INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send U M I a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.

University Microfilms International
A Bell & Howell Information C om pany
300 North Z eeb Road. Ann Arbor. Ml 48106-1346 USA
313/761-4700 800/521-0600

O rder N u m b er 9 4 3 1 2 5 0

F arm land leasin g and con tra ct choice in M ichigan: T h e influence
o f social d istan ce
Gwilliam, K ent R alph, Ph.D .
Michigan State University, 1993

UMI

300 N. Zeeb Rd.
Ann Arbor, MI 48106

FARMLAND LEASING AND CONTRACT CHOICE IN MICHIGAN
THE INFLUENCE OF SOCIAL DISTANCE
By
Kent R. Gwilliam

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Agricultural Economics
1993

ABSTRACT
FARMLAND LEASING AND CONTRACT CHOICE IN MICHIGAN:
THE INFLUENCE OF SOCIAL DISTANCE
By
Kent R. Gwilliam
Choice of the share crop (as opposed to cash rent)

form

of contracting in agricultural leasing has been the subject of
debate for years. Traditional economic theory has held that
the practice is (in theory) fundamentally flawed and leads to
inefficiency
Scholars

and

have

inappropriate

remained

puzzled

allocation
by

the

of

resources.

persistence

practice and have made numerous attempts

of

the

to produce models

that would explain its popularity.
This study explores the influence of social relationships
(termed social

closeness)

model

demonstrates

created

on the choice
that

of contract.

individuals

with

The
social

closeness will choose the share crop form of contract.

The

benefits of this type of agreement are cost savings and income
enhancement.
Data to test the hypotheses was collected by conducting
a mail survey of landlords and tenants in Michigan.

Testing

was done using a logit regression model.
It was

found

that

choice

of contract.

showed

evidence

that

social

Responses

relationships
of

interaction

landlords
on

a

influence

the

in particular

social

basis

and

interpersonal relationship were correlated with the choice of
the share crop form of contract.

In the case of tenants, the

Kent Ralph Gwilliam

attitude,
contribute

experience
more

than

and

willingness

just

land

to

of
the

the

landlord

agreement

correlated to the choice of the share agreement.

to

were

ACKNOWLEDGEMENTS
Sincere thanks
patience,

is

extended to Dr.

understanding

research project.

and

A Allan Schmid

encouragement

In particular,

throughout

for
the

gratitude is expressed for

willingness to listen and provide timely advice.
Dr. Ralph Hepp's faith and confidence in my ability to
complete

this

especially
resources

at

work

provided

times

when

to complete

the

a

needed

the vision

steadying

became

survey through

dim.

Dr.

influence
Financial

Miron

Kelsey

couldn't have come at a better time.
Appreciation is also extended to Michigan Agricultural
Statistics Service for extraction of the sample.
My tenure as a graduate

student received considerable

support from Michigan Agriculture Experiment Station funding
for which I am very grateful.
My deepest and most sincere appreciation goes to my wife
who patiently bore the enormous burden of primary breadwinner
while continuing as mother, wife, and homemaker.

I thank my

children for understanding through missed school plays, ball
games,

swimming events,

family has been stellar.

and so forth. The support of all my

TABLE OF CONTENTS
LIST OF F I G U R E S ........................................... viii
LIST OF T A B L E S ............................................. ix
CHAPTER I
....................
INTRODUCTION and PROBLEM STATEMENT
1.1 Introduction ...................................
1.2 Problem statement
............................

1
1
4

CHAPTER II
REVIEW OF LITERATURE
...................................
2 .1 I n t r o d u c t i o n ...................................
2.2 Historical setting and emergence of leasing
.
2.3 Areas ofF o c u s .................................
2.4 Terms and D e f i n i t i o n s .........................
2.5 Purpose: Why R e n t ? ............................
2.5.1 For the l a n d l o r d ......................
2.5.2 For the t e n a n t ........................
2.6 Contract C h o i c e ...............................
2.6.1 Early M o d e l s ..........................
2.7 The Traditional M o d e l s ........................
2.7.1 The efficiency q u e s t i o n ...............
2.7.2 Advalorem tax or share model
. . . . .
2.7.2 Lump sum m o d e l ........................
2.8 Problems with the Traditional M o d e l s .........
2.9 Elements that remain unexplained by the tax
model and its variations......................
2.10 Proposed Answers to the Theoretical Paradox
and Contract C h o i c e ..........................
........................
2.10.1 Input Sharing
2.10.2 Empirical and Practical Questions
. .
2.10.3 The Equal Efficiency Models
.........
2.10.4 Cheung Model ..........................
2.10.5 The Screening Approach ...............
2.10.6 Agricultural Ladder
.................
2.10.6 Linkage
......................
2.10.8 R i s k ...................................
2.10.9 Transaction Costs
....................
2.10.10 Service Extraction
. ...............
2.10.11 Exploitation of the l a n d ...........
2.10.12 Social Distance ......................
2.11 S u m m a r y .......................................

22
22
25
27
27
28
30
31
31
34
37
41
42
46

CHAPTER III
MODELING THE LANDLORD-TENANT RELATIONSHIP ............
3.1 I n t r o d u c t i o n ...............
3.2 The m o d e l .....................................
3.2.1 The Cash Rent A g r e e m e n t ..........

47
47
49
50

v

7
7
7
9
10
13
13
13
14
14
14
15
15
17
18
21

3.3
3.4
3.5

3.6

3.2.2 Summary of the Cash Lease Agreement . .
52
3.2.3 The share a g r e e m e n t ..................... . 52
Incorporating risk into the m o d e l ...........
54
The effects of social c a p i t a l ...............
57
3.4.1
Features
of
the
social
closeness
58
f a c t o r ...................................
H y p o t h e s e s .....................................
69
3.5.1 The
effect of risk aversion . . . . . .
69
3.5.2 The
effects of social closeness . . . .
71
3.5.6 The
effects of transaction costs
...
74
S u m m a r y .......................................
76

CHAPTER IV
78
DATA COLLECTION AND A N A L Y S I S ..........................
78
4.1 I n t r o d u c t i o n ...................................
78
4.2 Obtaining a s a m p l e ............................
4.2.1 Sample s i z e ................... .
79
4.2.2 Drawing the s a m p l e ....................
80
4.2.3 Questionnaires sent, received, different
........................
81
mailings, etc
4.2.4 Testing for non response b i a s .........
82
4.3 General Respondent information ...............
86
4.3.1 Distribution of respondents ...........
86
4.3.2
Number,
size,
and
distribution
of
parcels of land being r e p o r t e d .........
89
4.3.3 Length of L e a s e ........................
90
4.3.4 Formality of A g r e e m e n t ...............
90
4.4 Cash Rent A g r e e m e n t s ..........................
91
92
4.5 Crops grown share, cash ............. . . . . .
4.6 A d j u s t m e n t s ...................................
93
4.7 Share Rent A g r e e m e n t s ........................
94
4.7 Landlord participation in inputs .............
95
4.8 Changes from cash to share and visa versa
. .
97
4.9 Specifics of landlords and tenants . . . . . .
98
4.9.1 C o n t r a c t i n g ...............
99
4.9.2
Hypothetical
contract
choice
with
parties of varying degrees of closeness
and c o n f i d e n c e ............................. 100
4.9.3 Landlord qualification and participation
in the a g r e e m e n t ...........................106
4.10
General
rental
market
and
respondent
i n f o r m a t i o n ..................................... 108
4.10.1 Use, maintenance and care of assets
. 108
4.10.2 Improvements . . . .
.................
110
4.10.3 Self evaluation of risk preferences and
contract choice
............. . . . . .
110
4.10.4 Interaction and contract choice
. . . 112
4.10.5 Current relationship,
changes,
and
contract c h o i c e .................
114
4.10.6
Social
Interaction
and
Contract
C h o i c e ......................................114
vi

4.11

Testing of factors associated with contract
c h o i c e ..........................................
4.11.1 Landlord choice of contract
.........
4.11.2 Landlord logit regression results
. .
4.11.3 Tenant Choice of Contract
...........
4.11.4 Tenant regression results
...........
4.12 S u m m a r y ..........................................

115
116
118
122
124
126

CHAPTER V
RESULTS AND HYPOTHESIS T E S T I N G .............
127
5.1 I n t r o d u c t i o n ...................................... 127
5.2 Basis for hypothesis rejection ...............
127
5.3 Interpretation of the data from the logit
r e g r e s s i o n ...................................... 129
5.4 Testing of h y p o t h e s e s ........................... 129
5.4.1 R i s k ...................................130
5.4.2 Risk and tenant's choice of contract
. 130
5.4.4 Discussion of risk
. . . . . . . . . .
131
5.4.5 Transactions Costs
.............. 133
5.4.6 Social Closeness
......................
136
5.4.7 Further study of social closeness . . . 138
5.5 S u m m a r y .......................................... 139
CHAPTER VI
C O N C L U S I O N S ................................................. 141
6.1 R e v i e w ............................................ 141
6.2 Empirical findings . ........... . . . . . . .
142
6.2.1 Findings on r i s k ..................... 143
6.2.2 Findings on social closeness
. . . . .
145
6.2.3 Findings on transaction costs .........
14 7
6.2.4 Summary of f i n d i n g s ................... 148
6.3 I m p l i c a t i o n s ...................................... 149
6.3.1 Implications for landlords
...........
149
6.3.2 Implications for t e n a n t s ............. 150
6.3.3 Implications for farm management and
extension personnel
....................
151
6.3.4 Public policy implications
...........
152
6.3.5 Theoretical implications
.............
153
6.4 Recommendations for further research .........
153
APPENDIX I
THE Q U E S T I O N N A I R E .............

156

APPENDIX II
STATISTICAL DATA
FROM THE LOGISTIC R E G R E S S I O N S ............................. 164
BIBLIOGRAPHY

............................................

vii

170

LIST OF FIGURES
Figure II- 1 The tax m o d e l ............................

16

Figure II- 2 The effects of sharing input costs to
resolve the share lease paradox....................

23

Figure III- 1 Social closeness and the accompanying
attitudes have the effect of shifting the demand
for variable inputs out and upward................

61

viii

LIST OF TABLES
Table III.l Summary of Contract Choice Under Risk . . .
Table

Choice Under Consideration
of
...................................

65

III. 3 Contract
Choice Under Consideration
of
Management Resources................................

66

III.4 Contract
Choice Under Consideration
of
Investment Resources ...............................

67

IV.1 Testing for
non-response
bias
in
the
sample.
.
.......................................

84

IV. 2 Acreage Owned and Leased with Estimated
V a l u e ..............................................

87

IV.3 Profile of
respondents by income from
farming, age and s e x ................................

88

Table IV.4 Distribution of parcel holdings by number and
t y p e .............

89

Table IV.5 Lease Length by contract type, respondent and
f o r m a l i t y ..........................................

90

Table IV.6 Formality and renewal ofagreements

91

Table
Table
Table
Table
Table

Table

III.2 Contract
Social Distance

56

. . . .

IV. 7 Distribution
of Crops grown by contract
t y p e ................................................

93

Table IV. 8 Tenant's share of o u t p u t ....................

94

Table IV.9 Landlord participation in input cost sharing
by input
...................................

96

Table IV.10 Landlord experience and choice of contract
Table IV.11 Landlord prior cultivation of the land to be
leased and contract choice ........................
Table

99
99

IV.12
Respondent actual contract choice with
parties of varying degrees of closeness..............100

Table IV.13 Landlord contract preferences (hypothetical)
with parties of varying degrees of closeness;
assuming no risk in performance...................... 102
ix

Table IV.14 Landlord contract preferences (hypothetical)
with tenants of uncertain performance and varying
categories of social closeness....................... 103
Table IV.15 Tenants' contract preferences (hypothetical)
with landlords of varying categories of social
closeness...............................................105
Table

IV. 16 Tenant's view of the contribution of the
landlord; Crop-share Agreement ...................

107

Table

IV. 17 Tenant's view of the contribution of the
landlord; Cash Rent A g r e e m e n t ...................... 108

Table

IV. 18 Perceived interest in, and care for, the
a s s e t s ................................................. 109

Table IV. 19 Capital improvements reported.................. 110
Table

IV.20 Tenants' risk preferences and contract
choree . . . . . . . . . . . . . . . . . . . . . .

Ill

Table

IV.21 Landlords' risk preferences and contract
c h o i c e ................................................. 112

Table

IV.22
Degree
of
interaction
and
choice
of
contract . . . .................... . . . . . . . .

Table IV.23 Respondents Evaluation of Fairness
Table

IV.24
Contract
interaction

Table

IV.25
Variable
list
for
landlord
regression.........................

. . . .

113
114

choice
(actual)
and
social
....................................115
logistic
117

Table IV.26 Classification Table for Landlord Contract
C h o i c e .................................................119
Table IV.27 Result of the landlord regression equation
on choice of the share agreement- variables and
corresponding statistics
. . . . . . . .

121

Table

IV.28
Variable
list
for
tenant
logistic
regression........... ... ............................ 123

Table

IV.29 Classification Table for Tenant Contract
C h o i c e .................................................124

Table

IV.30 Variables
in the tenant equation and
corresponding statistics ..........................
x

125

Table

VI .1 Selected social closeness and interaction
variables
and
corresponding
statistics
for
landlords and tenants..............................

xi

146

CH APTER I
IN T R O D U C T IO N A N D PR O B L E M ST A T E M E N T

1.1 Introduction
Two in every five acres of U S farm land is u nder som e kind of lease
agreem ent. U nderstanding the issues and im plications of leasing can m ake an
im portant contribution to th e adaptability and well being of US agriculture. The
efficiency of share cropping agreem ent versus cash rent has long been the subject of
debate due to ap p aren t conflicts betw een theory and practice. This study adds a new
dimension. T he m otivation for leasing and the choice of contract will be exam ined
particularly with regard to the social relationship betw een the landlord and tenant.
W hen a landholder faces a change in circum stances such th at the farm ing of
the land is no longer consistent with his/her interests, a decision must be faced.
E ith er the land can be sold, left idle, let to a tenant or the ow ner can hire custom
w ork done. Leaving the land idle is rarely an economically sound choice. Choosing
to sell the land may pose oth er problem s. Land is not liquid. Land m arkets, though
traditionally stable, have recently been subject to considerable instability, interfering
with marketability. F u rtherm ore, the land ow ner may w ant to retain ow nership of
the land for investm ent, personal or tax reasons.
O n the other hand, an am bitious farm er may be m otivated to expand his or
her land base in o rd er to achieve a m ore efficient use of resources, expand

1

2
operations, or take advantage of m arket fluctuations w ithout taking ow nership of the
land. Lack of capital, desire for mobility, or interest in accom m odating the needs of
family m em bers no longer in a position to farm can m otivate a potential ten an t to
turn to leasing as an alternative to ownership.
F o r th e landlord, securing a good ten an t is no small m atter. In addition to
annual income, the landlord knows that soil m aintenance and conservation practices
depend on the attitude of th e ow ner and/or occupant of the land and the level of
com m itm ent to investm ent in the future. T he ow ner of the land is assum ed to have
a vested interest in the land and m aintenance of the value of that investm ent. T he
tenant on the other hand tends to be m otivated by current term profit potential. It
is not difficult to envision th a t such differences in interests can lead to som e conflict
and/or controversy in establishing and m aintaining lease agreem ents.
F urtherm ore, the scene is com plicated by the em ergence of two alternative
m odes of leasing.

F or one, there is the traditional share leasing agreem ent

(som etim es referred to as share cropping) u nder which the ten an t pays the landlord
a portion of the crop either in kind or the equivalent in cash after the crop is sold.
T he oth er alternative is a cash rent agreem ent w here the ten an t pays a fixed fee for
the use of the land, usually in advance of the cropping season.

T he different

agreem ents necessarily foster different sets o f econom ic incentives.
The share leasing agreem ent, in particular, has perplexed students of
econom ics for a num ber of years. The source of much of the confusion stem s from
the analytical approach of econom ists and the assum ptions underlying the traditional

econom ic m odel. Strictly speaking (in the language of econom ics), when a landlord
engages a tenant, the ren t paid for the use of the land partially o r wholly
recom penses the landlord for foregone returns from the use of the land. Provisions
m ay also be m ade to account for any depreciation, time and effort in servicing the
lease, and w hatever econom ic profits are forthcom ing given the conditions of the
m arket subject to the term s and conditions of the lease.
Conflicts arise betw een theory and practice which are generally acknowledged
in th e literature, but not resolved (C heung 1969, H eady 1947, Sutinen 1975). These
include: 1) the disincentives to production on the p art of the tenant, 2) the difficulty
associated with m onitoring the m aintenance of the soil (o r lack thereof), 3) the
difficulty associated with m onitoring of conservation practices, and 4) the inefficiency
associated with perform ance of m aintenance and conservation practices separately
from th e routine cultivation and farm ing of the land.
No line of thinking has em erged to dom inate the field. It is intended that this
study will g ath er valuable em pirical data to be used in evaluation of hypothesized
links in the developm ent of a theory of leasing. Concurrently, it is anticipated that
practical inform ation will be generated to contribute tow ard the p reparation of a
handbook on leasing. This docum ent should prove useful to those faced with the
prospects of leasing and desiring educational and conventional inform ation on the
subject.
T he author has observed that the interests of landlords and tenants may
conflict as each tries to optim ize returns under a lease agreem ent (see also Schickele

1941). O f particular interest are the long range effects on the fertility of the soil and
preservation of the environm ent (Ervin 1983, 1986, Dillman 1982). It appears that
th ere is a relationship betw een the attitude of the ten an t and the perform ance of
fertility m aintenance, soil conservation and environm ental protection investm ent.
This attitude may be directly related to the expected length of anticipated tenure or
specifically related to the interest the individual has in the welfare of the o th er party
in the leasing agreem ent, or both. U ltim ately the prevailing attitude may lead to the
kind of contract chosen.

1.2 Problem statement
Behavior exem plary of tem porary (versus p erm anent) land tenure will have
negative effects on efficient use of the productive assets, specifically the m aintenance
and care given to the soil and the environm ent. Though behavior of the p erm anent
ten an t (including landow ners) can be less than ideal due to constraints in time,
resources and knowledge, the behavior of the tem porary ten an t is expected to be
substantially worse. In principle, the term s and conditions of the lease agreem ent
may be drawn up in such a way as to encourage the short term ten an t to exhibit
behavior similar to that of a p erm anent tenant. However, in practice, the contract
is necessarily incom plete because of uncertainty and m onitoring (transaction) costs.
D espite it’s significance, detailed knowledge and understanding of agricultural
leasing in the U.S. does not exist at the present time. To the au th o r’s knowledge
only two books have been w ritten on the subject, one by Steven N. S. C heung in

1969 and an o th er by J. M. C urrie in 1976. Each attem pts to address the leasing
issue theoretically, but n either offers solid em pirical d ata to substantiate the
theoretical claims1. And, though num erous articles have ap p eared on the subject,
the broad observation by this author is th at the existing theoretical explanations have
failed to provide a consensus, or generally accepted basis, for the choice of the share
cropping agreem ent.
C urrent (traditional) models have limited capacity to explain and predict
tenant (and landlord) behavior.

F u rth er understanding of leasing and leasing

behavior is crucial to analyze policy issues concerning land ownership, land use, land
values and farm finance. T he traditional model needs to be supplem ented with such
things as the ten an t’s reputation, com m itm ent to family or friendships, personal
integrity, credible penalty th reats on the part of the landlord, and incentive program s
which may all influence the behavior of the tenant.
In anticipation of a b etter understanding of the attitudes and behavior
associated with tenure and contract choice this study was conceived. T he objectives
of this study are:

1) Seek a solution to th e conflict betw een theory and practice,
2) Im prove the data on leasing in Michigan, and

1

As this work is being con clu ded and prepared for binding it has com e to the author’s attention
that a n ew book on contract ch oice has com e out. The authors' work has b een cited herein
but it would b e appropriate to acknow ledge that the book w as not available at the time of this
study (se e Hayami and Otsuka 1993).

3)
agreem ents.

Exam ine the contribution o f the concept of social closeness in contractual

T he following pages will detail the w ork of reviewing previous studies,
conception and exam ination of a m odel of leasing, form alization o f testable
hypotheses, organization and execution of a survey of individuals involved in the
leasing of farm land in M ichigan (both landlords and tenants), collection and analysis
of the data, and exam ination of the results and ensuing conclusions.

