ocodnoooa .qouooos .g-uuu oouuuu auoauuoz no vooonsa “and; can gang no cocoon “coouaaumo «ago no gouuoo no nanny no uuoouuuoo.n.n gangs

 

rear. the case i:
rates of utiliz
explicitly intr
production func
now than would
of taxation. E
:2 work less i:
fore in the ab
fitted real c
that are fully

;rices of othe

:IGduce

Price.

u
“H
u

‘a I taxan WOu

The f

‘~.
a u
“:34?“

"rtions 0

lane: (fami 1']

tie periOd we

.axation polj

 

“ taxeS o .
.ixapton p01
"C

J‘Vundn‘u‘

J. an
LLA
§~.:Sa '
c two r,—
Jnl
sh .
'4 1:».
“ice c
fl
"1114

60

been the case in the absence of taxation. Insofar as
rates of utilization of family labor and hired labor are
explicitly introduced as factors of production in the

production function, labor is being used less intensively

 

now than would have been most profitable in the absence
of taxation. Farmers are being induced by taxes on crops
to work less intensively than they would otherwise have
done in the absence of these taxes. Insofar as there are
imputed real costs (rents, customary dues, etc.) of land
that are fully borne by farmers (Xi = 1), then for given
prices of other inputs, quantities of other inputs and
produce price, these taxes would induce the use of less
land than would have been the case without them.13

The following conclusions may be drawn using the
assumptions of the model. From Table 3.2, both land and
labor (family and hired) used in groundnut production over
the period were, on the whole, induced through government
taxation policies to be employed on a smaller scale than
the amounts that would have been employed in the absence
of taxes on groundnuts. Government and marketing board
taxation policies therefore have tended to limit the size
of groundnut farms in the Northern States measured in
terms of acreage or size of labor or intensity of use of
these two resources.

Only in one year (1958) was this taxation such as
to induce farmers to use more labor and land than they

would have done in the absence of taxes.

 

Table 3
marketing board
results are sir
these taxes inc
Lab: resource:
Period, relati
ergloyed in t‘n

in two years

 

cotton f arme r

 

 

 

 

61

Table 3.3 shows the effects of government and
marketing board taxation on the resources in cotton. The
results are similar to the groundnuts case: on the whole,
these taxes induced the under-utilization of land and
labor resources used in cotton production for most of the
period, relative to the amounts that could have been
employed in the absence of these taxation policies. Only
in two years in Table 3.3 was taxation such as to induce
cotton farmers to use more land and labor resources in
cotton production than would have been consistent with
profit maximization in the absence of these taxes. There
was no year in groundnut production when labor and land
resources were induced to be employed in the same amount
as would have been done in the absence of government and
marketing board taxation, i.e., there was no year in which
taxation policy was neutral with respect to the equili-
brium amounts of land and labor that could have obtained
in the absence of taxation. Only in two years (1958,
1961) was taxation neutral with respect to the equilibrium
amounts of resource use in cotton.

Our results on the allocative distortions of
government and marketing board taxes with respect to
resource use in the cotton and groundnut farms in the
Northern States do approximate reality and have increased
validity because only a very tiny proportion of these

farmers use fertilizers and other inputs that are

 

 

subsidized; her.

and chemicals b

   
  
 

iamers are cor.

gractical purp:.

ifgl-ortant poli

sf this chapte

Some 5

k I
(D
(1)
‘. )J

to be n01

:ias Qfi Out e
using a CObb—
7.22113

..‘arianCe

:f‘-
o E of the ‘

:§ F.~“
‘v \
.1 F,
“I ..
A‘N
‘N‘
‘FE‘AQ_
1“.t ‘J
E‘
.1v-\
‘1" ' Av- ‘
“is?”

62

subsidized; hence, any existing subsidies on fertilizers
and chemicals become insignificant when the aggregate of
farmers are considered and can thus be disregarded for all
practical purposes. As a practical matter therefore,
there have been no compensating subsidies in the past on
any significant scale, though this could become an
important policy variable in the future (see last section
of this chapter).

Elasticities and Their

Limitations

Some serious limitations of elasticity concepts
need to be noted before we illustrate the nature of the
bias of our estimates of effects of government policies
using a Cobb-Douglas function. Samuelson14 has listed the
noninvariance of elasticities to changes in origin and our
failure to treat the variable of interest directly as
some of the problems. Thus, the price elasticity of
demand for a product may be of some use in examining the
qualitative behavior of revenue, but then this is due to
our failure to examine the behavior of total revenue
directly.

Glenn Johnson15 has provided theoretical reasons
why elasticity concepts may not be very useful. He has
shown by implication that the percentage change in a
dependent variable for a one per cent change in an

explanatory variable is not invariant with respect to the

 

 

 

traction and
grevious histo
15:3.” variabl
:etween its 55
are held wi th
::e future, t}
:15 asset.
Let s

tiiit salvage

graduct price

'C"n
“°‘~~lrces
5'11. -
35’)
“‘3 va-
g
“mes.” ir
33.-n
wQPts
C
Pres
H

63

direction and size of the change; to this we add the
previous history of changes or movements in the expla-
natory variable. When MVP's of a durable input are

between its salvage and acquisition values and expectations
are held with certainty that these bounds would hold in

the future, then it does not pay to invest or disinvest in
this asset.

Let some durable inputs have their MVP's between
their salvage values and their acquisition prices. Let
product price history be such that in the opinions of
farmers crop price has been unusually low and they expect
a return to more "normal" prices in near future. If
farmers have regressive expectations, then they would
invest in additional resources even though calculations
using (current) actual product price will indicate that
it does not pay to invest or disinvest in additional
resources. The same considerations will apply in our
analysis of disinvestment in productive assets. All this
calls for caution in our response studies; it is a cry for
modesty in our claims for the usefulness of elasticity
concepts.

Comparative Statics and Biases in Our
Estimate of'Effects 0f_§overnment and

Marketinngoard Taxation and
Subsidies on Resource Use

 

One important use of comparative statics is to

predict the response of the dependent variable (direction

 

335 magnitude)

of analysis is

 

 

ire response of

resources to g:

  
  
 

:ax rates on c:
at their rest.

The na
government pri

F -

:....5T.d curve f

Ll;

191i constant

513*

E“.t1y S t E

64

and magnitude) for given parametric changes. A framework
of analysis is potentially more useful if it can predict
the response of equilibrium amounts of land and labor
resources to given changes in subsidy rates on inputs and
tax rates on crops that are imposed by State Governments
and their respective marketing boards.

The nature of the bias in estimates of effects of
government pricing policies using the MVP curve as the
demand curve for the resource needs to be ascertained.

In the agricultural economics literature, what is being
held constant has sometimes not been clearly and con-

sistently stated.16

The Bias in Estimates of Effects of
Government_and Marketing Board
Taxation of Crops and Subsidies

on Inputs on Resource Use
Using a Copy-Douglas
Production Function17

 

 

 

With the imposition of government and marketing
board taxes on a crop and a subsidy on an input used to
produce a crop, then for given prices of other crops and
other inputs, whether use of the subsidized input will
increase, remain the same or decrease depends on whether
the percentage fall in the net acquisition price of the
input times its own price elasticity of demand exceeds,
equals or is less than the absolute value of the percentage
fall in net producer crop price times the elasticity of

demand for the input with respect to net producer crop

_ _.,-.L

 

 

:rice.18 Estirr

rescurce use de

resource use a
board pricing

7‘ ‘_ ‘
vial. l

3 35-"? curve a

'11

N
(X
.L

65

18 Estimates of induced percentage changes in

price.
resource use depend on these elasticities and the magni—
tudes of the tax and the subsidy. Any bias in any of the
elasticities would bias estimates of percentage changes in
resource use as a result of government and marketing
board pricing policies.

What is the nature of the elasticity bias in using

an MVP curve as the recource demand curve instead of the

"true"19 demand curve? Let the Cobb-Douglas form be:

<1 , Bi > O (i = 1,2), (7)

from which the profits (N) are defined as:

B B
1x Z-Ap
2 l xla

TT=IJPAX

y y 1 X - A P X2 (7')

Solving the necessary conditions for profit maximization
simultaneously for X1 and X2, we get, in the case of X1,

its true demand function to be

8 + -1 B

2
[(81+82) u PYAB2 B

-__1_
B
2
1 1.xla Y

l 2)!&

a demand function that is linear in the logs of the price

variables. P P and Px are the market prices for

y' xla 2a
crOp Y and inputs X1 and X2 respectively in the absence

of taxes and subsidies. From the above, the elasticity

w. _--L

I
hum-n-

 

I
w ‘ "
:f :enanc for 2.

ire elasticity

‘
pr»

...p pIOC‘dCGI

 

66

of demand for X1 with respect to its net acquisition price
(llPx a) is:
l

n = <0;
Xl'(Allea) 1+ 2 1

the elasticity of demand for X1 with respect to the net
crop producer price (uyPy) is

”x (uP)='1'-'B'L-‘6“ ’0'
l' y y l 2

Now suppose we solve for X1 and X2 not simultaneously but

from each marginality (equilibrium) condition separately

(which is what Heady and Ulveling have done in their

analysis of leasing rights), then the demand function for

X1 which is defined for given quantities of X2 is

1 1
31.x 82 _ 61‘1 .
x1 = (Alpxla) [uyPyBle2 1 (7a)

from which these elasticities (whose meanings are clear

from the subscripts) are derived:

, 1
xl'(Allea) B1 1
n, =—l—)O'

from which

 

 

I I
I

"Xult’lpx a)
* l
implies
elasticity of
:Ef-farn acqui
;C'.'err.r:ent an:
extension ser'
when we use a,
true dehand C

Very

m curve was

not only ess e;

5133 be dete .

is also mpg.
.ntuiti'lely
stitching to
al.

a.30 the imp
«1.

...e C5139 .or

Catmgt e163:
ltles Of ‘1
e make Stat
Easting t
or the true
515‘Stlcit‘l
be deeeh

 

67

B182

) =

xl'(AlPx1a)

(n -n >0

 

which implies that there is a downward bias in the
elasticity of demand for labor with respect to its net
off-farm acquisition price (after deduction of such
government and marketing board subsidies like free
extension services, free adult education classes, etc.)
when we use an MVP curve as a demand curve rather than the
true demand curve.20
Very often we write and draw conclusions as if the
MVP curve was the true demand curve for a resource. It is
not only essential that the exact form of the elasticity
bias be determined for different production functions; it
is also important to determine on what its size depends.
Intuitively it would appear that this bias will vary
according to the importance of the input in production and
also the importance of the other input. This indeed is
the case for the Cobb-Douglas function. Larger values of
81 produce larger values of the bias; the smaller the
output elasticity of the input, for given output elasti-
cities of other inputs, the smaller the difference whether
we make statements on the effects of government and
marketing board prices using MVP curves as demand curves
or the true demand functions. Similarly, the bias in the
elasticity of demand with respect to produce price can

be determined.21

 

 

Another
II? of the inp
and its market
of demand for

:arket acquisi

 

failure to (iiS
grices. This
iisinvestment
:arketing boa.
intestment re

raismg Of ma

(f
y
(D
1:,
rd
0 _

:‘japL‘
‘ “isles re
““‘Ctiora, l
1332+

68

Another important source of bias arises when the
MVP of the input is between its off-farm acquisition price
and its market salvage value. In this case the elasticity
of demand for the input with respect to its off-farm
market acquisition price is zero. This bias arises from
.failure to distinguish between salvage and acquisition
prices. This has implications for our predicted investment-
disinvestment response to changes in the level of taxes on
marketing board crops. Suppose we want to predict the
investment response to a lowering of taxes, i.e., the
raising of marketing board producer prices. In Figure 3.4,

let the MVP of X with taxes at a given level be EE’. Let

1
taxes be reduced so that the MVP curve is EE. With initial
quantities on hand of K1, or K3, both the neoclassical
theory and its extension predict increased demand for X1

up to K3. However, through errors of organization, initial
quantities may be K2. With the same amount of tax
reduction, it does not pay the farm-firm to vary the
quantity of X1 as its MVP is still between its off—farm
market acquisition price and its salvage price. Thus, the
neoclassical theory which assumes equality between the
acquisition and salvage values will tend to underestimate

the investment response to a lowering of taxes on marketing

board crops.

69

 

 

 

 

 

 

 

 

 

l

l

l

" , ?
K. “3 K3 “1 E Xl/Xl
Figure 3.4

Aggregation Problems

 

Under normal circumstances when we try to aggre-
gate these individual farm-firm effects of government and
marketing board pricing policies based on the true demand
curves for resources, there would be another bias from the
downward sloping nature of the aggregate demand curve
facing all farmers. We shall not go into the nature of
this bias because of our assumption that all farmers in
any production period face an aggregate demand curve that
is infinitely elastic at the prices set by the marketing
boards. If we summed individual farm effects based on
individual MVP curves, we get another bias. The nature of

this bias will not be examined here.

 

r3
m‘ if
m

L

An :1;
15 that fame}
equilibriun e."

3‘. ,.‘ 1
lsuCl’ aao la:

5 .s;
Lfa$lLiOhal
NJ \‘. ‘ H‘-

v o... SL8“; 'W.

 

E'Iiéence of ,
49.: _ I
MSnell in?

T: .,
“ *6 had cia

was

I y‘

(T)
3

t:
(D
O
Q
c;
,J

'F"‘(V“v\§.
nu \
duke; at),
G:
‘
p
e-
‘Qrfi-i n
“‘L.‘g -\
G
5; ~
“g SUr
UI
R
)- s
.- PH
.sJ H
9uCc' .
50 I
a
9'»
"is
3:.
on
'71..
1‘s:

 

70

The Implicit Assumption of Allocative
Equilibrium: SchulEz‘s Hypothesis
as Our Point of Departure

An implicit assumption of the foregoing analysis
is that farmers in the Northern States achieve allocative
equilibrium with respect to such traditional resources as
labor and land which still constitute the bulk of value
added in agriculture in Nigeria. Time was when the need
was to empirically test Professor T. W. Schultz's hypothe~

22 with respect to the use of

sis of allocative equilibrium
traditional factors. Most of the evidence that is in is
consistent with his hypothesis. Norman's data show
evidence of allocative equilibrium in the use of tra-
ditional inputs like land and labor in Southern Zaria.23
If we had data to estimate the input demand elasticities
then we could use these to predict changes in resource use
for given changes in taxes on crop and subsidies on inputs
brought about by government and marketing boards.
Government and Marketing Board Taxation of

Crops and Subsidies on Ipputs:
A Second-BestfiiProblem

 

If labor produces the bulk of value added in
farming activities in the Northern States and it is not
being subsidized, in the presence of taxes on a crop
produced by labor, if you want to minimize distortions in
the amounts of labor employed, what is the compensating
amount of subsidy on the other factor(s) to keep the

quantity of labor employed constant? The theory of

sccné-‘oest sa
scre of your r

cannot say '3:

‘—
.'.,,'.', . , 6
u. cuer a“ dY .

swag; 5" my: v-
..tnal ..u..a I

For c.

:A 'o .
.:r:1.;zer re

'
“‘Ffi‘. I! s —
nouksoou' 'fl’a:

Li;“.3r ‘
U -S\
w I
ug‘a
"“
:.‘,’.C-' A -

‘ amt
i
..o

o . '_[_l “

\Va., “He r:
L. I
‘U (‘l-

U‘Ill t!

n _
t.“

L

h.

“He acorn-“ h

‘Q.1\a‘..
I

 

71

second-best says that given that you have a violation of
some of your marginality conditions for an optimum, we
cannot say apriori whether we would move towards or

further away from the optimum by violating more marginality
conditions.

For example, with X1 and X2 denoting labor and
fertilizer respectively in a Cobb-Douglas production
function, what are the required changes in the net
acquisition prices of fertilizers paid by farmers and in
the net crop price received by farmers to keep the profit
maximizing (equilibrium) quantity of labor employed on
farms constant? In many important policy making circles
in Nigeria, policy makers are wondering whether, given
existing taxes on crops imposed by State Governments and
their marketing boards, it might very well not be better
to just compensate farmers by increasing the subsidies on
inputs like fertilizers, chemicals, sprays, etc. For the
Cobb-Douglas form, let X1, X2, X3, X4 denote labor of a
given skill, fertilizer, chemicals, and land respectively,

i.e.,

4
Y = AX x x x , Z Bi'< 1; Bi > o, (i = 1,....,4). (8)
From the necessary conditions for equilibrium with respect

to only the off-farm acquisition price and the MVP, the

true demand functions for labor, fertilizer, chemicals,

 

 

art land hold

1
L

112,113, and Xil

 

i,[=l

”331 (36) abs

1109): =‘

l l-E

 

 

72

and land holding only the net acquisition prices of XI,

 

X2, X3, and X4 and the net crop price constant are, in log:
1
10 X. = 10 A + 10 P + X 10 + X -l
9 1 4 l 9 g (uy y) .Br gBr ( . Br )
r#i .r#i
1 - 2 B.
. 1
i=1

109(Aipx.a) - ( X. Br-l)logBi - X BrlongPx a)];
1 r#i r#1 r

i, r = 1, ..., 4. (8a)

From (8a) above,

— 1 —
d log x1 — 1—81-82-83-84 [(82+B3+B4 1) d log (Alexia) + d log (uyPy)

 

‘ 82 d log (XZsza) - 83 d log (XBPXBa) - 84 d log

(A (8b)

4Px4a)]

where we have set dK = 0, where

_ l
K - 1-8 [82 log 82 + 83 log 83 + 84 log 84

1'32'83'84

 

-(82+B3+B4-1) log 81], all constants, involving the
given output elasticities of the four inputs.

Now, set d log X1 = O and solve for d log (Aszza)

d 1 ' P
09 (By y)

 

I

d log (Allea)

l P
09 (uy y)

 

d

 

 

Setting dlog X

r;
“us-fee “
l 2 3 4

““31 says t‘r

‘
um,

""1 would he

 

 

73

Setting dlog X = O,

 

 

1
i:§1-B:-B3-B4 [82 d log (Aszza) + 83 d log (A3px3a) + 84
d log (A4Px4a) - (82+B3+B4-1) d log (11pxa)] = 1-81-82283-84
d log (uyPy) ; (8C)

then for given net acquisition prices of chemicals and land,

d log X1 = 0 implies that24

d log (AZPx a)
2 l
=—>O
dlog (uyPy) 82

 

which says that as the net crop price received by farmers
goes up by 1 per cent, the net acquisition price that

they would have to pay for fertilizers to keep the quantity
of labor constant goes up by the inverse of the output
elasticity of fertilizers. Fertilizers still contribute a
minor share of value added in Nigerian agriculture. It
would intuitively appear that the smaller the factor share
of fertilizers in Northern States agriculture, then the
larger would be the required compensating fall in the net
acquisition price of fertilizers for a given increase in
taxes on a crop if the quantity of labor is to remain
unchanged. For the above case we see that the smaller the

output elasticity of fertilizers, the larger the "required"

percentage re:
lizers to com
producer pric
labor. For e
elasticity of
producer pric

sating" fall

 

of 20 per cer
We d?

either at thl

do have the

from Noman'

-~ I
“rlsation :

 

2dashes al;
I: PIOductié
“mess

h A ‘
Mrvps' Kar‘

d

L'
urn“

 

74

percentage reduction in net acquisition price of ferti-
lizers to compensate for a l per cent fall in the net
producer price of a crop produced by both fertilizer and
labor. For example, if fertilizers have an output
elasticity of 0.05, then a l per cent fall in the net
producer price of,say,groundnuts would require a "compen-
sating" fall in the net acquisition price of fertilizers
of 20 per cent.

We do not have the output elasticity of fertilizers
either at the farm level or at the States level. But we
do have the output elasticity of land in Southern Zaria
from Norman's study.25 The government does undertake
irrigation projects, reclaims lands, takes conservation
measures all of which can be seen as subsidizing land used
in production. If we want to minimize distortion in the
amounts of labor employed as a result of taxation of
crops, what is the compensating amount of subsidy on land
through the above projects for given taxes on crops?

Using Norman's output elasticity for land, we find for
cotton/COWpeas/sweet potatoes crop mixture on ggga land
that a 1 per cent fall in the net producer price of

cotton would "require" a "compensating" fall in the net
acquisition price of land of 1.857 per cent; on cotton
(ggga land), it would require a compensating fall in net
acquisition prices of land of 2.422 per cent; in groundnut

production (gona land), a l per cent fall in the net

 

producer priC
sling" fall
cert. These
tie governme
groundnuts a
requires irr
estimates ar
each carketi
the levels <
cannot appn

can write

AIIV

as

 

 

 

7S

producer price of groundnuts would "require" a "compen-
sating" fall in net acquisition prices of land of 2.746 per
cent. These required compensating investments in land by
the government in the presence of taxation of cotton and
groundnuts are particularly relevant for ggga land which
requires irrigation for all season cultivation (quoted

26 If at the beginning of

estimates are for gona land).
each marketing period, the marketing board in determining
the levels of taxes and subsidies determines that it

cannot appreciably influence world market prices, than we

can write

d 1°9 (AZsza) d log A

_ 2 _ 1
d iog (uyPy) d log u

Y

 

_ 82

so that instead of dealing in net crOp prices and net
input acquisition prices after all taxes and subsidies
have been netted out, we can talk directly of percentage
changes in proportions of unit crop price received and
unit input price paid. Starting from equilibrium
employment of labor, then for given changes in taxes on

a crop and subsidies on land,

d log A
d log u

2
Y

Allv

 

0 according as

AHV

ax F; ,
2

where Axl represents a "change" in X

 

Consi

the largest f
States, then
to minimize c
the "compense
maximizing (z
Pain, the "
the compensa
sition price

reduction it

In t

iemand curs.

me Product

 

 

76

Considering a variant of the above, if labor gets

the largest factor share in agriculture in the Northern

States, then given taxes on, say, groundnuts, if "we" want
to minimize distortion in the employment of labor, what is

the "compensating subsidy" on labor to keep the profit
maximizing (equilibrium) quantity of labor unchanged?
Again, the "special elasticity" we are interested in is
the compensating percentage reduction in the net acqui—
sition price paid on labor for a given one percentage
reduction in the net producer price of, say, groundnuts.

In the case of using a single MVP curve as the

 

 

demand curve, from (8a), and assuming only X1 and x4 in
the production function,
logX=—l—-log(}\P )+ 1 log(uP)+B4 logX+K
1 81-1 1 xla l-Bl y y 1-81 4
_ l l
where K — -——— log 8 + ———— log A
1-81 1 1-81

from which after setting d log X1 = O and assuming d log

X4 = 0,

d log (A P ) _
l xla 81 l

d log (uyPy) = 81—1 = 1'

 

which says that a l per cent fall in the net producer

price of groundnuts or cotton requires a "compensating"

l per cent fall in the net acquisition price of all

labor employed if the quantity of labor employed is to

 

remain constar
in the case 0
tie price of

production f u

 

which Say 5 t‘

“a.

~.-..ton produ

fall in the

 

 

 

 

77

remain constant. The corresponding "special elasticity"27

in the case of the true demand curve for labor, assuming
the price of land fixed is, with only X1 and X4 in the
production function

d log (AlPX a)
l l
2 )0
d P l-
109 (uy y) 54

 

which says that as the board reduces the net groundnut or
cotton producer price by l per cent, the compensating
fall in the net acquisition price of labor would be the
inverse of one minus the output elasticity of land.
Table 3.4 shows some empirical results.

A parallel policy question is the amount of compen-
sating subsidy on fertilizers used on Northern State
farms for a given decrease in the net price of a marketing
board crop if the amount of land employed in these crops
is to remain the same. Policy makers are currently
concerned about the diversion of land from groundnut
production in the Northern States to other crops in the
face of the relatively heavy taxes on groundnuts in recent
years. There is an increasing competition between
groundnuts and the other food crops in Kano State (the
main producing state) with respect to the allocation of
land. A major problem in cotton production in North
Central State, North West State, and North East State is

the competition between cotton and food crops for land.

 

33L}: Mr"

2. Groundn:

m la:

 

 

78

TABLE 3.4.--Compensating (Percentage) Subsidies on Labor
for a One Per Cent Decrease in Net Price of a
Marketing Board Crop so as to Keep the Quantity
of Labor Employed on Northern State Farms

 

 

 

Constant.
Compensating Percentage
- a
Enterprise Sub51dy on Labor
Per Cent
1. Cotton Production on gona land;
84 = 0.4128b 1.702
2. Groundnuts Production on
gona land; 84 = .3641 1.57
3. Guinea corn/Groundnuts
Production on gona land;
84 = .6695 3.025
4. Millet/Guinea Corn/Groundnuts
Production on gona land;
84 = .6429 2.800
5. Cotton/Cowpeas/Sweet Potatoes 1
B4 = .5385 2.166

 

Source: For output elasticities of land, D. W.
Norman, An Economic Study of Three Villages in Zaria
Province: Part II Input-Output Relationships (Zaria:
Institute for Agricultural Research, Ahmadu Bello Uni—
versity, 1970), p. 126.

 

aCompensating percentage subsidy on labor is
defined as

d log (11px a)

1 _ 1

 

b84 is the output elasticity of land.

"'-

"'O i

 

Gwen 1

‘
JEC‘IEEE
4

IQ;?II ‘
'obu y.

' ‘r‘vgv
was

..e

79

Given that there are taxes on cotton and groundnuts, what
degrees of subsidies are required on these resources to
keep their quantities employed in the production of
marketing board crops constant? The question is important
because of the allocative distortions introduced by
taxation-~these compensating subsidies are a way of coping
with the adverse effects of these taxes with respect to
resource use in Northern States farming.

The application of chemicals in groundnuts and
cotton has great potential for high payoffs.28 If the
government wants to subsidize these chemicals (for seed
dressing and spraying) to compensate for taxes on cotton
and groundnuts so that the (equilibrium) quantity of land
employed does not change, then, from (8a) the required
percentage fall in the net acquisition price of chemicals
for a l per cent fall in the net price of the marketing
board crop is:

d log (ABPx a)

3 =1>o
d log (uyPy) F; '

 

i.e., the inverse of the output elasticity of chemicals.
Similarly if the government wants to subsidize
these chemicals to compensate for taxes on cotton and
groundnuts so that the (equilibrium) quantity of chemicals
used in these crops does not change, then, from (8) and

after setting d log X3 = O, the required percentage fall

 

‘n the m

.
cent in

dlc

3;
"J
n

,, 1
"”RI
.‘I.1'.O S
Rf‘bh
HV‘uv'Jn
' \
'9 ‘1‘»

on 3;“:

I‘n
...e (e

Q:

’I

D
H

v“.
.. ,
fi.!

1.“ ts

f!)

zastic.

.

V ..
“Main:
"3'

be

80

in the net acquisition price of chemicals for a l per

cent fall in the net price of the marketing board crop is:

d log (ABPx a)

3 :: 1 >0
d log (uyPy) 1-81-82-84

 

 

which says that as the board reduces the net groundnut or
cotton producer price by l per cent, the compensating fall
in the net acquisition price of chemicals required to keep
the (equilibrium) quantity of chemicals employed in these
crops unchanged is the inverse of one minus the sum of the
output elasticities of the other inputs in the production
function. In the case of fertilizers, the corresponding
required percentage fall in the net acquisition price of
fertilizers for a l per cent fall in the net price of the
marketing board crop is:

d log (AZPx )

12:. = 1

 

> O .

 

With the perfect competition assumption, the output
elasticities of inputs become their factor shares and
arguments about required compensating subsidies on inputs
can be restated in terms of factor income shares.

From the last two results, it is evident that the
less important a given resource is in production (as
measured by its output elasticity) the more it has to be

subsidized for a given percentage fall in the net producer

V

'1!

 

:rice of t

IESC‘QICE l

stirs the
13 D CREE
"is“ ]
:‘bivwo

V
3:56 fid'.’

large ta

have ti
3.“. hand

Orcari -

4 ”4-1

81

price of the marketing board crop if the quantity of this
resource employed is to remain constant.

Let us extend our results to include the case in
which the initial quantity on hand is fixed where its MVP
is between its off-farm acquisition cost and its salvage
price. From our earlier analysis, small taxes in this
case have no effect on total resource employment while
large taxes lead to salvaging and disinvestment of the
input. This implies that for small taxes, no compensating
subsidies are required to keep quantities of inputs on
hand at their initial levels. Large taxes lead to
disinvestment in resources. What is required in this case
is compensating policy action to prevent this disinvestment.

2’

In Figure 3.5, EE, EE’, EE , lea' and les all
have their previous meanings. With initial quantity of X1
on hand at K1 (as arises, for example, from errors of
organization) and taxes on the marketing board crop leading
to the adjusted MVP curve EE’, there is no need for
compensating subsidy on X1 for its quantity to remain
constant at K1 because there is no incentive for its
quantity to change even in the absence of subsidies.

With larger taxes leading to EE”, there is an incentive
to salvage Xl by the amount K2K1‘ To prevent this, policy

makers may "tax" the salvage value of X1: if it is off-

farm salvage value, they may lower this to P; S; if it is
1

on-farm salvage value, they may "tax" this to bring it

82

 

 

 

 

  

 

$ll
Ek‘
E’m\\\
\ \ 6,6
EllK ’
\ PXLS
L_.___ _. _ _ .’
I
I
l
I
K; K) Er ”In?
Figure 3.5

down to P; 3' However, we do not know why policy makers
1

would be interested in maintaining resources at production
levels where market acquisition costs are not being
covered!

Government and Marketing Board Taxation

o Crops Expressed as Taxes on
Resources Used in Farming

 

One possible alternative method of investigating
the effect of government and marketing board taxation of
crops and subsidy on inputs is to express all taxes as
equivalent taxes on factors of production--labor (family

and hired), land fertilizer, hoes, matchets as implied by

 

83

29 The owners of productive

Johnson many years ago.
resources can regard them as items of wealth, the returns
on which are the present value of net income streams
earned by these resources. The present value of the gross
income streams to a productive resource can be approximated
by the present value of the (gross) expected marginal value
I.

product of the resource where the expected prices used are

the potential producer (crop) price in the absence of all

’M‘--‘n—.__ 4
.-

taxation of crops. From this must be netted out the
present value of the per production period costs of using
this productive resource (the user costs which refer to the
costs incurred from the physical depreciation of the
assets which is a function of the intensity of use of
these resources, obsolescence costs, etc.). Then the
capital value of a productive resource over the relevant
horizon (which may be the economic length of life of the
asset or the length of time the farmer plans to remain in
business) can be represented as the discounted sum of the
gross marginal value product minus the present value of

per period costs of using the productive asset, i.e.,

 

u MVP; - T* + S

O t

HMS

K = u MVP - T +
o o o

t 1

 

 

II=Irfrt

*
n
n
1+ . n
( r3) =

3'1 j

where K0 is the capital value of the productive resource

now, TE is the cost of using the resource in period t,

 

I

 

s“

 

84

“t is the proportion of expected MVP of the resource
received by the farmer in period t (MVP*), rj is the per
period opportunity cost of funds tied up in this durable
asset, s; is the expected salvage value of the resource in
time n and the price used in calculating the MVP is the
potential producer price in the absence of all taxation.
The impact of taxation of crops that is not offset by
subsidies is to lower the capital value of resources.
Farmers engaged in the production of marketing board crops
suffer a loss in capital value of their initial resources.
This causes a redistribution of income: farmers
of marketing board crops lose on the capital value of their
resources; others who benefit from the use of this taxation

in social investments (health, education, roads, etc.)

gain. We have a non-Pareto better adjustment.

The Data

If we had adequate time series data on the use of
any resource (fertilizers, chemicals, insecticides, land,
labor, chemicals, etc.) we could estimate the elasticity
parameters. Such time series data just do not exist for
any reasonable stretch of time. Where some data exist for
short periods as in the case of fertilizers it is not
clear whether the figures reflect demand for these
resources or they reflect the efficiency levels of the
Extensions Department of the Ministry of Agriculture and

Natural Resources.30 The "frustrated demand" resulting

*“wMI-r

'M.-__
. .l

 

b,

 

 

I)

85

;he inefficiency of government officials does not

Ip in these figures. Even where the data on resource
rption are available, they are not likely to be

1 down according to crop use. Table 3.5 shows

.izer consumption in the Northern States, 1950-1967.

From the above model one would expect the demand
:rtilizers on Northern States farms to be a function
: acquisition price of fertilizers and the net
:er (crop) prices received after all subsidies and
of the government and the marketing board have been
I out. From theory we expect demand for fertilizers
a function of the real price of fertilizers. Even
had data on the subsidized prices of fertilizers, a
:te theory does not expect real fertilizer price to
.n most of the demand for fertilizers.

The amount of information about the mean of the
Iility distribution of the incremental yield from the
:ation of fertilizers plays a crucial role in the
,on of fertilizers. One can expect that in those
of the Northern States with demonstration plots,

, and other exposure to fertilizer use, the proba-
' distribution of the incremental yield of fertilizers
:s "tighter." Differential access to different types

31 will account for differences in fertilizer

‘ormation
. across the Northern States. In other words, a

‘te analysis would incorporate elements of the

A..-

 

 

86

 

:.oaQMaam>m uoc: mamme .m.c "ouoz

nMMWIdmm .ucmusmq .M .0 "mooaum umuaaauuom oouao
Iamndm co mocwoa>m owumuumom uom .o .m .aabma .wusuasoanom o0 ucoEuummoo amuoomm ”mommav muscocaouo
co mdouorwosum on» NO uuomom .nouvoma «vumlaa wands .nmmwldmm .uocaoaaom .mouomma "mooaum unconsoum
Mom .omumb .mm .Aanma .Eoumwm onmom mCauoxumz map so oocmnoocou amcoaumcuoucav couuou ocm muscocsouo
molmcauoxumz on» you mucosmmcmuua oaumosoo on» Ca mannoum on» «o mo>unm 4 .aamEmH .a .d can whoaauaa
.o .2 some 5.32 517: 032. ..uwlollmm .Hafimfiam $9.83 683m .888 not .2 828 £33m
cumnuuoz xam on» Ca Emumoum c0audnauumao Houaaauumm ocuImcaomm mannoum .ucousma .x .O nonooma nooma

.moonummw amoaumaumpm waummaz cnozuuoz Eoum monomma mousmao c0aumsdmcou umuaaauuom "oousom

 

 

 

 

 

 

 

oa maa hma mmm.o mom.o oom.hv homa
Na mma aoa omm.v mmb.v onm.mm ooma
.m.c oma aoa ama.N th.N oam.ma moma
.m.c mma oma mmm.a oon.a oam.aa coma
.m.c ama mva ama.a moN.a oom.m moma
.m.c ama mva mom mmm moN.N Noma
va mma mma omm mhm omm.a aoma
.m.c oma ova mma Nma omm ooma
.m.c mma Nba mma mva mao.a mmma
.m.c oma ooa moa oma mmm.a mmma
.m.c oma ova mm I moa Nob hmma
.m.c mma mma mha mwa 0am.a omma
.m.c oma boa vm aoa mob mmma
.m.c oma boa mm mm vao Gmoa
.m.c mva moa mm ooa woo mmma
.m.: oma ova mma hma mbm.a mmma
.m.c oma ova 50m 0mm oom.m amma
.m.c 00a 00a ooa boa om” omma
c09\zm ooauomma ooanomma ooauomma ooaummma mace

 

 

 

 

 

mHQNHHflUHMW MUfiHm woflhnw HGUDUOHQ COHflQEaamCOU HMUOB Ham»

 

87

decision-making processes of farm managers and factors like
the type of communal rights over land, the human elements
of their interaction with extensions men, the amount of
peak season labor available for application of fertilizer
and weeding of incremental weed—growth and any analysis of
the effects of government and marketing taxation and
subsidies has to reckon with these factors. Table 3.6
shows the consumption of fertilizers in Kano State 1960/61-
1969/70.

TABLE 3.6.--Fertilizer Consumption in Kano State, Nigeria,
1960/61-1969/70 (Tons).

 

 

Year Superphosphate Sulphate of Ammonia Total
1960/61 200 n.a. 200
1961/62 430 14 444
1962/63 750 196 856
1963/64 1,200 55 1,255
1964/65 1,200 200 1,400
1965/66 2,240 362 2,602
1966/67 3,500 350 3,850
1967/68 n.a. n.a. n.a.
1968/69 8,500 3,015 11,515
1969/70 1,815 508 2,323

 

Source: Kano State Survey of Agriculture, 1971.

Note: n.a. means "not available."

We do not have any reliable data on the main
resources used in farming in the Northern States either on
a time-series basis or on a cross-sectional basis covering

all the states.

88

Summary and Conclusions

The first part of the chapter provided a model for
analyzing the allocative distortions induced by government
and marketing board taxation, with respect to the relative
quantities of crops produced and resources used. With
differential taxation of cotton and groundnuts detrimental
to groundnuts, the relative quantity of cotton was higher
than would have been the case in the absence of all
taxation wherever cotton and groundnuts were substitutes
in production. Implicit in the model is the fact that
food crops, relative to cotton and groundnuts, were
produced in quantities larger than would have been the
case in the absence of all taxation since food crops bore
zero rates of taxation and cotton and groundnuts had
positive rates of taxation most of the period examined,
wherever food crops and these marketing boards crops were
substitutes in production. This is all the more surprising

33 The

given the current trend of rising food prices.
fundamental indictment of government and marketing board
taxation on this score is that it has affected the patterns
of specialization in the country: farmers of marketing
board crOps have been induced to produce more food than
would have been optimal in the absence of all taxation
while other areas with a comparative advantage in food

crop production alone have been neglected in government

policies, e.g., the Middle Belt.

89

The basic model was then extended to take account
of the possibility that it may not pay the farm-firm to
vary the initial quantity of the resource on hand, i.e.,
there is neither investment nor disinvestment in this
resource. The major analytical result was that investment-
disinvestment responses to changes in taxes and subsidies
were, in some cases, different from those predicted by the
neoclassical model. This was specially the case with
resources that were initially (economically) fixed in the
absence of taxes and subsidies. The model was able to
predict, in response to taxes on marketing board crops,
not only acquisitions of resources below what they would
have been in the absence of taxes but also to predict the
salvaging of some of the initial quantity on hand until,
in equilibrium, salvage value equals the MVP of the
resource.

The empirical evidence shows that land (acreage)
in cotton and groundnut farms in the Northern States are,
through the induced response to real input price changes,
smaller than would have been the case in the absence
of taxes on these crops. 'This expected reduced acreage
arises on two counts. First, taxes on cotton and ground-
nuts reduce land clearings or acquisitions below what they
would have been in the absence of taxes. This is signifi-
cant at a time when policy makers in the North-Eastern

State, North-West State, and North Central State are

90

concerned about the fact that farmers are not planting
suitable available land to these crops. Many production
expansion programs over the next decade for most of the
Northern States rely on expanded new acreage as well as
improvement of existing acres. The finding that these
taxes can be expected to adversely affect expanded acreage
in these crops has implications for these production
expansion programs. Second, these taxes lead to salvaging
and disinvestment of the initial quantities on hand. The
initial acreage is allocated to competing crops on the
basis of opportunity-cost on the farm. It is not surprising
that many farmers in the Northern States are known to have
diverted acreages away from groundnuts to food production.
Where cotton and groundnuts compete for land, this allo—
cation must have been influenced largely by the differ-
ential taxation of cotton and groundnuts.

These taxes have had the same effect on labor.
First, taxes on cotton and groundnuts have induced farmers
to acquire less labor than they would have acquired in the
absence of taxes. This reduces the rate of entry of new
farmers into the production of these crops and it decreases
the purchases of hired labor below what they would have
been in the absence of these taxes. Peak season labor is
a major constraint in the production of these crOps. Taxes
reduce purchases of hired labor during these peak seasons

below what they would have been in the absence of taxes

91

on these crops. Second, taxes have increased the rate of
exit of labor from the farm sector: as taxes become large,
not only are labor purchases below what they would have
been without taxes but some of the labor on hand in the
farm-firm as well as in the farm sector as a whole is
salvaged. Labor will continue to be salvaged until, in
equilibrium, the salvage value of the labor of a given age
and skill equals its marginal value product in cotton or
groundnuts. For the older farmers, their relevant salvage
values might be the present values of their expected
marginal value product in food production. Alternatively,
their salvage values might be the present values of their
expected marginal value products in off-farm rural activi-
ties. For the younger farmers and school leavers, the
relevant salvage value may be the present value of the
expected marginal value product in the urban areas. When
wages in urban areas rise from minimum wage legislation in
the presence of taxes on marketing board crops, there is
accelerated salvaging of young farmers and school leavers
from the rural areas until in equilibrium, their salvage
values equal their MVP's in cotton and groundnuts.

Our framework is potentially powerful for modeling
the (tremendous) off-farm migration experienced in recent
years. Unlike most existing models, this model's explicit
focus is on the economic activities of people in rural

areas as these are affected by national policies. Too

 

\V
I.

J.

S‘.

 

We. we I
n

I.

 

92

often, the focus has been on the urban centers. A frame-
work that is capable of explaining the impact of national
policies on labor absorption in rural areas is potentially
useful.

For those resources that were economically fixed
initially, the model predicts that there would be salvaging
of some of the initial quantity on hand as taxes increase.
Not only are the stocks of these resources employed on
these farms less than they would have been, but they are
also being worked less intensively than would have been
the case in the absence of all taxation.

Some resources may be fixed when off-farm acqui—
sition and salvage values are considered. But from our
apriori knowledge of the ease with which resources are
shifted between crops in the cotton-groundnuts-foods
economy of the Northern States, we infer that farmers must
be allocating these fixed resources on an on-farm oppor-
tunity cost basis. These allocative responses are known
to be sensitive to changes in taxes on marketing board
crops. This further strengthens our empirical results on
the allocative effects of these taxes on relative outputs
of cotton, groundnuts, and foods and on resource use.

Given the fact that these resources still produce
the bulk of value added in Northern States farming, these
taxation policies by implication have inhibited the
generation of private saving, investment and the formation

and use of farm-produced capital.

 

--.J

 

ll

(3

a
‘1'
ll

93

The third section of the chapter is more of the

nature of a planning tool: even though fertilizers,

chemicals and new inputs are insignificant in Northern

States farming today, if policy makers are conscious of

the instituitonal reality of marketing boards but are
concerned about the allocative distortions of these taxes
with respect to resource use, then what are the compen-

sating subsidies on these resources so that their quantities

and.their intensity of use remain unchanged? These

conmensating subsidies are shown, within the Cobb-Douglas

production function framework, to be related to the output

elasticities of these resources. Calculations are then

made using available estimated output elasticities of

land and labor. This "second-best" approach has potential

ftxr planning purposes--budget allocations to input

Stuosidies can rationally be related to expected tax rates

or: marketing board crops.

"n—oa— —_
v .

 

 

 

nun R.-

b-LJ

 

Rib

Clo

am DIP...

1.0-

-uu‘ c

«Q ‘.

HUM

a:

.‘h‘

IHU

R44 AV Vin

FOOTNOTES

1Insofar as marketing costs of government and
quasi—government agencies are unlikely to be lower than
competitive marketing costs that would have obtained in
the absence of these government agencies, then the marketing
board surplus is the lower bound of the marketing board
component of the taxes on farmers. This point has been
almost entirely neglected in discussions of marketing
board policies.

2One way of looking at this is to express specific
taxes as their ad valorem tax equivalents yielding the
same revenue. Their effects on outputs and prices are the
same under competition (see R. Musgrave, The Theor of
Public Finance (New York: McGraw Hill, 1959; C. Bishop,
"The Effects of Specific and Ad Valorem Taxes," Quarterly
Journal of Economics (1968).

3This formulation does not preclude positive
initial quantities of resources on hand, resources which
can then be presumed to have salvage values different from
acquisition values. It admits of positive initial quanti-
ties on hand up to the point where the acquisition price
of the resource equals its MVP. For cases in which the
MVP of the resource is between acquisition price and
salvage value and when it is below salvage value for the
initial quantities on hand, see p. 48.

4Some common combinations in Zaria Province have
been documented by Norman: millet/guineacorn; guineacorn/
groundnuts, guineacorn/cowpeas; groundnuts/cowpeas:
okra/pepper; cotton/cowpeas; cotton/cowpeas. See D. W.

Norman, op. cit.

5Behrman did the pioneering work in this area for
Thailand. See J. Behrman, Supplpresppnse in Under-

develo ed A riculture (Amsterdam: North-HOliand Publishing
Company, 19685.

 

94

I,“ _-7 ‘3.
O

 

NJ-

7.

 

-b

H"
'u

.
I .o

of!

EH!

.5

.rIl

95

6Alternatively, farmers may employ "certainty
equivalence" prices.

7We want to emphasize the assumption of initial
equilibrium, i.e., that for constant relative prices, it
does not pay to reorganize the farm. This point was not
given adequate emphasis in Clark Edwards' original good
treatment. See Clark Edwards, "Resource Fixity and Farm
Organization," Journal of Farm Economics (November, 1959).

8This may arise from errors of organization,
imperfect knowledge and foresight, etc.

91h this case, a resource that was initially fixed
at K1 becomes variable as a result of subsidies.

10 . .
In this sense, the neocla531cal theory over-
estimates the induced reduction in resource use in response
to taxes on crops.

11In the case of allocating a given quantity of the
resource among competing crops, this acquisition price
equals, in equilibrium, the MVP of the resource in all
uses (i.e., the on-farm opportunity.cost of the resource).
The two crops (e.g., cotton and groundnuts), one fixed
input (e.g., land) case is only a special case of the model
given on p. 36. The acquisition cost of an acre of land
to be used in cotton is the foregone marginal value product
of land in groundnuts and conversely for the acquisition of
land from cotton production for use in groundnut production.

12Both land and labor are nonhomogeneous. References
to resources are to resources of a given quality.

13Norman has both a lower and an upper bound for the
annual cost of services of an acre of land in Southern
Zaria. In gona land (upland field), this lower bound is
17 shillings and the upper bound is 100 shillings. On
gaudana land (lowland field) the lower bound is 47 shillings
and the upper bound is 370 shillings. In all likelihood
(in the absence of government irrigation and reclamation
projects, etc.), the farmer bears the full cost of this
land. See D. W. Norman, An Economic Study of Three Villages
in Zaria Province, Part II, Input-Oupput Relationships,
p. 125. Bearing the full real costs of land is the common
practice in the Northern States. He found the cost of non-
family labor to be 0.51 shillings/man hour which he also
uses to approximate the opportunity cost of family labor.
This is a simplication but it is convenient.

 

 

A:

est

N

Mir:

4"

96

14P. Samuelson, Foundations of Economic Analysis
(New York: Atheneum, 1965).

15Glenn L. Johnson, "The State of Agricultural
Supply Analysis," Journal of Farm Economics (1960).

16Heady, in an analysis of the effect of leasing

arrangements on farm size, derives demand curves for inputs
that are based on MVP curves. This introduces bias in the
estimation of effects. Such an analysis should be able to
predict the effects of changes in leasing arrangements on
farm size--but such estimates would be biased if we use
MVP curves as input demand curves. Ulveling, in a comment
on the Heady article, fails to bring out the basic
problem--the question of what is being held constant in the
derivation of input demand curves. He does not, for
example, make clear whether his equations at the bottom of
p. 142 are to be solved simultaneously to get the true
demand function or whether each marginality condition
(equilibrium) is to be solved to get the (special) input
demand function. Much of our own analysis of the nature of
the bias should be seen as a critique of Heady and Ulveling.
See E. Heady, "Optimal Farm Size," American Journal of
Agricultural Economics (February, 1971); E. Ulveling,
Comment, American Journal of Agricultural Economics
(February, 19723.

17For other sources of bias, see p. 68.

18This implicitly assumes input use is a function
of its acquisition price and crop price.

19What is "true" in some sense depends on your
assumptions. We use "true" here in the sense that it is
more general-~the demand curve that allows other input
quantities to vary is more general than one which does not.
"More general" could be used in place of "true"; use of
the latter here is the author's preference.

20There is consequently a tendency to under-
estimate the reduction in resource use that results from
an increase in taxes on marketing board crops using the
MVP curve as the demand curve for the resource.

21

n’(P)-n (IJP)=—L----1-—= -82
x ' “y y x ' y y 1-81 1-81-82 (1—81) (1-81—82)

1 l

N

'A—n- - -_...._

97

22See T. W. Schultz, T£ansforming Traditional
Agriculture (Yale University Press, i964?) Schultz's
hypothesis is general enough to include the case in which,
with fixed resources to be allocated among competing
crops, the relevant acquisition price is the on-farm
opportunity cost.

23Letting X1 and X2 denote labor and land
respectively in a Cobb-Douglas framework, Norman finds
these results:

Millet/guinea corn (gona land): 8 =M3§5925, 82 = 0.3189, 81+82 =

0.9104, MVP = 0.58, P = 0.51 => = 1.14; MVP = 66.11,
x x a x
l l x a 2

 

2 = .66 Fadama land (all crops): 81 = 0.2746,

 

 

 

0.90; MVP = 641.85, P
x x a
2 2 x a

ll
L»)
\I
0

ll

sweet potatoes (gona land): 81 = 1.05553 32 = 0°5385’ 81+82 =
MVPx
1 = 1.22

 

1.5941; MVP = 0.62; P 0.51;
x x P

a
l l xla

Source: For output elasticities, see D. W. Norman, op. cit.

24The same result holds if it is the quantity of
land that policy makers want to keep constant.

25D. W. Norman, op. cit.

26Note: These "compensating" subsidies on land in
the face of taxes on cotton and groundnuts are the inverse
of the output elasticity of land in each crop obtained
from Norman's study.

ICIA

Phi c-

Flu.

 

VII. I f\

98

27This "special elasticity," like the ones before
it, is not derived from any obvious behavioral postulate.
However, we could conceive of government officials who are
concerned about distortions introduced by government and
marketing board taxation in the amounts of labor employed
(including induced off-farm migration) but who are equally
conscious of the institutional reality of the marketing
boards reasoning along these lines: "at the going average
rates of taxation imposed by the government and marketing
boards, what would be the required compensating subsidies
on the price of labor that farmers pay or the price of
fertilizers, chemicals, etc., that they pay so as to keep
people on the farms and, say, stem the off-farm migration?"
Such implicit reasoning does indeed exist among Nigerian
policy makers today.

28Johnson states, " . . . Nigeria is unlike many
countries of the world in that it is not private land
owners who exploit the Nigerian farmer; instead, if the
Nigerian farmer is exploited by those who control income
from land, it must be the government which would be doing
so via marketing board levies on the economic rent of
land." See Glenn L. Johnson, "Factor Markets and the
Problem of Economic Development," in W. W. McPherson, ed.,
Economic Development of Tropical Agriculture (University
of Florida Press, 1968?. It Should be pointed out that
Johnson's point is valid for all resources engaged in the
production of marketing board crops receiving no net
subsidies.

29"Fertilizers and insecticides are now purchased
in bulk by State Ministries of Natural Resources and
distributed to farmers through the extension services and
local agents, but in some cases adequate supplies are not
available when the farmers need them." Re ort of the
Stud Grou on Cotton and Other Fibres (Lagos: e eral
Department 0 Agricu ture, , p. 8.

" . . . Nigeria’s agricultural output is being
severely limited by the non-availability of inputs.
No where is the effect of this greater or more forcefully
indicated than in the groundnut area . . . demand . . .
is so strong that it finally changes hands at full cost,
or higher, even though it is intended to sell for half
the actual cost." Repprt of the Study Group on Groundnuts
Lagos: Federal Department of Agriculture, 1971), p. 38.

 

99
30Glenn L. Johnson, Managerial Processes of Mid-
n Farmers (Iowa State University Press, 19615.

31Some of this is due to the poor transport
tructure coupled with poor storage facilities.

 

 

 

f1

vr‘

CHAPTER IV

SALES OF CROPS TO THE MARKETING BOARD
AND GOVERNMENT AND MARKETING BOARD

REVENUE FROM TAXES ON CROPS

Introduction

The second objective of this study is to examine
the quantitative impact on tax revenue from cotton and
groundnuts of paying farmers (generally) higher prices.
This will enable us to see to what extent, if any, the
resulting (crop tax) revenue loss can be offset by the
positive effect (if any) of higher prices on sales of
these crops on the marketing board.

The chapter is divided into two main sections.
The first section examines the determinants of revenue
from taxation of crops, or the relationship between taxes
on marketing board crops and the resulting revenue; the
second part argues that fundamental to the revenue potential
of any crop is the underlying investment-disinvestment
process.

It will be assumed in this chapter that compe—

titive conditions exist in the cotton and groundnut

100

 

1W.

1‘-

101

sectors, that farmers in the aggregate, in any marketing
period selling a crop to the board, face an infinitely
elastic demand curve at the producer price fixed by the
board.

The Determinapts of Government
Revenue From Taxation

of Crops

The immediate determinants of government and

 

 

marketing board revenue from taxation of a given crop are
the quantities offered for sale to the board and the tax
rates imposed per unit of the crop or on an ad valorem
basis.

The quantity of any crop offered for sale to the
board depends in some way on the total output of the crop
produced, the amount of the crop consumed by the farmers
themselves, the amount sold to the non-farm consuming
public or more generally the amount sold on the Nigerian
domestic market through private market channels and the
amounts sold to neighboring countries (e.g., Niger,
Cameroon, Dahomey) through smugglers.

The quantity of a marketing board crop that is
produced would depend on the net producer price of this
crop relative to the prices of closely competitive crops;
the relative availability of new and superior inputs in
the various crops and the amounts of 2933 normative and
non-normative information concerning the expected values

and variances of the probability distributions of yields,

 

 

 

‘V‘

4-

:0
ah

‘ v-

on

my
4.

'F

,‘1

"

In

I..

'6

o
I,‘

102

prices, costs, technological parameters, human behavior,
etc. In the presence of risk aversion, one would expect
the elasticity of output of a given marketing board crop
with respect to the relative standard deviation of its
price to be negative.

The consumption of a marketing board crop by its
producers (total or per capita basis) themselves would
depend on the relative price of the crop and the real
income of the producers.

The amount of groundnuts sold on the Nigerian
domestic market depend on net producer prices for ground-
nuts and/or groundnut products like locally made groundnut
oil relative to the net producer prices being paid by the
board and the ease with which the crop can be transported
to local markets in the country from surplus areas. The
domestic market has exercised a strong pull on marketing
board crops in recent years,1 an influence particularly
strengthened by some shortages created during the recent
civil disturbances. It has paid producers and middlemen
engaged in groundnuts and palm oil production and marketing
to divert large quantities of these crops away from the
marketing boards into the Nigerian domestic (internal)
market. If, for example, the consumption demand for
groundnuts is a function of income and population (for
simplicity, assume own-price constant), then the percentage

increase in its consumption over a year would depend on

 

 

u.

l

103

the income and population demand elasticities for ground-
nuts and the annual percentage changes in income of
consumers and population. Assuming that the income
elasticity of demand for groundnuts is zero and that the
population demand elasticity for groundnuts is unity, the
percentage increase in demand for groundnuts in the given
year will equal the annual percentage increase in popu-
lation.2

The strong influence of net producer prices in
neighboring countries relative to the net producer price
offered by the Nigerian Marketing Boards has become evident
in recent years.3 In spite of institutional obstacles
and poor transportation facilities, market forces can be
expected to bring about movements in the relative quantities
of a crop offered for sale in the respective markets (the
Nigerian domestic market, the market for smugglers and the
marketing boards) in response to net producer price differ-
entials. The smuggling of marketing board crops is subject
to some risk (the legal status of this economic activity is
still ambiguous in places). A marketing board crop will
tend to be smuggled across the border so long as the
expected net producer price exceeds the marketing board
net producer price received by farmers, producers and
middlemen, where the factors used to weight the net pro-
ducer prices ruling in neighboring countries are the

probability of not getting caught in the process of

104

smuggling and the probability of not getting a conviction
once caught.4 It is not clear whether policy makers
should drive these probabilities to zero or one.

In the short run, therefore, we expect sales to
the boards of any crop to be influenced by the net producer
price paid by the board, the net producer price of the
crop or its allied product in the Nigerian internal market,
the incomes of consumers of the crop, the stock of the
crop available in any given marketing period and the
relative net producer price for the crop in neighboring
countries--or more strictly the expected net producer
prices in neighboring countries.

In the long run, sales to the boards are also
affected by substitutions in production in movements along
given transformation curves as a result of differential
rates of taxation on crops or more generally on economic
activities and shifts to new transformation curves due to
research etc. Farm managers will in this case not only
be affected by relative expected crop prices but also
relative standard deviations of prices and yields.

Determinants of the level of taxes to impose will
include projected revenue needs of the marketing board for
its activities (of which stabilization is only one), State
Government revenue for development plans, and even political
considerations (the level of taxes on cocoa was a political
issue in the Western State in the 1950's and the early

1960 '5) .

105

Intuitively, it would appear that the response of
tax revenue from a given crop to changes in the board's
producer price for this crop is somehow related to the
response of sales (to the board) of this crop to changes in
the board's producer price. In the case of two crops,
given the same percentage change in their producer prices,
it would appear that tax revenue response to this price
change will be "larger" with the crop that has the "larger"
response in sales (to the board) to price changes. What
then is the link between the response of tax revenue from
a crop to changes in the board's price (or inversely to
changes in the level of taxes per ton on this crop) and
the response of sales of this crop (to the board) to
changes in the board's producer price?

Sales to the Marketing Board and

Government and Marketing Board
Revenue From;the Taxation

of Crops

Define revenue proceeds by the State Government

and its marketing board from taxes on a given crop as:
R. = t.M. (1)

Where Rj is total revenue from all taxation on a marketing
board crop, tj is the composite unit tax and Mj is the
quantity of the jth crop sold to the board. Total taxation
is composed of marketing board so-called trading surplus,

Produce sales tax and export tax. In such a regime of

 

s '1.
1:011

Gas

Nul-

’t)

a}
5.1

II“

106

unit taxes the elasticity of government and marketing
board revenue from taxes on a crop with respect to the net
producer price paid to the farmer is the sum of the unit
tax elasticity with respect to the net producer price5

and the elasticity of the quantity of the crop sold to the
marketing board with respect to the net producer price.