CHAPTER II
REVIEW OF LITERATURE

2.1 Introduction
The incidence of leasing accounts for betw een thirty-five and forty p ercen t of
the nation’s farm land and has rem ained fairly constant over th e years. Fluctuations
do occur with the m ost noticeable being a tendency for the am ount of farm land
under lease to rise when farm land prices are falling. Given th e m agnitude of the
practice of leasing of farm land, considerable interest is gen erated in leasing
agreem ents and the potential for im pact on national production of food and fiber.
In o rder to set the ground work for this study of contract choice in leasing
agreem ents, the following pages will exam ine previous w ork by oth er authors. O f
particular interest will be o th er attem pts to explain the behavior of landlord and
especially tenants.

2.2 Historical setting and emergence of leasing
E ver since Jacob contracted for the off color offspring of the cattle as
com pensation for his services in m anaging L ab an ’s herds and flocks, students of the
econom y have grappled with the conflicts of interest betw een landlord and tenant.
A dam Smith (1805) argued th at the te n a n t’s rent, paid to the landlord, discouraged
any investm ent by the ten an t in durable factors of production. H e com pared it with

7

a tax which interfered with productivity. A prevailing position of early w riters was
th a t tenure, particularly share tenan t agreem ents such as the widely adopted metayage
(share-cropping) system in France, constituted a curse to the land, the econom y and
th e citizens. A m ong others, R ichard Jones (as quoted by D. G ale Johnson, 1950)
attributed w idespread poverty of the rural citizens to this ten u re system which
discouraged input use and consequently low ered productivity.
Econom ists rem ained perplexed by the persistence of the share crop tenancy
and began to exam ine the reasons for its popularity. John Stuart Mill (1920) took
the position th at the metayage system was theoretically sound but that the concurrent
problem s w ere due to im perfections. Alfred M arshall (1920) encouraged the use of
cash, versus share, ren t (showing that an optim al solution could be achieved by
adjusting th e fixed and share proportions of the contract), in o rd er to avoid the
theoretical conflict o f interest. L ater Schickele (1947) H eady (1947) and D. G ale
Johnson (1950) began to propose conditions under which the share contract might
be considered as efficient as cash rent. N evertheless, these proposed solutions, and
others th at have followed, have not found overwhelm ing favor in the literature.
N um erous m odels have been proposed. However, for the m ost part, inconsistency
with em pirical d ata or use of assum ptions foreign to the environm ent found in the
field, have left the profession less than convinced.

9

2 3 Areas of Focus
Five m ajor questions have dom inated the discussion of tenure. T he first of
these receives disproportionately little attention. Q uestions 2-5 w ere identified and
listed by N ew berry and Stiglitz (1979) and again by Allen (1985):
1. Why doesn’t the landow ner farm the land personally, or sell it?
2. W hat is the reason for the persistence of sharecropping and coexistence of
types of lease contracts (share, cash, and mixed)?
3. Why does the share leasing agreem ent decline with developm ent?
4. W hat causes lower productivity under share leasing agreem ents? Or, does
the share agreem ent necessarily correspond with lower productivity?
5. H ow is share split determ ined?
T he second question has received m ost of the attention and continues to dom inate
the discussion in the literature. This au th o r takes issue with the third question since
this does not seem to be the case in U.S. agriculture (see Johnson et al 1987).
As with many others, the focus of this study will be on question num ber two.
In seeking answers to this question, it is expected that some understanding will be
brought to b ear on questions th ree through five. Q uestion one will be addressed only
briefly. Prior to addressing these questions, a brief review of term s and definitions
is provided.

10

2.4 Terms and Definitions
T he ten u re term s have applied to a broad spectrum of contractual agreem ents
and contracting parties with a variety of resources. Landlords consist of ow ners of
property with an interest in securing a ten an t that will occupy, m anage, care for, or
husband th e asset (usually land) in the place of the landlord and ren d er to the
landlord a com pensation for the use of the assets. T enants cover an even b ro ad er
category of individuals. T he medieval E u ro p ean description characterizes tenants as
peasant laborers. In m ore m odern US agriculture, the ten an t could be best described
as an entrepreneur, an ow ner of specialized productive assets, labor, and m anagerial
skills. In any event, the ten an t exhibited two com m on characteristics

1) he/she is

entrusted with a stew ardship over assets th at are owned by the landlord, and 2)
receives com pensation resulting directly from the proceeds of production rath e r than
a fixed wage.
Tenants will be distinguished from hired laborers by the latter characteristic.
W here a m ajority (over 50% ) of the engaged p arty’s incom e or com pensation,
relative to the contract or agreem ent, is derived directly from and varying with the
proceeds of production, the party is considered to be a tenant. O therw ise the party
will be considered a hired laborer. Both tenants and hired labors may share the
form er characteristic of stewardship, however, perhaps not to the sam e degree.
Because this depiction leaves the definition of th e ten an t open to a broad
range of individuals, researchers have been at liberty to further define and specify the
characteristics of the "tenant" according to the nature of the study or m odel at hand.

11
F or exam ple, Bogue (1964) rep o rts th at a prom inent m idw estern landlord in the late
1800’s classified tenants relative to personal holding of property.

O ttoson (1955)

distinguished betw een crop and livestock share-tenants, citing different incentive sets
for tenants involved in different assets. Lucas (1979) points out a distinction betw een
share tenants who m ake paym ent "in kind" and those who ren d er cash paym ent to
the landlord. H e points o u t th at the "in kind" paym ent elicits a focus on quantity
production while m oney paym ent encourages the ten an t to focus on quality as well.
O f course, it is not uncom m on for a distinction to be m ade betw een cash-lease and
share-lease tenants.
C heung (1969) in attem pting to m odel conditions in Taiwan, chose to classify
the ten an t as a landless, lab o rer w ithout assets.

O thers adopting this definition

include B aron (1981) and Boxley (1971). Several studies in India portray the ten an t
as a landless p easan t (see for exam ple, B ardhan 1979 and R ao 1971).

By contrast,

a recent study in N ebraska and South D akota (Johnson et al. 1987) found that the
farm er-tenant dom inating th e rental m arket in the US owned som e land as well as
having a substantial investm ent in farm ing equipm ent.
F or this study, the following set of term s will be used to identify and
distinguish tenure classifications:

12
O w ner O p erato r

O ne who farm s only the land he/she owns.

T e n a n tP art O w ner

O ne who owns som e land and rents additional land from
others.

Full T enant

O ne who farm s only the land he/she rents (does not own
any land).

Landlord:

Owns land and leases to one or m ore tenants. M ay be
part tim e landlord -still farm ing som e land, absentee
landlord - o n e who does not live in the sam e county as
th e land under lease. An institutional landlord usually
eith er a firm, governm ent body, or possibly com munity
organization such as an Indian tribe.

This set of definitions seem s to have been adopted as the norm in the current
literature, including the C ensus o f A griculture (1987) classifications, with only syntax
difference. A n exam ple of syntax variation would be Allen and Lueck (1990) who
em ploy the term farm er instead of ten an t in the th eir description of the party renting
the land.
The com m on factor is, however, th at the ten an t com es to the lease agreem ent
with som e capital. It may be hum an capital in the form of skills, time, or willingness
to incur risk, real capital in th e form of operating funds, or physical capital in the
form of m achinery and equipm ent, often a com bination of all three.

H e or she

contracts to com pensate the landlord for the use of leased assets while anticipating
a retu rn on the invested real, physical and/or hum an capital. A functional m odel of
leasing should therefore capture this aspect of retu rn on investm ent to the leasing
parties.

2.5 Purpose: Why Rent?
2.5.1 For the landlord
T he landlord may choose to retain ow nership of th e land but not wish to farm
it personally. R easons for this m ight include, but are not limited to the following:
a) Illiquidity
b) Tax incentives
c) H edge against inflation (B aron 1983)
d) Engage the professional services and/or skills of the tenant.
e) Share risk
f) Sentim ental value

2.5.2 For the tenant
O n the oth er hand, th e ten an t desires to lease farm land as an alternative to
ownership. R easons for this position might include:
a) lack of capital
b) cost of assets (cheaper to ren t than own)
c) mobility
d) gain experience u nder the guidance of a retired farm er
e) share risk
f) unavailability of land for purchase

14

2.6 Contract Choice
In o rd er to understand the traditional thinking on the various contracts and
the corresponding theoretical implications, the next section will be devoted to a
review of the characterization of ten u re in the literature. As m entioned above the
choice of contract question has seem ed to dom inate the literature.

2.6.1 Early Models
A dam Smith has b een credited with the form ulation o f w hat will be term ed
the traditional m odel which was later adopted and form alized by Alfred M arshall
(1920). T he simple depiction of the landlord ten an t relationship (discussed below)
together with the implications for behavior by the tenant and landlord, have persisted
for a long time. Still the theoretical predictions of the model result in paradoxical
outcom es w hen com pared to observed practices and em pirical studies. N um erous
variations and reform ulations have been proposed with no em ergence of a dom inant
line of thinking to supersede the traditional model. These will be exam ined below,
beginning with a discussion of efficiency in tenure, the driving force behind the
theoretical constructs.

2.7 The Traditional Models
In this section we will exam ine a num ber of models th at ap p ear with some
frequency in the literature.

15

2.7.1 The efficiency question
The vast m ajority of the literature on agricultural leasing is m otivated by the
theoretical issue of efficiency in determ ining contract choice.

T he hypothesis,

regarding the perform ance o f the ten an t with respect to em ploym ent of productive
inputs, is th at the share ten an t will choose variable inputs th at result in an inefficient
use of the available resources (see O tsuka and Hayam i 1988.) This characterization
of the landlord ten an t relationship is established theoretically, assum ing the te n a n t’s
incentive to equate m arginal revenue and his own m arginal product which is m erely
a share of total output. The tenant has control over variable inputs and chooses a
level of output that is below th at which the landlord would have chosen. This is best
portrayed by a careful exam ination of the traditional model.

2.7.2 Advalorem tax or share model
The traditional approach is to com pare the expected perform ance of the
ten an t under crop-share and lump sum rental agreem ents.

U n d er crop share

agreem ents the ten an t contributes to the landlord a portion of each unit of output.
T he result is similar to an ad valorem tax. And, as in the ad valorem tax m odel, the
profit maximizing point of output occurs (i.e. equates marginal cost and m arginal
revenue) at a lower level of output for the

m anaging (before tax) firm.

Consequently, theorists conclude that the profit maximizing sh are-ten an t will choose
a reduced level of inputs, lowering output.

16
The

graphic

re p re s e n ta tio n

Value of output
A

(S e e

Figure I I - 1) dem onstrates

Marginal value product (MVP)

that under th e crop share
agreem ent, the ten an t is
B

e x p e c te d

to

Tenant's MVP

e q u a te
Marginal Cost (HC)

D

m arginal

revenue

with

m arginal cost. If m arginal
value product (M V P) is
0
depicted
BQ3

by A Q 3,

r e p r e s e n ts

and
th e

0

0

Quantity of composit input X

F igure II- 1 T he tax m odel

te n a n t’s M VP, then given constant m arginal cost (M C) of inputs line DC, the ten an t
is expected to apply variable inputs X until the m arginal cost p er unit of X is equal
to the m arginal retu rn at input point Qj yielding o u tp u t E. This results in a lower
level of o utput than that which would be chosen by the landlord (Q 2 input level
yielding output C) and consequently an inefficient allocation of resources (see
Schickele 1941, H eady 1947, Johnson 1950, Issawi 1957, G eorgescu-R oegen 1960,
A dam s and R ask 1968, Boxley 1971, N ew berry 1974, R eid 1976, B raverm an and
Stiglitz 1986 and O tsuka and H ayam i 1988).
T o further d em onstrate the application of this model, the m otivation of the
ten an t is considered. U n d er a share lease agreem ent, the tenant supplies a variable
input, usually depicted by labor available in com pletely divisible, uniform units.

17
F aced with the reality of receiving only a portion of each unit of ou tp u t p er each unit
input of labor, and assuming diminishing m arginal returns to the variable input, the
ten an t will attem p t to equate the value of the m arginal product with m arginal cost.

2.7.2 Lump sum model
F o r com parison, let us consider the m odel often viewed as the efficient case:
a ten an t with a cash lease. First th e following variables are defined:
L

: proportion of te n a n t’s labor em ployed on leased land.
0 £ L <; 1

A

: num ber of acres rented (could also be viewed as all fixed inputs)

q : output which is a function of labor and land, q = q (L ,A )
c : p er acre cash rental
w : wage rate of labor (exogenous)

T he price of the output q is norm alized to unity. T he te n a n t’s incom e y is
then expressed as:

y = q - cA + w (l - L)

(!•!)

This leads to the fam iliar maximization result in which the first o rd er conditions
equate m arginal revenue (m arginal product of labor) with the wage rate:

7 =w
qL

(1.2)

Next the case of the share ten an t is considered. Incom e for the ten an t is given

18

as:
y = (1 - a)q + w(l - L)

(1*3)

w here a is th e landlord’s ren t (0 i a s 1) to be paid as a share of output. Using this
m odel, th e optim al level of labor input (L) by the ten an t would occur w here m arginal
cost o f L and m arginal retu rn to L are equated.

T he first o rd er conditions then

result in:

(1 - a)q'L = w

Since th e retu rn to the ten an t is now less than the m arket wage rate, and assuming
th a t the retu rn to labor is a declining function, the ten an t will choose to reduce the
am ount of variable input L . H e will attem p t to equate the cost of effort to returns
equivalent to the m arket wage rate w. O ptim al behavior by the ten an t would then
result in reduced output to the extent of the output elasticity of the labor input
(D rake 1952 and B raverm an and Stiglitz 1986). This result is considered to be an
inefficient use of resources, the efficient case being expressed in equation (1.2).

2.8 Problems with the Traditional Models
T he tax m odel as form ulated to this point, raises a num ber of questions. First,
is does not answ er the question of labor allocation betw een the technically equivalent
leasing agreem ent and working for the m arket wage. The implication is that there

19
is som e incentive for the ten an t to allocate all labor to the leasing agreem ent.
Consequently, it does not answ er the question of contract choice, implying that the
share tenancy agreem ent is illogical.
Second, there is a problem of labor allocation betw een parcels of land owned
and leased. M ost individuals that lease farm land in the US own some land (Johnson,
et a l 1987). Farm ers realize th at timing of planting and harvest are often extremely
crucial to the success of the crop. T he tax m odel assum es that all units of labor are
equal and are uniformly distributed. It would stand to reason that the tenant will
have an incentive to allocate prim e tim e to owned land w here the full value of the
crop is his to keep. Leased cropland, especially share-leased, would receive second
priority and therefore the value of crops from leased land would be relatively lower.
This should be especially evident when conditions are less than ideal, threatening
crop failure (H olstrom et al.

1985).

Third, the theoretical predictions are not supported empirically. O tsuka and
H ayam i (1988) have com piled an impressive list o f em pirical studies.

Their data

dem onstrates that, based on yields, there is no statistically significant difference
betw een o utput on ow ner versus ten an t o perated farms. H owever, they do show that
there appears to be a difference relative to the value of the output.
Further, the tax m odel raises the question of resource allocation. O n the one
hand, according to traditional econom ic theory, the (M arshallian) economically
efficient allocation is expressed in the model w here the rent is a lum p sum. M arginal
input use is equated with m arginal output.

The choice of the tenant is then to

20
evaluate th e returns to farm ing and then either pay the lum p sum rent or w ork for
the m arket wage. In a perfect m arket in equilibrium, one would expect to find the
earnings of the ten an t and wage earn er to be equated, assuming either no risk or
equivalent risk to the respective alternatives. This outcom e would then be evaluated
against the relative earnings from leasing a given parcel of land on a share basis. T he
te n a n t’s choice would equate all three possibilities in this simplified environm ent.
Similarly, the landlord is faced with finding the m eans of achieving the optim al
use o f th e asset, land. Shaban (1987) m aintains th at a m ajor failing of the traditional
approach is that researchers tend to neglect the fact that the landlord is an actor in
the leasing agreem ent. Specifically, if the landlord is not able to realize returns by
share leasing g reater than or equal to that expected under a cash lease, then the
landlord will choose the cash lease.
T he tax model predicts that the ten an t will equate the value of marginal inputs
and outputs. T he implication is that this is the sam e result that the landlord would
reach w ere he/she to farm the land personally. T he econom ic m odel assumes that
the contribution to production of the fixed asset, land, is known and constant yet
nearly all o f the authors consulted refer in some way to the differences in productivity
attributable to use and/or abuse of the land. H owever in the traditional m odel, the
change in output is attributed solely to the application of variable, inputs.
Early researchers, true to economic custom, held rigidly to the theory and thus,
the traditional model.

They attem pted to explain away the em pirical evidence

(H eady and K ehrberg 1952). But, m ore recent em pirical evidence has failed to reject

21
the hypothesis th at th ere was no reduction in production levels on farm s o p erated by
tenants (Johnson, et al. 1987).

2.9 Elements that remain unexplained by the tax model and its variations.
W hy does the share lease arrangem ent rem ain so popular (D illm an and
Carlson 1982)? Johnson et al. (1987) found th at 75% of N ebraska leases and 62%
of South D akota leases used the share rath er than cash leasing arrangem ent.

If

indeed it is inefficient w hat incentives prescribe its choice? Or, is it th at the
assum ptions that would lead to a theoretically efficient result are not borne out in
practice? B ernat (1987) concludes:
N either the traditional version of the tax-equivalent model, in which
the o p erato r is a landless tenant, nor the screening view ap p ear to be
very relevant to U.S. agriculture. . . T he traditional tax-equivalent
m odel requires m arket im perfections th at either do not exist in the
m odern U.S. econom y or are relatively insignificant.
F urtherm ore, w ere the landlord m otivated to participate in sharing all production
inputs in similar proportion to output share, why do we not observe the landlord
either contributing labor or com pensating the ten an t for a share of the labor
expenditure? O ne possible explanation proposed by B ernat (1987), which bears
further consideration, is the notion that many agricultural crops may not lend
them selves to marginal variations in labor inputs.

T hat is, the inputs are b etter

viewed as com plem ents. Indeed, a situation approaching an al! or nothing prospect
is com mon, causing the m arginal returns to labor to be either very great or
insignificant such that the term marginal returns has no meaning.

22
Em pirical evidence only partially supports the ten u re-lad d er variant o f the
screening hypothesis.

Full tenants tend to be younger than p art or full owners

(Johnson et al. 1987). H owever, the fact that many operators frequently ren t u nder
both cash and share contracts and go back and forth betw een the two alternatives,
frustrates the idea of a progression from one to the other.

2.10 Proposed Answers to the Theoretical Paradox and Contract Choice
As m entioned previously, the majority of the studies consulted in the literature
attem p t to provide answers to the seemingly paradoxical conditions th at rem ain
betw een the accepted theory and the limited em pirical evidence. These attem pts
include such things as revising the assum ptions, adding new assum ptions, adding
conditions or constraints, or attem pting to modify the theory. E ach approach has
som e m erit as well as som e shortcomings. This discussion is introduced by an input
sharing solution.

2.10.1 Input Sharing
Schickele (1941) and later H eady (1947) proposed an alternative to the results
of the traditional share model. They suggested that if inputs w ere shared by the
landlord and ten an t in the sam e proportion as outputs, the incentive problem leading
to reduced input levels could be avoided and the "perfect" or "ideal" socially optim al
lease would result (see also Castle 1952). A dam s and R ask (1968) d em onstrated this

23

e f f e c t

Value of total output

geom etrically with
a m odel th at has
gained acceptance
as the ap p ro p riate
depiction

of the

MVP
Tenant's
MVP

share-lease
paradox

(see

Figure II- 2).
T h e i r
m odel assum es a
50-50 ou tp u t share

Quantity of composite input X

F igure II- 2 T he effects of sharing input costs to resolve the
share lease paradox.

lease with no cost
sharing. T he line indicated as A Q 3 represents th e m arginal value product (M V P) of
output to th e firm as input X } (such as labor) is varied. Land and oth er factors of
production being held fixed. T he line B C represents m arginal factor cost (M FC ) of
input X } to the firm. T he ow ner-operator (or ten an t paying a lum p sum ren t) is
expected to choose Q 2 level o f variable input X , to eq u ate M VP with M FC.
Line D Q 3 then rep resen ts the effective M V P realized by the tenant. And, the
ten an t is expected to choose Q 3 level of input X ,, o r a reduced, inefficient output
level. T he argum ent is th a t if the landlord w ere to share input costs at the same
p roportion as the sharing of o u tp u t (the "ideal lease"), the te n a n t’s M FC schedule

24
would be H M . Now the incentive exists for the ten an t to choose Q2 input level which
maximizes profits with respect to X v
The m odel readily indicates th at the tenant has an econom ic incentive to
adopt th e Q2 level of variable input. However, the incentives faced by the landlord
are not so clear.