If policy makers implicitly assume that prices
paid to farmers exert no influence on sales of crops to
the marketing board, then the revenue elasticity with
respect to the net producer price equals the unit tax
elasticity with respect to net producer price. Since the
latter can be presumed negative as long as taxes are
positive, policy makers and some academicians implicitly
assume that the elasticity of government and marketing
board revenue from taxes on a given crop with respect to
the net producer price paid to farmers by the board is
always negative (i.e., that raising net producer prices
paid to farmers will always reduce government and marketing
board revenues from taxes on the crop). If the elasticity
of sales of a given crop to the marketing board with
respect to its net producer price is negative (i.e., sales
fall as farmers are offered higher prices), then the
revenue elasticity with respect to net producer price is
always negative which implies that government and marketing
board revenues will always fall if taxes on crops are

lowered or net producer prices are raised; if the elasticity

107

of sales to the marketing board with respect to net
producer price is positive and is greater than the
absolute value of the unit tax elasticity with respect
to net producer price, then the revenue elasticity with
respect to net producer price is positive, meaning that
government and marketing board revenues from taxes on a
crop will £i§g_when farmers are paid higher net producer
prices.

This possibility is potentially very significant:
revenue for State Governments has become a constraining
force6 in State development plans; this ability to raise
State revenues from internal sources affects the overall
economic and social development of the country. Higher
producer prices paid to farmers benefit them on many
counts: higher cash money incomes enhance the farmers'
ability to invest in new and superior forms of capital
(human and non-human), an investment ability that is
essential to the transformation of traditional agriculture;
or to create a potential market for the goods of the young
but rapidly expanding industrial sector, though the
importance of this depends on the percentage change in
farmers' income and the income elasticity of demand for
these products. Higher producer prices would slow down
the artificially induced exit of resources from the farm
sector as in the case of labor whose expected marginal
value products on the farms are drastically lowered by

heavy taxation of marketing board crops.

108

Tables 4.1 and 4.3 show the unit tax elasticities7
for cotton and groundnuts. The higher the taxes on crops
the larger is the (absolute value of) elasticity. These
tables form the basis for the revenue elasticity tables.

It is to be expected that if taxes on a crop are
very heavy and sales of this crop to the board are price
responsive, sales will fall and government revenues will

also fall. If sales are indeed very price responsive, it

is possible that increases in producer prices paid to
farmers will induce large increments in sales and possibly
lead to a rise in (crop) tax revenue. This indeed seems
to be the case for the levels of taxes imposed on cotton
and groundnuts for the period 1950-66, as shown in

Tables 4.2 and 4.4. It has been shown (footnote 4) that
the elasticity of government and marketing board revenue
(from taxes on a crop) with respect to the crop net
producer price paid by the board to farmers is the sum of

the unit tax price elasticity (E ) and the elasticity

t.P
j n
of sales of the crop with respect to the net producer

price of the crop paid by the board (E ). Under the

Man

assumptions (stated on page 100) above, the unit tax price
elasticities are presented in Tables 4.1 and 4.3 for

cotton and groundnuts. To get the elasticity of government
and marketing board revenue (from taxes on a crop), i.e.,

ER P , we assume different values for the elasticity of
' n

3
sales of the crop with respect to the net producer price

109

TABLE 4.l.--Unit Tax Elasticities for Cotton Using Producer Prices, Taxes and Potential
Producer Price, Nigeria, 1950-1966.

 

 

 

 

 

 

Potential Unit
Cotton Producer Tax

Year Prices Taxes Price Elasticities

Pn tj Pp
aN/Ton

1950 36.37 36.04 72.77 -l.01
1951 55.07 58.04 113.11 -0.95
1952 55.07 29.50 84.57 -1.87
1953 54.81 43.30 98.11 -l.27
1954 55.10 24.30 79.40 -2.27
1955 54.68 27.00 81.68 -2.03
1956 55.13 17.10 72.23 -3.22
1957 54.88 14.10 68.98 -3.89
1958 54.91 -0.04 54.87 1372.75
1959 55.07 -2.60 54.47 21.18
1960 55.07 -2.00 53.07 27.53
1961 46.67 .004 46.674 -ll66.75
1962 45.27 4.70 49.97 -9.63
1963 45.27 6.90 52.17 -6.56
1964 46.67 11.20 57.87 -4.17
1965 46.67 7.60 54.27 -6.14
1966 44.80 -l.30 53.50 -34.46

 

Source: Producer Prices, 1950-1959, Extended and amended Kriesel Series: H-
Kriesel, Cotton Marketigg in Nigeria, CSNRD-24, p. 73. The price series 1960-1966 are
our own constructed series resulting from our dissatisfaction with existing series.
Helleiner's Table 11-8-6 (pp. 474-475) usod only Grade 1 prices for all the years when
a weighted price index of prices is cloarly better. However, constructing a weighted
price index faces many problems. For weights for the price series 1960-1967, see H.
Kriesel, Cotton Marketing in Nigeria, CSNRD-24, p. 55; for net prices after deduction
of produce sales tax for different grades of cotton for 1960-67, sec M. o. Titiloye and
A. A. Ismail, §;Survey of the Trends and Problems in 322:00mestic Arrangements for the
Marketipg of Groundnuts and Cotton, International Conference on the Marketing Board
System, NISER, Ibadan, 1971, pp. 78-80. Prices for different grades of cotton 1968-71
from Report of the Study Group on Cotton 59d Other Fabrics, National Agricultural
Development Committee, Federal Department of Agriculture, Lapos, 1971, p. 7; weights for
prices of different grades 1968-70 from moan weights in above Report of . . ., p. 60
while weights for 1971-72 were assumed to be those of 1970—71.

Note: The unit tax elasticity is defined as the percentage change in taxes per
unit of a marketing board crop per unit porosntago change in the not producer price of
the crop paid by the marketing board (exact formula derived in footnote 4--the negative
of the not producer price divided by the composite tax per unit of the crop). Thus,
the unit tax elasticity is negative as long as taxes are positive (no subsidies). The
higher the not producer prices (Pn) relative to those taxes (t) the larger the (absolute)
value of this elasticity.

110

.eoaum neocooum uec e.oueon on» O» uuemneu new: canon ecu on common uo modem mo unwoauneao ecu can: eeneeuocw meowum Hoosooum

magmas“ no huwawnwmnom ecu .Hm>ea xeu co>wm s mom .couwoo uOu modem nauseous use n.0ueon on» ca emceno omeucmuuom eco e Ham

msce>eu use was» an euceco emeuceuuom ecu me pocmueo an .cm.mm .couuoo MOM modem mooscoum um: n.0uson ecu on uwwmnou nun: coupon
so none» Bonn nwace>om venom mcwuexuex use ucmscue>ou emueum cuecuuoz no huaoduaeam e59 .H.v manna eouu cm.ua "ouoz

.H.v canes an as seen nous» .meOMum couuou "ouusom

 

 

 

 

 

 

nv.n~- nv.Hn- ne.~n- n..mm- n..vnu n¢.nnu n..wnu on.mu an... noon
4H.H- «n.a- .H.¢- va.nn «n.a- hd.pu ea.nu on.p an.nv nnnn
mn.o 5H.H- ”n.a- an.n- an..- 5H.nu 5H.¢u °N.HH nn.nv eons
nn.~- nn.m- nn..- nn.nn nn.nu nn.nu nn.nu an.o su.nv nnnn
no..- nn.nu nn.as «n.a- nn.nu nn.ann no.na oa.¢ bu.n. «non
nn.nnnau na.mnHH- na.vnHHu nh.nnaau np.nnmnn ns.pnsnu na.nnnnn vo°.o an.nq nnnn
nn.nn nn.on nn.n~ nn.n~ nn.>~ nn.n~ mn.h~ oo.u- no.nn onnn
nH.na on..~. na.n~ na.- on.a~ na.cu «n.au on.nu no.nn nnna
nn.pann ns.nann nn.eann ne.nnnn np.~snm nn.n»nn nh.~ann vo.au an.qn onus
an.” no.o- no.n- an.~u no.nu no..- nn.nu on.¢n an..n snag
ne.a -.ou «n.a- «~.~- -.nu -..n «n.a- on.hn na.nn nnnn
an.” en. no.ou no.nu no.~u no.nu no.~u oo.pu on.vn nnnn
ms.“ nu. n~.on a~.au a~.~- n~.nu p~.~u on..~ oH.nn ennn
ma.n ne.n me. p~.o- ~N.H- n~.~- s~.H- on.nv an.vn «nan
nH.n MH.H ma. an.ou an.H- an.~- pn.nu on.n~ no.nn «nan
no.4 no.~ no.” no. nn.ou nn.nu nn.ou ec.an nc.nn Anna
na.n an.” an. Ho.on mo.nn Ho.~u Ho.nn vn.nn pn.nm anon
60
m I M I N I H I O I H I l .H.\2fl
c n c n c n c n c w c a c n n» on
a an a mu m an a mm a an a as a an new»
eoxea noowum
uceueuwwo haw noauwoauneau occu>om xea couuou

 

.nnnnuonnn .onuomnz .uouaun cumauuoz .uunon «a» 0» ocean no nonununuuonm
OUflHm ”Oman: Ugufluwfin HON GOUUOU HO OOfiHm H005muOHm n.6u08 Gnu Ou Hoamflfi 5%! IfldflfluuuudHfl 05G0>0¢ Run. «GOHHOUII.N.C an.

lll

fmxBLE 4.3.--Unit Tax Elasticities for Groundnuts Using Producer Prices,
Taxes and Potential Producer Price, Nigeria, 1952-1964.

 

 

 

 

Potential
Producer Composite Producer Unit Tax
Year' Prices Taxes Price Elasticities
P t. P
n J P
bN/Ton
1952 35.0 20.1 55.1 —1.741
1953 35.5 16.3 51.8 -2.177
1954 35.4 5.9 41.3 -6.000
1955 32.4 8.6 41.0 -3.767
1956 37.4 15.8 53.2 -2.367
1957 33.9 -1.7 32.2 19.941
1958 36.4 2.8 39.2 -13.000
1959 37.4 8.2 45.6 -4.560
1960 33.7 4.8 38.5 -7.020
1961 30.3 4.4 34.7 -6.886
1962 30.3 6.2 36.5 —4.887
1963 32.7 8.9 41.6 -3.674
1964 34.1 9.3 43.4 -3.666
Source: Producer Prices 1950-63, Helleiner, op. cit.,

Table ll-B-4; 1964-65, Report of the Studngropppon Groundnuts,

Federal Department of Agriculture, Lagos, 1971, p. 6.

Tax figures

from H. C. Kriesel, Marketing of Groundnuts in Nigeria, CSNRD-19,
Michigan State University, East Lansing, 1968, Table 14, p. 68.

Note:

The unit tax elasticity is defined as the percentage

change in taxes per unit of a marketing board crop per unit percentage
change in the net producer price of the crop paid by the marketing
board (exact formula derived in footnote 4--the negative of the net
producer price divided by the composite tax per unit of the crop).
Thus, the unit tax elasticity is negative as long as taxes are

positive (no subsidies).

The higher the net producer prices (Pn)

relative to these taxes (t) the larger the (absolute) value of this

elasticity.

1.1“2

: n .N - gamma non nn~.o n genes can H u amaze
you Hon.ou I m.mm .mmma cw cou\H.o~ 2m uo muscocsoum co nosey enmeomeou nun: ..u.e .euscocsOum no munuwueeae eowum mouse on» sun: eeueeuoca eeonun
mesons eueaueu acumen ma esce>eu vueon unauoxuss use ucoe=u0>om ocweeeuocn no >uwaanneeoa on» .euscvcsoum co nexeu no ae>ea ce>qu now so: eoauoz

.OUmua headache uec e.oweon ecu ou avenues can: oneon ecu on euncocsoum uo eeaee no

auwuaueeae on» sun: eeeeeuosa eouum nauseoun memeaeu an esco>eu can» mannweu uo mumawnwuaom ecu .He>ea xeu ce>am e mom .auscucdouo wow enema ueosooum

uec e.oueon on» c« emcenu eueuceuuem eco a now esce>ou menu cw eoceno emeuceouem on» as oesdueo ed euscocfiouo now euawm noose um 90: e.oueon ecu o»
uuemeeu can: successouu co nexeu souu neoce>em venom mcwuexue: use unsecue>oo aeueum chocuuoz no madame-eds one .n.v easea see .cm um you aeuoz

.o .m .usma .eomem .ewsuanowumc uo uceauuemea aeueoem .eouuwfifioo acoamoae>eo
ouo vsum ecu uo an em .mooauvoma .vumuaa vanes ..unu .mo .uocweaaom "momHIOan even neuaum someone «eonsom

 

Heunuasuaumd Hecoauez .euscocsouo co

 

 

 

 

 

 

enn.~ vnm.o noo.n- non.~- non.nu non.v- non.n- n.m n.vn «and
can." nun.o can.su ven.~- can.nu can..- .56."- n.o n.an mend
nnm.a enm.on enm.n- nmn.nu amm.e- nan.n- amn.vu «.n n.on «can
nnn.o- new.«. one..- nma.nu nmo.nn omn.a- nmm.n- v.v n.on Hons
ouo.nu omo.n- o~o.n- o~o.nu o~o.r- oao.o- oao.nu o.v a.nn coma
ovo.~ onn.ou onn.~- onn.n- onn.vu own.n- onn.vn «.n v.5n nnna
ooo.hu ooo.au ooo.nmu ooo.~«- ooo.nm- ooo.va- ooo.nn- o.« v.nn «nan
Hen.n~ an.n~ Hem.n~ Hen.o~ men.ns Hem.nm Hen.ma n.a- n.nn snag
nnn.n nnn.n ann.o- soM.H- enn.~- ann.n. snn.«n n.nm v.5n anon
"an." nn~.o hoe.au nnh.~- ene.nu an“..- sne.n- o.n v.~n nnma
. . ooo.«- ooo.v- ooo.nu ooo.o- ooo.~- ooo.n- n.n ¢.nn vnnn
nuo.n nan.H sen.on ahn.n- epn.~u aen.n- apn.~u n.an n.nn nnnn
nn~.e nnu.n nnn.o nvp.ou HvF.H- th.n- «vs.nu n.o~ o.nn «non
v n N u a n o n n- u eosxza

c n a n a n a n a n c n a n “a as
a an a an a an a re a xm a am a an

sexes eoowum use»

uceueuuwo Hen mequuuaueeam esce>ez mes assessouo

 

.vomnuanoa .anuumnz .nou-un :uoauuoz .eneom oau 0»
eeHem uo eewuaowueeau eouum assume: uceueuuwn Now sound moostum n.0ueom mcwuexue: ecu ou uoemmem nun: meauauwueeam esce>um xee “euncvcsououn.e.e mange

113

of the crop paid by the board (E ). The assumed values

M.P
j n
for EM.P are -1,0,1,2,3,4,5,6. From the calculated Et.P
J n j n
and the assumed values for EM P , ER P is calculated for
j n j n

1950-66. From Tables 4.2 and 4.4, the possibility of
increasing tax revenue from a crop for given levels of
taxes increases with the elasticity of sales of the crop
to the board with respect to the net producer price paid
by the board. This possibility increases with the level
of taxes, i.e., for given sales elasticity of the crop
with respect to its net producer price, the elasticity of
tax revenue from a crOp with respect to the net producer
price it pays to farmers increases in numerical value with
the level of taxes.8 From Table 4.2, in 1952, composite
tax on cotton was LN 29.50 per ton and with a sales price

elasticity of cotton = 2, E = 0.13; in 1953, the

R.P
3 n
composite tax was LN 43.30/ton and with the same sales

price elasticity, ER P = 0.73. Allowing only the unit
j n

tax elasticity to vary and setting E = 1, the corre—

M.P
J
sponding figures are: for 1952, ER.P -0.87; for 1953
3 n
ER P = -0.27. Thus as taxes on marketing board crops rise,
j n
the responsiveness of (crop) tax revenue to changes in net

":3

producer price also rises and conversely for a fall.
Resource Fixipy, Supplpresponse
and Government Revenue
If relative crop price changes in a given crop

enterprise are such that the expected MVP's of resources

114

are bounded by the salvage values and acquisition values
of these resources so that there is no motivation for
investment or disinvestment in these resources, the crop

supply (output) response to crop price changes will be

small. Building on the recognition that a durable

resource has two prices, the Johnsonian theory leads us

to expect, in principle, a difference between expansion

as... s

and contraction supply price elasticities corresponding,
respectively, to the rising and falling phases of crop
prices.9

Even though sales to the marketing board are of

Elrdnary interest for (crop) tax revenue purposes, insofar
as; we can presume positive correlation between total out-
EDtrt and sales to the board, we would still be interested
.ir1 the underlying production relations as such.

Starting from a production function, we can write
the output (supply) price elasticity of a crop as the
w€Eighted sum of elasticities of demand for the productive
reisources with respect to the crop price where the weights
Eilrea the output elasticities of these resources. From
tzk1£is, the more resources whose demands (and supplies of
t153€3d quantities on hand) are not affected by crop price
Cibiéinges, then the less likely it is that the supply of
tlrlee crOp will be affected by crop price changes. In

EtcIiCiition, for this to be strictly true, it must also not

pay to change resource service flows per unit of stock or

115

per unit of stock per unit of time through variations in
the rates of utilization of the durable resources.
Implied in the above is the following. The link
between the investment and disinvestment response of
durable resources to crop price changes and the supply
price response of crops would seem to be influenced in
some way by the "importance" of the resources in pro-
duction. If it does not pay to invest or disinvest in
some durable resources that are relatively unimportant in
the production of this crop but it pays to invest or
disinvest in a resource whose "importance" in the pro-
duction of this crop far outweighs these other resources,
then we may observe significant output response to crop
price changes on account of changes in this resource even
though it does not pay to invest or disinvest or vary the

rates of utilization of the other resources.10

The higher
the output elasticity of a resource, the stronger the link
between its investment and disinvestment response to crop
price changes and the supply price response of the crop
and vice versa. Under competition, the larger the factor
income shares of resources, the stronger the influence of
the factor's investment and investment response to crop
price changes on the output supply (price) relations of
the crop.11 Irreversibilities in the crop supply relation

derivable from irreversibilities in the investment-

disinvestment relations in productive facilities may enable

116

governments and marketing boards to maintain higher tax
rates on a given crop than would otherwise be the case in
the absence of these irreversibilities. The extent of
these irreversibilities will to some extent be influenced
by the ease of substitution of resources among enter-
prises.

For example, let the acquisition price of a
farmer of a given farming skill and age be what it would
cost a farm-firm to "acquire" a new farmer of the same
vintage now. This is the supply price of a farmer of
this vintage. Let his salvage value be approximated by
the present value of the net income streams that he would
earn in the next best alternative use. If government and
marketing board taxes are such that the MVP's of resources
(labor and land) are bounded by their off-farm market
acquisition prices and salvage values, where the relevant
salvage value is still within the farm sector, then
government and marketing board taxes induce farm reorgani-
zation that still leaves most of these resources within
the farm sector--a farmer may reorganize and quit groundnut
production in favor of food production as a result of
government pricing policy.12 If resources are not enter-
prise specific, then the degree of resource fixity within
a given crop enterprise may be much less than that for the

farm sector as a whole.

117

These propositions may then be derived: from our
aprior knowledge of the limited degree of resource fixity
within crop enterprises, and the transformation possi—
bilities between crops in the cotton-groundnut-foods
economy of the Northern States, the long~run revenue
potential from taxing any of these crops is not very
great. Second, given the much larger degree of resource
fixity within the agricultural sector vis-a-vis the non—
agricultural sector, there is some potential for revenue
from the agricultural sector as a whole given existing
demand functions for resources now in agriculture for use
in the non-agricultural sector. Public policies such as
minimum wage laws for unskilled workers in the urban areas
effectively raise the off-farm salvage value of people
currently engaged in the farm sector. Such policies have
the effect of making resources variable that might have
otherwise been fixed in the farm sector. Thus, if there
are substantial irreversibilitiesl3 in the investment-
disinvestment relations and in the output supply relation
in the agricultural sector as a whole, then there is some
potential for a tax on the agricultural sector. This is
the theoretical link between irreversibilities in the
investment-disinvestment relations as defined by Johnson
and problems of government and marketing board revenue and
supply response in developing agriculture. To the extent

that irreversibilities in the investment-disinvestment

118

relations and in the output supply relation are small when
the agricultural sector as a whole is considered, then the
revenue potential is small. If demand schedules for
resources (now in agriculture) for use in the non-
agricultural sector should shift rightwards for given
(upward sloping) supply curves of these resources, then
the salvage values of these resources in the non-
agricultural sector will rise and any government and
marketing board taxation which depresses the MVP's of
these resources below their non-agricultural sector
salvage values will lead to disinvestment in these
resources, the mobility of these resources from the agri-
cultural sector to the non-agricultural sector, and a
likely fall in agricultural output.

From our apriori knowledge of the extent of the
mobility of resources between crOp enterprises in the
cotton-groundnuts-food crops economy of the Northern
States, we suspect that irreversibilities in the investment-
disinvestment relations underlying the production of any
one of these crops (though not necessarily within the farm

sector as a whole) may not be extensive.

Summary and Conclusions
It has been shown in this chapter that for purposes
of government and marketing board revenues from taxes on
crops, the relevant variable is the sales and not the

total output of the crop (arguments are provided to show

119

why we cannot apriori presume that the correlation
coefficient between total output and sales to the board is
unity for any marketing board crop in Nigeria).14 Various
influences on sales of marketing board crops are discussed.
This discussion underlies the estimation effort in the
next chapter. Policy makers and some academicians are
concerned about the effects of changes in the net producer
prices the boards pay farmers on State Government revenues,
with the usual presumption that paying farmers higher
prices will always lead to a fall in State Government
revenues from taxes on these crops. A simple model is
presented to show the dependence of the responsiveness of
(crop) tax revenues on the price responsiveness of sales
of crops to the board, particularly when these crops have
large domestic and neighboring-country markets. Some
illustrative calculations are used with empirical data

on level of taxes and net producer prices for cotton and
groundnuts for the period 1950-66 to show how tax revenues
from these two crops would fall if net producer prices
paid to farmers were further reduced (and thus implied
taxes raised) for reasonably low elasticities of sales of
these crops to the board with respect to the net producer
prices paid by the board. These calculations illustrate
the eventual futility of heavy taxes on these crops with

respect to raising large crop tax revenues.

120

The last part of this chapter showed that questions
of the revenue potential of individual crops and the farm
sector as a whole depend on the underlying investment-
disinvestment process and its responsiveness to various
relative price changes.

The calculations of the government and marketing
board revenue elasticities with respect to both net
producer price and taxes in this chapter clearly point to
the relation of (crop) tax revenue to the price responsive-
ness of sales of cr0ps to the marketing board. In the
next chapter, this price responsiveness of sales of crops

to the marketing boards is estimated.

FOOTNOTES

1" . . . Palm oil is available for export only in
insignificant quantities . . . this could be due to
increased demand for food. " See Twelfth Annual Report of
the Western Nigeria Marketing Board, 1966, p. 7. In its
report a yearlater, the WNMB stated "The abnormal fall in
the quantity of palm produce purchased in the period was
due to the immediate effect of the Nigerian crisis. More-
over, domestic sales yielded higheryprices_per ton than
overseas sales for the farmer or the producerllicensed
buy1ng agent like the Western Niger1a Development Corpo-

rat1on (emphas1s my own). See Thirteenth Annual Re ort
of the Western Nigeria Marketing Board, 1967, pp. 6- g.

A correspondent for the Dail Times writes of
groundnut oil, "What used to cost 1793 a Bottle now costs

5/= . . ." See Daily Times, Lagos, February 3, 1972, p. 7.

 

2Assuming groundnut consumer prices are constant
between two years and making groundnut consumption (C) a
function of population (P) and measured income (Y), then
even if the income elasticity of demand for groundnuts
were zero and the population elasticity of demand for
groundnuts were unity, the percentage increase in domestic
groundnut consumption over the period will equal the
percentage increase in population:

C = C(Y,P)

from which

A—C..= .A—Y—+n AP
C ncy Y cp P

where AC is the percentage increase in groundnut con-
sumption over the period, “c is the income elasticity of
consumption of groundnuts, ncp is the population elasticity
of demand for groundnuts. Setting "cy = 0 and ”cp = 1
makes AC _ AP.

2:"?-

121

122

3A writer in West Africa writes "Available supplies
to the marketing board have in any case been reduced
because of the continuance of smuggling over the border
into Niger, which has been encouraged by the high prices
obtainable for groundnuts in that country. At the same
time in Nigeria, in the bush itself, it is believed that
larger quantities are being_pressed for oil for domestic

consum t1onfi—(empha51s my own). See West Africa, London,
52857, March 17, 1972. ‘—

4For the last groundnut season, "Farmers were
getting about LN 86 a ton on the border with Niger Republic
compared with SN 36 which they get in Nigeria." See Daily
Times, Lagos, February 3, 1972, p. 7.

5The sum of the three components of the tax per
unit of the crop when added to the net producer price paid
to farmers by the marketing boards (Pn) gives the potential
producer price (Pp ) that the farmer would have received
per unit of the crop in the absence of all State government
and marketing board taxation. Thus the derived value of
the unit tax on a crop depends on the net producer price
per unit of the crop that the marketing board decides to
pay the farmers. we can then write:

 

tj = f(Pn,Pp), or more specifically,
tj = Pp - Pn' Differentiating (3) with respect to Pn and
converting to elasticities, then:
ERan = Etan + EMan
P t. t.-P
_-Ep__-Lp-1]_12 .
where Etan - tj — tj — tj < 0, 1f tj < Pp

and ER pn is the State Government and marketing board
revenue elasticity with respect to the net producer price
paid out by the board, Etjpn is the unit tax ela ticity
with respect to net producer price and E (_ a—in Pn
Man Pn M

is the elasticity of sales to the marketing board witg)
respect to the net producer price.

Alternatively, suppose we write:

. = P - P
t] P n

. ..=P- M. P.-PM.
R3 tJMJ ( P Pn) 3 PM] n J

123

 

 

dR dM dP dM
P=PpP+d93 [Mj+P3_lP]
n n
dR. P dM. P M. M.P dM.
P E¥=dep Mg-(P-P)M.-[Pr-IP)M.+Pn Pugh-Pm.
n J n J P 3 P n J P n J
_ -22.
—EM.P t.
J n 3
But from t. = P - Pn, and assuming that P does
not change, 3 p p
..22
Et.P " t~
J n J
=> ER.P = EM.P + Et.P wh1ch 15 the same as 1n
3 n 3 n 3 n
the text.
6

In the 1972-73 Federal Budget, of the expected
gross revenue of LN 650 m., about LN 180 m. will be allo-
cated to State Governments and LN 160 m. will be set aside
for the development fund. The Federal Government is
repaying immediately the LN 5.5 m. loan given by the (then)
Regional Marketing Boards to the former Central government
in 1964. It is hoped that early repayment will help the
marketing boards and state governments to stabilize their
finances and pay higher producer prices. See West Africa,
#2861 (April 14, 1972), pp. 439, 465. In the -
Federal Budget, projected Total Revenue was LN 152 mill,
of which LN 54 mill was to be appropriated to the States.
The corresponding figures for 1969-70 were, respectively,
LN 188 mill, LN 73 mill, and for 1970-71, LN 279 mill,

LN 115 mill respectively. Source for these figures:
Annual Economic ReviewL Nigeria (London: Standard Bank
Ltd., 1968, 1970).

7The unit tax elasticity is not derived from any
choice--theoretic framework of an optimizing economic
agent. It is based more on definition of the tax per unit
of the crop being expressed as the difference between the
potential producer price and the net producer price per
unit of the crop after all taxes have been deducted.
Roughly speaking, this elasticity indicates the "required"

124

change in taxes if net producer prices paid to farmers are
changed--if prices are lowered, by definition, implied
taxes are raised, hence the negative elasticity.

8This is consistent with the expectations of the
Model: From footnote 4,

9Glenn L. Johnson, "The State of Agricultural
Supply Analysis," Journal of Farm Economics (May, 1960).

10Second order effects such as the effect of the
increase in the "important" resource on the marginal
products of resources with which it is complementary need
to be incorporated in a full analysis.

11Fortunately, empirical investigations of the link
between the investment-disinvestment relations in durable
resources and the output supply relation have focused on
resources whose factor shares loom large in the production
process. See Glenn L. Johnson, Overproduction Trap (The
Johns Hopkins University Press, 1972).

12Given a transformation curve between, say,
guinea corn and groundnuts, government and marketing
board taxation of groundnuts may so change relative crop
prices as to produce a corner solution in which the farmer
after reorganization produces only guinea corn. Allocation
of the fixed resources is then based on their on-farm
opportunity costs.

l3Bauer recognizes this point. He states, "the
presence of certain irreversible functional relationships
on the side of both production and consumption may also be
turned to good use in the framing of taxation. Though
very high prices may at times be required to stimulate the
establishment of productive capacity for certain crops,
once the capacity has been established, it may continue to
be operated at appreciably low prices; the establishment

125

of plantations (both estates and small holdings) of
cultivated crops in remote areas and their subsequent
operation are a convenient example." See P. T. Bauer,
Economic Analysis and Policy in Underdeveloped Countries,
Cambridge University Press, London, 1957, p. 95. Bauer
however, does not provide any theory to explain these
irreversibilities.

14Using seed issues and assuming different seed-
rate applications per acre and different yields per acre,
we have estimated total output of seed cotton for all the
Northern States for the period 1960/61-1970/71. For some
reason, the correlation coefficient between sales of seed
cotton to the marketing board and various estimates of
total output of seed cotton in the Northern States over
this period was found to be negative in all cases. For
all estimates of total output the correlation coefficient
‘was approximately -0.26. Probably this has something to
do with our estimates of output.

 

CHAPTER V

ESTIMATION OF IMPACT ON GOVERNMENT AND MARKETING

BOARD TAX REVENUE FROM GROUNDNUTS AND COTTON

OF CHANGES IN THE BOARD'S NET PRODUCER

PRICES FOR THESE CROPS

Introduction

The last chapter dealt with the relationship

laetween (crop) tax revenue responsiveness to changes in

'the board's producer price and the responsiveness of sales
(to the board) to producer price changes. Estimation of

tflie sales price responsiveness of marketing board crops
is; essential for a quantitative assessment of the effect

(crop) tax revenue of paying farmers higher producer

cum
This chapter attempts to estimate

Prices for these crops.
t1'lis sales price responsiveness and to assess the quanti-

tative impact on (crop) tax revenue of changes in the

board's producer prices.
This chapter is divided into four main sections.

.1115? first part sketches a simultaneous equation system in

vvrlidzh.we incorporate our apriori knowledge of the factors

influencing the sale of a crop to the marketing board. The

126

127

essence of this is the need to identify and specify a
structure of which the sales supply equation is only a
component. The second part of the chapter contains a
modified model in which all the regressors of the sales
supply equation are predetermined. Single sales supply
equations are then estimated for both cotton and groundnuts
by ordinary least squares. The third section contains
equations to predict sales when farmers are paid higher
prices. These predictions, together with the new higher
producer prices and the implied lower taxes on farmers, are
then used to assess the quantitative impact of higher
‘producer prices on tax revenue from these two crops.
The last section of the chapter contains a summary
c>f our findings in this chapter.
The simultaneous equation nature of the problem
“was recognized in Chapter IV. It is assumed that the
-féirmers, producers, and middlemen of any marketing board
(3113p allocate total output of the crop between domestic
nCurl-industrial consumption requirements (farm and non-
farm) and sales to the board. It is also assumed that
prOducers for any marketing board in Nigeria face an
infinitely elastic demand curve at the price set by the

board

 

128

The Model

 

The structural equations of the model are postu-

lated to be:

D.
J

u
o
75

I): domestic consumption demand;

D.’
3 J
= Mj(Yj, PN , PD ): sales to the marketing board
j j
= Y.(PN , PN , K)- supply relation;
3 r
= Yj - 0.: marketing board crop supply identity

is domestic consumption demand for the marketing
board crop by both the farming and non-farming

population, assumed endogenous:

is the domestic producer price of the marketing

board crop, assumed endogenous;

is real income of all domestic consumers of this

crop, assumed exogenous;

is the quantity of the crop sold to the marketing

board in any given year, assumed endogenous;

is the total output of the crop in a given year,

assumed endogenous;

is the net producer price of the crop exogenously
fixed by the board which determines the infinitely
elastic demand curve facing the farmers, producers,

and middlemen of this marketing board crop;

129

P is the net producer price of a closely competitive
crop (i.e., competitive in production). We assume
that the price of this crop is exogenously determined

outside this system;

K is the set of exogenous factors affecting production
(weather, new production functions from application
of fertilizer, spraying, and seed dressing, etc.);
in this study, this set is specifically represented

by time.

The crop supply identity says that total output is
allocated between non-industrial domestic consumption
(farm and non-farm) and sales to the marketing board.1

{The endogenous variables of the system then are Dj' PD ,
j

Y’ Mj while the exogenous variables are PNj' I, K, and

j'
PMQr‘ If we give the above structural equations their

llidiear approximations, we have

D - ‘ " - =

3 1: OLD (1ka 0:2 It alt (la)
”4- -Bo-B P - 81> -B Y. = 5 (1b)

3t 1 th 2 Djt 3 3t 2t

jt F w° ' m1Pth: ' wZPNrt ' (”3Kt = E:31: (1C)
gjt; ‘ Y. i-D = o ,t:= 1,...,11 (1d)

Vv - . . .
hlch can be written 1n matrix form as

BYt + I‘xt = st (1)

g: 1.5: nicer-was“... ~

130

 

 

 

 

 

 

 

 

 

 

where:
r 7 1
1 —a1 0 0 ~ FDjt
B = 0 -82 1 -B3 ; Yt = PDjt
O 0 0 1 th
l 0 1 -1
L J y.
-JtJ
1‘ = E7 ._ ‘1' = 3 =
no a2 0 0 0 r Xt [1 ' 8t Feltj
-80 0 -81 0 0 It EZt
-wo 0 -w1 -w2 -w3 Pde €3t
Lo 0 o o o 3 P L0 A
r
LKt 1

where we assume that the disturbances of the structural
equations satisfy the assumptions of the classical normal

linear regression model which in matrix notation are:

at ~ N(O,¢) E(ete;) = 0

Wherfi ¢ is the variance-covariance matrix of the structural
disturbances . 2

By the order conditions for identifiability3 we
have time following results: the equation for the domestic
COnSUHEJtion demand of the marketing board crop (1a) With
two endoqenous variables and two excluded predetermined
vari‘E‘bles is over-identified; the equation for the sale

C)
f a Ornap to the marketing board (lb) with three endogenous

¥

131

variables and two excluded predetermined variables is
exactlygidentified, while the equation depicting the supply
relation (10) with one endogenous variable and one excluded
predetermined variable is under-identified.

The equation of most interest to us is that
depitxing sales of the crop to the marketing board (lb),
the: sales supply equation. For the rank and order con-

ditjxans for identifiability of this equation, form the

system matrix:

 

‘1 -al 0 0 -ao —dz 0 O 0

(Bpl‘) = 0 -82 1 -B3 -80 0 -81 0 0
0 O 0 1 -wo 0 -w1 —w2 -w3
L1 0 1 -1 o o o o o 3

 

The :submatrix obtained by striking out the second row and
all (columns of (B,F) corresponding to variables that are

ificllnded in the sales supply equation is:

 

1
['1 -a2 0 o

A = z - _
,0 0 wz m3
51 0 o 03

By the .rank criterion, the sales supply equation is exactly
ldentified or over-identified if and only if rank (A) =
(3‘1 where G is the number of structural equations. A has

a. . .
t leasrt one 3X3 non31ngular square submatrix (A*)-

132

 

{'1 -a0 41%

_ I _ . _ _ .

A* — 50 w 0 , IA*I — azwo,
h. 0 0]

so that rank (A) = 3 = G-l which,4 together with the
order condition above, provides the necessary and sufficient
condition for identification of the sales supply equation.5

The reduced form equations can then be written in

 

 

 

 

 

 

 

unrestricted form as: Yt = th + Vt (2)
Ir 1 P ' ‘ '
1 Djt H11 H12 H13 H11 1 v1t
Y = , = = a
t PD3t ' Ht H21 H22 H23 H24 ' xt It ' Vt Vzt
M. n n n n v
3t 31 32 33 34 Pth ‘ 3t
y. n n n n iY
(_gtj _41 42 43 43 _kt J 4g
FIOHI (1) Y can be solved- Y = -B~1FX + 8.18 (3)
' t ° t t t

and from (2) above, (3) WOUld imply

_ _ -1
Ht_BI‘

_-1
Vt‘B at

from Which the variance-covariance matrix of the reduced

form d i sturbances is

l

w == E(VtV£) = E[(B’ at) (B_let)’]

9:-» ‘. '5 ““‘WJV

133

where w, being a linear conbination of the structural
disturbances, is assumed to obey normal classical linear

I . 6
regre331on assumptions.

 

 

 

 

 

 

 

,
B2 ”“1 0‘1"‘3‘183 0‘11
l
= -1 -1 1-8 1
al+32 3
-82 a1 82+a183 82
L0 0 82+a1 0‘
From
_ _ -1 -1
Yt _ B F Xt + B at,
F ‘1 "'_ _ __ __ . - .—
Dj a082+a18b w0a1(1 83) a282 a181 alw1(l 83)
p . 1 a +8 -w (1-8 ) a 8 -w (1—8 )
03 = - o o o 3 2 1 1 3
al+gz
Mj o‘082"°‘1Bo"*’o(82"0‘1‘33) “282 ’0181‘w1(82+3183)
ij L-wo(82+a1) o -w1(82+a1)
-w2a1(1-B3) -w3al(l-B3)q F1
I
—w2(1-83) -w3(l-B3) t

Pth
-w2(82+a183) -w3(82+a183)

 

 

 

- (B +a ) -w (B +a ) iLKt J
“2 2 1 3 2 1 .1

134

 

 

- _ _ . P 1
B2 “1 “1 “153 “1 8it

+ 1
a1+32 -l -1 1-83 1 €2t
'82 “1 B2 +“183 82 83t
L0 0 82 +011 OJ b0 J

 

 

Thus the endogenous variables are expressed in terms of all
the predetermined variables and the structural disturbances.

We would then estimate the sales supply equation by
two stage least squares.

Two stage least squares estimation of the sales
supply equation would require prior (first stage) esti-
mation of domestic price and total output of the marketing
board crop, the other endogenous variables in the sales
supply equation. As we have no figures on total output of
groundnuts and cotton over the last two decades,7 we cannot
estimate the sales supply relation in this model; as data
becomes available in the future, this equation could be
estimated within the context of this model. For ordinary
least squares estimation, we would need to operationalize
the model so that the OLS estimates will be unbiased and

consistent.

135

A Modified Model

We shall assume that the only price affecting the
quantities of any crop sold to the board are the board's
net producer prices.8 We would expect that the higher
this price, the less will be consumed on the domestic
market and the more that would be sold to the board. It
is also assumed that Nigerian farmers are unaware of the
true world prices for their products and that the price
affecting their production and sales decisions is the

board's net producer price.

Short Run Modified Model

The short run here is a given production-marketing
period, strictly speaking, a given marketing period, when
the total quantity of output from production of cotton or
groundnuts is fixed. The aggregate sales supply curve of
a crop to the board in this case of a partial monopsony on
the buying side is the difference between this available
total stock of the crop and the lateral sum of all indi-
vidual (domestic: farmers and non—farmers) demand curves
for the crop as shown on the following page, using ground-
nuts for illustration: Sijyj is the given stock of ground-
nuts within this period (zero price elasticity of supply or
production); CjCj is the lateral sum of all domestic
demand curves for groundnuts; Smj is the lateral difference
between total supply and total consumption; P . is the net

NJ
producer price for groundnuts offered by the board, which

 

 

 

 

 

 

 

I l

l I I

| I | 5?
l ' l

l l |

l i ' ' s

11
0 Me M, 2: I Co (:50ij

in equilibrium is assumed equal to the appropriately
adjusted home market price. With the demand curve CjCj
and a marketing board price of PNj' OCo is consumed and

OMo is sold to the marketing board. With a higher marketing

;

Nj'

marketing board.9 From the diagram,

board price P CO1 is consumed and OM1 is sold to the

10 the elasticity of
sales of groundnuts to the marketing board with respect to
the net producer price paid by the board is positive so
long as the elasticity of total output (supply) of the
crop with respect to this net producer price is non-

negative and the (money income constant) price elasticity

of demand for groundnuts is negative. Under normal

137

circumstances we expect these conditions to hold for most
marketing board crOps. We assume that the real income (I)
of the bulk of domestic consumers of groundnuts is not
influenced by the sales of groundnuts to the marketing

11 The producers themselves form a small proportion

board.
of domestic consumers of groundnuts. An increase in real
income leads to a rightward and upward shift in the
aggregate demand curve and a corresponding upward and
leftward shift in the sales supply curve if groundnuts are
a normal good while the converse result holds if groundnuts
are inferior goods. In the latter case, an increase in
real income of consumers leads to a downward and leftward

shift in the aggregate demand curve to CECE and a corre-

sponding rightward and downward shift in the sales supply

curve to SMjSMj' Thus if the sales supply function is
M). = Mijj, I, yj) , (4)

the plausible signs of the partials of the function are:

3M. > <

I < 0 as nch > 0

3M

8M.
5'19.) 0' Y

Nj

> 0,
j
where Yj is the given stock of the marketing board crop

and n is the income elasticity of demand for this crop.

ch
All the explanatory variables in the above relation are
assumed predetermined so that when given a specific

regression expression, the disturbances and the explanatory

138

variables are presummed to be independently distributed;
hence we can expect the OLS estimates to be unbiased and

consistent.

The Long Run Modified Model

If the period is long enough for production to
respond to price and other variables, then changes in sales
of groundnuts to the marketing board would then partly
reflect substitutions in production as producers move along
given or changing transformation surfaces in response to
relative price changes. For these long run relations,
theory leads us to expect some irreversibilities in the
underlying investment-disinvestment relationships. Then
the output response of Nigerian farmers depend not only on
the magnitude of the crop price change but also on the sign

of the price change.12

The theory of the investment-
disinvestment response to crop price changes (see Chapters
III and IV) leads us to expect a difference between
expansion supply price elasticities and contraction supply
price elasticities corresponding to the expansion and
contraction phases of crop prices respectively. These
elasticities are also influenced by the most recent history
of relative prices and farmers' expectations about their
movement in the immediate future. In the diagram on the
following page, if expected food prices rise relative to

the price of a marketing board crop, then with food crops

competing with the marketing board crop in production the

139

 

 

 

 

 

 

25A
c.
J
K5////////1
5»; l
SW, '
533‘ C
l 5 l J
1 fl. I >
M." 9'
supply curve of the latter sh1fts from Sonsto to
Slestl and the sales supply curve, for g1ven demand
curves, correspondingly shifts upwards and to the left,
to S’.S’ In this case, an increase in food prices leads

M3 M)”

to a fall in total sales to the marketing board from Mjo

to Mjl' From the above discussion, we would have a supply

relation

S . = S .(P

Y3 Y3 K) (5)

Nj' Nr'

and

2gljao,;TS:,-Y-j-<o and;IS<—Yi>o
Nj Nr

140

where PNr is the price of a competing crop and K stands
for those exogenous factors like technical change which

influence total Supply of the marketing board crop.

Empirical Results

The sales supply equations for groundnuts and
cotton were all estimated by ordinary least squares. It
became obvious at an early stage that there was auto-
correlation in our series. Whether this is a serious
drawback depends on the purposes to which we intend to put
our estimates. For prediction purposes Malinvaud maintains
that though the estimates of the coefficients may be
imprecise, predictions based on them "are nevertheless

13

fairly good." I decided to incorporate some knowledge

of the autoregressive structure in my analysis for the

purposes of prediction.14

I therefore ran first stage
ordinary least squares for all equations. For the ground-
nut equations, I assumed a first order autoregressive
scheme and estimated the autocorrelation coefficient from
the residuals yielded by my first stage estimates. These
autocorrelation coefficient estimates were then used to
transform the original variables; in the second stage,
ordinary least squares was applied to the transformed
variables. This transformation exercise produced

inconclusive results; it was therefore not done for the

cotton equations.

141

Most of the equations were of the linear and log
forms. The original equation is designated by its number,
like (6); the same equation reestimated with transformed
variables to take account of the autoregressive structure
is designated (6)'. The "short run" sales supply function

for groundnuts was postulated in two forms as:

Y1t = 8° + B1X2t + B2X3t + 6t (6)
log Y1t = 80 + Bllogx2t + leogx3t + 6t (7)
where:

Ylt is the quantity (tons) of sales of groundnuts
to the Northern States Marketing Board in year t; th is
the net producer price of groundnuts paid by the board in
year t (LN/ton) and X3t, standing for the real income of
consumers of groundnuts, is GDP at factor cost (LN), and
where the disturbance terms (at) are assumed to obey
normal classical linear regression assumptions. The
estimated equations and their transformed counterparts for
the Northern States (1950-51 through 1967-68), with

standard errors of estimates in brackets,are:

?lt = -440875.6 + 12729.7x2t + 588.6X3t, R2 = .83
(224368.5) (6267.2) (69.3) d = 2.25
6 = -0.1448 (6)
§lt = -354514.3 + 9224.9xzt + 577.2x3t, R2 = .86
(524678.5) (11945.3) (79.1) d = 2.17 (6)'

142

2

log vlt = .6013 + 1.3139 iogx2t + 1.0492 109X3t, R = .86
(.6023) (.3379) (.1255) d = 2.03
8 = -o.0230 (7)
log ilt = 2.0636 + .5901 1ogx2t + .9305 logx3t, R2 = .74
(1.4566) (.7323) (.1625) d = .2.22 (7)'

In equation (6) the sign of the groundnut price
coefficient is consistent with theory; it is significantly
positive at the 5 per cent level. The elasticity of sales
of groundnuts to the marketing board with respect to the
board's net producer price is .714, calculated at the
sample means. The real income coefficient is positive
and significant at the .5 per cent level. The sign could
be due to two reasons: an implication that groundnuts in

15 so that

the form sold to the board are an inferior good
with rising incomes less is consumed in the domestic
market and more is sold to the board, or it could be due
to the effect of the measure of income we used on the
ability of farmers, producers, and middlemen to produce
and sell more groundnuts to the board as incomes rise,
especially in a world of imperfect capital markets. The
fit of the model is quite good (R2 = .83); the calculated
partial correlation coefficient between groundnuts sales
and groundnut prices is .464 while the partial correlation
coefficient between groundnut sales and income is .909.
The null hypothesis of zero autocorrelation was not

rejected at the 5 per cent level (the Dubin-Watson

Statistic, d, = 2.25). Equation (6)' is the transformed

143

counterpart of (6): the autocorrelation coefficient is
calculated as -0.l4475 from the residuals of (6); this is
then used to transform the variables in (6) after which
ordinary least squares (OLS) is used to estimate the

16 Though (6)' improves the overall

transformed equation.
fit of the model (R2 = .86) and reduces the autocorrelation,
the statistical significance of the groundnut price
coefficient is drastically reduced.

In equation (7), the groundnut price coefficient
has the expected sign and it is statistically significant
at the .5 per cent level. A two tailed t-test finds the
coefficient statistically significant at the l per cent

level. From (7), the elasticity of sales of groundnuts to

the marketing board with respect to the board's net

producer price is 1.3139, meaning that if the board raises
groundnut producer prices by l per cent, groundnut sales to
the board will increase by 1.3139 per cent. The income
coefficient is positive and is statistically significant at
the .5 per cent level. The fit of the model is quite good:
R2 = .86; the partial correlation coefficient between
groundnut sales and its net producer price is .708, the

partial correlation coefficient between groundnut sales and

income is .907 and the Dubin-Watson statistic is 2.03.
"R2 deletes" for equation (7) show that 72 per cent of
the variation in sales of groundnuts is explained by the
model if we regress sales only on income and 21 per cent

is explained if we regress sales only on net producer

144

price. The transformation of (7) with the estimated
autocorrelation coefficient (8 = -0.0230) did not improve
my results: the price coefficient fell drastically in
statistical significance (significantly positive only at
the 25 per cent level); the income coefficient, though
statistically significant at the .5 per cent level, fell;
R2 fell to .74 and the Dubin—Watson statistic rose to 2.22.

From (4) above, total stocks of groundnuts in a
given marketing period should be one of our regressors.
We have no data or reliable estimates of total output of
groundnuts. It can be presummed that the marketing board
in deciding on the net producer price it pays to farmers
in any given year is to some extent influenced by the
expected output of the crop for that year. Thus, antici-
pation of a small crop would induce the board to pay
producers higher net producer prices in that year and
conversely for anticipations of large crOps. It is
plausible then to expect some negative correlation between
the output series and the net producer price series. To
this extent, the ommission of output from our short run
equations may underestimate the price elasticity of sales
of groundnuts to the board. We would expect the elasticity
of sales to the board with respect to stock output to be
positive.

To capture some long run influences on groundnut

sales to the board, the prices of competitive crops were

145.

included in some regression equations. X5 is the guinea

corn price relative to the groundnut price, and X6 is the
millet price relative to the groundnut price. If millets
and guinea corn are each competitive with groundnuts

in production, we would expect the elasticity of ground-
nut sales to the board with respect to X5 and X to be

6
negative.

The Institute of Agricultural Research, Samaru,
Zaria has been breeding and recommending new varieties and
practices for groundnut farmers over the last twenty years.
If groundnut farmers have been applying fertilizers, seed
dressing and spraying, new varieties and new cultural
practices, then there should have been a secular trend in
yields per acre which in turn should have given an upward
trend to groundnut sales to the marketing board. This
would render the coefficient of the trend variable, X4,
positive. On the other hand, transportation facilities
have steadily improved over the years, thus facilitating
the shipment of groundnuts from surplus to deficit areas.
This might render the trend coefficient negative in sign.

The estimated equations, with standard errors of

coefficients in brackets, are:

91t = 783789.1 + 12813.3x2t + 1003.8x3t - 27866X4t, R2 = .85
(997276.3) (6148.7) (336.7) (22132.9) d = 2.26
5 = -.135 (8)
91t a 1004923.2 + 7629.4x2t + 988.3X3t - 28323.9X4t, R2 = .80
(1306147.8) (11865.8) (355.8) (22569.6) d = 2.19 (8)'

146

163 Ylt = 3.096 + .899 iogx3t + .378 logxst - .101x6t_1 (9)
(.433) (.140) (.159) (.160)
R2 = .81
d = 2.72
s = —o.305
log Ylt = 2.892 + .956 iogx3t + .332 iogx6t - .082 iogx6t, (9)'
(.310) (.097) (.127) .125)
R2 = .91
d = 2.44

The trend coefficient in (8) and (8)' is significantly
negative at the 25 per cent level. The negative sign could
be due to several reasons: improved transportation and
market information networks that facilitate diversion of
the crop from producing areas to other domestic markets; a
probable upward trend in the demand for indigenously
crushed groundnut oil, or exogenous increases in popu-
lation.17
Relative to equation (7), the income coefficient
has increased substantially while the price coefficient
remains remarkably stable; both are statistically signifi-
cant at the 5 per cent level. The overall fit of the
model is good: R2 = .86; the calculated partial corre-
lation coefficients between groundnut sales and the
leX2 leX3 = .623 and
deletes" indicate that when

individual variables are:
2

= .486;
pY1X4 = -.318. The R
groundnut sales are regressed only on income and time,

80 per cent of the variation in groundnut sales is

147

explained; when sales are regressed only on groundnut
prices and time, 75 per cent of the variation in groundnut
sales is explained and when sales are regressed only on
groundnut prices and income, 83 per cent of the variation
in groundnut sales is explained. Transformation of (8)
into (8)' using the calculated autocorrelation coefficient
from the residuals of (8) did not improve the results:
the statistical significance of the price coefficient is
drastically reduced and R2 falls. The income coefficient
is fairly stable, falling from 1003.8 to 988.3 but still
statistically significant at the 5 per cent level.
Regressing groundnut sales on last year's groundnut
price yielded poor results. Regressing sales on last
year's groundnut price and this year's income improved the

overall fit of the model:

A
log ylt = 7.8426 + 1.5342 log x2t-1; R2 = .37
(1.7226) (.4994) d = 1.20 (10)
A
log Ylt = 3.9516 + 1.1084 logX3t + .4500 log X2t-13 R2 = .76
(1.3556) (.2249) .3873) d = 2.43 (11)

The coefficient of the log of groundnuts price is
significantly positive at the 0.5 per cent level in (10);
this is reduced to 25 per cent in (11). The coefficient
of the log of income is significantly positive at the
0.5 per cent level. The reduction in the elasticity of
groundnut sales with respect to last year's price when

income is included as one of the regressors suggests

148

positive correlation between income and lagged groundnut
price.18
To check for irreversibilities in the sales supply

curve, we postulated a model of the form

Ylt = 8° + 82X2t—1 + B3X3t + B4(X2t—1 ° th) + 8t
where:
zlt = 1 1f x2t-1 i x2t-2

= 0 otherwise; the other variables are defined as
before. If irreversibilities in the underlying investment-
disinvestment process are significant enough to transmit
their influence to the groundnut sales supply curve, then

84 is not zero. Alternatively, the model could be stated

as
= I
Y1t Bo + (82 + B4)X2t-l + 83X3t + s (when the board s groundnut
net producer prices are not
rising);
= u
ylt 80 + BZXZt-l + B3X3t + 8 (when the board 3 groundnut net

producer prices are rising).

A test of whether the elasticity of groundnut sales to the
board with respect to the board's net producer price
differs according to whether prices are rising or falling

19

is a test of the hypothesis that 84 = O. The estimated

equation is:

149

91t = -133373.9 + 3295.4x2t_1 + 597.7x3t - 1565.1(X2t_1 - 2t)
(186118.1) (6232.2) (101.7) (1940.6)
R2 = .76
d = 2.04 (12)

In (12), the lagged price coefficient is of the right sign
but is not significant at the 25 per cent level; the

income coefficient is significantly positive at the .5

per cent level; the coefficient of the interaction effect
between the dummy variable and the lagged groundnut price
is significantly negative at the 25 per cent level,
significantly non-zero at the 50 per cent level for a two-
tailed test. The fit of the model is good: R2 = .76 and
the Dubin Watson statistic is around 2. Thus, the influence
of these irreversibilities is not particularly strong on
the sales supply curve. They are likely to be important

if we focused solely on supply (output) considerations;

our model admits these long run influences through the

side door, as it were. A model focusing solely on output
(supply) response to producer price might produce differ-
ent results. A non-significant difference between expansion
sales supply price elasticities and contraction sales
supply price elasticities does not necessarily imply that
there are no irreversibilities in the underlying investment
and disinvestment response to groundnuts price changes.

The most recent history of the board's groundnut producer

20

prices, the factors influencing the way Northern States

150

farmers form and revise their expectations of the board's
prices are all factors that influence the investment-
disinvestment decisions of farmers and thus influence
groundnut sales to the board.

For sales of cotton to the marketing board, the

following equation was estimated by ordinary least

squares.
log Y2t = 80 + 81 log X2t + 82109 X3t + Et
where Y2t is the sales of cotton (tons) to the Northern

States Marketing Board in year t, X is the net producer

2t

price of cotton (LN/ton) based on the price series that I

have constructed (see p. 109), X is the income variable

3t
(LN) meant to capture the same effects as in the case of
groundnuts: in this case, it is the effect of income on
the demand for indigenously woven material and probably
the effect of the producer income effect on sales of

cotton to the board. The estimated equation for the

period 1950/51-1967/68 is:

A
log Y2t = .6768 + .9028 log X + .9212 log X : 2

2t 3t

R .73
(.4204) (.1454) d

2.55 (13)

The price coefficient is of the right sign; it is signifi-
cantly positive at the 2.5 per cent level. Since (13)
depicts mainly short run influences, the "short run"
elasticity of sales of cotton to the marketing board with

respect to the board's net producer price is .9028,

151

meaning that an increase in the board's net producer

price of l per cent would lead to an increase in sales of
cotton to the board of almost 1 per cent, all other vari-
ables remaining constant. The positive sign of the

income coefficient may reflect the fact that as incomes
have risen, the demand for most indigenous weaving (except
for specialty fabrics like Okene cloth) has declined and
thus has led to an increase in cotton sales to the board
or it may capture the effect of generally rising incomes
on the ability of cotton farmers, producers, and middlemen
to produce and sell cotton to the board, especially in a
world of imperfect capital markets. The fit of the model
is good: R2 = .73, the calculated partial correlation
coefficient between cotton sales and its producer price is
.48 while the partial correlation coefficient between
cotton sales and income is .85. Based on our results from
the groundnut equations, we found no compelling reason to
reestimate (13) with the calculated autocorrelation
coefficient from the residuals of (13). Our sales price
elasticity of cotton exceeds Oni's price elasticity of
supply of cotton (output) as would be expected since
output supply as such is more likely to be affected by
certain irreversibilities in the underlying investment-
disinvestment process.21

To capture some long run influences on cotton

sales to the board, we regressed cotton sales on lagged

152

producer price of cotton (XZt-l) and income. We obtained
poor results. The estimated equation (1951-67), with

standard errors of estimates in brackets, was:

2
— 2.4398 + .1714 log X2t-l + .7492 log x3t, R

(.4859) (.1795) d

log Y .56

2t 2.54 (14)

The lagged price coefficient is of the right sign but is
insignificant at the 25 per cent level; the income
coefficient is significantly positive at the 0.5 per cent
level. The null hypothesis of zero autocorrelation was
rejected at the 5 per cent level.