Participation in variable costs (X,) requires the landlord to

contribute the equivalent of area BG K H , receiving in retu rn increased revenues of
FEGK. The conclusion is that unless the latter area exceeds the form er, the landlord
would be unwilling to participate.

A dam s and R ask qualify this by adding th at

em pirical verification would be required to establish preferences.

Otsulca and

Hayam i (1988) are much m ore aggressive in their attack of this result, claiming th at
for the m ost p art it would require that the landlord be ". . . an altruist, who, as a
benevolent patron, supports the income of po o r clients at the expense of his income."

A disturbing im plication of the cost sharing model as proposed is the fact th at
in order to realize the optim al point indicated, Q2, all variable costs would need to
be shared, including the cost of the te n a n t’s labor. It is not clear to this au th o r why
previous studies have failed to address the implications of the landlord paying for a
portion o f the te n an t’s labor. F or example, H eady (1947 p.670) for reasons unknown
to this author, omits labor from his list of variable inputs. Only Boxley (1971) and
Schickele (1941) and m ore recently Shaban (1987) acknowledge this point in the
literature.

E ach

discusses it briefly but neither offers a solution.

Boxley and

Schickele seem to take the position th at this is an exception to the m odel that must

25
be simply taken into account. In any event, labor rem ains a m ajor variable input, a
choice variable u nder the ow nership of the tenant.

2.10.2 Empirical and Practical Questions
Interestingly th ere has b een little em pirical work done to verify the predictions

of th e A dam s and R ask m odel. A prior study by Berry (1962) which fielded opinions
of farm ers in South D ak o ta revealed that tenants and landlords both expressed
willingness to share the cost of fertilizer and, under som e conditions, seed. However,
b o th groups strongly resisted the idea of sharing the cost of things like fuel, hired
labor, and repairs to machinery. Berry offered the explanation that further sharing
tended to imply a p artnership rath er than rental agreem ent, som ething that was less
appealing to the agents.

O th er studies indicate a willingness by landlord to

participate in sharing fertilizer, seed and chemical costs (Shaban 1987, Johnson et al.
1987).
F u rth er reflection on the n atu re of the inputs considered by Berry, suggests
th a t there is a direct correlation betw een th e willingness of th e parties to share input
costs and the degree of either liquidity or substitutability of the actual inputs shared.
F or exam ple, fertilizer is purchased in large quantities involving one or possibly two
expenditures which are easy for the landlord to observe and verify. W hen purchased
in bulk it is n o t uncom m on for it to be prem ixed to the p urchaser’s specifications
according to land fertility and intended crop.

H andling is expensive.

These

conditions tend to limit or at least reduce the liquidity of the commodity on the resale

26
m arket and thus the incentive to steal.
F urtherm ore, fertilizer application to the land and or growing crops takes
place as a relatively infrequent event.

O nce applied, the fertilizer can n either be

rem oved nor transferred to an o th er purpose. H ence it has no m arketable value. By
com parison, fuel is purchased frequently, has many alternative uses, and has, as well,
a resale value.
In connection with th e issue of sharing of inputs and outputs one m ust also
consider the proportion or division of shares. The A dam s and R ask m odel (above)
adopted the conventional theoretical M arshalian approach of the time, which
preced en t has persisted in the literature. However, H eady (1947) noted:

It is com m on knowledge, of course, th at share rents vary betw een
certain wide areas. Even then, the discrete breaks in [proportion of]
share rentals betw een b road regions cannot conform accurately to the
gradual decline in econom ic rent resulting from increased
transportation costs or less favorable w eather. H owever, num erous
studies have indicated th at uniform rental shares tend to exist within
areas which are fairly hom ogeneous in respect to prices and w eather
but widely heterogeneous in respect to soil resources.

W hat causes the variation?
Newberry and Stiglitz (1979) proposed an argum ent to support the traditional
50-50 share split.

Allen (1985) dem onstrated a similar condition.

In the section

under risk (below) a result by R obison and Barry (1987) is exam ined which further
supports the 50-50 split.

However, the empirical work by Johnson et al.

(1987)

shows decided support for H eady’s regional differences. This author will argue that
the relative contribution of th e parties will influence the contract choice and th e share

27
split. It is a question of retu rn on invested resources. Brown and A tkinson 1985 base
their conclusions relative to contract choice on the entrepreneurial ability of th e
tenant, a type of resource contribution.

2.10.3 The Equal Efficiency Models
In answ er to the problem s relative to the share tenancy m odel discussed
above, C heung (1969) posited that th e root of the problem could be found in the use
o f partial equilibrium analysis. The traditional model does not account for the supply
side in th e economy. T he argum ent is that if potential tenants are plentiful and are
com peting for limited available land, there will be a tendency to agree to contractual
agreem ents th at specify labor inputs beyond the marginal equation of the tenant.
P rior to

Cheung, D.

G ale Johnson (1950) challenged the notion that

equilibrium o utput would occur at Q t (see Figure II- 1 page 15) since the landlord
would not be satisfied with this solution.

M ore recently Shabin reiterated this

argum ent m aintaining that th e landlord would not tolerate this level o f ou tp u t and
w ould settle for the cash lease instead.

This model resolves the problem of

com paring all form s of tenancy in general equilibrium.

2.10.4 Cheung Model
Steven C heung assum es (1) the firm ’s output level is uncertain due to som e
random factor of production such as w eather, (2) the contracting parties are averse
to risk, (3) the contractual alternatives are either a pure wage contract, a pu re rental

28
contract, o r a pu re share contract, (4) transaction costs are zero or the sam e for all
forms of contracting, and (5) all oth er m eans o f shifting th e risk are too costly (1969

P68).
By adopting these assum ptions C heung takes the position that equilibrium
conditions, in a world of p erfect com petition, will result in equal retu rn to the
landlord from cash and share leased land. Similarly, tenants will be indifferent to
share leasing or wage rent since each will g enerate m erely the m arginal product of
labor (A dam s and Raslc 1968, Boxley 1971 & 1972, G isser 1969, Kim 1972, and Scott
1970). C om petition am ong ren ters will ensure th a t labor supplied would be equal to
the labor engaged under a wage contract with no shirking. Indeed, C heung assum es
that if the level of labor expenditure is specified in th e agreem ent th at no shirking
will occur.

2.10.5 The Screening Approach
T he screening approach attem pts to explain contract choice by incorporating
into the tax model additional assum ptions.
differentiation of the variable inputs.

It also recognizes a degree of

F or exam ple, labor is divided into skilled

(m anagerial) and nonskilled. T enants are assum ed to possess nonskilled labor and
varying degrees of skilled labor. Landlords on the oth er hand have only skilled labor.
F urtherm ore, output is a function of both types of labor.
Asym metric inform ation about the ability of the tenant, on the p art of the

29
landlord, m otivates the landlord to offer a variety of wage, share, and cash contracts.
T he tenant, by choosing the contract to maximize net returns, reveals skill level and
resources.

In turn the landlord enjoys optim al returns by com plem enting the

contribution of the ten an t with m anagerial skills w here app ro p riate (H alagan 1978).
The prediction, therefore, is that screening will lead to wage contracts by
tenants possessing no m anagerial skills, cash rent contracts by highly skilled tenants,
and share contracts by tenants with limited m anagem ent skills (Allen 1984). Brown
and Atkinson (1985) surveyed landlords and tenants in Indiana and dem onstrated
that under share agreem ents significantly m ore decisions w ere shared by landlord and
ten an t than under cash ren t agreem ents. They suggested th at this finding supported
this idea of self selection proposed by H alagan (1978) as well as the agricultural
ladder theory.

W hat they did not disclose is the location of the landlords (present

versus absentee), degree of m anagerial ability possessed by the landlord, and the
relative disposition for cooperation. T he theory thus assumes that the landlord has,
and is willing to provide, the necessary m anagem ent skills. W here the landlord is not
able to contribute, output will be reduced on land un d er share and wage contracts
(A llen 1985). T he practice of use of contract types to induce self selection resulting
in screening of tenants is not readily observed in U S agriculture.
F urtherm ore, the screening approach assumes at the outset that skill levels are
not observable. Johnson et al. (1987) found that the most prom inent category of
agricultural tenants in the US w ere part owner operators, a group that by occupation,
necessarily possess highly developed m anagerial skill levels. N evertheless, this group

30
participates in bo th share and fixed rent contracts, a fact th at rem ains unexplained
by the screening viewpoint.
F u rth erm o re the fact rem ains that skill levels are not necessarily invisible to
landlord. This is fu rth er com plicated by the fact th at th e self selection m odel is static
o r would p ro p o se that tenants re-negotiated contracts in each period, with past
history becom ing invisible once again. Em pirical evidence, again from the N ebraska South D ak o ta study reveals th a t the parties to a given contract tend to establish long
o r semi p erm an en t though unw ritten agreem ents. Very little land is let to strangers.
B oth of these conditions suggest that the landlord will have the m eans to determ ine
skill level in advance of contracting and thus by-pass the screening problem
altogether.

2.10.6 Agricultural Ladder
A n o th er theory of land ten u re posits the existence of an agricultural ladder.
T he perception is that young beginning farm ers would tend to have limited skill and
resources. T he logical initial step would be to engage in a wage contract, consistent
with the predictions of the screening theory discussed above. Successful tenure would
lead to the accum ulation of skill and assets leading to share and ultimately cash rental
agreem ents with ow nership th e final goal (K loppenburg and G eisler 1985).

The

incentive for the ten an t would be to receive com pensation for accum ulated
m anagerial skill and asset.

31
T he problem with the ladder m odel was that, although it enjoyed substantial
ideological appeal, empirically it was never substantiated.

D ue to lack of hard

evidence the theory fell from academ ic grace as a viable theory of land ten u re around
1950. The contribution to th e understanding of tenure in the context of this study is
simply to note th at this approach has been em braced, scrutinized and discarded.

2.10.6 Linkage
A ttem pts have been m ade to incorporate tenant behavior into a model of
m ultiple incentive variables, a practice term ed linkage (B ardhan 1984, Bell and
Zusm an 1976, B raverm an and Srinivasan 1981, and Braverm an and G uasch 1984).
The concept centers around the idea that the landlord plays a role of not only
landow ner but also as financier of com m odities for the tenant much the sam e as the
com pany store for coal miners. T he occurrence of this kind of leasing agreem ent is
limited to developing countries and has little application to leasing in the U nited
States.

2.10.8 Risk
A num ber of econom ists have undertaken to show that the sharing of risks
provides an incentive for selecting the crop-share lease as opposed to oth er types
(C heung 1969, Stiglitz 1974, Reiss 1984, Sutinen 1975, N ewberry and Stiglitz 1979,
R eid 1973, R obison and Barry 1987, and O tsuka and Hayami 1988). C heung and
Sutinen argue th at the presence of risk is a necessary condition to the choice of the

32
share contract.
Reid, however, reached the conclusion th at dispersion of risk is not a sufficient
incentive to m otivate choice of share contract over wage or fixed rental contracts
(R eid 1976 & 1983). M oreover, Stiglitz (1974) and Newberry (1977) dem onstrated
that given constant returns to scale, the risk sharing attributes of share contracts
could be achieved through a mixture of wage and fixed rent contracts. The relative
significance of risk sharing as an incentive to establishing share crop agreem ents in
the US p ro m p ted New berry (1973) to conjecture:

" . . .
I agree th a t risk sharing is not the only motive for
sharecropping, and, in U S agriculture, possibly the least im portant
motive, I think the evidence from developing countries suggests that it
has an im portant role."

In Taiwan, C heung (1969) found th at the risky crop (w heat) was most
frequently the target of sharecropping contracts. Less risky crops such as rice tended
to attract cash rental agreem ents. As a result C heung proposed that the sharing of
risk dom inated the transactions cost result on crops of high risk. Sutinen raised some
im portant questions about the C heung data, pointing out that there was no attem pt
to account for (hold constant) differences in technology or changes in relative crop
values over the period of th e study.
By contrast, R ao (1971) found that tobacco was m ore often cultivated under
the cash ren t agreem ent while the less risky Indian rice was dom inated by share
agreem ents, exactly opposite to w hat the risk sharing argum ent would predict. R a o ’s

33
conceptual explanation was th e individual’s entrepreneurial ability th at governed the
choice of contract. This argum ent may have been seen as a source o f support the
A gricultural L adder theory of tenure that was popular in the early p art of the
century. H owever, one is left to ponder w hether tenants in India acquire skill in rice
production m ore easily than tenants in Thailand learn to produce w heat?
Allen and Lueck (1990) tested several risk related hypotheses using the
N ebraska-South D akota data. They found no support for the notion th at the share
agreem ent was p referred un d er conditions of m ore risky crops.
Sutinen (following C heung’s basic argum ent) showed th at risk could be used
to argue for a mixture of contract type. R obison and Barry dem onstrated th at the
Sutinen and H ieb ert models could be used to establish the crop-share split w here the
percentage is derived from a ratio of a m easurem ent of the landlord’s risk aversion
to the com bined risk aversion of both landlord and tenant. T he usefulness of this
result can be challenged on the grounds that em pirical m easurem ent of the risk
aversion variable X has yet to be established.

F urtherm ore, the contract choice

question rem ains unansw ered.
R obison and Barry (1987) dem onstrated that the models proposed by Sutinen
and H iebert, although slightly different in basic construction, yielded a result
indicating th at the relative risk aversion of the participants provides an explanation
for the predom inant 50:50 share lease:

34

K

w here k L and k T rep resen t th e m easure of risk aversion for landlord and ten an t
respectively. W here the landlord and tenant are equivalently risk averse, crop share
a will be equal to one half.
W orking backw ard from this result they propose that w here the ten an t is risk
neutral the cash ren t agreem ent will be chosen. Similarly, w here the landlord is risk
neutral w orkers will be hired as laborers. Though the argum ent is plausible, the
author is not aw are of a study that attem pts to sort tenants and landlords into risk
averse and non risk averse groups. T he intention is to explain the choice of the share
contract by agents that are largely of com parable relative risk aversion.
T he literature seem s particularly inconclusive on the issue of risk and the role
of relative risk aversion in contract choice. It would seem ap p ro p riate to conclude
that, given a prospective leasing situation w here the dom inate crop is subject to
substantial risk, risk averse parties would tend to choose the share contract to help
defray the effects of crop or price failure, all else being equal.

2.10.9 Transaction Costs
Typical transaction costs in a lease agreem ent are the costs of establishing the
contract and ensuring the provisions of the contract are carried out.

M ore

specifically, the contracting costs will include things like search costs, legal costs, and

35
tim e spent in bargaining and negotiation. Enforcing the contract involves such things
as exacting paym ent, supervision, and m easurem ent and division of the ou tp u t in the
case of share contracts.

A n o th er cost in this category would be the cost of

m onitoring the use of the asset, in this case land.
V ery little attention is given to the costs of establishing the contract in the
literature.

Even those addressing the issues of contract choice give only passing

acknow ledgem ent of this topic.

F urtherm ore, alm ost no attention is given to the

bargaining aspect of reaching a contract and the associated costs of negotiation.
Alston et al. (1984) assum e that contracting costs are similar for all of the various
types o f agreem ents and th erefo re regard these as an unim portant constant.
Costs associated with enforcing the contract vary greatly with the type of
agreem ent. W here the paym ent is fixed, the relevant cost may be minimal in the
case of the reliable tenant; or, conversely, substantial if the situation requires eviction
of a perfidious tenant. [U nder the cash ren t agreem ent the ten an t bears not only the
risk associated with crop failure and price fluctuations, but also the costs associated
with the minimization of these potential problem s. O n the o th er hand, a pure wage
contract may provide some facility and flexibility in reaching, or w here necessary,
term inating agreem ents, but, the landlord m ust b ear the risks and accompanying
costs.

In addition, due to the incentive and potential for the w orker to shirk,

perform ance supervision of the w orker adds an o th er cost.]
Alston et al.

(1984) d em onstrated th at on cotton plantations in G eorgia,

supervision costs of hired labor provided grounds for the choice of share contracts.

36
O tsuka and H ayam i (1988) regarded the supervision costs of labor and land as being
independent and som ew hat inverse. They m ade the assum ption that the incentive
for gains by the ten an t from land and capital abuse increased with the share
(percentage) received by the tenant. U n d er the wage versus share contract set of
possibilities, this may be true. But, under the cash lease agreem ent the ten an t has
the greatest incentive and potential to maximize extraction of w ealth from the fixed
asset since he has control of all variable inputs, including labor effort and production
inputs. It would stand to reason that the landlord would therefore prefer the contract
choice th at would minimize this loss or provide adequate com pensation.

F or

exam ple, Allen and Lueck (1990) found evidence that the share contract was m ore
often chosen w here the potential for exploitation was the greatest, indicating that
asset abuse could be dim inished under the share contract.
Finally, in the case of the share tenant, the parties face the problem of
m easurem ent and division of the output. Again, the literature is virtually silent on
this aspect of the subject, with the noted exception of Allen and Lueck (1990). They
hypothesized and found support for the notion that w here the costs of m easurem ent
and valuation w ere high (with crops such as pasture and hay) the parties would
p refer the cash lease.

Similarly w here these functions could be perform ed with

m inimal cost, the choice of share agreem ents would prevail. F or example, w here the
product such as w heat is m arketed in its entirety at the local elevator and is of
uniform quality, the problem is minimized.

Indeed, it is not uncom m on for the

elevator to provide the service of dividing the proceeds of the delivered grain and

37
paying th e landlord directly. H ow ever, crops such as hay and p asture are much m ore
difficult to m easure. Quality may also vary significantly (th at is, the crop may not be
fully hom ogeneous). This could have a bearing especially if the crop is delivered "in
kind" to th e landlord o r purchased by the ten an t for feed. In som e cases there is
substantial m arketing to be carried out which may or may not be included in the
share agreem ent.
Transaction costs, especially the costs of supervision identified by Cheung,
provide defensible grounds for choosing the share contract. T he question is, does this
issue have sufficient im pact in the leasing environm ent to dom inate the opposing
forces? Can we point to the transactions costs as the m otivation behind the choice
of the share contract in spite of the disincentives discussed above?

2.10.10 Service Extraction
Schickele (1941) in discussing efficiency conditions, alluded to the costs
associated with the durable assets, as being relevant to the maximizing conditions.
His point was th at th ere m ust be som ething missing in the traditional m odel that
overlooked the im portance of the contribution of the durable assets. Alston, in a
discussion of supervision costs, com m ented on the landlord’s incentive to supervise
the use of durable assets in o rd er to avoid ". . . careless or excessive use.

. .

[which would] . . . result in the depreciation of such assets."
Scott (1982) addressed the issue of costs in a slightly different way.
Recognizing the problem of equating the m arginal costs and returns as in the Adams

38
and R ask m odel he states,
"Also, A dam s and R ask do not consider the returns to the landlord and
ten an t from fixed resources. A lthough the theory is correct about the
decision on th e am ount of variable resource to use being determ ined
by m arginal returns, this is tru e only within limits determ ined by fixed
resource costs and returns. This is one case w here fixed costs and
returns are relevant in the decision process, because the range in
shares u nder the ideal lease which either the landlord or the ten an t can
profitably accept is not only determ ined by m arginal costs and returns
but also by returns to fixed costs both before and after the change."

U nfortunately Scott does not take this issue any fu rther and does not elaborate on
the perceived problem .
Robison and B arry exam ined the problem in their discussion of general
efficiency of the leasing agreem ent prior to considering the effects of risk.

They

proposed a m odel th a t incorporated the change of the asset value in the landlord’s
maximization problem . T heir m odel is included here with notation changes to be
consistent with the earlier discussion:

<1

A

w here q is output, A a unit o f land, and z fixed inputs other than land.
They then incorporate a term th at represents the change in value of the land
over the production period. T he landlord’s income would then be:

y L = (i - u )\p A ^ , z) -

A

A

+

dVlML)]
dL

39

Here V represents the value of the land Robison and Barry maintain that the term

dF[/l(X)]
BL

is negative as production extracts value from the asset unless reinvestm ent is m ade.
This point deserves som e discussion.
Raw land alone has a basic m arket value. However, for agricultural purposes,
the cultivation, fertilization, incorporation of organic m atter into the soil, and/or
application of m inerals (such as lime) can im prove the productivity and in turn, value,
of the land. In addition, these steps require time for the working of biological and
chem ical transform ations to occur— som etim es several years. Similarly, neglect of
fertility m aintenance by failing to perform these functions will result in a diminishing
of the productive potential or value of the land.

Consequently, the cultivation

practices perform ed have the potential to either im prove or diminish the value of the
land. T h ere is a great potential for com plem entarity in som e cultivation practices to
enhance not only the current crop, but the long term value of the land. O n the oth er
hand there is also considerable potential for substitution of long term m aintenance
for current crop needs or enhancem ent. T herefore, the sign of the final term in the
equation will depend on the practices of the cultivator of the land and can be
positive, negative or zero.