To capture the substitution effects in production,

we introduced the lagged price of groundnuts relative to

 

 

the lagged price of cotton (§%E3%), the lagged price of
millets relative to the lagged cotton price (:::::) and
the lagged price of guinea corn relative to the lagged
cotton price (:::::). To the extent that a crop is

competitive with cotton in production, we expect the
elasticity of sales of cotton to the board with respect

to this crop's (lagged) price relative to cotton's (lagged)
price to be negative. The estimated equations for the
period 1951-67, with standard errors of estimates in

brackets, were:

 

x
log‘Y2t = 1.04828 + .2088 log XZt-l - 2.0905 log ( 41:“1) +
(.4401) (1.0324) 2t-l
1.0687 log x3t; R2 = .67
(.2264) d = 2.97 (15)

153

 

 

X
1og"):2t = 2.2157 + .0992 log x2t__1 - .4059 log (XSt J‘) + .8196 log x3t;
(1.1764) (.2359) 211-1 (.3870)
R2 = .51
d =3 3.16 (16)
A X6t-1
log Y2t = 2.4222 + .0975 109 X2t-1 - .3109 109 (X ) + .7589 log X312;
R2 = .48
d = 3.18 (17)

In (15) the lagged own price coefficient (XZt-l)
is of the right sign but is insignificant at the 25 per
cent level; the price of groundnuts relative to cotton is
of the right sign and is significant at the 5 per cent
level while the income coefficient is significantly

22 The overall fit of

positive at the 0.5 per cent level.
the model is fair (R2 = .67).

In (16) the own lagged price coefficient is of the
right sign but is insignificant at the 25 per cent level;
the lagged millet price relative to lagged cotton price is
of the right sign and is significantly negative at the
10 per cent level while the income coefficient is signifi-
cant at the l per cent level. In (17), the own lagged
price is of the right sign but is insignificant at the
25 per cent level; the lagged guinea corn price relative
to lagged cotton price is of the right sign and is signifi-
cant at the 10 per cent level; the income coefficient is

significant at the 10 per cent level. Both (16) and (17)

are fair in overall fit; in both, the null hypothesis of

154

zero autocorrelation was rejected at the 5 per cent level.
There appears to have been competition in production
between cotton, groundnuts, millet, and guinea corn.

Our attempt to include a trend variable (X7) in
equation (14) changed the sign of lagged cotton price.

The estimated equation was:23

log Y = .3772 - .1602 log X

2t + 2.8052 log X - .0603 X

(.4046) Zt’l (.7058) 3t (.0202)7t

d = 2.80 (18)

The lagged cotton price has the wrong sign; the
trend variable is negative and is significant at l per
cent. The negative sign may reflect structural changes in
the transportation and market information networks over
time, thus facilitating the shipment of cotton from surplus
to deficit areas particularly for the specialty woven
fabrics.

To test for the effects of certain irreversibilities
of the underlying investment-disinvestment process on sales

of cotton to the board, we postulated a model of the forms:

log Y2t = Bo + 81 log X + 82 log X3t + 83 (Z t log X ) + 8

2t-1 1 2t-1 t

0<
II

80 + 81 log x + B x3t + 83 (x - s ) + 8

2t 2t-1 2 2t-1 1t t

155
where:

B = 1 if X

1t X

2t-1 i 2t-2

0 otherwise; the other variables are defined as
before. If irreversibilities in the underlying investment-
disinvestment process are significant enough to transmit
their influence to the cotton sales supply curve, then 83
is not zero.24 Alternatively, the model could be stated

as

log Y

2t = Bo + (81 + 83) log X

+ 82 log X + at (when the

2t-1 3t

board's cotton net producer prices are not rising);

log Y = 80 + 81 log X + 82 log X + 6 (when the board's cotton

2t 2t-1 3t t

net producer prices are rising);
and

+E3)

1 3 x2t-1 + B

th = so + (B (when the board's cotton net

2X3t

producer prices are not rising);

Y2 + B + B X (when the board's cotton net producer

t = 8o 1 X2t-1 2 3t

prices are rising).

A test of whether the elasticity of cotton sales
to the board with respect to the board's net producer
differs according to whether prices are rising or not is

a test of the hypothesis that 83 = 0. The estimated

156

equations, with standard errors of estimates in brackets,

were:
106 Y2t = 2.5885 + .0855 log x2t_1 + .7544 log x3t - .0167
(.5274) (.1847) (.0328)
(Z logX ) 82 = 57
1t 2t'1 d = 2.50 (19)
? = 7262.4 + 445.7 x + 72.1 x - 175.1 (x - z )
2t (1131.4?t l (19.1)3t (247.1) 2t 1 1t
R2 = .54
d = 2.59 (20)

In (19), the coefficient of the lagged cotton price is of
the right sign but is insignificant at the 25 per cent
level; the income coefficient is significantly positive at
the 0.5 per cent level, while the coefficient of the inter—
action term between the dummy variable and the lagged
cotton price is not significant at the 25 per cent level.
In (20) the coefficient of the lagged cotton price is of
the right sign but is insignificant at the 25 per cent
level; the income coefficient is significantly positive

at the 0.5 per cent level while the coefficient of the
interaction term between the dummy variable and the lagged
cotton price is significantly negative at the 25 per cent
level. The fit of both models is fair; the null hypothesis
of zero autocorrelation was rejected at the 5 per cent
level. Though our results do not indicate any particularly
strong influence of irreversibilities on the sales supply

curve, two things must be noted: first, our test is weak

157

insofar as it fails to take into account the influence of
magnitudes of price changes on the investment-disinvestment
decisions of farmers and second, we are not directly
focusing on output (supply) and all its relevant long run
considerations.

The following conclusions can be drawn. In all
the equations for both groundnuts and cotton, sales to the
marketing board are very price responsive on the average;
both the price coefficient and the income coefficient are
stable with respect to exclusion and inclusion of several
other variables. We tried to cope with the autoregression
problem in the following way. We estimated the groundnut
equations in the original variables and calculated the
autocorrelation coefficient from the residuals of our
first stage OLS estimates. The calculated autocorrelation
coefficient was then used to transform the original vari-
ables after which OLS was used to estimate the transformed
equation. The results of this exercise are mixed: in
some cases, the overall fit of the model improved and
autocorrelation was reduced; in others, things got worse.25
That is why no transformation was performed for the cotton
equations. Our efforts to detect irreversibilities in the
underling investment-disinvestment relations produced
rather inconclusive results. This was probably partly
due to our rudimentary knowledge of the factors influencing

farmers individually and in the aggregate with regards

158

to their formation and revision of expectations about the
board's net producer prices and partly because we were not
mainly concerned with the long run factors.

The multicolinearity problem was not as serious as
might be expected; for example, the simple correlation
was-0.165.

coefficient between X and X

6t 6t-1

Prediction

Our efforts at correcting for the autoregressive
structure yielded mixed results. We, therefore, predict
sales of groundnuts to the board using both the original
estimated regression equations as well as some reestimated
regression equations that do incorporate knowledge of the
autoregressive structure and predict sales of cotton using
the untransformed estimated equations.

For the 1948-67 period, groundnut net producer
prices paid by the board had a mean percentage of world

26 Three differ-

groundnut market prices of 55.01 per cent.
ent series of annual groundnut net producer prices were
then constructed on the assumption that the board paid
groundnut farmers the following percentages of world
market groundnut prices for each of the years 1950-1967;
55 per cent, 60 per cent, and 67.8 per cent. We stay on
one side of the actual mean percentage that obtained
during the period 1948-67 (= 55 per cent) because we want

to investigate the sales and government and marketing

board revenue consequences of paying farmers higher

159

producer prices. It is assumed that the other variables
in the estimated equations used for predicing sales stay
at their initial values: there is only a new value for
groundnut price for each of the years 1950-67. The
results are shown in Tables 5.1, 5.2, 5.3, and 5.4, with
the standard errors of predictions in brackets.

It is evident from these tables that in terms of
receiving a percentage of world prices, Northern States
groundnut farmers did much better in the 1960's than in
the 1950's: in the 1950's, they received less than
55 per cent of world prices while in the 1960's they
received more than 55 per cent of world prices. Thus,
predicted sales, assuming farmers received 55 per cent of
world prices, tend to be higher than actual sales to the
board in the 1950's and lower than actual sales to the
board in the 1960's. This same trend is exhibited even if
we suppose that farmers were paid 60 per cent of world
prices as shown in Tables 5.3 and 5.4.

To avoid cases in which our new hypothetical
prices are lower than the actual prices farmers received,

we suppose the board paid farmers a price:

Pth = max(Pth, .6 Pwt)
where Pth is the board's actual price that was paid to
farmers in year t and Pwt is the world price of groundnuts
27

in year t. The constructed price series and the resulting

predicted sales are shown in Table 5.5.

160

 

 

 

 

.HHH .m mom .moamm was museum uwooooum assess mom "condom
momeom AaommHHVmOhbbo Hoemhm m.mm o.om noma
mmammm Ahmemaavmammom snowmoa m.mm o.mm mama
Homhmb Amvmmoavmmamam ommwnm m.nm ~.vm moma
censor Amwamoavommohb hmhmhw m.mm H.vm voma
ommmmo Aboamoavmmmmmm hmhmmb m.vm >.~m moma
memoam Avoemoavmmammo vumabm m.mm m.om mood
maammm Ammmmoavmmbvmm mvmmmm H.vm m.om Hood
oemamm Amvamoavaeamvm Hmomao m.mm >.mm coma
Hoavom Ameomoavmohmmm Heemvv o.Hm v.bm mmma
mommmm Ammmomavammamo ommmmm m.me v.0m mmma
ovomvm Aaoomaavmmaaao mmovah m.o¢ m.mm omma
mwwmmv Amhmooavvvoamv mmmhmm o.mm ¢.bm mmma
emomam Ammohaavo>hmmm mHNOmm N.~v ¢.mm mmma
mmommv Ammmoaavhamaam mhhmmhm H.mm v.mm vmma
bbmmav Annemaavmaomme mvmvmv m.~v m.mm mmma
mbmame AmmmeHVvamom movemv h.mv o.mm mmma
vmmmhm Aoamvmavmmmnvv mmmmmv m.me 0.0m Hmma
mommae Aavmmmavmemaam MthvH m.am N.H~ omma
mace coa\zm
Amusuosuuw Amusuosuum oueom moaum ousom use»
e>nmnonmouous< m>wmmmumouous< on» tauos no any hp
mo moccasocu mo emoeazocx on meaem ucmo mom mm name
mcaueuomuooch mcwosHuxmv amouo< owed ousom moowmm
.Amv cowumswm “my cowuesvm we mooamm umuuooum
advanced Hmsuod

 

so venom moamm omuoaomum

 

.nnoauonoa .mauoonz .mmumun cumsuuoz .Ho>oa
pauoz «0 ammo mom mm as manned pumom mcwuoxumz mcwssmmd mmamm usaocsouw couoapoumll.a.m mqmda

.a.m manna ca coaumswm Hoocaa on» coma Hmsoa maasumcem one scaumsvm man» spas meaMm oeuoaomum .wnmc
camuno omas a.m wanes Ca we mwamm owuoaoeum one meoaum omnammnuommn ca ocmuu 056m one ”0902
.aaa .m mom .meamm was mooaum monotone assuom mom "meadow

161

 

 

 

 

momomo bmmoam aovmmm m.mm o.m~ hmma
mommmm womovh havomoa m.m~ o.m~ moma
amaomm momemo ommmmm m.m~ ~.vm moma
manoma mnoaen noose m.m~ a.vm onoa
mmoomm maammm smoomm o.v~ m.mm moma
mmmmmm mommmv vmmamm m.m~ m.om moma
moomme mmmmme mommmm a.vm m.om aoma
mmmmem mmmmmv amomam w.mm m.mm omma
mmoovm mmevam aevmev n.am «.mm mmma
mmmmmm mmmmmm emmmmm m.mv v.0m mmma
ommmmm memmmm mmovam m.ov m.mm mmma
mmmmmv mmvmhv Nmmbmm o.mm v.mm mmma
mmmovm mmmmmo mamomm «.mv v.mm mmma
Nwmomv mommmm common a.mm «.mm vmma
mmoovv ommoom mememv m.~v m.mm mmma
mmmvmv mmmoam mmoomv m.me o.mm mmma
enmemm emamve mmmmmv m.mv 0.0m amma
mmnamm hmmmmv memmva m.am ~.am omma
mcoe coa\zm
amusuosuum ammsuosnum oumom moans oumom use»
e>ammeum0uous¢ m>ammommmuousm emu ou oauoz mo ecu an
no emoeasocx mo emcoazocx mmamm ucmo mom mm camm
mcaumuomuoocav mcaosaoxmv assuom came oumom mooamm
.amv coaumswm any scaumsom ma meoaum monotone
Monotone assuom

 

so venom mmamm touoanwum

 

.noxenoauan\onoa .maumoaz .mouoon
cumcuuoz .am>ea panos mo ucmu new mm um meoaum ousom mcassmmd mmamm usaocsouw emuoapwumll.~.m mamda

162

.ue30a maamuecem ens eusuosmum e>ammeumenousm ecu mo emoeasocx mcaumuomuouca meaMm oeuoaoeum
.Aomma udeuxev meuaum oauos mo uceo uem oo cmcu mmea oe>aeoeu mnesumm cec3 m.omma ecu ou ummuucou
ca .meoamd oauoz mo uceo nee oo oeuMasumom ecu cmcu uecmac ones meoaum neocooum amouom .ooma now umeoxe

m.ooma ecu CH

.~.m use a.m meacme ca cmcu meoaum Hecmac cuas necmac ens meamm oeuoaoeum
.aaa .m eem .measm use meoaum meosooum amsuom mom

 

 

 

 

aamommavmmmemo amaomaavmmmmom aovmho a.o~ o.o~ moma
avooaeavvvmmmm mammaaavooommm mmeomoa m.om o.mm ooma
ammmmaavmmmoam amvmmoavooonvm ommmmm m.mm ~.vm moma
aaoomaavmeoaom ammamoavammom ommono m.om a.vm voma
aommmmavewamom aooaooavhvmomm ommomm m.om >.~m moma
ammmoeavvmvmmo avovooavmaoamo vmmamm o.m~ m.om aoma
aemmovavmammom ammmmoavmoommm mvmmmo m.om m.om aoma
onmmmavvmommm amvamoavomommm amomao «.mm m.mm ooma
aoommoavmm>0mm amvoooavomvoom avvmvv m.vm «.mm mmma
commaoavmooomo amomevavmomvmh vmmmmm o.mv v.om mmma
avvmamavomommm avmmmmavamammo mmovam m.vv m.mm mmma
Ammvaaavveaomv commooavmmmmmm Nmmmmm o.om v.hm omma
ammemavooamvm aammomavmmommo m.m0mm a.ov v.~m mmma
aaavmaavammmmv avmmmaavooammm ohmmbm o.me ¢.mm vmma
aobvmmavmmommv Ammmmmavmmommm meoome N.oe m.mm mmma
ammmvoavammoov amaomvavmmmmmm omoomv m.mv o.mm mmma
covavvavavmmav ammvmmavmaawmv mommmo m.mv o.om amma
amoaommvboommv awmmmmavaaooom mommva w.mm n.am omma
mcoa coa\zm
censuosuum censuosuuw ousom eoaum venom noes
e>ammeumeuousc e>ammeumeuous¢ ecu ou cauoz mo ecu mc
mo emoeasocx mo emoeasocx meaem uceu Hem oo oamm
mcaumuomuoocav mcaosaoxmv aesuoc oamm oueom meoaum
.aov GOaumsvm aov c0aumsvm ma meoaum neosooum
neusooum amsuoc

 

co oemem meamm oeuoaoeum

 

.ao cOausDwm co cememv wo\homalam\omma .sauemaz .meumum cuecuuoz
.ae>ea oauoz mo uceu Hem oo um meoaum oueom mcaESmmc meamm unconsono Ueuuaoeumll.m.m mamce

.163

.m.m manna.
ca scaumsoe umecaa ecu cmcu nesoa maamuecem ems scaussoe macu cuas meawm oeuoaoeum "euoz
.aaa .m eem .meaem one meoanm neocooum aesuom Mom ”eousom

 

 

 

 

 

mammoo hmommm aovmmo a.om o.o~ moma
momovm momomm mueomoa m.0m o.mm ooma
emoomo whomon ommmmm m.mm ~.vm moma
maamob hommmm mmooho m.om a.vm voma
ommvmo oommoo baboon m.o~ b.mm moma
mmmmao mmmmam ommaom o.mm m.om moma
mmomnm aOOMmo memmmo m.om n.0m aoma
ovoanm mommom amomao «.mm h.mm ooma
memmom mmmomm aovmvo m.em e.mm mmma
mmmooo amomom vmmmmm o.me v.om mmma
ommvao vmommm mmovam m.vv m.mm omma
vemamm mmmmmm mmmbmm o.om «.mm omma
mvmmmm omoeo> manomm a.ov «.mm mmma
emboam ommmmm ommmmm o.~v v.mm emma
ammomv momaom ovoomv m.ov m.mm mmma
mmvmmv mmvmmm omo0mv m.mv o.mm mmma
ommmav ooommv mmmmmv m.>v o.om amma
aovuav mvomvm mommva m.mm n.am omma
woos coa\zm
censuosuum censuosuum oumom eoaum oumom nee»
e>ammeumeuousc e>ammenmeuousc ecu ou oauoz mo ecu an
no emoeasocx mo emoeasocx meamm uceu Hem oo came
mcausnomnoocav mcaosaoxmv assuuc cane oumom meoaum
.amv coaumsom any coaumsvm ma meoaum ueosooum
neocooum amsuoc

co oemmm meamm oeuuaoeum

 

..n coaumsom co commas mo\oooauan\onma .mauomaz .moumun
cuecuuoz .ae>ea cauoz mo uceu uem oo um meoaum Unmom mcaesmmc meaMm useocsouu oeuoaoeumll.v.m mamce

.uesoa ewe ensuosuum e>ammeumeuousm ecu mo emoeasocx mcaumu
nomnouca macauoaoeum .meoaum oases mo uceo new oo ens use» zone ca meuaum cec3 emocu cscu .emmue>o
ecu co .uecmac euooeuecu eum euec meamm oeuoaoeum eca .oe>aeoeu ec meoaum amsuom ecu ou aesve
no cecu ueuweum one meoaum mm moo emuos uoc ueEMMM ecu mebmea moaoaum amoauecuommc ece "euoz
.aaa .m eem .meamm use meoaum neosooum aesuos Mom ”eousom

164

 

 

 

 

 

mmmemo mmmmOb aovmmo a.om o.om boma
evmmmm oovmmm hmvomoa m.om o.mm ooma
Nmaomw ooaaom ommmmm N.vm N.om moma
omaaom mmmomm mmmomo a.em a.vm voma
mommmm mammmm mmhomh m.mm m.~m moma
ommvmo oaombo emmamm n.0m n.0m moma
mmooao mmmmvo mvmmmo m.om m.om aoma
momvoo aommvo amomao m.mm >.mm ooma
ammmmm mmmmao avvmvv v.mm v.om mmma
Nooomo momeo vmmmmm o.mo ¢.om mmma
omommm aoammo mmoeam m.ve m.nm mmma
emOmmv omonvm ammomm v.mm v.hm omma
ooamvm «mommo mam0mm a.oe v.mm mmma
ammmmv moahmm omhmmm o.~e v.mm vmma
mmommv mmommm ovovmv ~.oo m.mm mmma
ammooe moommm omoOmv m.mv o.mm mmma
avmmav maammv mmmmmv m.hv o.om amma
moommo aaooom mvmmva m.mm n.am omma
mcoe coa\zu
3 . eflz
censuosuum censuusuum oueom a mo nmv oumom umew
e>ammeumeuousc e>ammeumemou54 ecu ou xms u .wm ecu mu
mo emoeasocx mo emoea3ocx meamm came ounom ease
mcaueuomuoocav mcaosaoxmv amsuoc ma meuaum meoaum
.aov scaumsmm cod scaumsvm ueusooum ueosooum
aesuoc

so oemmm measm oeuoaoeum

 

.mo\homauam\mmma .eauemaz .meumum cuecuuoz .eoaum pauoz ecu ma 3m use eoaum ueosooum aesuoc
ecu ma .zm euecs asmo. .anv xmsnnam us eoaum oueom mcassmmc meamm unconsouo oeuuaceumlu.m.m mamce

165

A pricing policy of paying farmers 55 per cent of
world prices would have benefitted farmers most of the
1950's but would have hurt them most of the 1960's. This
fact is also reflected in the predicted sales relative to
actual sales. Preference for this policy over the actual
one that was followed could be justified thus: first, if
there are any substantial irreversibilities in the under-
lying investment-disinvestment relations in groundnut
production, then paying farmers higher net producer prices
in the 1950's would have expanded productive capacity with
the result that in the presence of taxes in the 1960's
there would have been much more over production relative to
the present state of affairs as there would have been no
tendency towards investment or disinvestment in these
resources because of these irreversibilities; second, for
long run allocative purposes marketing board prices that
are linear functions of world prices that should ultimately
guide farmers' production decisions are preferrable to
prices that either have no functional relationship to
world prices (as when election year politics and the
political party in power determine board prices) or prices
that may be nonlinear functions of world prices. These
comments are applicable to subsequent tables.

Tables 5.6 and 5.7 show the effects of hypothesized
new groundnut prices on government and marketing board tax

revenue from groundnuts. Over the period, there are more

.lf56

TABLE 5.6.--Tax Revenue: Estimated Changes From Higher Groundnut Prices, Larger Vblume,
Individually and in Total, Nigeria, l950-l965.‘

 

 

 

a Change Fromb

Change From Higher Prices Net Change in
Year Higher Prices and Larger Volume Tax Revenue:

(PNj pNj)Mj (Pp -P Nj)(Mj- Mj) a + b

EN

1950 -4,938,907.8 513,44l.6 -4,425,466.2
1951 -4,809,144.4 43,518.0 -4,765,626.4
1952 -6,4l7,370.4 50,033.2 -6,367,337.2
1953 -4,543,733.6 1,009,028.6 -3,534,705.0
1954 -2,683,987.2 l,695,845.2 -988,142.0
1955 -7,263,945.5 -49l,697.5 -7,755,643.l
1956 501,104.8 859,505.0 l.360,609.8
1957 -7,432,859.2 -503,090.3 -7,935,949.5
1958 -7,040,272.8 -3,501,07l.4 -10,933,930.6
1959 l,29l,778.9 615,629.5 l,907.408.4
1960 3,404,780.5 -7l3,3l3.0 3,691,467.5
1961 2,742,196.0 -l,133,160.4 l,609,035.6
1962 4,619,077.2 -213,215.7 4,405,861.S
1963 4,563,016.6 -574,848.0 3,988,168.6
1964 2,233,430.l 1,43l,086.4 3,664,516.5
1965 4,202,476.0 -l,759,225.5 2,443,250.S

 

Note: For PNj , PN , Mj and M5 figures, see Table 5. 3. The potential producer
price figures were obtained by jjadding the composite tax per ton on groundnuts for each
year (see Table 4.3: p. 111) to P“, the actual net producer price. For derivation of
the three columns, see p. 168. In the first column the change is negative (a reduction
in the revenue from groundnuts) for most of the 1950's when hypothetical prices exceed
actual prices: the converse holds for the 1960's: the 1950's thus correspond to a gain
to farmers while the 1960's correspond to losses to them due to the new prices, relative
to actual prices and actual sales. The second column corresponds to neither gains nor
losses to farmers. The last column is defined as the Eg£_change in government and
marketing board tax revenue from groundnuts as a result of the new prices. It is the
second column plus the first, i.e.,

Pl ) (MI

- - C ’ - - ’ O ’ I
(pp Nj j Mj) (PNj Psi)": Mij M p ’p + M p + p Mj

i p "3 Ni 3 Ni Mi
9 M’ - M + p M - M “P
9‘3 3’ N33 3N3
Now, we can derive the last column from first principles. It is the difference between
total government and marketing board tax revenue from groundnuts using the new hypothetical
prices and total government and marketing board tax revenue from groundnuts using the
actual producer prices paid by the board, i.e.,

M (pp - P ) - M (p - p ) - M’p - M p’ -M p + M p P(M’ - M

j u: jp u: jp in: :p 3N3' Pp: 3"“ '"P

j N) 1 Ni

which is exactly the same expression for the third column derived above. *For H; see
Table 5.1, equation (6).

167

years in which the hypothetical prices exceed actual
prices than those in which they fall below them. In
Tables 5.6-5.8, the first column shows the change in
government and marketing board tax revenue from groundnuts
due to the new prices (Pfij)° The change involves a
reduction in tax revenue if the new price exceeded the
actual price and an increase in tax revenue if new prices
were below actual prices. The second column shows the
change in government and marketing board tax revenue due
to the new prices when account is taken of the effect of
the new prices on sales to the board. It is defined as
the "tax" using the new price multiplied by the difference
between the predicted sales to the board using the new
prices and sales to the board under actual prices. An
increase in tax revenue occurs when sales to the marketing
board increase. A reduction obtains when predicted sales
are less than actual sales. A positive change in one year
in the second column is not the negative of the corre-
sponding change in the first column because the two columns
are derived differently. The third column shows the net
change in government and marketing board tax revenue from
groundnuts when everything is considered: actual price,
(new) hypothetical price, potential producer price, actual
sales, and predicted sales. With the definition given to
the elements of column one, the third column is the sum

of the first and second columns.

168

Advocates of the revenue role of marketing boards
have implicitly or explicitly assumed that all elements of
the second column are zero: that sales do not change as
producer prices rise or fall, so that M3 - Mj = AMj E 0
(see Table 5.6). Our entries in this column, particularly
in the 1950's when hypothetical prices exceed actual prices
show that these sales effects of price changes are not
definitionally zero. The significance of the entries in
the second column is that they partially offset the change
in tax revenue from groundnuts due to the new prices
relative to the actual prices and actual sales. How big is
this offset?

The offset is very significant in some years. In
1954 for example, as a result of higher producer prices,
farmers would have gained LN 2,683,987.2, the government
and marketing board would have gained LN 1,695,845.2 from
the positive price sales effect, leaving the government and
marketing board a net (groundnut tax) revenue loss of
5N 988,142.0. Total tax revenue from groundnuts in the

28 thus still leaving

same year was LN 5,125,384,424,
government and marketing board revenue at LN 4,137,242.4.
Thus, even though taxes could have been drastically cut
down in this year leading to extra income for farmers of
over two and a half million pounds,29 total tax revenue
from groundnuts would have been little affected. Simi-

larly, in 1953, as a result of the postulated higher

169

producer prices, groundnut farmers in the Northern States
would have gained 5N 4,543,733.6, the government and
marketing board would have gained LN 1,009,028.6 from the
positive price sales effect, thus leaving the government
and marketing board with a net revenue loss of

LN 3,534,705.0. With total tax revenue from groundnuts

of LN 4,788,395.830

in the same year, the government and
the marketing board would still have been left with a
handsome positive balance of well over one and a quarter
million pounds from groundnuts alone. In 1952, with
higher producer prices, farmers would have gained in
increased income LN 6,417,370.4: the government and
marketing board from the positive price sales effect
would have gained 5N 50,033.2, thus leaving a net tax
revenue loss of LN 6,367,337.2. But with a total tax

31 this

revenue from groundnuts of LN 7,375,800 in 1952,
could have still left a positive balance with the govern-
ment and marketing board. In 1950 and 1951 would the
government and marketing board have ended up in a net
total revenue deficit. For most of the other years (see
Table 5.6), the government and marketing board would have
been better off in revenue terms because on the average,
60 per cent of world price was less than actual producer
prices. Roughly the same trends in terms of the degree

to which revenue losses are offset by the positive price

sales effect can be observed in Table 5.7.

170

TABLE 5.7.--Tax Revenue: Estimated Changes From Higher Groundnut
Prices, Larger Vblume, Individually and in Total, Nigeria,

1950-1965.**

Change Fromb

Change Froma Larger Vblume Net Change in

 

 

Year Higher Prices and Higher Price Tax Revenue:
(PNj PNjH‘Ij (Pp PNj) (M3. M3.) a + b
EN
1950 -4,938,907.8 485,882.4 -4,453,025.4
1951 -4,889,l44.4 421,068.0 -4,388,076.4
1952 -6,4l7,370.4 228,995.2 -6,188,375.2
1953 -4,543,733.6 1,218,588.0 -3,325,l45.6
1954 -2,683,987.2 2,025,453.6 -658,533.6
1955 -7,263,945.5 -837,700.8 -8,101,646.3
1956 501,104.8 898,115.0 l,399,219.8
1957 -7,432,859.2 213,47S.4 -7,219,383.8
1958 -7,040,272.8 -5,894,023.8 -12,934,296.6
1959 l,29l,778.9 617,819.7 l,909,598.6
1960 3,404,780.5 -1,933,975.2 1,470,805.3
1961 2,742,196.0 -2,349,085.6 393,110.4
1962 4,619,077.2 -3,474,404.2 l,l44,673.0
1963 4,563,016.6 -l,760,256.0‘ 2,802,760.6
1964 2:233o430.l 822,744.0 3:056:174-1
1965 4,202,476.0 -2,824,767.0 l,377,709.0

 

Source: For Pfij, Per Mj, and PP' see Table 5.4. For
derivation of columns, see p. 168, and note to Table 5.6. **For MT
see Table 5.4, equation (7). 3

171

With the price series PNj = max (PNj,

overall extent to which lost tax revenue is offset by the
32

6Pw), the
positive price sales effect is even larger. Because the
new hypothetical prices never fall below actual producer
prices, farmers are not worse off. In Table 5.3, it is
evident that government and marketing board are better off
in the 1960's while the farmers are worse off. In Table
5.5 with a new pricing policy, the farmers are not worse
off at any time.