Obviously, w here this term is g reater than zero, the

landlord stands to gain.
In any event, their first o rder condition for the landlord:

40

^viAim
8I2( 1 - a)

reveals th at th e landlord will have an incentive to choose a share cropping agreem ent

w henever the term

is negative, causing the m arginal product to be
d L \ 1 - a)

greater than under cash rent.
Allen and Lueck (1990) encounter this issue in attem pting to develop a model
of contract choice betw een crop share and cash ren t leasing agreem ents.

They

recognize the potential for th e ten an t to adjust variable inputs to alter the rate of
service extraction from the durable asset, stating that, particularly in the case of cash
rent, th e ten an t will have the potential and incentive to "overutilize any inputs
supplied by the landowner". Still they overlook the source of their argum ent or at
least fail to recognize the nature of the services from the durable asset. For, in the
following paragraph, they state th at "landowners supply just land and no other
services".
This author will contend th at the land ow ner will, and does, engage in actions
to both m aintain and enhance the value o f the durable asset, land. F urtherm ore, the
landowner, recognizing the potential for extraction of services (and therefore w ealth)
from the land, takes this into consideration in reaching the term s of the contract and
the particular individual to do business with.

41

2.10.11 Exploitation of the land
D ue to the potential for the short term ten an t to extract w ealth from the land
by either, (1) not replenishing soil nutrients, (2) not practicing erosion control, or (3)
excess use of the asset through intense use of variable inputs, it is widely
acknowledged and often hypothesized in the literature that the ten an t will do so
(Schickele 1941, Lee 1980).

O f course this assumes that there is no internalized

m oral standard that prevents the ten an t from pursuing myopic self interest.
S haban (1987) found th a t input and output intensity varied with contract type.
Specifically, tenants who both owned and rented land applied non shared inputs, such
as labor, m ore intensely on owned land.

Fertilizer application, however, was less

intense (approxim ately 10%) on owned land, but yields w ere higher. Assuming the
land was of com parable inherent fertility, this suggests a degree of substitutability
betw een inputs and a willingness of the tenant to allow fertilizer (shared on rented
land) to substitute for other (unshared) inputs.
M ore recently Allen and Lueck (1990) proposed that the potential for
exploitation of the land (landlord) played a focal role in governing the choice of
contract. Using data from the N ebraska-South D akota 1987 survey they showed that
cash ren t contracts w ere m ore com m on under conditions w here the exploitation
potential was low. F urtherm ore, share contracts prevailed w here conditions were
m ore susceptible to exploitation.

Som ewhat disturbing is th at they propose and

defend these findings using a model that assumes joint profit maximization which

42
raises the question, "If the agents are maximizing jo in t profits, would they not choose
a contract to this end rath e r th an exploiting each o th e r’s interests?"

2.10.12 Social Distance
The traditional ap p ro ach to contract enforcem ent employs the use of
supervision with accom panying penalty for shirking.

Disciplinary action such as

dismissal or contract cancellation provide m otivation for satisfactory perform ance.
In several of the works consulted, the authors point out that one of the tools available
to the landlord to ensure the perform ance of the ten an t is the short term agreem ent
(Johnson 1950, C heung 1969, R eid 1976 and Sutinen 1975). T he argum ent is that
w here the agreem ent is short-term and com petition exists am ong potential tenants
for the sam e parcel of land, the th reat of dismissal or refusal by the landlord to
renew the contract will provide an incentive for the ten an t to perform .
Again the evidence from the N ebraska-South D akota study suggest that, at
least in US agriculture, m ore often than not, the ten an t encounters little or no
com petition when negotiating a lease.

T he question then arises, "Is the te n an t’s

perform ance really m otivated by the short term agreem ent or is(are) th ere some
o ther factor(s) involved?" F u rtherm ore, approxim ately one third of the tenants and
landlords rep o rted their agreem ents as "multi-year" contracts.
A n alternate explanation has long been acknow ledged but not incorporated
into the theory. T h ere is a definite tendency for relatives and close friends to act
altruistically.

Schickele (1941) observed,

43
"... the closer, the m ore harm onious the personal relation, the m ore
evenly balanced the m anagerial contributions of landlord and tenant
are.
If this harm ony betw een the two partners is perfect, the
"entrepreneur" u nder stock share tenancy can theoretically be
considered as one person who behaves according to the p attern
outlined for the ow ner-operator."

B aron (1983) cites R eid (1975) and W inters (1974) as evidence that a m ajor
m otivating factor in the prevalence of the share leasing agreem ent was the com mon
interest of the parties. Joint profit maximization constituted the grounds for using the
talents of ten an t and landlord to achieve an altruistic result. While addressing the
issue o f m oral hazard, A rrow (1968) proposed that the m ost effective way to prevent
adverse behavior would be to develop " the relations of trust and confidence betw een
principal and agent . . .

so th at the agent will not cheat even though it may be

’rational econom ic behavior’ to do so."
R eid (1976) expands on the intricacies of the crop production process,
identifying an o th er interesting peculiarity. First, he identifies the three m ajor stages
of crop production (planting, cultivating and harvesting) and the fact that there is
potential for (at least partial) substitution of productive inputs betw een stages. F or
exam ple, intensive w eed control efforts during cultivation will diminish or possibly
even elim inate som e cleaning steps at the tim e of harvest. H e notes that conditions
in one stage will m otivate input appropriations in later stages. An exam ple of this
would be w here ideal growing conditions early in the season necessitate em ploym ent
of special crop drying techniques at harvest time.
As the production of th e crop progresses both the landlord and the ten an t

44
have a vested interest in the outcom e under a share agreem ent.

Each has an

incentive to continuously evaluate the rem aining variable inputs in light of past
conditions and anticipated future events to optim ize the value of the developing crop.
O ptim al application levels of the rem aining inputs will vaiy depending on the p a s tand anticipated fu tu r e - conditions.
T he share agreem ent offers the kind of flexibility th at allows the ten an t to
adjust with the changing conditions. T he theoretical stipulation of variable (labor)
input quantity and intensity proposed and defended by C heung (169) and Newberry
(1974) does not account for this im portant aspect of the agreem ent.

Indeed the

simplicity of the existing theory leaves no room for such ad ju stm en ts- assuming that
the production process is known, fixed, and

optim ization is merely a m atter of

following a recipe. T he features of the share agreem ent allow both the ten an t and
landlord to benefit from the graces of nature as well as share in the hardships that
may occur. This is not m erely a m atter of variation in quantity. The quality of the
crop is subject to num erous m anagem ent decisions.

Furtherm ore, the level and

intensity of labor and other variable inputs required to salvage a failing crop or
maximize the value of a bum per yield, may change from day to day. Interestingly,
Allen and Lueck (1990) assum e joint profit maximization in their model but offer
very little in the way of grounds to support this assum ption. Johnson et al. (1987)
found a correlation betw een the use of the verbal contract and family ties.
Alston (1984) conjectured that supervision costs might be reduced if those
supervising w ere relatives of th e landlord, thus diminishing the cost of supervising the

45
supervisors. His findings su pported this notion.
Robison (1987) investigated the im plications of social distance on the outcom e
of contractual agreem ents. H e considered investm ent in public goods, externalities,
and horizontal integration.

T he synergistic benefits of social closeness w ere

dem onstrated using a variety of social investm ent problem s. O ptim al investm ent and
outcom e results w ere derived, showing th at social closeness is a necessary condition
for app ro p riate investm ent in goods o f this type. O f interest to this study was the
derivation of a condition showing that the determ ination of th e crop share depends
(in part) on the degree of social closeness of the contracting parties. This concept
will be incorporated into the developm ent of a m odel in C h ap ter 3.
T he im portance of social relationships was fu rther exam ined by M arcelo Siles
in a study of lending behavior by small town banks in M ichigan (Siles 1992). H e
clearly dem onstrated attitudes by lenders to offer both lower rates and a willingness
to lend m ore money to individuals considered to be socially close. T he scope of the
social closeness issue has also been addressed by R obison and Schmid (1991) in a
p ap er finding that social relationship and identity of the buyer and seller affects price.
It appears that the literature is beginning to docum ent an area of study with
potentially im portant econom ic implications.
done.

Very little em pirical work has been

O ne aspect of this study will be to exam ine the role of socially close

relationships in establishing contractual agreem ents in leasing of land.

46

2.11 Summary
V arious works have been exam ined representing th e views of researchers’ in
explaining the workings of the leasing agreem ent under a variety of circumstances.
T he majority of the effort has focused on the paradox of the share agreem ent and
the seem ing conflict faced by the tenant. Essentially the problem s stem from the
complexity of the lease agreem ent and incom pleteness of the contractual agreem ent.
Costs arise as alternative m odes of enforcem ent are considered. The potential for
social closeness to alter incentives or otherwise reduce these costs appears to m erit
further investigation.
In the next ch ap ter the ideas gleaned from the literature will be incorporated
into a m odel of leasing. The m erits of the different approaches will be exam ined and
w eighed relative to appropriateness for the study at hand.

CH A PTER III
M O D ELIN G T H E LANDLORD-TENANT RELA TIO N SH IP

3.1 Introduction
In th e previous ch ap ter the land ten u re literature was exam ined. Conflicts
w ere observed relative to established, traditional theory and em pirical evidence.
R esearchers seem to be som ew hat perplexed by the prevalence and persistence of
share leasing. T he authors consulted used a variety of m eans to point out that the
crop share agreem ent is inefficient and not a logical choice of profit maximizing
landlords. F urtherm ore, tenants would be m otivated to m ore efficient use of labor
and o th er resources with the cash agreem ent.

N um erous attem pts were m ade to

offer theoretical alternatives, b u t no clear line o f reasoning with supporting evidence
has em erged. Specifically, th e question rem ains, "Why does the crop share leasing
agreem ent persist?"
M ost prom inent in th e argum ent against the use of the crop share lease is
m otivation of the ten an t to eq u ate marginal use of inputs with the te n an t’s marginal
output.

T he ensuing result is an inefficient use of assets both durable and

expendable. Theoretically, the landlord receives less retu rn than otherwise would be
expected and may suffer exploitation of the land through mining of the soil or neglect
with respect to m aintenance and conservation. In the short run the ten an t may do

47

48
well as w ealth is extracted from the land, but this is not a perm anently tenable
position. Sharing of variable inputs provides only a partial answ er to this problem
since th ere are some variable inputs such as labor or fuel th at are virtually never
shared.
M uch of the literature results from those who have taken up the challenge to
provide theoretical alternatives. O f the alternative explanations offered, th e sharing
of risk seem s to provide the most fundam entally sound argum ent for th e choice of
the share contract. U nfortunately, the em pirical evidence has not been convincing.
Conflicting results suggest th a t some other phenom enon is dom inating the ch o ic e at least in som e cases. H ence the problem rem ains unsolved.
The area showing the greatest potential for further research in answering this
question is th at dealing with social distance of the parties involved in the contract.
A nticipation of potential gains from cooperative use of resources, particularly when
those resources are seen to be com plim entary betw een the landlord and tenant, may
provide sufficient motivation for the choice of the crop share contract.
In the following pages a simple m odel of leasing is presented. T he effects of
risk will be incorporated. A case will then be m ade for the inclusion of social capital
as an im portant motivating factor in modeling the choice of tenure contract. Finally,
transaction costs will be considered.

49

3.2 The model
T he choice to lease, o r not to lease is ultimately the landlord’s, how ever the
alternatives are som etim es n o t very agreeable. N evertheless, it will be assum ed that
the landlord is prim arily responsible for offering the contract and therefore decides
on the type of contract to offer. However, it is understood th at the offer is m ade in
anticipation of the acceptance of the sam e and subsequent behavior of the tenant.
This latter aspect will influence the landlord’s decision or choice of contract type. In
any event for our purposes, th e landlord will choose the contract type and the tenant
chooses the levels of production inputs, which crop is to be grown and farming
practices to be followed.
M odeling contract choice involves exam ination of the alternatives faced by the
ten an t and th e landlord and weighing the respective advantages and disadvantages.
It is assum ed th at both parties will be m otivated by the desire to satisfy wants and
needs, w hether directly or indirectly. Specifically, the potential exists for either party
to experience satisfaction through enhanced well being of the other is allowed.
R epresen tatio n of satisfaction will be through the greek symbol pi ( it) but will have
this b ro ad er m eaning than the traditional "profit" although this latter term will also
be used in the discussion.
In M ichigan the contract types have precipitated to th ree main categories: cash
rent, share leasing with som e production inputs shared (usually 50:50, landlord and
ten an t respectively), and share leasing with no inputs shared (usually l/3:2/3, landlord
and ten an t respectively).

F arm m anagers hired on a salary or wage basis are

50
sufficiently rare th at this kind o f contract will not be considered empirically in this
study.
T he problem will then be to

com pare

contract types and evaluate

circum stances that will precipitate the choice of one over the other. T he cash rent
agreem ent will be exam ined first since it will be used as a benchm ark to com pare
alternative results.

3.2.1 The Cash Rent Agreement
U n d er the cash rent agreem ent the landlord is faced with a fairly simple
problem . First, find a ten an t willing to pay the rent. Second, m aintain the net value
of the asset. T he landlord profit function is:

n L = $ - &La - T c

W here re is the landlord’s profit (superscript c refers to the cash lease and subscript

L to the fact th at this is the landlord’s profit). Also, p is the cash rent, 6, represents
the landlords expenditure on a the cost of m aintaining the assets, and T represents
transactions costs associated with the cash agreem ent which includes such things as
contract negotiation, collecting paym ent, and m onitoring the use of the assets. The
landlord’s profit is constrained, subject to being able to secure a ten an t willing and
able to pay the rent. (L ater the ability of the ten an t to pay ren t as a function of the
landlord behavior will be investigated.)
T he te n an t’s profit function in the cash agreem ent is:

51

%C
T = p j{L Z A J ) - wL - vZ - 6 ydi - T c - p
and includes variables:

p = output price
L =

inputs in production for which the cost is not shared

Z =

inputs in production for which the cost is shared

A =

the contribution o f the assets to production

I = contributions from the landlord such as inform ation transfers, shared
resources, and shared financing of inputs costs.
w =

the price of th e non shared inputs, and

v = the price of the shared inputs.
P = cash rent

In o rder to conform to the tenets of m athem atical optim ization we assum e that the
function is concave, fLL < 0, fzz < 0, and that f, z = fZL = 0.

The tenant maximizes profit choosing the variable inputs (which have a secondary
effect on the use of the asset(s)).

52
FO C

f L(L'c ,z ; A ,D

=p

UKXcAJ) = -

Obviously, if none of th e inputs are shared (a com m on condition under a cash
rental agreem ent) the second equation disappears and the optim al level o f inputs
chosen by the ten an t maximizes profit. If the m arket for land to ren t is com petitive,
the landlord can adjust th e ren t up to the point w here the ten an t is just willing to
rent the land.

3.2.2 Summary of the Cash Lease Agreement
In summary if we assum e th at the m arket for tenants is com petitive such that
the profit of the ten an t is zero, then by solving the te n an t’s equation for f) and
substituting into the landlord’s profit equation we get:

n eL = [pf(L'c,Z'c A ,r) - w L - v Z - V

~T c(Cl t )] - bLa - T c(Cti)

C ontract choice will then depend on w hether profit from o th er agreem ents will
exceed or fall short of the gains under the cash agreem ent.

3.2.3 The share agreement
U n d er the share ag reem en t the landlord’s profit depends on the outcom e of
the crop:

53

ic‘L = (1 - a)pf(L,ZA,D ~ ( l - a) vZ - 6 La - T s

w here th e variables are defined as above with the addition that the landlord now
bears a portion of the cost o f the shared inputs.

T he te n an t’s profit is similar:

n sT = u p f(L ,Z A J) - wL - avZ -h ^a - T s

F O C (tenant) are:

f L{ L : x s A ,r) =

—

ap

U L lX s A J ) = -

p

Since 0 < a r < l then w/ap>wlp which implies t h a t / l (Lc‘,Zc‘,A,I) < / l (Ls*,Zs*,A,I) which
in turn implies that the em ploym ent of non shared variable inputs will be less under
the share agreem ent than under the cash agreem ent: Lc > Ls, the norm al result.
W ith output reduced, the ten an t would be unable to pay rent at the sam e level as
under the cash agreem ent.

Result 1: Under the share agreement the tenant has an incentive to apply variable
inputs at a level that is less than that observed under the cash agreement. This
results in an inefficient use of the productive assets.

54
T he landlord would refuse to accept a share agreem ent of this type, opting for the
higher ren t from the cash contract.

3.3 Incorporating risk into the model
T h ere are a variety o f different kinds of risk that could be considered in a
m odel of leasing arrangem ents. T he m ost com m on are listed as follows:

* O utput~ crop failure or bu m p er crop. May be the result of farming
practices, growing conditions or harvest conditions.
T r i c e - unexpected rise or fall in output prices. May be the result of
favorable (or unfavorable) growing and or harvest conditions nationally or
internationally. M ay result from changing dem and.
*Input— unexpected change in input costs. O ne might include the prospect
of changes th at would require the application of a costly input in o rd er to
salvage a crop.
*Incom e~ inability to to lerate downward fluctuations due to dependence on
incom e for living expenses or debt servicing, including inability to tolerate crop
failure on the p art of th e ten an t under a cash rent agreem ent.
T he com mon elem ent in each of these is the fact that the risk averse individual would
be willing to give up so m eth in g - a risk p re m iu m - to avoid the cost of the
unfortunate event. From our basic model we form the te n an t’s certainty equivalent:

r i - a p f iL J M - wL - avZ -|i -

- jy

The term p takes on a b ro ad er meaning, representing all transfers o th er than the
sharing of the crop. If a is zero then p is the cash rent, but if a is g reater than zero
then P can be zero or some o th er am ount. The corresponding certainty equivalent

55
for the landlord would be:

Y ce = (

i

-

f

+ p -

T he landlord maximizes certainty equivalent incom e subject to being able to attract
a tenan t that will pay the rent. If we solve the te n an t’s equation for p and substitute
into the landlord’s equation we get the landlord’s constrained m axim ization problem :

Yce = (1 - « ) M ) + «M -) ~ w L - avZ
~ y ^ M - ) ] 2^

- Tt - TL - F

- ^ [ ( l - « M - ) ] 2a 2

D ifferentiation with respect to a gives:

dY^
= - X j a M - ) ? o 2p + ^ ( l - c O f o O 2* ; = 0

T hen solving for a we get H ieb ert’s result:

XT +
From this the following is inferred. W here the ten an t is risk neutral (Xx = 0)
and the landlord is risk averse, a cash lease ( a = 1) is chosen. W here the landlord
is risk neutral (XL — 0), a is 0 and a wage contract will result.

Evidence of risk

aversion on the p art of both parties results in the choice of a share contract. This
condition is expected to prevail as most individuals are risk averse.

The

56
interpretation is facilitated by the following table:

Table III.l Sum m ary of C ontract Choice U n d er Risk
R isk Aversion Com binations
Al = 0

*L > 0

XT = 0

undeterm ined

cash

X1 > 0

wage

share

represents the m easure of the landlord’s risk aversion.
Ax represents the m easure of the te n an t’s risk aversion.

57
W here th e landlord and ten an t are both risk averse, the certainty equivalent
profit of each is enhanced by th e sharing of risk. F urtherm ore, w here the parties are
equally risk averse, th e optim al way to share risk is half and half. This result gives
the optim al alpha, or division of o u tp u t u nder risk.

Result 2: Where both the landlord and tenant are risk averse, sharing the risk
reduces the cost. The cost of avoiding risk is minimized if equally risk averse
landlord and tenant share this cost equally.

B oth of the above results are simply rew ritten versions of w ork that has
appeared previously in the literature. And, as noted above, empirically both results
have been questioned.

A ttention will now turn to focus on the attributes of

incorporating social capital into the model.

3.4 The effects of social capital
The approach to form ulate a m odel that incorporates the effects of social
capital will be p attern ed som ew hat after that used to introduce risk. Som e of the
features of social capital have effects similar to those observed in the risk model.
This is to be expected since reducing risk has the sam e kind of benefits as building
social capital, nam ely transaction cost reduction through sharing. H owever, the social
capital m odel must also cap tu re the feature of income enhancem ent through the
sharing of resources, inform ation, and experience. F urtherm ore we expect a cost
savings to occur with a reduction in costs associated with such things as monitoring,
bargaining, and exacting paym ent.

58

3.4.1 Features of the social closeness factor
R obison and Schmid (1992) identified four potential m otivators in addition to
the self interest principle endorsed by the economics profession. They argued that
satisfaction could be derived from success or happiness of an o th er for whom one
cares on a personal basis. T he conjecture would then be th at augm entation of the
o th e r’s success and happiness would lead to greater satisfaction or personal utility.
Individuals could therefore be expected to invest in things that would enhance the
w elfare of close friends and family.
They further observed that contribution to community tends to raise the level
of satisfaction of the contributor when community services are used.

The

businessm an who contributes to the construction of the downtown park may feel
g reater satisfaction in enjoying the resulting aesthetic environm ent than the one not
contributing.
Similarly, efforts to invest in establishing im proved relationships with those
m ore distant may enhance well being. A case may therefore be m ade that the one
who succeeds in befriending an enem y may enjoy the outcom e as much or m ore than
the one who conquers a foe. A nd finally, they discuss briefly an issue of oneness with
a perceived ideal of self—being true to oneself, that provides a degree of satisfaction.
E ach of these constitutes a form of social capital.
It is the first item, the issue of deriving satisfaction from the welfare of friends
and family, that will be considered here. However, in addition to the vicarious utility
draw n from the well being of th e other, this author would argue that the static model

59
does not tell the whole story. T h ere is an o th er dynamic aspect to this condition that
was not addressed by R obison and Schmid. It is the prospect o f rep eated interaction
betw een individuals and a resulting in terd ep en d en t confidence.

Active, m utual

exchange of social capital generates a synergistic result. R eciprocity prom ises an even
greater degree of satisfaction. It may be th a t the propensity to socialize is greater
than A dam Sm ith’s propensity to truck and barter.
T he econom ic outcom e of th e relationship is evidenced in two ways. Cost
savings occur as the resources of two individuals are joined to avoid mistakes.
Confidence in the m otivation and integrity of the o th er elim inates policing,
supervision, or m onitoring costs.

O th er transactions costs may be reduced by

avoiding duplication of effort in counting and dividing or verifying the quantity and
quality of output.
T he second benefit is observed as income enhancem ent.

Exchange of

inform ation, resources, and experience increases the productivity of the respective
inputs and in turn the potential profitability of the endeavor.

The crop share

agreem ent allows for a dual incentive to participate in this kind of exchange. Not
only might one party be able to augm ent the income of the other, but in so doing,
helps him self (or visa versa depending on m otivation).
Furtherm ore, the landlord and ten an t will likely enjoy some diversification of
resources.

T he benefits of a g reater pool of resources for productive skill and

adaptability may also enhance production.
In summary, the socially close landlord and ten an t would seem to have a

60
potential to gain from the jo in t use of assets, skills and oth er resources. T he share
agreem ent would tend to provide the environm ent for the m ost productive use of the
same.

61

3.4.2 Conceptualization of the Social Closeness Factor

Total Cost

MC Tenant's Share

MC of Landlord or Tenant
with Cash Rental Agreement

Composite input L

Figure 111,1 Social closeness and the accompanying attitudes have the
effect of shifting the dem and for variable inputs out and upward.

In Figure 111,1 the concept is presented geometrically. Two cost structures are
conceptualized. Line A represents the marginal cost of inputs w faced by the share
tenant and corresponds to w/ap calculated above. Line B represents the m arginal
cost of inputs w faced by the ten an t renting for cash and corresponds to w/p

62
calculated earlier. T he reason th at A is above B is because 0 < a < 1. O ne might
otherw ise explain the relationship by stating that since the ten an t only receives a
portion ( a ) of the output, the m arginal retu rn to the unshared inputs is lower. W hen
equated marginally this com pares to a higher cost structure faced by the share tenant.
Line D L represents the dem and for inputs. Suppose that in the presence of
social closeness the inputs can be used m ore effectively or efficiently. This would
result in a shift of the production function, causing a change in the value of the
variable inputs and increasing dem and.

This, in turn, would be equivalent to an

outw ard shift of the dem and curve to D L’. Essentially w hat this m eans is th at in the
presence of social capital, the value of the inputs in the production function change
and are no longer hom ogenous with the inputs in the function w ithout closeness.
L ester Thurow (1983) envisioned this sam e effect in a discussion of the labor m arket.
U n d er various circum stances the w ork effort and corresponding productivity of the
laborer may vary with such things as the perception of fairness, m aking it impossible
to assum e that all labor hours in the production function are hom ogenous and
independent of the social relationship.
Levels of output by cash and share tenants are rep resen ted by Q, and Q 2
respectively.

D epending on the m agnitude of the shift in dem and for inputs, or

increase in productivity of the inputs, due to the presence of social closeness, it is
conceivable that total output under the share agreem ent (represented by Q 3) could
exceed that achieved under the cash agreem ent. The precise way in which the level
of use responds to increased dem and is not known, but the graph helps to envision

63
how even with the higher cost, the level of use could increase. This is precisely the
effect th a t the social capital com ponent is proposed to have on the model. Such an
effect would satisfy th e landlord’s expected retu rn for the use of the assets and the
ten an t would benefit from increased income.
T he term th at will be introduced into the m odel to represent social capital
m ust have com ponents to correspond to attitude, ability, and degree of interest or
willingness to devote effort.

T he following form ulation em erged after trying a

num ber o f alternatives:

= (l-cc)p[f(L 2M

+ G J . W A l ~ c c ) p T] - v Z + p - TL

where:

=

the social closeness the ten an t feels tow ard the landlord,
O <0<1

M-A.)

=

the m anagem ent skill or ability possessed by the tenant, being
a function of the productive inputs

(1-a)

=

the perception of the tenant relative to his ability to help the
landlord

pT

=

the contribution of the tenant to the production activity (this
value is viewed as a percentage), (p T + p, = 1).

Essentially what is being portrayed is a term that im pacts upon the production
function. T he core com ponent is M 7> the m anagem ent skill, ability, effort in labor,
or inform ation that the ten an t can contribute to the production process to augm ent
the incom e of the landlord. T h e closeness coefficient (0 ) represents his willingness

64
or propensity to do so.

T he extent of effort is fu rth er w eighted by the te n an t’s

perceived incom e enhancem ent to the landlord (1-ct). This m eans th at the ten an t will
weight his actions by th e perceived (as opposed to actual) benefit they have to the
landlord.

Finally, the entire contribution is also weighted by the corresponding

resources the ten an t has com m itted to the process (p).
In a similar fashion th e landlord can contribute to the incom e of the tenant:

tzt

= ap[f(L,Z^iJ) + ®LM L(.)a p L] - wL - vZ - p - T t

T he natu re o f the contribution is som ew hat different for the landlord.

Transfers

might include such things as inform ation relative to the land and local environm ent.
Such transfers are normally virtually costless in term s of expense to the landlord but
may be useful in enhancing production or avoiding costly errors. T he landlord may
also have at his disposal capital or physical resources which he may m ake available
to the ten an t at little or no additional cost. The landlord’s decision to participate or
help again depends on the closeness coefficient (®L), his ability to do so, and
possession of resources. F u rth erm o re his actions will be w eighted by the perceived
benefit th e contribution will have to the ten an t ( a ) and will also depend on the
resources the landlord has invested in the production process ( p L).
T he sam e p attern (used earlier for risk) of solving the te n a n t’s equation for
P and then substituting into th e landlord’s equation yields:

@jMTp T
a = --------------------------&7MTp t + ®LMLpL
This result has num erous meanings. First, suppose th at M r = M L and th at pT = p L.

65
T hen, plugging in various values for 0 gives the following:

Table III.2 C ontract Choice U n d er C onsideration of Social D istance
C om binations of Varying D egrees of Social D istance
o
II

Landlord

®L = 1

0 < &L < 1

T enant
o
II

undeterm ined

wage

wage

&, = 1

cash

share

share

0 < &T < 1

cash

share

share

By exam ining the results using the extrem e values for the closeness factor, it is
evident th at the share agreem ent will be preferred by participants that experience
som e closeness. T he benefits ensuing from the closeness factor are maximized when
shared equally. Supposing th at this factor dom inated the contract choice decision,
we obtain the following results:

Result 3. The share agreement will be chosen by parties that have some degree of
closeness.

Result 4. Cash rent will be chosen under circumstances where the landlord has no
closeness toward the tenant.

Result 5. A wage agreement will occur where the tenant has no closeness toward the
landlord.

66
Now, supposing that &r = 0L and that pT = p D the result of plugging in values for
M will b e examined.

Table III.3 C ontract Choice U n d er C onsideration of M anagem ent R esources.
C om binations of Varying D egrees of M anagem ent R esources
o
II

2«

Landlord

Ml = 1

0 < Ml < 1

T enant
o
II

£

undeterm ined

wage

wage

Mt = 1

cash

share

share

0 < Mt < 1

cash

share

share

Again, the value of the share contract is evident. F urtherm ore, if skills are similar,
the optim al a will be one half. And, assuming dom inance of this factor, we obtain:

Result 6. The share agreement will be chosen where both parties have some degree
of management resource to offer.

Result 7. Cash rent will be chosen under circumstances where the landlord has no
management resource.

Result 8.
resource.

A wage agreement will occur where the tenant has no management

Finally, an exam ination of the results w here various values for p are considered. In
this case &j• = &L and M T = M L.

67

Table III.4 C ontract Choice U n d er C onsideration of Investm ent R esources
Com binations o f Varying Levels of Investm ent R esources
o
II

Q.

Landlord

Pl =

1

0 < Pl < 1

T enant
pT =

0

undeterm ined

wage

wage

pT =

1

no agreem ent

share

share

no agreem ent

share

share

0 < pT < 1

In this case the in terpretation is slightly different. If n either party has resources to
invest, no agreem ent will be reached. For exam ple, if the landlord’s investm ent is
zero, a = 1 m eaning that the ten an t receives all the o u tp u t from no land.
F urtherm ore, the landlord would not be paid any rent, the equivalent to no
agreem ent. O n the other hand if the tenant m akes no investm ent of resources, a
wage agreem ent would still be an alternative.

Logically, contractual agreem ents

betw een parties with nothing to invest in the agreem ent have little m eaning. The
m ore interesting part of the outcom e is the potential for sharing the ou tp u t at rates
relative to the corresponding level of investm ent.

C onsequently, optim al share

arrangem ents could conceivably be som ething other than the traditional 50:50. This
reasoning yields the following results:

68

Result 9. The share agreement will be chosen where both parties that have some
degree of investment resources.
Conversely, w here the landlord is w ithout resources (no land) result ten

addresses a m ute point.

Result 10. No contract will be chosen under circumstances where the landlord has
no investment resources.

Result 11.
resources.

A wage agreement will occur where the tenant has no investment

Result 12. Optimal share arrangements will reflect the comparative level of
investment resources attributable to each party.

T he next step is to consider com binations of the above results. It is readily evident
th a t total absence of any one of the com ponents on the p art of either of the
participants will cause the extrem e results (cash rent or wage contracts) to be chosen.
W ithout m eans to m easure and calibrate both closeness and m anagem ent ability,
interpretatio n of levels betw een zero and one results in m ere conjecture.
It is further understood that the landlords and tenants will bring significant
variety in com bination of resources to any particular agreem ent. In some cases there
will be conflicting com binations.

F or example, a landlord may have a wealth of

knowledge and skill and a willingness to participate with the tenant but be
constrained by time, health, or distance. Cumulatively it will not always be readily
evident which factor may dom inate the choice of contract.
The scientific approach prescribes that m eans be devised to control for all

69
variables except the one being studied and then exam ine the way the one variable
behaves under m easurably changing conditions. T he above m odel and ensuing results
suggest th at num erous variables are influencing the choice of contract. T he task then
is to determ ine under w hat conditions certain choices tend to dom inate others. In
o rder to do this we will first narrow the range of variables to be studied and then
develop testable hypotheses from the predictions of the model.

3.5 Hypotheses
The preceding results obtained from the m odel will now be exam ined. Those
lending them selves to testing with em pirical data in the scope of this study will be
transform ed into testable hypotheses. In the case of the first result:

Result 1: Under the share agreement the tenant has an incentive to apply
variable inputs at a level that is less than that observed under the cash agreement.

it is recognized th at this results in an inefficient use of the productive assets;
and will be taken as given.

3.5.1 The effect of risk aversion
In the prelim inary version of the m odel, it was observed th at risk was the
variable u nder study. From the initial calculations the following result was derived:

Result 2: Where both the landlord and tenant are risk averse, sharing the risk
reduces the cost. The cost of avoiding risk is minimized if equally risk averse
landlord and tenant share this cost equally.

70
T he issue of risk aversion m ust be divided according to landlord and tenant interests
respectively. Simply put, the risk averse landlord will p refer that the ten an t b ear as
m uch risk as possible. Similarly the risk averse ten an t will p refer that the landlord
b e a r as m uch risk as possible.
In this case the reference is to risk associated with stochastic events that affect
price and production. As previously discussed, a cash agreem ent dictates that the
ten an t bears all the risk, providing the landlord with a cash rent w hether the crop
succeeds or not. By contrast the share agreem ent would necessarily involve both
parties in the success of the crop since income would be tied to revenues from the
production and sale of the same.

H ence the risk averse ten an t would p refer the

share agreem ent w here the landlord shares in production and price risk reducing the
te n a n t’s exposure.

Conversely, the risk averse landlord would prefer the cash

agreem ent deferring all the risk to the tenant.
F rom this reasoning we generate two hypotheses:

(1)

He:

Risk, relative to the tenant’s preference for secure income, makes no
difference in the choice of contract.

(2)

H0:

Risk, relative to the landlord’s preference for secure income, makes
no difference in the choice of contract.

It is anticipated that both of these hypotheses will be rejected.

Testing of these

hypotheses will result from survey data gathered. R espondents would be asked to
self evaluate their aversion to risk on a scale of 1 to 10, 1 being high risk associated
with potentially higher, but insecure, income; 10 being low risk with the potential for

lower m ore secure income.

T he data will serve as observations for a variable

included in a logistic regression.

3.5.2 The effects of social closeness
In term s of social relationship from the m odel we obtained th e following three
results:

Result 3. The share agreement will be chosen by parties that have some degree of
closeness.

Result 4. Cash rent will be chosen under circumstances where the landlord has no
closeness toward the tenant.

Result 5. A wage agreement will occur where the tenant has no closeness toward the
landlord.

Since the scope of th e study is to investigate the existence and persistence of
the share agreem ent, im plications of R esult 5 will not be investigated. Results 3 and
4 will be com bined into one hypothesis:

(3)

Hc:

Social closeness makes no difference in the choice of contract type.

It is expected that this hypothesis will be rejected. Proxies for genuine closeness will
include:

1. Family relationships.
2.

Stated close friendships.

3. Evidence o f social interaction.

72
A problem lies in the perception of degree of closeness. Since for many, a friend
may be closer that a relative, no attem p t will be m ade to scale the responses. It will
simply be assum ed th at friends and family are considered close. Strangers, at the
outset of the agreem ent, will thus be considered distant. F urther, it is assum ed th at
individuals with antipathetic relationships will not en ter into any agreem ent. This
causes some problem in m easuring the full dim ension of the variable since we d o n ’t
have data on those who may have considered the leasing contract but chose n o t to
engage in o rder to avoid aggravation, irritation or hostility.
T he range of the social distance variable may be perceived to have positive,
neutral, and negative regions.

D ue to limitations, only the positive and neutral

ground may be examined.
T he responses to question 7 will be transform ed so that options 1, 2, and 4 will
be considered "close" and options 3 and 5 "distant". This variable will then be used
with others in a logit regression with 10a as the dep en d an t variable.

3.5.3 The effects of m anagem ent skills and resources
The farm ing of land is especially susceptible to ap p ro p riate use of
m anagem ent inputs. T he complexity of production requires a constant m onitoring
of the progress of the crop and adjustm ent of timing and application of variable
inputs to coincide with growth and maturity of the crop for optim al results.
Production can thus be viewed as an art as well as a science.
The skills associated with the art of farming can be envisioned as valuable

73
contributing factors th a t a landlord or ten an t brings to a tenure agreem ent.

The

ability and propensity to use these skills constitute a resource th at enables the
individual to en ter the non m arket region of negotiation. T he share crop agreem ent
provides the environm ent for the optim al use of these skills (w here they are
possessed by both parties).
E arlier we obtained th e following results from the model:

Result 6. The share agreement will be chosen where both parties that have some
degree of management resource to offer.

Result 7. Cash rent will be chosen under circumstances where the landlord has no
management resource.

Result 8. A wage agreement will occur where the tenant has no management
resource.

Following the sam e reasoning expressed above, R esult 8 will not be considered
in this study. Further, it is assum ed that for a ten an t to take responsibility of the
m anagem ent of the farm, the ten an t will necessarily have m anagem ent skills and
resources.

H ence, the variable influencing the choice of contract will be that

representing m anagem ent resources of the landlord.

Following this reasoning we

hypothesize:

(4)

Hc:

The managerial resources of the landlord and the disposition to
provide such make no difference in the choice of contract.

74
It is expected that this hypothesis will be rejected. Proxies for resources, ability and
disposition of the landlord will include:

1. Farm ing experience of the landlord
2. S tated participation in decision m aking
3. Willingness of the landlord to participate.

3.5.6 The effects of transaction costs
This topic, though discussed at length in the literature review, does not lend
itself well to the behavioral (m athem atical) m odel p resented above. Included in the
m odel is a

simple

additive

transaction

costs term ,

differentiation of a constant elim inates the term

how ever

m athem atical

from the result obtained.

N evertheless, transactions cost are real and logically im pact the choice of contract.
H ence, the following hypothesis will be tested:

(5)

H0:

Transactions costs (defined as the costs associated with carrying out
the terms of the agreement— contracting, bargaining, ensuring
performance, exacting payment, etc.) make no difference in contract
choice.

It is expected th at this hypothesis will be rejected.
Sub-hypotheses will be tested in this case.
Since some crops tend to be m ore difficult to m easure and divide (as required
under a share agreem ent), it is expected that share cropping will be m ore com m on
w here field crops are grown. Land m ore suitable for forage crops such as hay and
pasture will tend to be cash rented.

75

(5.1)

He:

Type of crop makes no difference in the choice of contract.

It is expected th at this hypothesis will be rejected.
D uring the course o f the study ano th er indicator of the im portance of
transaction costs in contract choice was discovered.

N um erous individuals

com m ented on the difficulty o f keeping records w here multiple parcels of land were
being leased. H ence it was decided to test the following hypothesis:

(5.2)

H0:

Number of parcels makes no difference in the choice of contract.

It is expected th at this hypothesis will be rejected.
A prim ary factor affecting the ability of the landlord to contribute to the
agreem ent as well as m onitor the activities of the tenant is the frequency visits by the
landlord to the farm. In keeping with a num ber of oth er studies w here this issue
form ed the focal point o f the transaction costs, the following hypothesis will be tested:

(5.3)

H„:

Facility of the landlord to observe and participate in the
measurement and division of the crop makes no difference in the
choice of contract.

It is expected th at this hypothesis will be rejected.
A n aspect of the agreem ent in the share lease has to do with the idea of
uncertainty, not the risk associated with stochastic variables but the uncertainty of the

76
perform ance of the tenant.
It would stand to reason th at w here the landlord has doubts about the ability
or willingness of the ten an t to perform , the landlord would tend to choose the cash
ren t agreem ent.

Obviously, this would cause the ten an t to b ear the b urden of

uncertainty bo th with respect to his own perform ance as well as the stochastic events
of nature.
O n the oth er hand, w here the landlord has confidence in the te n an t’s
perform ance, em pathy could m otivate the landlord to share som e of the risk
associated with production and price.

This may well be envisioned as "fair".

F urtherm ore, w here the skills, ability and effort of the ten an t have proven to be
outstanding, the landlord may wish to "share" in the expertise of the ten an t and the
anticipated gains from his/her labors. O f course this brings us back to the question
w hether the ten an t will exert him- or herself knowing th at a portion will ultimately
go to the landlord.

3.6 Summary
In this ch ap ter a m odel of contract choice was developed incorporating the
influences of risk, social closeness, m anagem ent resources and transaction costs. The
effects of risk have been exam ined previously and th ere is no new outcom e either in
the m odel or in the expected d ata to be collected. Transaction costs effects have
been addressed by other authors and in fact form the basis of several m ajor works
including that of Steven Cheung. Empirically, however, there is little data either to

77
support or refute th e im portance o f this variable. The concept of social closeness
was also introduced and the anticipated effects set forth.
T he prim ary effects of social closeness w ere identified to be two fold. First
th ere is an anticipated effect on the production function resulting in increased
productivity through m ore effective use of the production inputs.