In Table 5.8 the net change in tax revenue from
groundnuts due to the new hypothetical prices relative to
actual producer price, potential producer price, actual
sales, and predicted sales to the board is shown in the
third column. The zero entries in the last seven rows of
the first column reflect the fact that the new prices were
the same as actual prices in those years.

Relative to the price series in Tables 5.6 and 5.7,
the prices underlying Table 5.8 are higher in the 1960's.
Farmers therefore gained. The net losses in tax revenue
as a result of these higher prices in the 1960's are
insignificant (third column): LN 104,553.2 in 1963. With
this new pricing policy, government and marketing board
policy would not have been substantially affected, on the
average. Farmers would have gained without the government
and marketing board being significantly worse off in

revenue terms.

TABLE 5.8.--Tax Revenue From Groundnuts:
Higher Prices, Larger Volume, Individually and in Total,
Nigeria, 1950-1965.

172

Estimated Changes From

 

Change Froma

Change Fromb
Larger Volume

Net Change in

 

 

 

Year Higher Prices and Higher Price Tax Revenue:
-(PNj-PNj)Mj (Pp-PNj .)j(M -Mj ) a + b
EN
1950 —4,938,907.8 513,441.6 -4,425,466.2
1951 -4,809,144.4 43,518.0 -4,765,626.4
1952 -6,4l7,370.4 50,033.2 -6,376,337.2
1953 -4,543,733.6 1,009,028.6 -3,534,705.0
1954 -2,683,987.2 l,695,845.2 -988,l42.0
1955 -7,263,945.5 -49l,697.6 -7,755,643.l
1956 . . 682,984.8 682,984.8
1957 -7,432,859.2 -503,090.3 -7,935,949.5
1958 -7,040,272.8 -3,501,071.4 -10,933,930.6
1959 . . 302,214.6 302,214.6
1960 . . 352,835.0 352,835.0
1961 . . -346,220.4 -346,220.4
1962 . . -844,395.2 -844,395.2
1963 . . -104,553.2 -104,553.2
1964 . l,305,000.0 l,305,000.0
1965 . . 701,224.0 701,224.0
Source: For P’, , Mj see Table 5. 5. For P see P. 168

For derivation of columns sea P. 3111. Tablep5.5,

equation (6).

For M.
J

173

In the case of cotton, only a brief description is
given as the procedures for assessing government and
marketing board (tax) revenue effects of changes in the
board's net producer prices of cotton are the same as those
used for groundnuts above.

Actual producer prices paid to Northern States
cotton farmers (1950-67) had a mean percentage of potential
producer price (world cotton price minus all marketing
costs) of 84.6 per cent.33 Three new hypothetical price
series were then constructed on the assumption that the
board paid cotton farmers the following percentages of
world prices minus all marketing costs (i.e., potential
producer price)34 in each year 1950-67: 80 per cent,

85 per cent, and 90 per cent. As with groundnuts, we stay
on one side of the actual mean percentage since our primary
interest is in the sales and tax revenue consequences of
paying cotton farmers higher prices. However, the linear
regression model is capable of handling the case of paying
cotton farmers lower prices too. This new set of prices is
then used to predict cotton sales, assuming other variables
remain at their sample values. The results are shown in
Table 5.9.

From Table 5.9, our estimates indicate that if
farmers were paid 85 per cent of potential producer price
in each year, they could have been better off in the

1950's but worse off in the 1960's, relative to the actual

174

TABLE 5.9.--Predicted Cotton Sales Assuming Marketing Board Prices at
85 Per Cent of Potential Level, Northern States, Nigeria,
1950/51-1967/68.

 

Predicted
Actual Sales

Producer Price
if Board Paid

 

 

 

Actual Producer 85 Per Cent of Sales to Using New
Year Prices Potential Price the Board Prices
hN/Ton Tons
1950 36.37 47.48 42,240 48,550
1951 55.07 71.90 63,679 78,341
1952 55.07 71.90 50,825 83,526
1953 54.81 83.34 75,366 102,639
1954 55.10 67.49 98,663 97,596
1955 54.68 69.47 80,791 106,514
1956 55.13 61.42 72,951 99,869
1957 54.88 58.60 123,906 99,704
1958 54.91 46.67 87,374 82,357
1959 55.07 44.58 85,892 83,569
1960 55.07 45.01 148,962 95,314
1961 46.67 39.70 83,373 89,590
1962 45.27 42.10 144,004 103,865
1963 45.27 44.43 128,585 115,729
1964 66.67 49.59 128,684 132,304
1965 46.67 46.07 127,296 130,301
1966 44.80 37.04 146,368 111,141
1967 42.93 35.50 75,519 93,272

 

Source: For actual prices and sales, see p. 109. For most of the
1950's, the hypothetical prices were higher than actual prices
received by cotton farmers in the Northern States. Consequently, the
predicted sales tend to be higher than actual sales from the positive
price sales effect. The converse holds for most of the 1960's. For
source of actual producer prices, see page 109. The hypothesized
price series make producer prices a simple linear function of world
prices and marketing costs. This is to be preferred to a situation
in which producer prices are either not related to world prices (as
when election year politics and the party in power determine producer
prices paid by the board) or one in which producer prices are, in some
(unknown) way, nonlinear functions of world prices.

175

prices they received. That is why we have constructed a
new price in which we suppose the marketing board paid

cotton farmers a price:

P = max (P .85P

th th' pt)

th th

is actual producer price in year t and Ppt is potential
35

producer price of cotton in year t. The predictions

where P is the hypothetical new price in year t, P

th

The resulting changes in tax revenue from cotton

based on P above are shown in Table 5.10.
with the two hypothetical pricing policies are shown in
Tables 5.11 and 5.12. The elements of the three columns
in each table are arrived at in exactly the same way as
with groundnuts. The first column shows the change in tax

revenue from cotton due to the new prices, P’ relative to

Nj'

the actual producer price, and actual sales to the

PNj ,
board, Mj; it is the difference between the new price and
the actual price multiplied by actual sales, being negative
(and thus a reduction in tax revenue) when the new price
exceeds the actual price, positive (and thus an increase

in tax revenue) when the new price is less than the actual
price and zero when the new price equals the actual price.
When the change represents a reduction in tax revenue, the
entries represent a gain to farmers in monetary income;

when the change represents an increase in tax revenue,

farmers lose, relative to the actual prices received.

176

TABLE 5.10.--Predicted Cotton Sales Assuming Marketing Board Prices
at P .-max (P ., .85P ), Northern States, Nigeria, 1950-
1967?] N3 P

 

 

 

 

Producer Price Predicted
if Board Paid Actual Sales
Actual Producer P§.=max Sales to Using New
Year Prices (PNj? .85Pp) the Board Prices
bN/Ton Tons

1950 36.37 47.48 42,240 48,550
1951 55.07 71.90 63,679 78,341
1952 55.07 71.90 50,825 83,526
1953 54.81 83.34 75,366 102,639
1954 55.10 67.49 98,663 97,596
1955 54.68 69.47 80,791 106,514
1956 55.13 61.42 72,951 99,869
1957 54.88 54.60 123,906 99,704
1958 54.91 54.91 87,374 95,360
1959 55.07 55.07 85,892 101,200
1960 55.07 55.07 148,962 114,300
1961 46.67 46.67 83,373 103,800
1962 45.27 45.27 144,004 119,000
1963 45.27 45.27 128,585 117,600
1964 66.67 66.67 128,684 179,000
1965 46.67 46.67 127,296 131,800
1966 44.80 44.80 146,368 131,999
1967 42.93 42.93 75,519 110,997

 

period.

Because farmers are never worse off in this regime as
hypothetical prices are never below actual prices, predicted sales
are, on the average, higher than actual cotton sales most of the

. P
NJ' 9

see Table 5.9.

177

TABLE 5.ll.--Tax Revenue From Cotton: Estimated Changes From Higher
Prices, Larger Volume, Individually and in Total,
Nigeria, 1950-1967.

 

Change Fromb

Change Froma Larger Volume Net Change in

 

 

Year Higher Prices and Higher Price Tax Revenue:
(PNj PNj)Mj (Pp PNj) (Mj Mj) a + b
EN

1950 -468,864.0 159,479.9 -309,284.1
1951 -1,071,717.6 608,033.1 -463,684.4
1952 -855,384.8 414,321.7 -441,063.1
1953 -2,150,192.0 402,822.2 -1,747,369.8
1954 -l,222,434.6 -12,707.9 -1,235,142.5
1955 -1,194,898.9 314,077.8 -880,821.1
1956 -458,861.8 290,983.6 -167,878.2
1957 -460,930.3 -251,216.8 -712,l47.1
1958 719,961.8 ~41,139.4 678,822.4
1959 901,007.l -18,328.5 882,678.6
1960 1,498,557.? -432,402.9 1,066,154.8
1961 581,109.8 43,332.5 624,442.3
1962 456,492.7 -315,893.9 140,598.8
1963 103,011.4 -99,550.5 8,506.0
1964 2,197,922.7 29,973.6 2,227,896.3
1965 76,377.6 24,641.0 101,018.6
1966 1,135,815.? 227,566.4 1,363,382
1967 561,106.17

 

Source: For PNj' Mj and M- see Table 5.9. For derivation of
columns see p. 175.

178

TABLE 5.12.--Tax Revenue From Cotton: Estimated Changes From Higher
Prices, Larger volume, Individually and in Total,
Nigeria, 1950-1967.

 

Change Fromb

Change Froma Larger Volume Net Change in

 

 

Year Higher Prices and Higher Price Tax Revenue:

(PNj PNj)Mj (Pp PNjHMj Mj) a + b

EN

1950 -468,864.0 159,643.3 -309,220.7
1951 -l,071,7l7.6 608,033.1 -463,684.5
1952 855,384.8 410,900.77 -444,484.0
1953 -2,150,192.0 402,822.21 -1,?47,369.8
1954 -1,222,434.6 12,707 -1,209,727.6
1955 -1,l94,898.9 314,077.83 -880,821.1
1956 -458,861.8 290,983.58 -167.8?8.2
1957 -460,930.3 -251,216.?6 -712,147.1
1958 . . -34,649.04 —34,649.04
1959 . . -39,800.80 -39,800.80
1960 . . -159,445.20 -159,445.20
1961 . . . . . .
1962 . . -11?,518.80 -117,518.8
1963 . . -75,796.50 -75,796.5
1964 . . ~442,780.8 -442,780.8
1965 . . 34,230.40 34,230.4
1966 . . 18,679.70 18,679.?
196? . .

 

Source: For Pfij, PNj, see p. 175. For derivation of

columns see p. 176. For M3 see Table 5.10.

179

The second column shows the change in government and
marketing board tax revenue from cotton due to the new
hypothetical prices and new predicted sales relative to
potential producer prices and actual sales. It is the
difference between the potential producer price and the
new price multiplied by the difference between predicted
sales and actual sales. It is positive (and thus an
increase in tax revenue from cotton) so long as these two
differences are both positive or both negative,36 and
negative (and thus a decrease in tax revenue from cotton)
so long as these two differences are of opposite signs.
The third column represents the BEE change in government
and marketing board tax revenue from cotton due to the new
prices and the predicted sales relative to actual producer
prices, actual sales and potential producer price: it is
the second column plus the first. Derived from first
principles, it is the total tax revenue from cotton with
new prices and predicted sales minus total tax revenue
from cotton using actual producer prices and actual

37 Tables 5.11 and 5.12 show these results for the

sales.
two hypothetical price series. From Table 5.12 in 1953,
with the new higher prices for farmers, farmers would have
gained LN 2,150,192.0 (a loss to the government and
marketing board), the government and marketing board would

have gained LN 402,822.2 as a result of the positive

price effect on sales, thus leaving the latter a net

180

revenue loss of LN 1,747,369.8. Total tax revenue from
cotton in 1953, incorporating the price effect on sales,
was calculated to be LN 4,788,395.8.38 Thus, tax revenue
is not significantly affected. In 1957, as a result of
the higher prices, farmers would have gained LN 460,930.3,
the government and marketing board would have lost

LN 251,216.76, leaving a total loss of LN 712,147.1.

Total tax revenue from cotton in 1957 incorporating the
positive price effect on sales would have been

LN 5,857,721.9.39

The loss from paying farmers higher
prices still does not substantially affect tax revenue

from cotton. In 1954, with the new higher prices, farmers
would have gained LN 1,222,434.6, tax revenue would have
increased by LN 12,707 as a result of the positive price
effect on sales thus leaving a net revenue loss to the
government and the board of LN l,209,727.6. The calculated
total tax revenue from cotton in same year is LN 5,125,384.4.
A 20 per cent reduction in revenue with farmers gaining
twice what the government and marketing board lose is
certainly not a drastic reduction. For given new higher
prices and marketing costs, the lower the potential
producer price, the lower the gain in tax revenue

resulting from the positive effect of tax increases on

sales. This gain is in fact zero when potential producer

price equals the new higher price as in 1961.

181

The central point is to show that the elements
of the second column are not identically zero as has been
implicitly assumed by those concerned about the revenue
functions of the marketing boards. It is reassuring that
for many years when farmers received higher producer
prices, government and marketing board tax revenue from

cotton was not substantially affected.

Summary and Conclusions

In the first part of the chapter, I presented a
simultaneous equation model in which the sales supply
equation of a marketing board crop was one of four
structural equations of the system. Identifiability of
the sales supply equation was ascertained after which the
reduced form solutions of the endogenous variables were
derived. The reduced form solution for sales of a crop
to the marketing board could not be estimated for lack of
data--avai1ability of data on the other endogenous vari-
ables in the sales supply equation would have enabled us
to estimate the sales supply equation by two stage least
squares. The intent here was to provide a model that
approximates our apriori knowledge about the variables that
influence the sales of a crop to the marketing board.
This simultaneous equation approach should prove useful
as more data become available.

The model was operationalized to remove obvious

sources of biasedness and inconsistency from our ordinary

flip

1’7“"‘7'737'. ;.-

182

least squares estimates. In this second section of the
chapter, coefficients of regressors in the sales supply
equation were estimated. For groundnuts, the sales supply
equation reflecting more or less "short run" influences in
linear form gave a sales supply elasticity of .714 with
respect to the board's net producer price, calculated at
the sample means of observations on groundnut sales and
net producer prices. The log equation of groundnut sales
gave an elasticity of sales of groundnuts to the board
with respect to the board's producer prices of 1.3139.

My efforts to correct for autocorrelation produced
inconclusive results. That was why the transformation of
variables with the autocorrelation coefficient calculated
from the least squares residuals was done for only the
groundnut equations and not done for the cotton equations.
For cotton, the sales supply elasticity with respect to
the board's producer price was estimated to be .90278.

I tested for a difference between the expansion
sales price elasticities and contraction sales price
elasticities corresponding to the rising and falling
phases of the marketing board's producer prices
respectively. I found no significant difference. This
does not necessarily mean that there are no significant
irreversibilities in the investment-disinvestment relations
underlying the production of groundnuts and cotton: our

test was not able to capture some important relevant

h -._-.a._______
Q. '

183

factors that are known from apriori knowledge to influence
farmers' investment-disinvestment decisions.
In the third section of the chapter, I provided
quantitative evidence on the effects on tax revenue from
cotton and groundnuts of paying Northern States farmers
higher prices. Predictions of sales of cotton and !
groundnuts (to the board) under different hypothetical 1
prices were made. It was found that when hypothetical ,
prices were higher than actual prices, predicted sales
exceeded actual sales of these crops to the board. This
implies that other things equal, taxes on cotton and
groundnuts reduced sales of these crops to the board.
These predicted sales together with the implied
lower taxes on farmers were used to compute the revenue
effects of (generally) higher prices to groundnut and
cotton farmers in the Northern States. When the board
paid farmers higher (hypothetical) prices, farmers gained
in monetary income. Some of this gain however must be
offset by their loss in consumer surplus as they cut down
their own home consumption of these crops. We did not
estimate this loss in consumer surplus. Our estimates
show that the positive effect of higher prices on sales
(to the board) partially offsets the loss in tax revenue
due to the higher prices. Our main point has been to show
that this offset of tax revenue loss is not identically

zero as has been implicitly or explicitly assumed by

184

previous writers. With higher prices than the actual
prices received by farmers, this tax revenue loss offset

is larger, the larger the difference between predicted and
actual sales. This offset is also larger the larger the
difference between the potential producer price (i.e.,
world price minus all marketing costs) and the hypothetical
price. The higher the sales price elasticity,40 the higher
the predicted sales and thus the larger the offset in tax
revenue loss from the positive effect of higher prices on
sales to the board.

The following conclusions are drawn. Sales of
cotton and groundnuts to the board are very responsive to
changes in the board's producer prices. The positive
effect of higher (hypothetical) prices on sales to the
board was able to substantially offset the crop tax revenue
loss that results from paying farmers higher prices in
many years in the 1950's and 1960's; in other years there
would have been sizeable reduction in crOp tax revenue.

There is an inevitable shortcoming in the
analysis: we have not been able to account for the
production effects of higher prices directly because of
lack of data or even reliable estimates of total output or
production of any of the crops studied. It is hoped that
as more data becomes available, this factor will be

incorporated in the analysis.

FOOTNOTES

1We have not included smuggling in this formulation.

2Jan Kmenta, Elements of Econometrics (New York:
Macmillan Co., 1971).

r~***V”wsEEI
‘4‘
p.

3Ibid., pp. 543-45.

4With no compelling a riori reason to believe that
either the elasticity of suppEy (output) of any marketing
board crop with respect to its net producer price is unity
or that the marginal propensity to consume this crop is
zero, we shall, respectively, presume e and c2 to be non-
zero.

0

5Charles R. Frank, Jr., Statistics and Econometrics,
(New York: Holt, Rinehart and Winston, Inc., 1971), p. 332.

An alternative statement of the necessary and
sufficient condition for identifiability is that we can
form at least one nonvanishing determinant of order G-l
from the coefficients with which the variables excluded
from our equation appear in the G-1 other structural
equations. By this criterion,

1 -a2 0 0
A* = O 0 -w2 -w

1 O O 0

with at least one largest nonsingular submatrix with rank =

3 = G-l, again showing that the sales supply equation is

identified. For this alternative statement see G.

Tintner, Econometrics (New York: John Wiley and Sons,

Inc-)0 p0 13,- .

6Kmenta, op. cit., p. 551.

185

 

186

7For evidence which may or may not cast doubts on
estimates of total output from existing data in the case of
cotton, see footnote 14 on p. 125-

8This more or less corresponds to an assumption
that in equilibrium, the price per unit of the crop in the
domestic market equals the net producer price paid by the
board net of transportation and other marketing costs. If
the argument were extended to include markets in neighboring
countries serviced by smugglers, then this equilibrium
price also equals the net producer price in these countries
properly adjusted for transportation and normal marketing
costs, weighted by the probability of not getting caught
(arrested) in the process of smuggling and the probability
of not getting a conviction after arrest.

9This increased sale to the marketing board and
reduced consumption as a result of higher marketing board
prices need not be restricted to the usual substitution and
income effects describing the response of consumers to
price changes in their capacity as consumers per se.
Indeed, in the Nigerian case, producers of most marketing
board crops are consumers as well as producers of their
crops. So there is a producer income effect as well as a
consumer income effect. If Nigerian farmers consume a
significant proportion of their crop and the (implied)
income elasticity of demand for this crop is very low or
negative, then an increase in the net producer price paid
by the marketing board will lead to less consumption and
more sales to the marketing board.

10For treatment of the dominant firm on the selling
side, see G. Stigler, "Notes on the Theory of Duopoly,"
Journal of Political Economy (1940).

11If this were not the case, the disturbance term
and the income variable would be jointly distributed which
would make the OLS estimates inconsistent.

12Glenn L. Johnson, "The State of Agricultural
Supply Analysis," Journal of Farm Economics (1960).

13E. Malinvaud, Statistical Methods of Econometrics
2nd revised ed.; Amsterdam-London: North-Holland Publish-
ing Co., 1970), p. 567.

‘m‘“ 0‘4 ..—‘-—?
a s
1 ..w“' t

i "'"Ma.
t
. A--

187

14For possible usefulness of this exercise, see
J. Johnston, Econometric Methods (New York:_ McGraw Hill
Book Company, 1960).

15This indeed may be the case. A major form in
which groundnuts are eaten domestically (and thus diverted
away from the marketing board) is through roasting or
frying of the nuts. It is not unlikely that consumption
in this form falls as income rises.

16We assume that the disturbance, st, follows a E
first order autoregressive scheme:

at = pet_1 + Ut (t = 2,...,n)

so that the transformed equation which is estimated with
OLS to give (6)' above is:

sa...-.-.—4q____
1. . .4 1
.

Y - pY _ = Bo(1-p) + 81(x2t - px2t_l) + 82(x3t - ) + Ut.

px3t-l

where 6 is the estimated autocorrelation coefficient from
the residuals of (6). This method is only a special case
of the iterative procedure suggested by Cochrane and Orcutt
(see D. Cochrane and G. H. Orcutt, "Application of Least
Squares Regressions to Relationships Containing Auto-
correlated Error Terms," Journal of the American Statistical
Association [March, 1949]). Our two stage procedure which
does not go through the whole iterative procedure until
convergence of the values of the estimates is attained
does, however, possess the same asymptotic properties as
the maximum likelihood estimators. Little is known about
their small sample properties. See Kmenta, op. cit.,

p. 288.

 

17See the discussion on p. 103, especially
footnote 2.

18The simple correlation coefficient between income
and lagged groundnut price is 0.5280.

19This test is rather weak. The magnitude of the
price change, the most recent history of relative prices .
and the process by which Northern States farmers form and
revise their expectations about future board prices are
all relevant factors that are not captured by this test.

20See discussion on p. 63.

188

218. A. Oni, "Econometric Analysis of Supply
Response Among Nigerian Cotton Growers," Bulletin of Rural
Economics and Sociology (Ibadan: University of Ibadan,
1969). Oni's two estimates were .2069 and .2593 (see
Ibid., p. 217). Assuming the change in sales per unit
change in the board's price exceeds the corresponding
change in output (production), then we expect our sales
price elasticity to exceed the output (production)
elasticity. This is likely to be so for most marketing
board crops that have multiple market outlets.

 

22For possible reasons for the sign see p. 151.

23For reasons for including the trend variable,
see p. 145.

24

See qualifications on p. 149, especially
footnote 19.

25Not all cases of the transformation exercise on
the groundnut equations are reported here.

26This mean figure was computed from the annual
percentage figures reported in S. O. Olayide and Dupe
Olatunbosun, "The Effect of the Marketing Boards on the
Output and Income of Primary Producers," (NISER, 1971),
p. 17.

27One alternative would have been to go further
right of the mean percentage of world prices that was
actually paid (mean [PNj/Pw]100 = 55.01%). The highest
percentage of world prices farmers actually received in the
sample period was 79.7 per cent in 1961. I could have then
hypothesized a new price series with farmers receiving a
percentage of world prices in each year in the neighborhood
of 79.7 per cent. 1, therefore, constructed a price series
on the assumption farmers received 70 per cent. .The
maximum value in this series was LN 110.52 in 1951 and the
minimum value was LN 27.61 in 1962. The mean of actual
producer prices that farmers received (1950-67), the price
series used in estimating the regression equations, was
LN 33.38. There were only four years in which the hypo-
thetical prices would have been below this mean; most of
the other years, they were much higher (e.g., LN 62.14 in
1959, LN 56.58 in 1953, LN 55.14 in 1952, LN 110.52 in
1951, etc.). Thus, deviations from the mean of our
original sample would have been very large. This would
have made our predictions of sales to the board more

I! ' fie -wu—1-. ._
,. «nEZ!
_l

189

imprecise. That is why I hypothesize a new price series
in which we stay at 60 per cent of world prices but in
which we avoid subperiods in which our hypothetical prices
are lower than the actual prices farmers received.

28The total tax revenue figures are derived by
multiplying the new predicted sales for a given year by
the difference between the potential producer price (P )
and the new hypothetical Pfi° (= 60 per cent of world pfice).
For the derivation of potential producer price in each
year, see Table 4.3. There is an implicit assumption here
and in all calculations of the revenue impact of new hypo-
thetical prices that the change in sales induced by the
new hypothetical prices does not affect the world price.
Within the relevant price increases permitted with the new
hypothetical prices, it is unlikely that the induced
increase in sales will significantly affect world prices.

29This represents only the monetary income gain to
farmers: real gains may not rise in the same proportion as
incremental sales to the board lead to a reduction in the
farmers' own home consumption.

30See footnote 28 for method of calculation.

31See footnote 28 for method of calculation.

32For the rationale behind this hypothesized
pricing policy, see footnote 37.

33This mean percentage of potential producer price
was calculated from H. C. Kriesel, "Cotton Marketing in
Nigeria," CSNRD 24 (East Lansing: Michigan State Uni-
versity, 1968), p. 73.

34For potential producer price figures see p. 109.

351 rejected the alternative of going further to
the right of the actual mean (PN-/Pp) (mean (PNj/P ) =
.846 for the period 1950-67) by choosing a higher 5Nj/pp
ratio because this would have led to large positive
deviations between our new hypothetical prices, Pfi° and
the mean of the actual prices received (calculated to be
PNj = 50.68) that were used to estimate the regression
equation and this in turn would have increased the
imprecision of our predictions.

 

190

36Though "both negative" is mathematically correct,

we doubt its economic importance.

37When the third column is defined as the second
column plus the first, it is:
P -P’.
‘ p N:
When the third column is derived from first principles,
it is: (P — Pfij)M§ - (P - PNj)Mj. The two expressions
are equiva ent. See p. 1 6.

) (M5 - Mj) - (Pfij - PNj) Mj

38These are calculated by multiplying the differ-
ence between potential producer price (Pp) [for figures
on this see Table 4.1] and the new higher price (Pfij) by

the estimated sales (M5).

39See footnote 38 for method of calculation.

4OSee p. 144 and pp. 68-69 for reasons that may
possibly lead to an underestimate of the sales price

elasticity.

111:9

Ill. -.ln-’ .—
., -
I

CHAPTER VI

MARKETING BOARDS AS POTENTIAL STABILIZERS

OF STATE GOVERNMENT REVENUES

Introduction

Stabilization of producer prices or producer
incomes was the original raison d'etre for the establishment
of the marketing boards.

With the emphasis shifting to the fiscal role of
these boards, a new rationalization is emerging: that
they can help stabilize State Government revenues.

First, are (crop) tax revenues "stable?" What
does our apriori knowledge tell us about the nature of
factors influencing the supply and demand for these crops?
What do these influences lead us to expect with respect to
"stability" in volume of sales of crops to the boards?
Second, can fluctuations in (crop) tax revenues offset
fluctuations in State Government revenues from other
sources apart from taxes on marketing board crops?

The third objective of this study is to examine

the extent to which reliance on revenue contributions from

191

192

marketing boards and allied statutory corporations can be
trusted to stabilize State Government revenues.

This chapter is divided into four sections. The
first section examines total variance of revenue from taxes
on marketing board crops; the second section looks at the
relationship between numbers of crops and stabilization of
average revenue yield per crop: in the third section,

total variance of (crop tax) revenue is minimized subject

"“"*“‘*"“‘m
’ J

to certain constraints while the fourth section looks at
variances and covariances of total State Government
revenues with and without revenue contributions from
marketing boards and allied statutory corporations.
Marketing Boards as Potential
_Siabflizers gffiate
Government Revenue
It has recently been suggested that "one important
fiscal implication of the regionalization of the marketing
boards is their conversion from potential stabilizers of
producer prices into potential stabilizers of Regional
Government revenue.1 This potential is implicitly
recognized in Helleiner's strong advocacy for the fiscal
role of marketing boards. This is a very important issue
because in it lies a potentially powerful new justification
for widening the taxing powers of the marketing boards.
Many State Government Agricultural Officers did suggest
widening the powers of the marketing boards at the (summer)

1971 Seminar of the National Agricultural Development

193

Committee held at the University of Ibadan. Is the revenue
stabilizing potential of marketing boards real?

Suppose stabilization means minimizing deviations
from the mean revenue of the State Government over a given
plan period so that a factor will be stabilizing if it
reduces the variance of total State Government recurrent
and capital revenue over a projected plan period of some
years into the future. Let total revenue from taxes on m
marketing board crops over the given plan period be the

sum of revenues from taxes on the m individual crops, i.e.,

m
R= 2: R. = 2t.M. (1)

where R is total revenue from taxes on all marketing board
crops within a given State over the projected State
Development plan period; Rj is the total revenue from
taxes on the jth crop over the plan period; tj is the unit
tax rate (e.g., LN per ton) on the jth crop assumed constant
to simplify arithmetic; Mj is total sales of the jth crop
to the board over the plan period. From the point of view
of the State Government and the marketing board, sales of
any crop to the board can assume any positive value in any
given year or when the whole plan period is considered.
For all practical purposes then, given the uncertainty
about the volume of sales of a given crop over the plan
period, this volume of sales to the marketing board can be

treated as a random variable. Therefore, total revenue

9» ~ 51"“1-‘1‘E'
_ ‘ . I

194

accruing to the State Government and its marketing board

from taxes on marketing board crops is also a random

variable.2
The variance of total (crop) tax revenue is:3
m
V(R) = Z 0.. t. t. (2)
i.j=1 ‘3 1 3

where Oij is the covariance between the sales of the ith
crop and sales of the jth crop to the marketing board and
where we have, for simplicity, regarded the unit taxes as
constants. Thus the variance of revenue from taxes on
marketing board crops depends on the levels of taxes, the
variances of sales of individual crops and the covariances
between sales of all the crops. For given positive corre-
lations between sales of crops, the higher the variances
of individual crops, the higher the variance of total crop
tax revenue.

When the sales of all crops to the marketing board
of a given State are independent of one another, then the
covariance matrix of sales of these crops is the diagonal
matrix with the ith diagonal element being the variance of
sales of the ith crop to the board over the plan period
and the variance of total revenue from taxes on all m

crops is:

‘1‘“,- . h.

1:. '

195
m
V(R) = Z t. c. (3)

where a; is the variance of sales of the jth crop to the
board. Thus in the case where the covariance matrix of
sales of all crops under marketing board control to the
board is the diagonal matrix either because the sales are
mutually uncorrelated or because they are independent, then
it follows that the more crops are brought under the
control of the board the greater the (absolute) variance

of total revenue from taxes on these marketing board
crops.4

The same result follows if all the off-diagonal
elements of the covariance matrix were positive. If some
or all of the off-diagonal elements cf the covariance1
matrix of sales to the board were negative, then it is
possible to reduce total variance in (crop tax) revenue
by bringing more crops under marketing board control.

With non-independence among sales of crops to a
marketing board (or with negative correlation coefficients),
stability of State Government revenue from revenue con-
tributions of marketing board crops is therefore determined
by the magnitudes of the correlation coefficients and the
size of the diagonal elements.

Suppose public authorities are interested, not in
the variance of total crop tax revenue, but in the variance

of some "average tax revenue yield per crop."

r-‘x- nu n—‘~_--._-

196

Let the State Government expect from its marketing
board and allied statutory corporations a sum equal to the
arithmetic mean of total revenue from taxes on m marketing

board crops equal to R:5

- l
R=-<2R)
m j=1 j

where R. is total tax revenue from the jth marketing board

I

- i1.

'7‘ gym.

{91:1}... e—-—-—.
‘ .
l
. 4

J
crop.6
The variance of this is:
-— 12m2
V(R) = (a) .5 OR. + 2 Zi<j Z Cov (Ri' Rj (4)
-1 l
where a; is the variance in revenue from taxes on the ith
i

crop and Cov (Ri, Rj) is the covariance between revenue
from taxes on the ith crop and revenue from taxes on the
jth crop. The variance of the given contribution expected
from marketing boards and allied statutory corporations
then depends on the number of marketing board crops, the
variances in revenues from individual crops and the co-
variances between revenues from the ith and jth crops
(i # j, i, j=1,..., m). The effect of the number of
marketing board crops in reducing the variance of this
given sum (R) is limited under present Nigerian circum-
7

stances. But more significant in the Nigerian case are

likely to be the variance in sales of individual crops to

197

the marketing board and the correlations among sales.
Marketing boards were set up as export monopsonies. In

the past, for most of their crops, either there was no
domestic demand (cocoa) or transport and storage facilities

were so poor that surplus producing areas could not

 

profitably send much of their products to deficit areas
so that the marketing board was the effective "dominant
firm" on the buying side (palm produce, beniseed, ground- 1
nuts, etc.). There is now a large and expanding internal
market for these crops (except cocoa) as population and

real incomes have risen and transport and storage facilities
are expected to improve in the years ahead. These internal
market possibilities are even greater for any new food

crops marketing boards may be contemplating to bring

under their control. On the supply side, the usual un-
certainties with respect to yield and weather exist,
uncertainties which will increase with the transformation

of agriculture from one depending mainly on traditional

and farm-produced inputs to one depending increasingly on
new inputs from the non-farm sector. With these demand

and supply considerations it is only realistic under
Nigerian circumstances to expect the variances of sales

of present (and potential) marketing board crops to the
boards to increase and thus increase the variance in
expected revenue from the board, i.e., destabilize State

government revenue. General economic conditions in

198

Nigeria will exercise influence on sales of some crops to
the marketing board in the same direction, e.g., if
expected relative food prices are rising in the Northern
States, farmers will reduce their production of both cotton

and groundnuts and conversely for a sufficient fall in

W

expected relative food prices. These substitutions seem

“IF—'7.

..I.‘

to have occurred in the Northern States in recent years.

‘. sus—.-._ ..

In such cases we expect sales of crops to the marketing
board to be positively correlated. In the Northern States
for the period 1947/48-1969/70, the correlation coefficient
between all sales of seed cotton and groundnuts to the
Northern States Marketing Board is 0.5713, that between
groundnut sales and beniseed sales is 0.5475, that between
groundnut sales and soyabean sales is 0.1344, that between
seed cotton sales and beniseed sales is 0.2037, that between
beniseed sales and soyabean sales is 0.4094 and the least
correlation coefficient is that between seed cotton sales
and soyabeans, 0.0567. An implied influence here is the
common influence of supply factors on crops that can be
grown in the same area; for some pairs of these crops the
random variables influencing production as such are
probably jointly and identically distributed.

In North Central State for the period 1956/57-
1970/71, the correlation coefficient between sales of
groundnuts and sales of seed cotton to the Northern States

Marketing Board is 0.2737. On the other hand, in Katsina

199

Province, the correlation coefficient between sales of
groundnuts and sales of seed cotton to the Northern States
Marketing Board is -.3263 while in Zaria Province this
correlation coefficient for the same period was -.1105.
These negative coefficients most probably reflect the
competition between these two crops in production in the
two provinces. Large positive correlations among sales

of crops to the board increase the variance of a given
expected revenue contribution from the marketing board and
allied statutory corporations, i.e., they destabilize
total State Government revenue over a given plan period.
Thus, if crops are taxed by the board whose variances in
sales to the board are progressiVely larger than the
variances of the present marketing board crops and which
have large positive correlations, State Government revenue
would be destabilized even if the number of marketing
board crops becomes large.

The strongest argument to justify widening the
taxing powers of the marketing boards so that they can help
stabilize State Government revenues lies in the fact that
the variance of the amount the State Government expects to
receive from them and allied statutory corporations over
a given plan period (R) approaches the average covariance
of the individual tax revenues from all the crops as the
number of crops taxed by the government and the marketing

board becomes very large. From (4) above,

.1 “WW

200

 

2 m
— _ 1 2 2 (m-l)m
V(R) — (5) [i=1 0Ri + ———§—— .2 iEiCov(Ri, Rj)] (5)
(m-l)m
2

+ ——— . (average covariance)

lim var (R) = average covariance (6)

m + m*

-m-—___.~_...,.___e .
w , ‘y. HT?

where m* is a very large number of marketing board crops.8

However, from our earlier arguments about the nature of
the variances and covariances and the evidence below,
there is no room for comfort. The most favorable circum-
stance is when this average covariance is zero. Our
argument is that the average covariance in Nigerian
circumstances is unlikely to be small. Using the co-
variances of sales of all crops to the Northern States
Marketing Board as a proxy, we find the average covariance
with four crops (cotton, groundnuts, beniseed, and soya-
beans) for the period 1947/48-1970/71 to be l,676,292,ll8.28.
If the average covariance of revenues from these four crops
comes close to this, then the potential for stabilizing
State Government revenues through reliance on revenue
contributions from marketing boards and allied corporations
is small.

Let us now take account of the institutional set

up. The Northern States Marketing Board is responsible

201

mainly for five crops: groundnuts, cotton, beniseed,
soybeans, and palm produce.9 The board in each year
(implicitly) decides how much of desired total crop tax
revenue shall be borne by each crop. We could then ask
the question: how should desired total crop tax revenue
be distributed between these crops, so as to minimize the
variance of total crop tax revenue subject to the con-
straint that the desired crop tax revenue is indeed
realized?

In this case, we want to minimize:

2 2 W2 2 .
0 = . joj subject to
R j=l
5 5 *
(i) Z W.E(R.) = Z W.u = R , and
i=1 3 3 i=1 3 R3
5
(11) Z W = l
j=l 3

where “R is the expected value of the tax revenue from

J

the jth crop, Wj

revenue borne by the jth crop and R* is the desired total

is the share of total desired crop tax

crop tax revenue the board (implicitly) wants to raise
over the given period. Form the Lagrangian:

I S 1*
L = T . Win Oi. + Alt; WjE(Rj) - R ] + 12 (EWj-l) (7)
193 3 J 3

The necessary conditions for minimum variance are:

202

3L _ 2 I W. 0'

w; - 2WiO'Ri + i<j j ij + A1E(Ri) + A
5

3L == 2 vv.E(iz.)

== 2W5
3

where Oij is the covariance between tax

ith and jth crops. The seven equations

unknowns can be written as

FB = K
where
' 2
201 29120102 20130103 20140104 291501
2
2p120102 202 20230203 2p240204 2p2502
2
U
2013 103 2p230203 203 29340304 293503
2
F: 20140104 2"240204 2"340304 204 294504
2
29150105 2p250205 zp350305 2p450405 205
E(R1) E(R2) E(R3) E(R4) E(R5)
L3. 1 1 1 1

 

O

O'

O

0'

5

5

5

S

2

ll
0

1,...,5

revenues from the

in the seven

'__”

E(R1) 1

E(R2)
Eng)
E(R4)
Bag)

0

O

1

1

 

 

 

203

and
r01
0
K = 0
0

 

 

Pa O
l-
ffi on _A-u—mT—W

B=I‘K (8)

From (8) it is evident that the shares of total (crop tax)
revenue borne by any crop depends on the variances and
covariances of individual crop tax revenue yields as well
as the expected (tax) revenue yield of each crop. Expected
tax revenue from a crop depends principally on expected
sales of the crop to the board. To use weights derived
from historical data for future planning requires an
assumption that past trends continue into the future.

We expect the variance of (crop) tax revenues to
be large on account of the large variances and positive
covariances of sales of crops to the marketing boards.lo
Let us now look at the relationship between revenue ‘
contributions from the boards and allied statutory

corporations and State Government revenue from other

sources e

204

Total State Government Revenues With
and Without Revenue Contributions
From the Marketing Board

 

Suppose RT is total State Government revenue over
the plan period inclusive of revenue contributions from
the marketing board and allied statutory corporations made
possible by continued taxes on marketing board crops over
the plan period, R is revenue contribution from the board

and allied corporations to State Government revenues over

flirim— ‘-

the plan period and z is total revenue of the State
Government over the plan period exclusive of contributions
from the marketing board and allied statutory corpo-

rations.11

For our measure of the "instability" or
"fluctuations" in State Government revenue over the plan
period, we use the variance of the revenue series. Thus,
the variance of RT will measure fluctuations in State
Government revenue inclusive of contributions from the
marketing board and allied statutory corporations (loans
and grants) and the variance of Z will measure fluctuations
in total State Government revenues exclusive of con-
tributions from the marketing board and allied statutory
corporations during the plan.

Within the context of this model, a suggestion
that the boards will stabilize State Government revenues
over a given period would mean that the variance of RT

will be less than the variance of Z. If the variance of

RT(O§T) exceeds the variance of Z(o§), then State

205

Government dependence on revenue from the marketing board

and allied statutory corporations will actually destabilize

 

State Government revenue. We know from

RT = Z + R, that

02 02-9-0

_ 2 .
RT ’ z R + szz 0R 0z (9) E

.4. _

2 I I e e
where OR is the variance of revenue contributions from the

‘ '1! was!“ ..-,
A v

board and allied statutory corporations and pRZ' the 5?
correlation coefficient, measures how well the revenue
contribution from the board and its allied statutory
corporations help to stabilize State Government revenue
over the plan period.

When pRz = 0, then the net (average) effect of
revenue contributions from the marketing board and allied
statutory corporations on deviations of Z from its mean
would be zero, a neutral result with respect to the
objective of stabilizing State Government revenue.

When 9R2 = 1, then dependence on contributions
from marketing boards and allied statutory corporations in
the form of loans and grants would be perfectly destabi-
lizing for State Government revenues over a given plan
period; in this case fluctuations in State Government
revenues would be amplified the most as a result of
dependence on revenue contributions from the marketing

board and allied statutory corporations. As government

206

revenues from other sources tend to fall, those from its
marketing board and allied statutory corporations would

tend to fall and similarly for a rise in government revenues
from other sources.

Dividing (9) through by 0:, we find that if
2

RT

and allied statutory corporations in the plan period have

0 /c§ < 1, then the revenue contributions of the board '

indeed helped to stabilize State Government revenue as

suggested explicitly by Teriba and Olakanpo and implicitly
2

RT

from the marketing board and allied statutory corporations

by Helleiner; if c /o§ > 1, then revenue contributions

during the plan period would actually destabilize State
Government revenue; if OR: /c§ = 1, then revenue con-
tributions from the marketing board and allied statutory

corporations would contribute nothing to the stability of

State Government revenue in the plan period, i.e.,

2
O
I S - 5 _ 0
——§ 3 1 according as 0R2 3 2 R (10)
o 0
Z Z
f h' h 'f o - 0 th ch/o2 5 1 d' S l
rom.w 1c 1 R — Z' en, Z 3 accor ing as 0R2 3.5

From equation (9) civil servants in state governments and
academicians who believe or advocate the revenue stabi— -
lizing role of marketing boards and allied statutory

corporations would be most delighted if OR = oz and 9R2 = —1

because this would be saying that the variance of state

207

government revenue over the plan period inclusive of revenue
contributions (loans and grants) from the board and allied
statutory corporations is zero. Factors causing fluctu-
ations in their revenue contributions include the many

random factors that influence planned output of crops

I

(weather, technological uncertainty, etc.) and the many
factors which influence planned sales to the marketing

board (demand conditions in the country, demand conditions

in neighboring countries, expected relative crop prices in
these markets, etc.). The factors influencing State
Government revenue exclusive of revenue contribution of
boards and allied corporations include those factors that
influence the revenue yield of personal income taxes and
other direct taxes, Federal Government allocations that
actually materialize, etc. It is only by accident that
these sets of factors will operate in such a way as to

= o

render a The more responsive sales to the board of

Z.

a marketing board crop are to changes in demand conditions

R

on the home market and in neighboring countries, the
larger would be the fluctuations of the revenue con-
tributions of the marketing board and allied statutory
corporations and therefore the more likely that State
Government revenue would be destabilized.

There is growing evidence that reliance on revenue
contributions from marketing boards and allied statutory

corporations is actually destabilizing State Government

208

revenues and plan budget operations. For the 1971/72
Budget Year for the Western State,

Capital receipts of some LN 7 mill fell LN 5 mill

short of estimates because the state of world markets,

particularly for cocoa, made it impossible for the

marketing board to make a contribution. . . . The

Governor, Brigadier C. O. Rotimi, said that prospects

for the coming year were better, although cocoa

prices remained low and they still had to deal with

smuggling, food shortage and inflation. . . .12

For most other marketing board crops including
cotton and groundnuts, the factors making it "impossible
for the marketing board to make a contribution" and thus
leading to capital receipts falling short of estimates
include an increasing diversion or sale of large quantities
of these crops to the Nigerian domestic market through
private marketing channels, the increasing amount of
smuggling across Nigerian borders as expected relative
crop prices in neighboring countries have been consistently
higher than Nigerian marketing board prices and also
increasing on-farm consumption by the farmers themselves.
Reliance on marketing board revenue contributions

has also destabilized revenues of statutory corporations
especially when marketing boards made so-called trading
losses. Thus, in the Eastern States in the 1950's,

In order to enable the Production Development Boards

to plan their programmes more effectively the Oil

Palm Produce Marketing Board guaranteed them a minimum

allocation of LN 800,000 for each of the years 1950-55.

. . . In order to fulfil this obligation during its

two years of trading losses it was necessary actually

to reduce reserves which were not originally to be
employed'for developmentgpurposes at all.13

 

 

209

From our earlier analysis, positive correlation
between a State Government's revenues exclusive of any
revenue contributions from its marketing board and allied
statutory corporations and revenue contributions of these

agencies would indicate that these contributions do indeed

7'

tend to destabilize total State Government revenue, i.e.,

a—
‘

571:9”-
). i 9

when State Government revenues from all other sources

apart from the marketing boards and corporations tend to

tn! Um‘-—-—-—.

I

fall, the revenue contributions of the board and corpo-
rations also tend to fall and conversely for a tendency
towards a rise. The empirical evidence supports the view
that reliance on revenue contributions from marketing
boards and allied statutory corporations has destabilized
State Government revenues for all the states in the 1950's
and 1960's. For the Northern States, this correlation
coefficient was .6286 (pRz = .6286); for Western Nigeria,

= .0171; for the Eastern States, = .7049.

0R2 pRz

Implied in our earlier analysis is the fact that
reliance on revenue contributions from the marketing
board and allied corporations has the greatest potential
for destabilizing State Government revenue when: first
the crop has a large (and fluctuating) domestic market
and/or markets in neighboring countries and secondly, the
crop does not have significant irreversibilities in the
investment-disinvestment process underlying its pro-

duction.14 We would expect, therefore, that the State

210

Government relying on tax revenues from a crop subject to
the most uncertain influences on the demand side and the
least irreversibilities in the underlying investment—
disinvestment process to be subject to the greatest danger
of destabilization of State Government revenue, and the
State Government relying on tax revenues from a crop
subject to the least uncertain influences on the demand
side and the greatest irreversibilities in the underlying
investment-disinvestment process to be least exposed to
the danger of destabilization of State Government revenue.
This indeed is the case. The Western State, relying on
revenue contributions from taxes on cocoa, had the least
potential for destabilization (the most potential for
stabilization) with, pRz = .0171; the Eastern States,
relying on revenue contributions from mainly palm oil and
palm kernel, had the most potential for destabilization

(the least potential for stabilization), with 9R2 = .7049,

while the Northern States, relying on revenue contributions

from cotton, groundnuts, beniseed, and soybeans, also had
high potential for destabilization (low potential for
stabilization), with pRz = .6286. Table 6.1 summarizes

these findings.

Summary and Conclusions

This chapter has been mainly concerned with the

problems of relying on revenue contributions from marketing

boards and allied statutory corporations for purposes of

211

. u. . I»! m
rill-19‘-.. Iqu’fl‘y;nl.l.u . %

.mcowueuomuou pewaae use mpueoc
mcwuexuee ecu scum coHuscuuucoo esce>em ecu mu m use mcoHuscHuucoo emecu mo ebumsHoxe
mesce>em eueum Heuoe mu N .mcoHueuomuoo huousueum peuaae use mpueoc mcwuexuefi

ecu Eoum ecouuscwuucoo esce>eu mo e>wmsaocu mesce>em eueum Heuoe mu ace

 

 

oooo.H meow. Hmmb. m HNNN.N mmme.ou omem.ma m
oooo.H swam. N vmmv.mn v~mm.mm N
Ammawmmav
oooo.e em eoom.ooe em neuenm sneunem
oooo.H Heao. «amp. m mmsm.mm bmvm. «Hma.em m
oooo.a Home. N mnem.me vmmo.me N
Amoummmav
oooo.H 9m mvmo.ooa em euuemuz cueumez
oooo.a meme. bmmn. m mmwv.a monm.¢ nmmm.m m
oooo.a comm. N mmmm.mm vamm.hm N
Avonmmmav
ooo.H am mmma.mv em meueum cuecuuoz
m N am m N am

 

eueum

 

xuuuez cOuueHeuuoo eHmEHm

xuuuez eoceuue>ouaeoceuue>

 

e.mmmalwmma

.ewuemuz .mcowueuomuoo pewaas use mpueom mcuuecuez Scum mucuuscwuucou
esce>em unocuug use cuuz mesce>em eueum mo meoceuue>ou use meoceuue>ln.u.o memes

212

stabilizing State Government revenues. It was shown that
when total crop tax revenue is considered, the variance of
this revenue depends on variances and covariances of sales

of crops to the board and the level of taxes. For the four

main crops under the Northern States Marketing Board, it
was found that for the period 1947/48-1969/70, sales to
the board were all positively correlated. This positive

correlation was probably mainly due to supply and demand

. :11

-.._"..

'““‘W

vI

considerations and other common influences within the
general economy. These positive correlations and the
large variances must have contributed to instability in
expected total crop tax revenue from these crops. If
there had been negative correlations between sales to the
board at the all-states level, the additional taxation of
new crops, particularly those with smaller variances in
sales (to the board) than those now being taxed, might
have reduced total (crop tax) revenue variance.

The next step was to take account of the insti-
tutional set up. The Northern States Marketing Board is
responsible for five major crops: groundnuts, cotton,
beniseed, soybeans, and palm produce. At the beginning
of each season, it decides how much to be paid to farmers--
how much tax to levy on the producers of these crops. The
board is perceived to have a desired level of total (crop
tax) revenue. When it decides on annual producer prices

for the various crOps, it is implicitly deciding the

.——._c
‘k v A
v

.

213

proportions in which the desired total (crop tax) revenue
will be shared by the crops. We then asked the question:
how should this desired total (crop tax) revenue be shared
between these crops so as to minimize the variance in crop
tax revenue subject to the condition that the desired total
(crop tax) revenue will indeed be raised or realized? E
The weights (shares) for the five crops under the Northern f
States Marketing Board (including here palm produce from :
Idah, Ankpa, and Dekina Divisions) were found to be e-
functions of the elements of the Jacobian matrix: the
variances and covariances of individual crop tax revenues
as well as the expected crop tax revenue from each crop,
and the desired total (crop tax) revenue from the five
crops.
The final section of this paper looked at vari-
ances and covariances of total State Government Revenues
with and without revenue contributions from the marketing
boards and allied statutory corporations. Our intuitive
expectations were borne out by the empirical results.
The Western State relying mainly on a perennial crop--
cocoa-~that is likely to be most subject to irreversi-
bilities in the underlying investment-disinvestment process
and also least subject to domestic demand fluctuations
experienced the least destabilization of total state
revenue through reliance on revenue contributions from the

marketing board and allied statutory corporations. The

214

Eastern States relying on palm produce picked mainly from
the wild trees--a crop least likely to be subject to
irreversibilities on the supply side, but having a large
and fluctuating domestic market--experienced the greatest
destabilization in total state revenue through reliance on
revenue contributions from the marketing board and allied
statutory corporations. The Northern States relying
mainly on annual crops--groundnuts, cotton, beniseed, and
soybeans-~occupied middle ground between the Western and

Eastern States.

7mm. .

FOOTNOTES

10. Teriba and O. Olakanpo, "Fiscal, Monetary and
Investment Implications of the Marketing Boards" (Inter-
national Conference on the Marketing Board System, NISER,
Ibadan, 1971), p. 10.

2 O O O O O
R, being a linear combination of random variables,
is also a random variable.

3S. S. Wilks, Mathematical Statistics (New York:
John Wiley and Sons, Inc., 1962).

4 . . .
There may be some reduction in tax revenue vari-
ance from reductions in some taxes.

5We can give meaning to the concept of an "average
tax revenue yield per crop," R. First, different states
have different numbers of marketing board crops and the
concept is useful for interstate comparisons of crop tax
yields. Second, an individual crop's tax revenue yield
can be compared with the average for all crops. Third,
the concept may be useful in decisions as to whether more
or fewer crops should be brought under marketing board
control (what does a marginal change in number of crops do
to average cr0p tax revenue yield?).

6R, being a linear combination of random variables
(the Rj's) is also a random variable.

7That increasing the number of crops tends to
reduce V(R) can be seen in the case of mutual zero corre-
lation from

- m
%_%i_1 = .53 2 02R < o (6')
m i=1 i

215

216

Even in this case, the contribution of the ith new crop
to V(R) which also depends on the variance of sales and
therefore of revenue from taxes on this crop is given by:

R.
1 0
m2

 

and given the present nature of demand and supply of
Nigerian agricultural crops, any new crop to be taxed is

likely to have large OZR- than those now being taxed.
i

8This is only an analytic result with no obvious
practical significance.

9The palm produce is mainly from Idah, Dekina,
and Ankpa Divisions of Kwara State.

10This becomes evident from the following variance-
covariance matrix of sales of cotton, groundnuts, beniseed,

and soybeans to the Northern States Marketing Board,
Nigeria, 1947/48-1969/70:

Groundnut Sales 66,805,459,l38
Seed Cotton

Sales 8,590,375,363 3,384,704,988
Beniseed Sales 1,091,769,796 91,429,158 59,513,615
Soybean Sales 239,623,846 22,762,873 21,791,670 47,608,633
Groundnut Seed Cotton Beniseed Soybean
Sales Sales Sales Sales
11

Friedman, in his analysis of countercyclical
policies has used this method. See M. Friedman, "The
Effects of a Full-Employment Policy on Economic Stability:
A Formal Analysis," in his Essays in Positive Economics
(Chicago: University of Chicago Press, 1953). Don
Patinkin has argued in the February 1969 issue of the
Journal of MoneyL_Credit and Banking that Friedman's 1956
essay, 1""'I‘rhe QuantityTheory--A Restatement" in M. Friedman,
ed., Studies in the uantity Theory of Money (Chicago:
University of Chicago Press, 1956), is mistitled, on the
grounds that Friedman's essay is an eloquent réiormulation
of Keynesian liquidity theory that borrows heavily from
the modern asset portfolio school headed by James Tobin.
Patinkin's argument is convincing. Similarly, the title
of Friedman's book, Essays in Positive Economics is mis-
leading in the sense that there is hardly any unified

(and meaningful) body of economics without substantial
normative content. For a modern development of these

 

 

217

themes, see Glenn L. Johnson and Lewis K. Zerby, 152
Plates of Spaghetti (East Lansing: Michigan State Uni-
versity); and Glenn L. Johnson, et al., A Generalized
SimulationjApproach to Agricultural Sector Analysis With
Special Reference to Nigeria (East Lansing: Michigan
State University, November, 1971), Chapter X.

 

12See West Africa, #2865 (Apapa: Times Press,
May 12, 1972), p. 587. (Emphasis in passage my own.)

13G. K. Helleiner, "The Fiscal Role of the Marketing
Boards in Nigerian Economic Development, 1947-61," in
C. K. Eicher and C. Leidholm, eds, Growth and Development
of the Nigerian Econom (East Lansing: Michigan State
UniverEity Press, 1970 , p. 129. (Emphasis in passage my
own.)

14A third possible condition is the degree to which
the marketing board is committed to the stabilization of
the producer price of this crop.

CHAPTER VII

SUMMARY AND CONCLUSIONS

This study had three main aims: first to provide
a framework for analyzing the effects of government and
marketing board taxation on cotton and groundnuts in the
Northern States on resource use in these crOps: second,
to examine the quantitative impact of paying groundnut and
cotton farmers higher prices on tax revenue from these
crops and third, to examine the extent to which these crop
tax revenues can be trusted to stabilize State Government
revenue for planning purposes.

It was shown (see Table 3.1, Chapter III) that
government and marketing board taxes on cotton and ground-
nuts introduced allocative distortions in the relative
quantities of these crops produced in the Northern States
during the period 1950-66, as farmers have substituted (in
production) along given transformation curves in response
to changing relative (crop) net producer prices induced by
differential tax rates. With differential taxation of
cotton and groundnuts detrimental to groundnuts, the

relative quantity of cotton was higher than would have

218

219

been the case in the absence of all taxation, wherever
cotton and groundnuts were substitutes in production.
Implicit in the model is the fact that food crops,
relative to cotton and groundnuts, were produced in
quantities larger than would have been the case in the
absence of all taxation since food crops bore zero rates
of taxation and cotton and groundnuts bore (and still bear)
positive rates of taxation most of the period examined,
wherever food crops and these marketing board crops were
substitutes in production. Taxes on marketing board crops
have distorted patterns of specialization in Nigerian
farming, either in terms of the production patterns for
food and marketing board crops on the one hand or in terms
of the production patterns of marketing board crops
bearing different rates of taxation.1

A model was developed in Chapter III to explain
the effects of these taxes on the scale of farming in the
Northern States (1950-66), measured either in terms of
acreage or labor. When the model was extended to account
for the differential between off-farm acquisition and
salvage values, the investment-disinvestment responses to
changes in taxes and subsidies differed, in many cases,
from those predicted by the neoclassiCal model that
assumes equality between acquisition and salvage values
for durable inputs.