Second it is

expected th a t social closeness will interact with transaction costs resulting in a
reduction of m onitoring costs, and costs associated with errors in m anagem ent,
production and m arketing.
H ypotheses w ere developed in o rder to test the contribution of these issues
in the choice of contract. In th e next chapter the em pirical steps of this study will be
set forth, including the results of data collection.

78

CHAPTER IV

DATA COLLECTION AND ANALYSIS

4.1 Introduction
T he process of data collection and analysis is described.

T he m ajor steps

include drawing a sample, preparing the survey instrum ent, mailing, coding the data,
testing for non response, and then compiling and analyzing the data collected,
including running tests of the hypotheses developed in the previous chapter. The
following pages will detail these steps and provide reasons for decisions in some
cases.

4.2 Obtaining a sample
T he object of the study is to learn specific points about the characteristics of
leasing of farm land in Michigan. Since it is im practical to gather inform ation from
all landlords and tenants in the state, a scientific sam ple will be drawn.

The

inform ation generated from the data collected can then be used to m ake
generalizations about the population as a whole.

79

4.2.1 Sample size
D eterm ination of the sam ple size was done using the form ula for estim ating
a proportion within a certain erro r tolerance, since the goal of the survey is to study
the proportion of farm ers choosing the share contract as opposed to a cash ren t
agreem ent.

Initial investigation indicated that cash rent would be expected to

dom inate at a rate of approxim ately three or four to one. T herefore, reliability of
statistics on groups within the sam ple would dem and that the sam ple be large enough
to provide for ad eq u ate size of these groups, specifically tenants and landlords.
H owever, since the sources consulted for lists had no m eans of selecting only those
tenants or landlords involved in leasing agreem ents, over sampling of the population
would be necessary. Census data estim ates suggested that 40 to 50 p ercen t of the
farm s and/or farm ers w ere involved in leasing.
T he form ula to calculate the required sample, adopted from M arzillier (1990)
is as follows:

z 2LN
n = ------E2

W here n is the required sam ple size, z corresponding to the 95% confidence
level is approxim ately 2, and L and N are the proportion of farm ers leasing and nonleasing in the population. E is the erro r level allowable (.05). T he result, using this
m ethod, indicates that the required sample be 400. By relaxing the confidence level

80
to 90% th e sam ple size could be reduced to 176. Assuming a rate of retu rn of at
least 25% , it would be necessary to mail out approxim ately 1000 questionnaires to
assem ble enough d ata to fall within the acceptable range. H owever, since there was
no way to preselect those involved in leasing, it was decided th at 2000 questionnaires
would be sent, anticipating th a t approxim ately half of the respondents would not be
involved in leasing.

4.2.2 Drawing the sample
O btaining a representative sam ple of Michigan landlords and tenants proved
to b e a fairly difficult task. Initially an attem p t was m ade to follow the procedure of
the N ebraska-South D akota study and seek a sam ple from the Agricultural
Stabilization and C onservation Service (ASCS). However, the ASCS was reluctant
to offer any assistance. They cited changes in the federal rulings on the Freedom of
Inform ation Act making them liable for the release of personal data in their
possession. T he request was consequently denied.
A second request was m ade to the ASCS in such a way that they would be
able to preserve the confidentiality of their list.

It was proposed that the

questionnaire and mailing m aterials would be delivered to their office leaving their
own em ployees to affix labels and mail the survey.

The returned questionnaires

would have no m eans of attaching the responses to an individual.

However, the

ASCS office did not change th eir previous position. Again the request was denied.

81
Failing this approach, help was sought from the Michigan Agricultural
Statistics Service (M ASS). T he MASS proved to be quite helpful and willing to work
with us. They required $3,000 for the list of 2,000 nam es m ade available for two
mailings. T he contract also provided for a follow up telephone consultation with a
sam ple of non-respondents to check for non response error.
Some concern rem ained how ever in relation the quality of the list m aintained
by MASS. T he list was designed to include producers of agricultural com m odities
throughout th e state. By contrast the ASCS list included landowners with an interest
in the farm program s but not necessarily involved in production. It was felt th at the
MASS list might be deficient in landlord listings since these individuals would no
longer be involved in active production. Consequently, a third request was m ade to
ASCS for a listing of corporate or institutional entities in four of the m ost prom inent
agricultural counties of the state since these would not be restricted under the
F reedom of Inform ation Act. They agreed to provide this list for a cost of $90.
The reasoning was th at this sm aller but im portant ASCS list should help
com pensate for deficiencies in the MASS list particularly with respect to landlords.
H ow ever it was acknow ledged that the data would of necessity rem ain sep arate in
order to preserve the integrity of the scientific sam ple drawn by MASS.

4.2.3 Q uestionnaires sent, received, different m ailings, etc.
A pproxim ately 2000 questionnaires w ere m ailed to the MASS sam ple in July
1992. A second mailing to this sam e list took place in August 1992. Also in August

82
a sep arate mailing was sent to the selected list from ASCS. Finally, in late August
a second mailing w ent out to the ASCS list.

In S eptem ber the MASS office

com pleted th e telephone follow up of non respondents.

No m eans was m ade

available to follow up on non respondents from the ASCS list.
Five hundred and eleven of the questionnaires mailed to the MASS list w ere
retu rn ed of which 244 claim ed they were not involved in any leasing agreem ents. O f
the rem aining questionnaires 17 were not filled out properly or w ere unusable. T he
rem aining 250 constituted the actual data sample.

These were coded and the

following results w ere determ ined using the ensuing data.

4.2.4 Testing for non response bias
T h e telephone contact of non respondents by MASS gathered an additional
148 questionnaires 77 of which represented individuals not involved in leasing. O f
the rem aining 71 questionnaires all were sufficiently com plete to be used in analysis.
T hese w ere coded separately from the main data set. Subsequently the two data sets
w ere m erged, using a dummy variable to facilitate identification of the parent data
set, to test for differences.
T he bench m ark variables chosen cam e from the first q u estio n - size of
personal holdings— which should indicate w hether there was a difference between
individuals; question se v e n - num ber of years le a se d - which should indicate any
differences in the duration o f leases and/or experience of the respondents; and
questions eleven and s e v e n te e n - estim ated p er acre value of parcels le a se d - which

83
should indicate any differences in the kind of operation. T able IV. 1 shows the results
of testing of these variables for significant differences in the m eans.

84

Table IV.l Testing for non-response bias in the sample.
V ariable

G roup

Personal
Holdings

M ain
R esponse
D ata
NonR esponse
D ata

Lease term
(yrs.)

E stim ated
M arket V alue
Share
A greem ent

E stim ated
M arket V alue
Cash
A greem ent

M ain
R esponse
D ata

M ean

246.7

Standard
Deviatio
n
312.9

Standard
E rro r

20.1
-.24

256.2

9.45

279.1

7.1

33.84

.466
-.39

Non
R esponse
D ata

9.88

8.2

M ain
R esponse
D ata

875.30

684.8

.996

96.9
-1.2

Non
R esponse
D ata

1092.31

558.9

155.0

M ain
R esponse
D ata

892.3

468.8

36.7

Non
R esponse
D ata

tvalue

1.2
814.3

362.3

Source: 1992 M SU Survey of Landlords and T enants

53.4

85
In each case the /-test shows no significant difference betw een the value from the
m ain response d ata and the phone d ata collected from non responding individuals.
T he phone data carried one interesting peculiarity. T he telephone operators
w ere m ore successful in ensuring th at all the questions w ere answ ered. Q uestion
thirty-six in particular had b een frequently m isinterpreted such that many of the main
d ata set respondents had skipped the question. T he author felt that it would be
useful to include the d ata collected over the phone in o rd er to increase the num ber
of responses available for analysis.

D r.

Les M andersheid was consulted on the

advisability of this idea. H e cautioned that the testing of four bench m ark variables
(appearing in T able IV .l) was not conclusive evidence that there w ere no differences
in the d ata sets but m erely th a t th ere was strong reason to believe that there were
no differences. F urther, to m erge the data would w eaken the pow er of generalization
of the findings.

H owever, he explained that as long as w hat was done was fully

disclosed and docum ented, to m erge the two sets of d ata would not be breaking any
cardinal rule. It was consequently decided th at since it seem ed th at th ere would be
m ore gained than lost by merging, the two data sets would be m erged in o rd er to
provided a larger set of data for testing and analysis.
T he response from th e ASCS list was disappointing. Only 89 w ere returned
of which 48 w ere not involved in leasing and four were unusable. T he data collected
from the 37 useable questionnaires was very similar to that from the MASS mailing.
T he proportion of landlords in the ASCS returns was not significantly different from
the MASS d ata set.

F urtherm ore, oth er variables tested showed no differences

86
com pared to the main d ata set. It was decided th at since this d ata was not p art of
the scientifically draw n sam ple by MASS, that the 37 cases would not be m erged with
the m ain d ata set and hence was not included in any of the following analysis.
F o r the purposes o f th e following analysis, the term respondent will refer to
landlords and tenants w hose questionnaires w ere actually deem ed acceptable for
evaluation.

Similarly, the term responding landlord or responding tenant will be

restricted to the group whose questionnaires w ere considered satisfactory for analysis.

4.3 General Respondent information
43.1 Distribution of respondents
T he data sam ple used in analysis consisted of 97 landlords and 213 tenants
and included the responses of th ree individuals not identified as eith er landlords or
tenants.

These

three

provided

sufficiently com plete inform ation

regarding

preferences under the general inform ation sections that the questionnaire was
deem ed satisfactory to add to the pool. T he effective sam ple for analysis, hence
included the com piled responses from 313 individuals. The acreage totals, reported
by those responding to the questionnaire, are shown in T able IV.2.

87

Table IV.2 A creage O w ned and L eased with Estim ated Value
Acreage

Estim ated
p er acre
value

T otal A cres O w ned by R espondents

72429

$1021

T otal A cres U n d er Leasing A greem ents

66272

$ 971

R ep o rted A creage

Source: 1992 M SU Survey of Landlords and T enants

T hese figures give a sense of orientation with respect to the
num ber of acres rep resen ted by the respondents.
T o fu rth er establish a profile of respondents by income from farming, age, and
sex for landlords and tenants, the corresponding data appears in the following
T able IV.3.

88

Table IV.3 Profile of respondents by income from farming, age and sex.
P ercent in com e from farm ing
R esp on d en t
T enant

less than
30%

50 to 80%

m ore than
80%

totals

101

31

32

41

205

69

13

9

5

96

.170

44

41

46

301

Landlord
T otals

30 to 49%

ag e category in years
R esp on d en t
T enant
Landlord

< 25

25-34

35-44

45-54

55-64

> 64

1

22

64

52

54

17

3

9

13

29

42

sex
R esp on d en t
m ale

fem ale

T enant

201

8

Landlord

83

13

Source: 1992 M S U Survey o f Landlords and T enants

In a few cases (no m ore than four) the respondent was both a landlord and
a tenant. However, the questionnaire was designed to gather the d ata as either one
or the other.

In one case the inform ation was filled out in such detail that two

records w ere created, one for the individual as a tenant and one as a landlord.

89

4.3.2 Number, size, and distribution of parcels of land being reported
T he majority of respondents reported only one parcel of land involved in
leasing. T h ere w ere substantially m ore parcels rep o rted by those involved in cash
ren t agreem ents. Also, as the num ber of parcels p er individual increased, cash rent
dom inated as the contract type chosen. T able IV.4 gives a breakdow n of parcels and
contract type chosen.

Table IV.4 D istribution of parcel holdings by num ber and type
N um ber o f
resp ond en ts by
Contract type

N um ber o f parcels per resp ond en t
1

%

2-3

%

4-5

%

6 & up %

T otal

Share crop

59 6 6 %

21 2 3 %

9

10 %

1

1%

90

Cash rent

127 5 1 %

68 2 7 %

31

12%

24

10%.

250

T otals

186 5 5 %

89 26%

40

12%

25

7%

340

Source: 1992 M S U Survey o f Landlords and T enants

As discussed in C hapter 2 there is a transaction cost factor at issue here. The
share agreem ent requires m ore tim e and effort when it com es to counting, weighing,
dividing, or otherwise m easuring the crop and calculating the rent. W here inputs are
shared the effort investm ent increases. Parties with m ore than one share agreem ent
face another potential cost of keeping the crops from different parcels sep arate until
the ren t can be calculated. This cost may or may not be significant depending on the
n atu re of the crop and the tim e period over which it is harvested. T he data would
suggest th at this issue is a real concern for those with m ore parcels of land.

4.3.3 Length of Lease
T he average length o f lease rep o rted was 9.5 years indicating that th ere is
considerable stability in the leasing of land.

Table IV.5 Lease Length by contract type, respondent and formality

Category

Lease Type

Respondent

Share

Cash

Landlord

Tenant

Oral

Written

Formality

Number

less than 5
years

84

23

61

50

34

56

29

5 to 9 years

76

22

54

18

58

49

26

10 to 19 years

97

15

82

16

81

72

25

20 years or
more

42

6

36

6

36

34

8

Totals

299

66

233

90

201

211

88

Source: 1992 M SU Survey o f Landlords and Tenants

T able IV.5 categorizes the length of the agreem ent com pared to the kind of
lease and the role of the respondent. T h ere appears to be some tendency for cash
agreem ents to continue for longer periods of time.

Also, proportionately

m ore

landlords (than tenants) are reporting leases that have been in existence for short
periods of time.

4.3.4 Formality of Agreement
A majority of lease agreem ents in M ichigan are reached rath er informally as
evidenced by the num ber of oral contracts reported. F urtherm ore, those engaging

91
in crop share leasing are m ore likely to use the oral agreem ent. A large nu m b er of
the leases are renew ed annually. T able IV.6 sum m arizes these details.

Table IV.6 Form ality and renew al o f agreem ents
R enew al Frequency
C ontract
formality

O ral A greem ent
W ritten
A greem ent
Totals

C ontract Choice

R enew ed
Annually

R enew ed
for a Multi
Y ear
Period

C rop
Share
A greem ent

Cash R ent

171

42

63

155

43

43

7

81

214

85

70

236

Source: 1992 M SU Survey of Landlords and T enants

4.4 Cash Rent Agreements
Each respondent was asked to choose one rental agreem ent, or parcel of land
if m ore than one, th at could be considered typical or most representative of his/her
holdings. Those responding to the questionnaire rep o rted on 71 share agreem ents
and 236 cash ren t agreem ents. Clearly the cash rent agreem ent was m ore popular
am ong the responding M ichigan farm ers and landowners.
T he average size of th e parcel of land under a share rent agreem ent was 88
acres com pared to 126 acres for cash agreem ents.

Those with share agreem ents

92
averaged slightly higher estim ated value on the land, $920 p er acre, with a range of
$100 to $5,000, com pared to $875 ranging from $27 to $3,000. The highest land
values cam e prim arily from land being used for orchards or vineyards. Some rath er
unique share cropping agreem ents w ere reported by those with fruit farms. M ore
discussion of this below.
Cash ren t averaged $44.76 p er acre, ranging from a low of $2 to a high of
$175 p e r acre. Overall the p e r acre rent averaged 6% of estim ated land value.

4.5 Crops grown share, cash
T he distribution of crops grown betw een the two groups was alm ost identical.
Field crops including corn, soybeans, wheat, sugar beets, as well as other small grains
accounted for approxim ately 75% of the contracts. Hay and pasture dom inated the
balance with roughly 20%, while specialty crops and fruit filled the rem aining 5%.
T able IV.7 dem onstrates the similarities and differences in crop choice betw een cash
and share crop agreem ents.

93

Table IV.7 D istribution o f C rops grown by contract type
C rop
G roup

Share Crop
N um ber

T otal

Cash R en t

P ercent

N um ber

P ercent

N um ber

Percent

F ield C rops

55

77

172

74

227

75

H ay &
Pasture

12

17

49

21

61

20

Fruit

2

3

9

4

11

4

Specialty
C rops

2

3

3

1

5

2

71

100

233

100

304

100

T otals

Source: 1992 M S U Survey o f Landlords and T enants

4.6 Adjustments
R espondents reporting cash rent agreem ents w ere asked to rep o rt conditions
w here rental adjustm ents might occur.

Specifically if provisions existed in the

agreem ent to adjust the am ount of the rent up or down if growing and m arket
conditions turned out to be especially favorable or unfavorable respectively. Very
few of the respondents rep o rted any kind of adjustm ent.

O f the 21 total cases

acknowledging such adjustm ents, nearly half (10) cited rent reduction in the case of
especially low yields.

Seven reported rent increases associated with substantially

higher than norm al yields.

Only two rep o rted adjustm ents associated with price

fluctuations, one each with higher and lower prices respectively. T he rem aining two
cited oth er unspecified reasons.

94

4.7 Share Rent Agreements
T he num ber of agreem ents by te n a n t’s share of o utput is shown in
?, T able IV.8. T he author is suspicious that the one case showing the te n a n t’s share
of 15 percent is an error. M ost of the respondents fall into the traditional categories
of V2 or % of the crop going to the tenant.

O f interest is the lack of a single

respondent choosing the 60:40 split commonly rep o rted in the N ebraska-South
D akota study.
The four cases reporting 100

Table IV.8 T enant s share of output

percent of the crop going to the
tenant are a bit perplexing.

This

T en a n t’s Share
of O utput

N um ber of
Cases R ep o rted

15%

1

50%

19

67%

28

ten an t reported th at the landlord had

75%

13

engaged him to husband a vineyard

90%

3

100%

4

Total

68

would not seem to be a share crop
agreem ent. In one of these cases the

and that all the landlord required was
th at the tenant upgrade the condition
of the fixtures and vines over a

Source: 1992 M SU Survey of
Landlords and T enants

period of five years. In other cases there may be some confusion on the part of the
respondent.

F or example, it is not uncom m on for the ten an t to purchase the

landlord’s share after the calculation of the split. Consequently a ten an t (or landlord)
may respond that the tenant is retaining the entire crop, ignoring the fact th at the
question is asking w hat percent does the tenant receive for services rendered.

95

4.7 Landlord participation in inputs
In C h ap ter II some discussion was presented with respect to the sharing of
input costs.

It was shown that, in theory, if all inputs w ere shared in the same

proportions as the sharing of the output that the conflict of the share agreem ent
could be resolved. H owever, no exam ple of com plete sharing has been found or
dem onstrated.

Arguably, sharing of some of the inputs may provide some

im provem ent in the agreem ent.
Surprisingly few of the respondents involved in share lease agreem ents
participated in the sharing o f inputs. Specifically, in only twenty-six of the seventy
share agreem ents was there any sharing of input costs. F urtherm ore, the sharing of
inputs in most cases was limited to only one or two items. M ost commonly shared
inputs w ere fertilizer, seed, and chemicals (including liming, insecticides and
herbicides) as costs of production.

T able IV.9 outlines the frequency with which

som e of these production costs w ere shared.
T he m ajority of respondents reporting contracts which included the sharing of
inputs showed no change in the sharing ratios since the initial establishm ent of the
lease agreem ent. T h ere is, however, some evidence that the landlord’s contribution
to inputs in share agreem ents is diminishing.

Nine of the twenty-six respondents

reported th at the landlord’s share of inputs had decreased. Similarly nine (though
not necessarily the sam e nine), respondents reported that the num ber of inputs in
which the landlord participated decreased. In neither of these categories was there
a rep o rt of increased participation by the landlord.

96

Table IV.9 Landlord participation in input cost sharing by input
Totals

L andlord’s share of costs
Input
100%

50%

33%

Seed

7

14

2

23

Fertilizer

6

17

5

26

12

14

0

26

H erbicide &
Insecticide

2

14

1

17

chemical application

3

12

3

18

irrigation

2

0

0

2

harvesting

0

9

1

10

transportation

0

3

0

3

drying

1

7

2

10

other

1

0

0

1

Lime

Source: 1992 M SU Survey of Landlords and T enants

T h ere seem s to be som e m otivation for the choice of traditional, simple ratios
for splitting the crop and the input shares such as half and half or one third - two
thirds. O ne may w onder why splits such as 47% - 53% are never observed. O ne
possible explanation is that tradition prevails.

It may be that sharing of inputs

provides a m eans of fine tuning a share agreem ent. F or exam ple, w here the te n an t’s
share of the crop is less than sufficient to equate invested time, effort and resources,
there is a greater tendency for the landlord to participate in the sharing of costs of
m ore inputs.