The empirical evidence on the effects of these

taxes on resource use provided in Chapter III is revealing.

220

It was shown that cotton and groundnuts farms (acreage) in
the Northern States are smaller than they would have been
in the absence of taxation of marketing board crops. At

a time when state governors and other policy makers are
very concerned about the diversion of acreages away from
marketing board crops to other crops, this is significant
evidence indeed. As a result of these taxes, farmers have
not found it profitable to expand acreages under cotton
and groundnuts as fast as they would have done in the
absence of these taxes. Strong logical reasons were also
provided, based on the levels of taxes on cotton and
groundnuts (1950-66), to show that the amounts of labor
(family labor and hired labor) employed on Northern States
cotton and groundnut farms during the period 1950-66 were
lower than would have been the case in the absence of
taxes on these crops. Not only were the stocks of these
resources employed on these farms less than they would
have been, but they were also worked less intensively than
would have been the case in the absence of taxes on these
crops. These taxes thus directly affect resource use as
these relate to the underemployment and unemployment of
human resources in rural areas. Marginal farms have been
forced to close as a result of these taxes; people
released from these marginal farms are induced to migrate
to the towns by higher expected present value of life

time earnings in the urban areas like metropolitan Kano,

Zaria, Kaduna, Sokoto, and Lagos.

221

These adverse effects on resource use operate at
two levels: first, taxes reduce acquisitions of labor and
land below what they would have been in the absence of
taxes and thus reduce the rate of entry of new resources
into the farm sector and second, taxes induce salvaging
and disinvestment of used resources on hand and thus
increase the rate of exit of resources from the farm
sector.

These adverse effects implied by our analysis can
be inferred from available empirical evidence for three
reasons. First, even if resources are fixed with respect
to off-farm acquisition and salvage values, they are
allocated among competing crops according to their on-farm
opportunity costs. Farmers are known to be reallocating
these resources in response to the effects of taxes on
these on-farm opportunity costs.2 Second, these resources
are heterogeneous. Older farmers in response to taxes on
marketing board crops may simply reallocate their skills
between these crops and non-marketing board crops on the
basis of their on-farm opportunity cost while younger
farmers and rural farm youth may quit farming altogether
and move to the urban centers3 when their off-farm market
salvage values exceed their on-farm opportunity costs.
Third, farms of different sizes are differently endowed
with resources and have different initial quantities on

hand.

222

With less resources being used as a result of
these taxes, less cotton and groundnuts were produced (and
presumably less was therefore sold to the marketing
board) relative to the (profit maximizing) quantities of
these crops that would have been produced in the absence
of any taxes on them. This has obvious implications for
those states that have production campaigns for marketing
board crops in their development plans. Suppose a state
wishes to increase its production (output) of a marketing
board crop (e.g., cotton, groundnuts) by a given annual
percentage for each year of the development plan. Starting
with a production function, we can express the percentage
increase in the output of the marketing board crOp as the
weighted sum of the percentage increases in the resources
used to produce this crop, where the weights are the output
elasticities of these resources.4 For given output
elasticities of these resources, the annual percentage
increase in the output of the marketing board crop is
higher, the higher the percentage increases in the
quantities of resources used to produce this crop. To the
extent that taxes on marketing board crOps have impeded
annual percentage increases in resources used in the
production of marketing board crops, they have consequently
impeded the rate of growth of output of these crops. The
evidence in Chapter III implies that this must have been

the case (see pp. 58-59.

223

The last part of Chapter III was of the nature of
providing policy makers with a policy variable: given
that labor and land produce the bulk of value added in
Nigerian agriculture and if we are willing to assume the
institutional reality of the marketing boards, then, if
State planners wish to minimize the allocative distortions
with respect to the use of land and labor introduced by
these taxes, what are the compensating subsidies on
fertilizers, chemicals, land, labor, etc.? It was shown
that these compensating subsidies are related to the out-
put elasticities of these resources. Using a Cobb-Douglas
production function and the available estimates on the
output elasticity of land, it was calculated that to keep
the acreage in groundnuts production from falling, given
a l per cent decrease in the net producer price of ground-
nuts received by farmers, then, the net acquisition price
of land that groundnuts farmers have to pay has to fall by
2.931? per cent. In cotton production, the corresponding
compensating subsidy is 1.8860 per cent. It was also
calculated that to keep the quantity of labor employed in
groundnuts production from falling, given a l per cent
decrease in the net producer price of groundnuts received
by farmers, then the net acquisition price of land that
groundnut farmers have to pay has to fall by 2.746 per
cent. In cotton production, the corresponding compensating

subsidy is 2.422 per cent. In this way, programs for

224

subsidies on resources used in Northern States farming in
the future can be rationally related to expected tax rates
on the crops in which these resources are employed.

If resources are fixed where their MVP's are
between their off-farm market acquisition and salvage
values, small taxes have no effect on resource employment
and thus would require no compensating subsidies. Large
taxes will lead to disinvestment which may be necessary
until resources are being employed at levels where
acquisition costs are being covered.

The link between the response of tax revenue from
cotton and groundnuts to producer price changes and the
response of sales of these crOps to the board to changes in
the net producer price was established in Chapter IV. The
more reSponsive sales of a crop (to the board) are to
changes in its price, the more responsive (in absolute
value) is tax revenue to these producer price changes
(see pp. 105-113). The illustrative calculations in
Chapter IV using actual data on taxes and net producer
prices for the period 1950-66 bring out clearly the link
between the elasticity of (crop tax) revenue with respect
to producer prices paid out by the board and the elasticity
of sales (to the board) with respect to the board's net
producer prices: the more elastic sales to the board are
of a crop with respect to its net producer price, the more
elastic (in absolute value) is the tax revenue from this

crOp with respect to the board's net producer price for

225

the crop (see Tables 4.1-4.4). Attempts to raise more
(tax) revenue from a crop by paying producers lower prices
might become self-defeating if sales of this crOp to the
board are very price responsive or even moderately price
responsive. In the last section of Chapter IV, it was
argued that when long run influences are considered,
questions about the (tax) revenue potential of marketing
board crops and the farm sector as a whole depend impor-
tantly on the crop price response of the underlying
investment-disinvestment process.

In the first section of Chapter V, we sketched a
simultaneous equation system in which the sales supply
equation was one of four structural equations. Previous
econometric treatments of sales (to the board) of marketing
board crops have not provided any system structure of which
the sales supply equation was only one component. The
system incorporates our apriori knowledge of the factors
influencing the sale of a crop to the board. Identi-
fiability of the sales supply relation was then investigated
after which reduced form solutions for the endogenous
variables of the system were derived. Availability of
data would then have permitted two-stage least squares
estimation of the (structural) sales supply equation.

A modified model which removes obvious sources of
biasedness and inconsistency from our ordinary least

squares estimates of the parameters of single sales supply

226

equations for groundnuts and cotton was then estimated.
The elasticity of sales of groundnuts to the Northern
States Marketing Board with respect to the board's
producer price was estimated to be 1.3139 and that of
cotton to be .90278.

It was then postulated that farmers were to be
paid (on the average) higher prices than they actually
received (1950-67). Predictions of sales were made, on
the basis of the estimated equations, using the postulated
hypothetical prices. It was found for both cotton and
groundnuts that when the postulated prices were higher
than actual prices the board paid the farmers, predicted
sales were higher than actual sales. The predicted sales
incorporating the "price effect" on sales and the implied
lower taxes were then used to show the quantitative impact
of (generally) higher producer prices on tax revenue from
these two crops. It was shown that in many years with the
new higher producer prices, farmers would have gained
substantially in monetary income without the government
and marketing board being significantly worse off in
revenue terms (where "significantly worse off" is defined
in terms of net (tax) revenue loss relative to new total
tax revenue). For cotton, with the postulated new higher
prices in 1951, farmers would have gained LN l,0?l,7l7.6
in monetary income; total (tax) revenue incorporating the

positive effect of higher prices on cotton sales would

227

have been LN 3,248,801.3.5

The net loss in (crop tax)
revenue would have been LN 463,684.4 (see Table 5.12), thus
leaving the government and marketing board a positive net
(crop tax) revenue balance of LN 2,785,116.9. Thus, total
net revenue would not have been significantly affected.6
Further, when this is compared with the total tax with—
drawals of LN 1,742,2007 from cotton in 1951 using actual
producer prices and actual sales that were both lower than
the hypothetical price and the predicted sales respectively
that were used in the above computations, it is evident
that the government and the marketing board were not
significantly worse off revenue-wise paying farmers higher
prices than the prices actually paid. In 1957, with
farmers receiving (the new) higher prices, they would have
gained LN 460,930.l in monetary income while the government
and the marketing board would have lost LN 712,147.l (see
Table 5.12). Total tax revenue from cotton in this year
after incorporating the positive effect of higher prices
on sales was calculated to be LN l,034,927.5,8 thus leaving
a net (crop tax) revenue of LN 322,780.4. Relative to the
actual total tax revenue of LN 77,6009 using actual prices
and actual sales, the government and marketing board were
better off, revenue-wise.10
For groundnuts in 1954, with the postulated new

higher prices, farmers would have gained LN 2,683,987.2 in

monetary income. The government and the Northern States

228

Marketing Board would have suffered a net revenue loss of

LN 988.142.0 (see Table 5.8): in the same year incorporating
the positive effect of higher prices on sales, total tax
revenue from groundnuts was calculated to be LN 5,125,384.4,
thus leaving a net tax withdrawal from groundnuts of

11

LN 4,337,242.4. The net loss in total (crop tax) revenue

12 In 1950 and 1951, the net

is not very significant.
losses in (crop tax) revenue were significant. For the
other years for both groundnuts and cotton, there are
varying degrees of offsets of revenue loss by the positive
effect of higher prices on sales. On the average, the net
revenue losses do not appear to be very substantial,
relative to the new total tax revenues and the gain in
monetary income to farmers.

Any assessment of the total effects of taxation of
marketing board crops must take into account the effects
on production (output) of the crop and the effects on its
sales to the board. Chapter III provided indirect
evidence on production by examining the effects of these
taxes on resource use. To the extent that resources were
being employed on a smaller scale than would have been the
case in the absence of these taxes, then production (out-
put) was less than would have been the case in the absence
of taxes on these crops. The estimates and predictions of
Chapter V showed that sales of crops to the board were

lower than would have been the case if the board paid

229

farmers higher prices. We were not able to take the
production effect of these taxes into account directly
because of lack of data on total output of any of the crops
treated. This production effect of these taxes would then
have been incorporated in our analysis of the impact on
(crop tax) revenue of paying farmers higher (hypothetical)
prices. It is hoped that this shortcoming will be remedied
as data on total output of these crops become available.
The CSNRD series were a landmark in studying the
effects of marketing board taxation of Nigerian agriculture.
We have extended these studies by explicitly specifying,
identifying and estimating the sales supply relation for
marketing board crops as a basis for estimating the
quantitative impact on tax revenue from these crops of
paying farmers higher prices and also by explicitly
treating the resource employment effects of these taxes.
The major justification for the past and present
roles of marketing boards has been their potential ability
to stabilize producer prices and producer incomes. Their
success in this regard has not concerned us in this study.
Their potential for stabilizing State Government revenues
is the newly emerging rationale for not only retaining
these boards but also for widening their taxing powers.
Treating sales to the board of any cr0p as a random vari-
able, it is shown in Chapter VI that the (State Government)

revenue stabilizing potential of marketing boards and

230

allied statutory corporations is very small. In the case
of total revenues from taxes on all crops within marketing
board control, it is shown that the variance of total
revenue depends largely on the elements of the covariance
matrix of sales to the board of these crops. Arguments
and empirical evidence are used to show that not only are
the diagonal elements of this covariance matrix large, but
that the off-diagonal elements are likely to be positive.
In this case, the larger the number of crOps under the
board's control, the larger the variance in tax revenues.
Using crops under the control of the Northern States
Marketing Board as an example, it was found that the
correlation coefficient between groundnut sales and cotton
sales to the board over the period 1947-1970 was .5713,
between grounduts sales and beniseed sales .5475, between
groundnut sales and soybeans sales .1344, between seed
cotton sales and beniseed sales .2037, between beniseed
sales and soybeans sales .4094 and the least correlation
coefficient (still positive!) was that between seed cotton
sales and soybeans sales, .0567. The variances of sales
of groundnuts, cotton, beniseed, and soybeans were
respectively: 166,805,459,138.30, 3,384,704,988.6l,
$9,513,615.71, and 47,608,633.87. Thus, the variances of
sales to the board are large; and their covariances are
also large and positive. Thus, the potential for stabi-
lizing total revenue by increasing the taxing powers of

the boards is small for the present composition of

231

marketing board crops in the Northern States. Most of the
variation in tax revenues from these crops is likely to
come not from variations in tax rates on these crops but
from variations in sales of these crOps to the board.
Using sales as a proxy for revenue, we found that the
average covariance of sales for the four marketing board
crops of the Northern States (1947-40/1969-70) was
1,676,292,118.28. If the average covariance of (tax)
revenues from these crops is some (nondecreasing) linear
function of this, then prospects for stabilizing State
Government revenues from reliance on marketing boards and
allied statutory corporations are very small indeed.

Using actual budget figures for the Northern States,
the Eastern States and the old Western Region, it was also
shown that reliance on revenue contributions from marketing
boards and allied statutory corporations destabilized State
Government revenues. It was argued that such reliance was
potentially destabilizing when, first, the crop has a
large (and fluctuating) domestic market and/or markets in
neighboring countries and second, when the crOp does not
have significant irreversibilities in the underlying

13 The old Western

investment-disinvestment process.
Region, relying on revenue contributions from taxes on
cocoa--a perennial crop--had the least potential for
destabilization (the most potential for stabilization) of

State Government revenue. The correlation coefficient

between State Government revenues in the absence of

232

contributions from the marketing board and allied statutory
corporations and revenue contributions from them for the
period 1956-66 (i.e., PRZ) was .0171. The Eastern States
relying on revenue contributions from taxes mainly on palm
oil and palm kernels had the most potential for destabili-
zation (the least potential for stabilization) of State
Government revenue. The corresponding correlation coef-
ficient (PR2) for the Eastern States for the period 1956-65
was calculated to be .7049. The Northern States, relying
mainly on revenue contributions from taxes on cotton, soy-
beans, groundnuts, and beniseeds--all annual crops--occupied
the middle ground between the Eastern States and the old
Western Region. Such reliance had high potential for
destabilization (low potential for stabilization) of State
Government revenues. The corresponding correlation coef-
ficient (PRZ) for-the Northern States for the Period

1956-64 was calculated to be .6286.14

Implications of Findings

For the past two decades, some form of price or
income stabilization role for producers has been used to
justify the existence of marketing boards. There is still
no conclusive evidence that marketing boards have succeeded
in this regard. A new emerging role is that of state
revenue stabilization. Our analysis in Chapter VI indi—
cated that the potential of marketing boards and allied

statutory corporations as stabilizers of State Government

233

revenues is almost nil. Unless a new cogent reason is
found, we find it difficult to justify the present marketing
structure.15

Unemployment is currently a major problem in
Nigeria. It was shown in Chapter III that taxes on
marketing board crops have distorted the relationships
between on-farm and off-farm opportunity costs of labor,
and, by implication, that the ratio of the present values
of the latter to the former is higher than would have been
the case in the absence of these taxes. This has induced
the drift of people from the rural to the urban areas.
Any attempt to arrest this drift must seriously consider

16 The high social costs

the elimination of these taxes.
of artificially induced unemployment of people in the urban
areas can then be minimized or avoided.

The existence of these taxes is one major reason
why cotton and groundnut farms are small. It is possible
for most farmers in cotton and groundnut production to
increase their acreages in these crops if it is profitable.
In the Report of the Study Group on Groundnuts it is
stated: "In some parts of the North, particularly in the
North Eastern State, substantial areas of land are unused
at present and could be available for expansion of
groundnuts."17

If the increased acreages are to be forthcoming,

these taxes must be drastically reduced or eliminated.

234

A current problem is the transformation of Northern
States agriculture from one relying mainly on traditional
inputs to one that employs new and superior inputs. This

18 The demand

transformation is proceding rather slowly.
for new inputs has been adversely affected by taxes on
marketing board crops. A minimum requirement in the
presence of these taxes is the urgent need for compensating
subsidies on resources to minimize allocative distortions
introduced by these taxes. It was shown in Chapter III

that these compensating subsidies are related to the output
elasticities of these resources used in production. In

this way, programs in State Development Plans and production
campaign projects for subsidizing chemicals, fertilizers,
insecticides, land, skilled labor, etc., can rationally be
related to expected taxes on the crOps in which these
resources are used.

We do expect structural changes like improved
transportation network and improved marketing information
about crop prices in different parts of the country.

These changes will raise the elasticity of sales (to the
board) of a marketing board crop with respect to the
board's net producer price for the crop. Such changes may

indeed render policies that seek to raise (tax) revenue by

paying farmers lower prices self-defeating.

FOOTNOTES

1See pp. 26-27 for evidence and discussion on the
policy of self-sufficiency in cotton being pursued by the
Western State.

2From our apriori knowledge of the ease with which
labor and land are shifted between crops in the cotton-
groundnuts-foods economy of the Northern States, we infer
that the allocation of fixed resources must be on the basis
of their on-farm opportunity costs. Let us see the effects
of taxes on this opportunity costs. Suppose labor of a
given age and skill is fixed in the sense that it does not
pay to invest or disinvest in it. Let this fixed labor
(L ) be allocated between groundnuts (G) and food pro-
duction (F). The farmer's production function is, in
implicit form,

f (L, G, F) = 0 (1)'

which can be written explicitly as

L0

= f(G, F)

Let the farmer then maximize his revenue from food and
groundnut production subject to the constraint that he has
this fixed amount of labor to allocate between them. Form
the Lagrangian.

— o-
L — PFF + (uGPG)G + alL f(G, F)]
where PF and PG are food and groundnut prices respectively
and “G is the proportion of potential producer price of
groundnuts paid by the board to farmers. The necessary
conditions for equilibrium are

a:._ _ af_
fi-% “fi-0

3L at
$3

235

 

 

236

EL _ o

from which
a _ Pf _ uGPG
“W“?
6'5

from which it follows that a = PF %% = “GPG %%

where 3F and as are the marginal products of labor in food

and groundnut production respectively and a, the unde-
termined Lagrangian multiplier, turns out to be the on-
farm opportunity cost of labor. Without any taxes, pg = 1.
With an increase in taxes on groundnuts, o < “G < l, and
there is induced reallocation of labor away from groundnut
to food production until, in equilibrium, the opportunity
cost of labor in food and groundnut production is the same.
Fixity of a resource does not therefore preclude these
allocative responses in resource use to taxes on marketing
board crops.

3In a June issue of West Africa, it is stated: "A
total of 72,000 tons of grouhdnuts were produced in the
North Eastern State in 1970-71 compared with 204,000 tons
the previous season. Cotton production fell from 86,000
tons to 40, 000 tons. The State Commissioner for Agri-
culture and Cooperatives, Alhaji Muhammadu Mai, attributed
the decline to the drift of farmers to the towns. . .

See West Africa, #2871 (Apapa: Times Press, June 23, 1972).
CompIaints about the "drift of farmers" to the towns are
very frequent these days among state governors. It is
evident that taxes on marketing board crops have distorted
the relative magnitudes of the mean present values of
marginal value products of these farmers on Nigerian farms
and the expected present values of their off-farm oppor-
tunity costs in the urban areas. It is perfectly rational
for these farmers to migrate in the face of these taxes on
marketing board crops.

 

4Writing the production function as:

Yj = f(Xl' eeee' Xn)’

it is evident that

AY. n AX.

__1 z e __1

Y O I Y O X I x
3 i=1 j i i

237

where Y is the quantity of the jth marketing board crop
produced with the resources X1, ..., Xn, - is the
elasticity of output of the jth crop with Eraspect to the
ith resource, AYj is the percentage change in Yj and AXi

Yj Xi
is the percentage change in xi! (1 = l, ..., n).

5For the method used to calculate total tax revenue
incorporating the positive effect of higher prices on
sales, see footnote

6Net (tax) revenue loss is 14 per cent of the new
total (tax) revenue.

7For this figure on total tax revenue using actual
producer prices and actual sales, see G. K. Helleiner,
Peasant Agriculture, Government and Economic Growth in
Nigeria (Irwin, 1966), Table V-F- 5.

8See footnote 38.

9For this figure, see Helleiner, o . cit.

10However, with net (tax) revenue loss 68 per cent
of the new total tax revenue, government and marketing
board tax revenue is significantly affected.

11For method of calculating this figure, see p. 189.

12The net loss in tax revenue is 19 per cent of new
total (crop tax) revenue.

13See discussion on p. 221, especially footnote 2.

14These relative performances of the states are

consistent with our apriori expectations. Cocoa is a
perennial crop and therefore most likely to have sub-
stantial irreversibilities in the investment—disinvestment
process underlying its production; it also has very little
domestic demand in the form sold to the board. The bulk
of the palm oil and kernels sold to the board comes from
wild palm trees, thus making it possible to adjust rapidly
to price changes. Palm oil has a large (and fluctuating)
domestic market and neighboring country market demand.
Cotton and groundnuts occupy middle ground between cocoa
and palms both in terms of expected irreversibilities

238

underlying their production as well as the size of the
domestic and neighboring country market conditions.

15It might be tempting to play down on the need
for the extended discussion here given to this issue, on
the grounds that petroleum revenues will do the job of
stabilization. The present value of petroleum revenues is
certainly bounded by the nonreplenishable nature of the
resource. Even in the face of the present petroleum
revenue boom, not all revenue needs of all states can be
met. State Governments continue to seek sources of revenue
independent of the Federal Government. Marketing boards
are a good source of revenue and given the exposure of the
myth that these boards are there to stabilize prices and
incomes and help farmers, marketing board advocates are
searching for new justifications for the existence of
marketing boards. We must avoid replacing old myths with
new ones. It should also be recognized that we have not
examined other sources of State Government revenues in
terms of "stability." Stability might not be important for
some purposes, but is vital for state budgetary operations
and is important as a potential justification for the
existence of the marketing boards.

16The Governor of Lagos State, Colonel M. Johnson
recently called on other State Governments to help him
stem the outflow of people from their states into his own
state. The Governor will go a long way toward solving
this problem if he seriously encourages these State Govern-
ments to abolish the present marketing board system. See
West Africa (Apapa: Times Press, May-June, 1972).

17See Report of the Study Group on Groundnuts
(Lagos: Federal Department of’Agriculture, 1971), p. 13.

18Ibid.

BIBLIOGRAPHY

BIBLIOGRAPHY

P. T. West African Trade. London: Cambridge

University Press, 1954.

Economic Analysisgnd Policy in Underdeveloped
Cambridge University Press,

Bauer,

Countries. London:

 

Markets, Market Control and

, and B. S. Yamey. __
Weidenfeld and Nicolson, 1968.

MarketinggReform.

Supplngesponse in Underdeveloped Agriculture.
North-Hoiiand Publishing Company, 1968.

 

Behrman, J.
Amsterdam:

Bishop, C. "The Effects of Specific and Ad Valorem Taxes."
Quarterly_Journal of Economics (1968).

Cochrane, D., and G. H. Orcutt. "Applications of Least
Squares Regressions to Relationships Containing

Autocorrelated Error Terms." Journal of the
American Statistical Association (March, 1949).

Statistics and Econometrics. New York:

Frank, C. R.
Inc., 1971.

Holt, Rinehart and Winston,

"The Effects of a Full-Employment Policy on

A Formal Analysis." In
Chicago: University

Friedman, M.
Economic Stability:

Essays in Positive Economics.
of Chicago Press, 1953.

E. "Optimal Farm Size." American Journal of Agri-

Heady,
cultural Economics (February, 1971).

Peasant Agriculture, Government and

Helleiner, G. K.
Richard Irwin, Inc.,

Economic Growth in Nigeria.

T966.

. "The Fiscal Role of the Marketing Boards in

Nigerian Economic Development," In C. K. Eicher
and C. Liedholm, eds., Growth and Development of

the Nigerian Economy. East Lansing: Michigan
State University Press, 1970.

 

239

240

Johnson, Glenn L. "The State of Agricultural Supply
Analysis." Journal of Farm Economics (May, 1960).

. "Supply Function--Some Facts and Notions," In
E. O. Heady, H. G. Diesslin, H. R. Jensen, and

G. L. Johnson, Agricultural Ad'ustment Problems
in a Growing Economy. Ames: Iowa State CoIlege
Press, 1958.

 

 

. "Capital," In International Encyclopeida of the
Social Sciencee. The MacMilian Company aha the
Free Press, 1968. ' '

 

. "Factor Markets and the Problem of Economic
Development." In W. W. McPherson, ed., Economic

Develo ment of Tro ical Agriculture. University
of FIoriaa Press, 1968.

. "Review of Nerlove." Agricultural Economic
Research (January, 1960).

 

 

 

 

, and Leroy Quance. The Overproduction Trep in
American A riculture. The Johns Hopkins University
Press, 19 2.

 

, O. J. Scoville, G. K. Dike, and C. K. Eicher.
Strategies and Recommendations for Ni erian Rural
Development, 1969-1985. Consortium or the Study
of Nigerian Rural Development (CSNRD) 33, Michigan
State University, East Lansing, 1969.

 

, and L. K. Zerby. What Economists Do About Values:
Case Studies of and Answers to Questions They Don't
Dare Ask. East Lansing: Michigan State UhiverSity,
1971.

 

 

, et a1. Managerial Processes ofMidwestern
Farmers. Iowa State University Press, 1961.

 

, et al. A Generalized Simulation Approach to
Agricultural Sector AnalySis With Special Appli-
cation to Nigeria. East Lansing: MiEhigan State
University, 1971.

 

 

Johnson, H. G. "Towards a Generalized Capital Accumulation
Approach to Economic Development," In M. Blaug,
Economics of Education 1. Penguin Books, 1968.

Johnston, J. Econometric Methods. New York: McGraw Hill
Book Company, 1960.

241

Kmenta, Jan. Elements of Ecopometrics. New York:
MacMillan Company,_197l.

Kriesel, H. The Marketing of Groundnuts in Nigeria.
East Lansing: CSNRD l9,Michigan State University,
1968.

. Cotton Marketing_in Nigeria. East Lansing:
CSNRD 24, Michigan State University, 1968.

 

Laurent, C. K. Problems Facing the Fertilizer Distribution
Program in the Six Northern States. East Lansing:
CSNRD 27, Michigan State UniVersity, 1969.

Malinvaud, E. Statistical Methods of Econometrics. 2nd
revised ed. Amsterdam-London: North-Holland
Publishing Co., 1970.

Musgrave, R. The Theor of Public Finance. New York:
McGraw Hill, 1 59.

Nigeria, Federal Ministry of Information. Second National
Development Plan 1970-74. Lagos: 1970.

, Federal Department of Agriculture. Report of the
Studngroup on Cotton and Other Fibres. Lagos:
1971.

 

 

. Report of the Study Group on Groundnuts. 1971.

. Report of the Study Group on Oil Palms. 1971.

 

, Federal Office of Statistics. Annual Abstract
of Statistics (various issues); Digest of_Sta-
tistics (variOus issues).

 

 

Norman, D. W. An Economic Study of Three Vilieges in Zaria
Province. Vol. 1, Part II. Zaria: Samaru Miscel-
laneous Paper, Ahmadu Bello University, 1970.

 

. "Initiating Change in Traditional Agriculture."
Zaria: R.E.R.U., Ahmadu Bello University, 1970.

 

Northern Nigeria, Ministry of Economic Planning, Kaduna.
Northern Nigeria Statistical Yearbook 1966.

Olatunbosun, D., and S. O. Olayide. "The Effects of the
Marketing Boards on the Output and Income of
Primary Producers." Ibadan: NISER, 1971.

242

Patinkin, Don. "The Chicago Tradition, the Quantity Theory
and Friedman." Journal of Money, Credit and
Banking (February, 1969i.

Samuelson, P. Foundations of Economic Analysis. New York:
Atheneum, 1965.

Schultz, T. W. Transforming Traditional Agriculture. Yale
University Press, 1964.

Standard Bank Ltd., Nigeria. Annual Economic Review
(various issues).

 

Stigler, G. "Notes on Duopoly." Journal of Political
Economy (1940).
Teriba, O., and O. Olakanpo. "Fiscal, Monetary and

Investment Implications of the Marketing Boards."
Ibadan: NISER, 1971.

Tintner, G. Econometrics. New York: John Wiley and Sons,
Inc., 1952.

Titiloye, M. O., and A. A. Ismail. "A Survey of the Trends
and Problems in the Domestic Arrangements for the
Marketing of Groundnuts and Cotton." Ibadan:
NISER, 1971.

Ulveling, E. "Comment." American Journal of Agricultural
Economics (February, 1972).

West Africa. Apapa: Times Press.
Western Nigeria, Ministry of Economic Planning and

Reconstruction. Western State Development Plan,
1970-74. Ibadan: I971.

 

Wilks, S. Mathematical Statistics. New York: John Wiley
and Sons, Inc., 196 .

"711111111171111.1111711“