97
4.8 Changes from cash to sh are and visa versa
T h ere was very little evidence of m ovem ent betw een agreem ent types. The
average duration rep o rted for share and cash agreem ents was 8 and 10 years
respectively. Eight respondents reported changing from a cash agreem ent to share
cropping. By contrast twelve respondents reported changing from a share agreem ent
to cash rent.
R espondents w ere subsequently asked to identify their position with respect
to willingness to change and give reasons. Those with share agreem ents responding
to this question w ere evenly divided; thirty-three said they w ere willing while thirtyfour w ere not. By contrast, o f those with cash rent agreem ents only sixty-four w ere
willing to change while one hundred and sixty w ere not.
A variety of reasons w ere cited, and, since many of the com m ents w ere cryptic
and incom plete, interpretation is difficult. O f those with share agreem ents expressing
a willingness to change the com m on answer was that they would accom m odate the
wishes of the other party. Those unwilling to change cited such things as poor ground
and risky yields, lack of capital, and tradition.

Interestingly, those with cash

agreem ents unwilling to change often replied to the effect that the agreem ent would
be cash or nothing, implying that if they could not rent on a cash basis they would not
be willing to rent the land at all. O thers simply noted th at share agreem ents required
too much effort in dividing and calculating the rent.
A nother common answ er that ap p eared here (and elsew here w here com m ents
w ere requested) was that the cash agreem ent ensured that both landlord and ten an t

98
knew in advance th e am ount of the rent. This implied an unwillingness to incur the
cost of disagreeing or haggling over the actual yields or crop division.

F urther,

num erous tenants cited a strong preference for cash and an unwillingness to change,
claiming th at with a cash ren t agreem ent the landlord would have no right to
interfere or have anything to say about how the land was farm ed.
Finally, those with cash agreem ents and willing to change to share again
expressed in many cases simply a willingness to accom m odate the wishes of the other
party. A few indicated th at they did feel that the share agreem ent would be m ore
fair.

4.9 Specifics of landlords and tenants
T h ere is som e evidence from landlords th at the choice of contract is related
to personal farm ing experience. Landlords with less than 10 years farm ing experience
rep o rted choosing the share agreem ent about one third as often as cash rent. By
contrast, those with m ore th an forty years experience exhibited a som ew hat reverse
choice.

T able IV. 10 displays the choices and years of experience of responding

landlords.

99

Table IV.10 Landlord experience and choice of contract
Share Crop

Cash R ent

10 years or less

5

14

1 1 - 3 9 years

12

22

40 years or m ore

11

5

Totals

28

41

Landlord Experience

Source: 1992 M SU Survey of Landlords and Tenants

Further, if the landlord actually farm ed the land in question at one time (not
simply farm ed in the past),
he/she was much m ore likely

Table IV.ll L andlord prior cultivation of the
to

choose

the

share

ren t

land to be leased and contract choice

agreem ent (see Table IV. 11).
Prior Cultivation
Those who inherited the land,
or

possibly

farm ed

under

different circum stances might
have less experience to offer
in a share lease relationship.

Share

Cash

Landlord once farm ed
the land

35

44

Landlord did not farm
the land

2

12

Totals

37

56

Source: 1992 M SU Survey of Landlords and
Tenants

4.9.1 Contracting
Very few of the responding landlords (only 11) rep o rted asking for references
w hen they engaged the current tenant. T h ere was no distinguishing p attern betw een

100
share and cash contracts and the checking of references.
W hen asked about the perform ance

of the ten an t in business and

m anagem ent, only rarely did the landlord respond th at the ten an t was not perform ing
well. (It would seem that for this question a range of choices would have been m ore
useful in gathering the inform ation rath er than simply yes and no responses.)

4.9.2 Hypothetical contract choice with parties of varying degrees of closeness and
confidence
In this section choice of contract is presented using a com parison of stated
initial closeness and actual choice of contract (see Table IV.12). This is followed by
data collected using hypothetical situations with corresponding choice of contract.

Table IV.12 R espondent actual contract choice with parties of varying degrees
of closeness.
Share
Lease

Cash
Lease

Very close friend or family

27 (39%)

56 (24% )

Friendly acquaintance

31 (42%)

155 (65% )

Individual unknown

12 (17% )

24 (10% )

R espondent R elationship

1 (.5% )

Fam iliar institution
U nfam iliar institution

1 (1% )

Source: 1992 M SU Survey of Landlords and T enants

1 (.5% )

101
In recognition of the potential for unanticipated distortions in the collection
of actual d ata for testing the hypotheses, several hypothetical questions w ere asked
o f the respondents. It was hoped th at the pream ble to the question would provide
the controlled environm ent for the response. It was understood that th ere is some
m argin for erro r in in terp retatio n since w hat the respondent claims would be chosen
in a given situation, versus w hat would actually happen should the situation arise,
m ight be different. N evertheless, by asking the question in this m anner, it was hoped
that respondents would be able to provide useful inform ation regarding contract
preferences.

102

Table IV.13 Landlord contract preferences (hypothetical) with parties of
varying degrees of closeness; assuming no risk in perform ance.
P rospective
Tenant:

no
agreem en t

Cash
L ease

Share
L ea se
num ber

colum n
%

num ber

colum n
%

num ber

colum n
%

V ery close
friend or
family

41

51

45

17

6

5

Friendly
acquaintance

27

34

67

26

0

Stranger

8

10

66

25

18

15

B u sin ess or
institution

3

4

64

24

25

21

Individual or
firm with
w hom you
have had a
seriou s
disagreem ent

1

1

20

8

70

59

80

100

262

100

119

100

T otals

Source: 1992 M S U Survey o f Landlords and T enan ts, q u estion # 2 4 .

Landlords were asked to indicate w hat kind of contract they would p refer
given a variety of potential tenants. T able IV.13 shows the distribution of responses.
In this first instance, the landlords w ere told that the perform ance of the tenant, with
respect to farming ability was certain. T he variable to be exam ined was w hether the
closeness of the individual would m ake any difference in the choice. T he table shows
clearly a diagonal relationship.
friends and family.

The share lease finds m ost acceptance with close

Strangers tend to p refer cash leasing agreem ents.

N ot

103
unexpectedly, individuals with previous disagreem ents simply p refer not to engage in
any contract with such a person.
R esponding landlords w ere then asked to consider the sam e question b u t this
tim e they w ere instructed th at the perform ance of the ten an t was not to be
considered certain.

T he w ording of the question suggested that th ere was

considerable uncertainty to be taken into account. Accordingly the landlords adjusted
their responses as shown in T able IV. 14.
T able IV.14 Landlord contract preferences (hypothetical) with tenants of
uncertain perform ance and varying categories of social closeness.
P rosp ective
Tenant:

Share
L ease

Cash
L ease

no
agreem en t

num ber

colum n
%

num ber

colum n
%

num ber

colum n
%

15

63

59

23

17

10

Friendly
acquaintance

7

29

72

28

13

8

Stranger

2

8

56

21

32

19

B u sin ess or
institution

0

0

54

20

36

22

Individual or
firm with
w hom you
have had a
seriou s
disagreem en t

0

0

20

8

69

41

24

100

261

100

167

100

V ery clo se
friend or
family

T otals

Source: 1992 M S U Survey o f Landlords and T enants, question # 2 5 .

104
T enants w ere also asked to indicate contract preferences given choices of
landlords in various categories of social closeness.

Though the tenants w ere less

inclined to choose the share agreem ent, com pared to landlords, the tenants also
dem onstrated a diagonal tren d of responses similar to those of the landlords. O ne
possible explanation for th e difference in preferences betw een landlords and tenants
is th a t th e tenants place m ore value on being the sole decision m aker in the
agreem ent. By contrast a retired landlord after having farm ed the land for a long
period o f tim e may feel th a t he or she has som ething to contribute in a share
agreem ent. T he landlord may even feel som ew hat attached to the land and, as is
especially peculiar to farming, may really enjoy being involved though not able to
rem ain fully responsible for all the farm ing activities.

Furtherm ore, the bonds

associated with ow nership may influence the landlord’s desire to be involved.
T able IV.15 provides a breakdow n of the ten an ts’ responses.

105

Table IV.15 T en an ts’ contract preferences (hypothetical) with landlords of
varying categories of social closeness.
Share
L ease

P rospective
Tenant:

no
agreem en t

Cash
L ease

num ber

colum n
%

num ber

colum n
%

num ber

colum n
%

V ery clo se
friend or
family

45

40

157

21

9

5

Friendly
acquaintance

33

29

170

23

3

2

Stranger

18

16

182

25

5

3

B u siness or
institution

13

12

168

23

26

14

Individual or
firm with
w hom you
have had a
seriou s
d isagreem ent

4

3

59

8

143

76

113

100

736

100

186

100

T otals

Source: 1992 M S U Survey o f Landlords and Tenants, question # 3 2 .

U ncertainty relative to the perform ance of the landlord does not play the
sam e role in the agreem ent as with the tenant.

In some cases the landlord may

provide m anagem ent skill, resources or assist with production. By contrast, it is not
uncom m on

for the

landlord

to

have

no

involvement

in

crop

production.

Consequently, it was decided that a question suggesting uncertainty in relation to the
perform ance of the landlord would not be applicable.

106

4.9.3 Landlord qualification and participation in the agreement
T he te n a n t’s view of th e both the ability and willingness of the landlord to
contribute to the agreem ent was solicited in question thirty-one. U nfortunately the
wording of the question caused many of the potential respondents to skip over it
rath e r than respond.

A m ajority of the responses analyzed h ere com e from the

questionnaires filled out during the testing for non-response bias as discussed in that
section earlier. T he MASS op erato rs w ere instructed with slightly different wording
and asked to record th e responses to this question if applicable.

107

Table IV.16 T en an t’s view of the contribution o f the landlord: C rop-share
A greem ent
T he L andlord is
T he Landlord has

Well Qualified

Willing to H elp

yes

no

yes

no

24

5

22

7

Knowledge of
M arkets and
M arketing

20

9

19

10

R esources (assets and
skills)

18

11

16

13

Knowledge of
Farm ing

Source: 1992 M SU Survey of Landlords and T enants

T he responding tenants w ere asked to evaluate the contribution of the
landlord, both with respect to skill or knowledge as well as willingness to help or
participate in the production process. T he responses w ere sorted by contract type
and provide a rather revealing sketch of the perceived attitudes and skills of the
landlord and the type of contract chosen. T enants with share lease agreem ents and
cash rent agreem ents responses are tabulated in T able IV.16 and T able IV.17
respectively.

Com paring th e responses it is evident th at tenants involved in a share

agreem ent perceive the landlord as willing to help. F urtherm ore, those with cash
rent agreem ents are much m ore doubtful about the landlord’s qualifications.

108

Table IV.17 T en a n t’s view of the contribution of the landlord: Cash R ent
A greem ent
T he L andlord is
T he L andlord has

Well Q ualified

Willing to H elp

yes

no

yes

no

47

54

24

76

Knowledge of
M arkets and
M arketing

36

66

18

84

R esources (assets and
skills)

31

69

19

81

Knowledge of
Farm ing

Source: 1992 M SU Survey of Landlords and Tenants

4.10 General rental market and respondent information
In th e final section of the questionnaire respondents w ere asked to provide
further general inform ation th a t applied to all respondents.

4.10.1 Use, maintenance and care of assets
R arely did the respondents list assets m ade available to the ten an t by the
landlord in addition to the land. Those items th at were listed included for the most
p art the use of a shed or bin for storage. In a few cases cultivation equipm ent and
or tractors w ere listed. In each of these cases the parties w ere involved in a family
o peration and a share lease agreem ent.
Responsibility for m aintenance of land rested primarily with tenants according
to respondents. In less than ten percent of the cases this responsibility was shared

109
and a very few cases rem ained the responsibility of the landlord.

By contrast,

m aintenance and up k eep of th e buildings and oth er assets rested with the landlord.
In an effort to evaluate the perception of the respondent tow ard the other
party in the lease agreem ent, with respect to perceived interest of the other party in
upkeep of the assets involved, question thirty-six was included in the questionnaire.
T he vast m ajority indicated th a t the o th er party had a great deal of interest in the
upkeep o f the assets. Those with share agreem ents tended to have a higher regard
for the interest of the o th er party but this tendency was not sufficient to show any
sharp contrast. T he d ata collected for this question is sum m arized in Table IV. 18 by
contract type.

Table IV. 18 Perceived interest in, and care for, the assets
o th e r party’s dem onstrated care for assets
Type of
agreem ent

a great deal

som ew hat

very little

not at all

Share

41 (61% )

22 (32% )

2 (3% )

4 (6% )

Cash

119 (51% )

83 (36% )

20 (9% )

10 (4% )

Source: 1992 M SU Survey of Landlords and Tenants

110

4.10.2 Improvements
Eighty-one of the respondents indicated that physical im provem ents had been
m ade som etim e during the duration of their present agreem ent. T he cost of these
im provem ents was divided rath e r evenly betw een landlords and tenants. In thirtyfour cases the landlord b o re the cost of the im provem ent and likewise in another
thirty-four cases the ten an t bo re all the cost. In thirteen cases the costs were shared
betw een the landlord and

tenant.

Table

IV. 19 summarizes the

kinds of

im provem ents m ade. Tiling o f the land
was the m ost com m on im provem ent.

4.103 Self evaluation of risk preferences

Table IV.19 Capital im provem ents
reported.

and contract choice
Im provem ents to:

N um ber

In an effort to evaluate aversion
Buildings
to risk and the role played by the same

Land

in the choice of contract, respondents
w ere asked to evaluate personal risk
preferences

in question

questionnaire.

D ata

40

of the

collected

for

landlords and tenants w ere evaluated
separately. T enants responding to the
e v a lu a tio n

of

risk

p re fe re n c e

9

Clearing

14

Ditching

10

Fencing

11

Tiling

31

O ther

6

Total

81

Source: 1992 M SU
Landlords and Tenants

Survey

of

Ill
dem onstrated the expected distribution though there was not a com m anding contrast
(see T able IV.20).

Table IV.20 T en an ts’ risk preferences and contract choice
T enants responding to the
evaluation of risk preferences

contract chosen
share

cash

high income, high risk

9

56

m edium income, m edium risk

10

86

low income, low risk

13

32

Source: 1992 M SU Survey of Landlords and Tenants

112
Landlords on the other hand w ere som ew hat polarized in th eir choices. Those
preferring less risk w ere m uch m ore ap t to be involved in cash ren t agreem ents.
Similarly those willing to accept m ore risk chose share agreem ents (see T able IV.21).

Table IV.21 Landlords’ risk preferences and contract choice
Tenants responding to the
evaluation of risk preferences

contract chosen
share

cash

high income, high risk

15

4

m edium income, m edium risk

13

19

low income, low risk

9

30

Source: 1992 M SU Survey of Landlords and T enants

4.10.4 In teractio n and con tract choice
In the majority of cases the ten an t farm ed the land independent of
participation on the p art of the landlord. H owever, as T able IV.22 reveals, landlord
participation in m anagem ent decisions was clearly m ore com m on am ong those
engaged in share cropping agreem ents.

113

Table IV.22 D egree of interaction and choice of contract
D egree o f interaction

Share C rop

Cash R ent

M ost production and m anagem ent
decisions are m ade jointly

12

5

Only the m ost im portant production and
m anagem ent decisions are m ade jointly

16

8

R arely are production and m anagem ent
decisions m ade jointly

6

7

Decisions are m ade solely by the tenant
with som e advice from the landlord

18

48

Production and m anagem ent decisions are
m ade solely by the ten an t with no input
from the landlord

19

167

Totals

71

235

Source: 1992 M SU Survey of Landlords and T enants

114

4.10.5 Current relationship, changes, and contract choice
W hen asked to evaluate th e current leasing agreem ent on the basis of fairness,
the majority of respondents viewed their leasing arrangem ent positively. Though
those with share crop agreem ents showed m ore relative strength in the "excellent"
category, overall th ere did n o t seem to be a significant difference in the type of
contract and the satisfaction o f the respondent (see T able IV.23).

Table IV.23 R espondents Evaluation of Fairness
Satisfaction R ating By R espondents
C ontract
Type

Very
Bad

Poor

A dequate

G ood

Excellent

Totals

Share

1

1

6

32

31

71

Cash

1

3

55

119

57

235

Source: 1992 M SU Survey of Landlords and Tenants

4.10.6 Social Interaction and Contract Choice
It was not uncom m on for the parties engaged in a leasing agreem ent to be
acquainted socially. H owever, our interest in this study was to determ ine if social
closeness had any affect on the kind of contract chosen. H ence respondents w ere
asked to indicate which category or categories of social interaction existed betw een
the leasing parties.

115

Table IV.24 C ontract choice (actual) and social interaction
C ategories of social interaction of the
contracting parties

contract
share

%

cash

%

M em ber of the sam e club, church, or
oth er organization

15

22

44

19

Enjoy leisure activities
together

19

28

45

19

6

9

3

1

No interaction

27

40

142

60

Totals

67

100

235

100

W ork (off farm ) at the sam e location

Source: 1992 M SU Survey of Landlords and T enants

Though the responses indicate that social interaction is proportionately m ore
com m on am ong individuals th at choose the share leasing agreem ent, the evidence
here is not overwhelming. T he com piled findings are shown in Table IV.24.

4.11 Testing of factors associated with contract choice
Testing of the factors associated with choice of contract will be divided into
two sections. Landlords and tenants will be considered separately as the choice of
contract is exam ined using a logit regression model. In the logit model the dependent
variable, "CHOICE", is not continuous. T he choice is an either-or proposition with

116
no continuum of partial alternatives in betw een. The logit m odel provides a m eans
of analyzing the relative influence of the independent variables on the choice of
contract. W e will begin by examining the choice of contract by landlords.

4.11.1 Landlord choice of contract
O f the total 313 responses, 97 w ere from landlords. O f this group, twenty-two
w ere rejected by the program because of missing data in one or m ore variables being
used in the logistic regression. Thus 75 landlord responses w ere included in the data
evaluated in this section.
T he dep en d en t variable "CHO ICE" (of share lease) was taken from the data
collected in response to question ten2 which asked w hether the contract about which
the respondent had chosen to reply, referred to a share-crop or cash rent agreem ent.
T he independent variables cam e from questions designed to elicit responses that
would test the hypotheses previously stated. Table IV.25 contains the variable list,
the question eliciting the variable, and corresponding descriptions.
T he anticipated sign of each of the independent variables has been included
in parentheses with the name. T he interpretation is straight forward. A positive sign
would m ean that the variable is expected to contribute positively to the choice of a
share agreem ent, and visa versa.

A n exam ple can be noted in the case of the

variable labeled RISK. T he negative sign indicates that the risk averse landlord

2 Q uestion ten w a s se le c te d instead of question three sin ce actual c h o ic e s avoid the
d iscrep ancy b etw een stated preferences and real c h o ic es. However, in this c a s e there w a s
only o n e resp on d ent exhibiting this conflict.

117

Table IV.25 V ariable list for landlord logistic regression.
Variable name and
expected sign

Source*

Description

Choice of Share

Q10

This is the dependent variable and contains the
choice of contract by the respondent.

Crop (-)

Qllc and Q17c

Type of crops grown on the leased land.

Parcels (-)

02

Number of parcels of land under leasing
contracts.

Socializing (+)

Q42

Social interaction exclusive of farming
activities.

Joint (+)

Q41

Degree of joint decision making in production
and management.

Risk (-)

Q40

Self evaluation of risk preferences.

Social Closeness
(initial) (+)

Q8

Social closeness at the outset of the
agreement.

Oral (+)

Q4

Proportion of oral versus written agreements.

Fair (+)

Q43

Respondent's perception of fairness of the
agreement.

Assets (+)

036

Landlord's perception of how well the tenant
takes care of the assets.

Continue (+)

037

Anticipated continuation of the agreement.

Experience (+)

026a

Landlord farming experience.

Visit (+)

Q26d

Landlord visits the farm frequently.

Upkeep (-)

09

Respondent's appraisal of the general condition
of the farm at the time of entering into the
leasing agreement.

£ource: 1992 HSU Survey of Landlords and Tenants
Questionnaire numbers. See Appendix I.

118
would not p refer th e share agreem ent since this kind of contract would necessarily
increase exposure to risk. In the case of C R O P, the negative sign m eans th at the
values recorded for the choice of hay, pasture and specialty crops (which would tend
to be m ore difficult to divide) decrease the probability of the choice of the share
agreem ent.

4.11.2 Landlord logit regression results
T he results of the logit regression are given in several stages. First, the overall
perform ance or fit of the equation is evaluated. T he equation, as a m odel of contract
choice, estim ates the log of th e odds that a share agreem ent or a cash rent contract
will by chosen by the leasing parties given the rep o rted circum stances and then in
turn predicts the anticipated choice. The word predict may be a bit misleading in that
th ere is a considerable am ount of uncertainty as to causality. F o r exam ple, the fact
th a t a respondent reports frequent joint collaboration in decision m aking may be
either contributing to or resulting from the kind of agreem ent chosen. N evertheless,
the equation evaluates the correlation of the dep en d en t variables with the
independent variable and predicts which contract will be likely.
T he equation is p resented with the variables described in Table IV.25.

C hoice of Share

=

B0C onstant - B ^ R O P - B2PARCELS + B3SOCIALIZING + B4JOINT
- B5RISK + BgSOCIAL CLOSENESS + B?ORAL + BgFAIR
+ BgASSETS + B10CONTINUE + B ^EX PE R IEN C E + B12VISIT

+ b 13upkeep

119

T h ere are several ways to assess goodness of fit of the model. T he first will
be to com pare the predicted results with the actual data. The accuracy of the m odel
is evaluated in T able IV.26.
T he

first

n u m b e rs

column
show s

of

Table IV.26 Classification
C ontract Choice

Table

for

Landlord

Predicted

Observed

Percent
C orrect

Share

24

15

63%

Cash

51

44

86%

th e
Choice

num ber of share and cash
a g r e e m e n ts
w hen

each

e x p e c te d
set

of

respond en t characteristics
is

plugged

r e g r e s s io n

into

Source: 1992 M SU
Tenants

Survey of Landlords

and

the

e q u a tio n .

W ere the m odel one hundred percent accurate, there would be no difference
betw een predicted and observed.

The "observed" column in the table shows the

num ber of contracts th at w ere accurately predicted. This is followed by the percent
correct.

Obviously the model is m ore accurate predicting cash rent agreem ents

betw een individuals than share-crop agreem ents.
T he second step in evaluating the outcom e of the regression involves analyzing
the contribution of each of the variables. As discussed earlier the variables used in
the regression are for the most part without scale and hence the size of the m eaning
of the size of the coefficient if difficult to derive. Some com parison may be m ade
betw een variables that have com mon values such as the dummy variables in the

120
m odel. A listing of the variables in the regression and statistics p ertin en t to this
analysis are provided in T able IV.27. T he full printout of these statistics is located
in A PPE N D IX II.
A n alternative m ethod of establishing goodness of fit is to exam ine the
im provem ent of the log likelihood statistic betw een the regression m odel with only
a constant and then with th e variables added.

This statistic is rep o rted in the

A ppendix (II) together with the significance level.

In this case we reject the

hypothesis that the coefficients of the independent variables are significantly different
from zero.
Exam ination of these results reveals th at the variables all have the anticipated
sign, with respect to the model, except for C R O P, PA R C ELS and E X P E R IE N C E .
N one of these are significant at the .05 level though C R O P would be considered
significant at the .1 level. T he variables SO C IA LIZ IN G , SO C IA L C LO SEN ESS
(initial), and U PK E E P are all significant at the .05 level with C R O P, JO IN T, RISK,
FA IR , and VISIT betw een the .05 and .1 range. Implications of these findings and
corresponding discussion can be found in C h ap ter V.

121

Table IV.27 R esult of th e landlord regression equation on choice of the share
agreem ent- variables and corresponding statistics
V ariable

(J Value

Statistical Significance

-1.44

.09

.48

.63

SO C IA LIZIN G

2.31

.02

JO IN T

1.55

.10

-1.70

.06

2.84

.03

.07

.96

2.22

.07

.21

.78

1.88

.27

-1.52

.45

4.09

.10

U PK EEP

-1.71

.04

CO N STA N T

-2.00

.52

CROP
PA R C ELS

R ISK
SOCIAL
CLO SEN ESS (initial)
ORAL
FAIR
ASSETS
C O N T IN U E
E X P E R IE N C E
V ISIT

Source: 1992 M SU Survey of Landlords and Tenants

122

4.11.3 Tenant Choice of Contract
A lthough it was assum ed that the landlord would be the party primarily
responsible for choosing the contract type, reason suggests that landlords will also
sensitive to the preferences of prospective tenants. In the survey, tenants w ere also
asked to respond to questions about contract type and preferences for share-crop or
cash-rent agreem ents.
O f the 213 responding tenants, only 121 w ere sufficiently com plete to include
in the regression. This was prim arily due to the confusion over responses to question
31 regarding the te n an t’s view of the landlord participation, which many of the
respondents skipped. T he m ethod of evaluation was very similar to that shown above
for landlords.

However, a few variable changes are to be noted.

The variables

VISIT, E X P E R IE N C E and ASSETS w ere replaced by K N O W L E D G E and
W ILLIN G th at refer to the tenants perception of the landlord’s potential contribution
to the agreem ent. Also, assuming that tenants would be m ore sensitive to the current
social relationship in the choice of agreem ent, variable SO C IA L CLO SEN ESS
(current) was substituted for SO C IA L C LO SEN ESS (initial) in the regression. The
list of variables used is contained in T able IV.28.

123

Table IV.28 V ariable list for ten an t logistic regression.
V ariable nam e

Source*

D escription

C h oice o f Share
Contract

Q 10

T his is the d ep en d en t variable and contains
the ch o ice o f contract type by the
respondent.

C rop (-)

Q l l c and Q 17c

T ype o f crops grown on th e leased land.

C ontracts (-)

Q2

G enerated from q uestion 2 indicating the
num ber o f parcels o f land under leasing
contracts.

Socializing ( + )

Q 42

Social interaction exclusive o f farm ing
activities.

Joint ( + )

Q41

D e g r e e o f joint d ecision m aking for
production and m anagem ent.

R isk ( + )

Q 40

S e lf evaluation o f risk preferen ces.

K n ow led ge ( + )

Q 31a

T en a n t’s evaluation o f th e landlord’s ability
help with m anagem ent d ecision s.

W illing ( + )

Q 31b

T en a n t’s evaluation o f the landlord’s
w illingness to help with m anagem ent.

Social C losen ess
(current) ( + )

Q 38

Current social clo sen ess.

Oral ( + )

Q4

Proportion o f oral to w ritten agreem en ts.

Fairness ( + )

Q43

R esp o n d en t’s evaluation o f the fairness o f
the agreem ent.

C ontin ue ( + )

Q 37

R esp o n d en t’s exp ectation o f th e num ber o f
years the agreem en t w ill con tinu e.

Source: 1992 M SU Survey o f Landlords and Tenants.
* Q u estion num ber. S ee A pp en dix I.

124

4.11.4 Tenant regression results
As before, the results of the regression are given in several stages. First, the
overall perform ance of the equation is again evaluated by com paring how accurate
the m odel was in predicting the choice of contract. The ten an t equation is p resented
with som e differences com pared to the landlord equation:

C h oice of Share

=

B0C onstant - B ^ R O P - B2PARCELS + B3SOCIALIZING + B4JOINT
- B5RISK + BgKNOWLEDGE + B7WILLING + BgSOCIAL CLOSENESS
+ BgORAL + B 10FAIR + B „ LENGTH

As with the landlord model, the choice of the share agreem ent is not as
accurately determ ined as the cash rent agreem ent. A lthough it may be w orth noting
th a t th e ten an t m odel is slightly m ore accurate overall than the landlord model.
T hese results are shown in T able IV.29.
O nce
m odel

again

ap p ears

to

the
be

Table jy .2 9 Classification
C ontract Choice

Table

for T en an t

m uch m ore successful in
Predicted
predicting the cash rent
agreem ent than the share
crop agreem ent. As with
the landlord equation, the
results

of

the

im provem ent of the log

Observed
Share

Cash

P ercent
C orrect

Share

13

11

54%

Cash

8

86

92%

Source: 1992 MSU Survey of Landlords
Tenants

and

125
likelihood statistic fail to reject the hypothesis th at the coefficients of the variables
are significantly different from zero.
Now we will exam ine the outcom e of the regression with respect to the
relative contribution of the variables. As with the landlord m odel above, most of the
variables carried the anticipated sign. Only C R O P exhibited a sign different than
expected. JO IN T, W ILLIN G and L E N G T H , w ere significant at the .05 percent level
with K N O W L E D G E and O R A L on the border. PA R C ELS is on the border of being
considered significant at the .1 percent level. T able IV.30 lists the outcom e of this
regression.

Table IV.30 V ariables in the ten an t equation and corresponding statistics
V ariable
CROP

P V alue

Significance

.82

.14

-.27

.10

.02

.98

JO IN T

1.70

.03

RISK

-.14

.85

KNOW LEDGE

1.91

.06

W ILLING

2.04

.01

SO C IA L CLO SEN ESS
(current)

1.49

.18

ORAL

1.92

.06

FA IRN ESS

1.06

.22

LENGTH

3.53

.01

-11.05

.003

PARCELS
SO C IA LIZIN G

CO N STA N T

Source: 1992 M SU Survey of Landlords and Tenants

126

4.12 Summary
In this chapter the procedure of d ata collection has been set forth in detail.
T he collected data was analyzed and described in tables and prose. Also as p art of
the analysis a logistic regression was run using the m odel described in C h ap ter 3.
T he resulting statistical data will provide the basis for the discussion of testing of
hypotheses in the following chapter.

CHAPTER V
RESULTS AND HYPOTHESIS TESTING

5.1 Introduction
In the previous ch ap ter the d ata collected was rep o rted with only limited
discussion. In this chapter th e previously conceived hypotheses will be discussed in
light of the findings of the survey data.

A rgum ents will be presented to support

rejection of, or failure to establish rejection of, the various hypotheses as the case
may be.

5.2 Basis for hypothesis rejection
As discussed in C h ap ter III, using the scientific approach we do not confirm
a belief, notion, or hypothesis, we merely reject or fail to reject as the case may be.
T he concept is that, given the num erous limits in being able to examine, m easure,
and test a given phenom enon, there is always the possibility that new inform ation,
tools, m ethods, or techniques may be discovered to m ore accurately evaluate the
proposed relationship. By failing to reject a hypothesis, we are saying that given the
inform ation currently available, our m ethods, tools, and so forth, we are unable to
determ ine that a given relationship does not hold.
T h ere is always the risk th at the m ethods used lead to erroneous conclusions.

127

128
Consequently, it is im portant to analyze the potential of com m itting w hat statisticians
refer to as Type I and Type II errors. These are to fail to reject a hypothesis when
in fact it should have been rejected or to reject a hypothesis when it should not have
been rejected.

These kinds o f errors are normally evaluated with respect to the

anticipated corresponding cost of failure. T he discussion usually leads to choosing
a level o f statistical significance that will be considered acceptable in testing. M any
fields of science have ad o p ted a tradition of significance levels of .1, .05, or .01
corresponding to 90%, 95%, or 99% levels of confidence, respectively.
T h ere is little doubt th at the decision context will im pact on the choice of a
significance level. A study th at examines fatal exposure to radiation will not have the
sam e cost structure relative to hypothesis rejection as one that exam ines plant growth
response to exposure to music. For the purpose of this study, especially given the
lack of em pirical data and the relatively low cost of error, the .1 level will be
considered significant. This m eans th at w here we can be confident that at least nine
out of ten tim es th at erro r did not occur, resulting in rejection of (or failure to reject
as the case may be) the hypothesis in question. W e can thus be reasonably certain
that the tested relationships did not occur by chance.

This will be considered

sufficient to claim a significant finding and, as is com m on to science, to recom m end
further research to determ ine that the condition holds in oth er populations and
possibly under oth er circumstances.

129

5.3 Interpretation of the data from the logit regression
Prior to examining the results of the regression it may be useful to note som e
of the characteristics of the logit m odel.

First, the logit m odel approximates a

cum ulative norm al distribution and is primarily used when examining either/or
situations such as the contract type chosen in our model. It exam ines the probability
of an event occurring. T he relationship betw een the d ep en d en t variable and the
coefficients of the independent variables is not linear as in ordinary least squares
regression models. In addition, the unit changes in the data values for the variables
in the regression equation lack a corresponding scale to allow in terp retatio n of the
m agnitude of the actual change. Flence, interpretation of the size of the coefficient
will be limited to ordinal com parisons and relative changes.3 Consequently, the
majority of the discussion will focus on the sign of the coefficient and the level of
statistical significance.

5.4 Testing of hypotheses
T he hypotheses to be tested deal with three main areas of interest. First we
will exam ine the effects of risk on contract choice.
com plex transaction costs issues.

This will be followed by the

Finally, we will consider the effects o f social

closeness and the choice of contract type.

3Strict interpretation of a ch a n g e in the coefficient translates into a ch a n g e in the log of
the probability of the occurrence of an event. For exam ple, in Table IV.26 the coefficient for
RISK is -1.7. This would m ean that for every unit c h a n g e in the landlord’s aversion to risk, the
log (or logit) of the probability of the ch o ice of the share agreem ent w ould d e c r e a se by 1.7.

130

5.4.1 Risk
T he concept of risk in leasing situations is complex. T he scope of this study
allowed only a limited exam ination, focusing on self evaluation of risk preferences in
question forty. T he d ata coded from the responses was tested in the logit regression
which exam ined the influence of the response to the risk question with respect to the
choice of contract.

T he hypotheses w ere tested separately with seemingly mixed

results.

5.4.2 Risk and tenant’s choice of contract
T he risk averse ten an t would be expected to p refer the share agreem ent since
this would minimize exposure to crop failure especially in cases w here the cost of
inputs w ere shared.

Consequently, the expected result was that the following

hypothesis would be rejected:

(1)

Hc:

Risk, relative to the tenant’s preference for secure income, makes no
difference in the choice of contract.

N ot only were we unable to reject this hypothesis, but in addition, the sign of the
variable in the outcom e of the regression was negative instead of positive, m eaning
th a t aversion to risk on the p art of the ten an t encouraged the choice of the cash rent
agreem ent.

However, exam ination of the significance level (.84) reveals that the

variable did not contribute significantly in determ ining the probability of the choice
o f a share agreem ent (see T able IV,29). Further, it was noted by the author that as
variables w ere added to the regression equation, that the sign of this variable changed

131
frequently. L earner (1983) would term this a fragile variable. T he implication is that
for the tenant, the probability of the choice of the share agreem ent depends much
m ore heavily on variables oth er than risk.

5.4.3 Risk and the landlord’s choice of contract
In contrast to the tenant, the risk averse landlord would be expected to p refer
the cash rent agreem ent since this would provide secure income with the am ount
known in advance and not d ep en d an t on the crop. H ence, it was expected that the
following hypothesis would be rejected:

(2)

Hc:

Risk, relative to the landlord’s preference for secure income, makes
no difference in the choice of contract.

The regression results shown in Table IV,27 reveal that the variable RISK
carried the expected negative sign and would be considered significant at the .1
percent level. This is not an overwhelming rejection of the hypothesis. Nevertheless,
these findings indicate that for the landlord, risk is an im portant consideration and
m erits a consideration. This author would recom m end that a m ore thorough study
of all the im plications of risk be undertaken.

5.4.4 Discussion of risk
The traditional approach to studying risk is to exam ine a stochastic event
considering the costs or benefits of various outcom es of known or estim ated
probability. The discussion generally leads to considerations of insurance against the

132
occurrence of the undesirable event. In the case of leasing of land, the stochastic
results o f crop failure are certainly a consideration. T he aspect of risk tested was
evaluated in term s of sharing or not sharing crop failure or success as the case may
be. T h ere is certainly room for discussion and exam ination regarding the approach
of self evaluation used.
F o r one, it was pointed out that self evaluation may be d ep en d en t on
circum stances such as the relative security of the p resent position and the cost of
failure. F o r example, if a ten an t is facing financial ruin or bankruptcy, there is a
greater incentive to risk all. This might be considered similar to "going for it" on
forth down in a football gam e with only m inutes to play and an o th er score could
m ake the difference betw een victory and defeat. U n d er such circum stances even an
historically risk averse person may behave contrary to the aversion. T h ere rem ains
the question of the accuracy of the individual’s self evaluation as well.

F u rth er

exam ination of this phenom enon is w arranted.
T he author is som ew hat m ore interested in the less stochastic kinds of risk
faced by landlords and in som e cases tenants. T he perform ance of the ten an t has a
significant im pact on the health of the soil or well being of the land. Such things as
the potential for leaching of nutrients from the soil, failure to replenish the same,
failure to perform conservation practices, poisoning of the soil and ground w ater
through inappropriate use of chemicals, and allowing the land to becom e infested
with w eeds or other pests lists som e of the risks faced by the landlord. T he ten an t
may have to contend with m isrepresentation with respect to the fertility of the land,

133
potential for flooding, or o th er pre-existing conditions. Risks of this kind are subject
to knowledge, trust, and understanding on both sides of the agreem ent. F urtherm ore,
th ere is always the potential for a change in disposition. T h ere are events th at can
lead to changes which may b e influenced by attitudes and behavior of both parties.
Exam ination of these kinds o f risk provides an aren a of diverse potential study.

5.4.5 Transactions Costs
The issue of transactions costs is also very complex, particularly since this term
tends to encom pass a wide variety of different considerations in leasing agreem ents.
Since the share leasing agreem ent is expected to necessarily require m ore interaction
with regard to determ ining and dividing both input costs and output values it was
expected that the following hypothesis would be rejected:

(5)

H0:

Transactions costs (defined as the costs associated with carrying out
the terms of the agreement- contracting, bargaining, ensuring
performance, exacting paym ent, etc.) make no difference in contract
choice.

D ue to the complexity of the above hypothesis, it is necessary to test sub
hypotheses in o rd er to attem p t to separate some of the issues involved. T he type of
crop im pacts on the issue of transactions costs. Since some crops lend themselves
m ore readily to m easurem ent, these crops are easier to divide and identify the
respective shares and the corresponding values.

W here the crops are sold

im m ediately after harvest and/or as a single large lot, identification and splitting of
shares is facilitated. Corn grain, w heat, beans, and most of the small grains of lesser

134
im portance fall into th e category o f easy to divide.
O n the o th er hand, w here the crop is of widely varying quality, or w here
m easurem ent is difficult, the problem of equating shares fairly may becom e
significant. F u rth er, if th e crop is retained as feed for livestock, m easurem ent may
dep en d on facilities on th e farm at the disposal of the landlord or tenant. O n several
occasions a com m ent was en tered by a respondent indicating that share rental was
not an option since th ere w ere no scales available and the ten an t retained the crop
as feed for livestock.

H ence, one null hypothesis proposed for testing was the

following:

(5.1)

H„: Type of crop makes no difference in the choice of contract.

T he results are m oderately in favor of the rejection of this hypothesis. F o r the
landlord, th e variable C R O P has the anticipated sign and the significance level (.09)
is within th e bounds set above. H owever, the ten an t results show the significance
level (.14) marginally outside the bounds for rejection. It is evident that the choice
of crop is im portant but not overwhelmingly so.
A n o th er issue in transactions costs th at was not considered in the creation of
the theoretical m odel is the num ber of parcels of land or different contracts to be
taken into consideration. M ultiple share leases result in a com pounding of the costs
of dividing both o u tp u t and production inputs and keeping these separate or
accounted for m utual until agreem ent and verification has been accom plished. This

135
point was raised by respondents in the com m ents sections as well. Some indicated
th a t during seeding tim e they did not w ant to have to m easure partial contents of
seeding or fertilizer equipm ent after com pleting one parcel and moving to the next.
Similarly at harvest time, crops from several different locations may be m ost
efficiently stored in one bin. Consequently, the variable P A R C E L was added to the
regression. T he following hypothesis was tested:

(5.2)

H„: Number of parcels makes no difference in the choice of contract.

T he response in this case was quite meaningful. In the landlord results we fail
to reject this hypothesis. This is understandable since the majority of the landlords
ren t out only one parcel of land. However, the results of the tenants choices show
th a t the variable is im portant.

The significance level (.10) falls just within the

boundaries set earlier and thus for the tenants, this hypothesis will be rejected.

T h ere was also some concern with respect to the potential for the landlord to
m onitor the behavior of the tenant. It was felt that if the landlord could visit the land
frequently that this would m ake him or her m ore com fortable with the share leasing
agreem ent. H ence, the following hypothesis was proposed:

(5.3)

Hc: Facility of the landlord to observe and participate in the measurement
and division of the crop makes no difference in the choice of contract.

136
T he variable VISIT, corresponding to the frequency of landlord visits to the
farm , was included in the landlord regression with results indicating that indeed the
issue must be taken into consideration. Based on the .09 level of significance the
hypothesis is rejected.
To sum m arize the transactions costs results, the evidence in each case is
adequate b u t less than overwhelming that these considerations are being taken into
account in choosing the type of agreem ent. T he three sub-hypotheses tested were
rejected based on the statistics from the logit regression and data presented above.
F urtherm ore, th ere is no evidence th at transactions costs as defined and examined
here are not being taken into account when choosing the type of agreem ent.
T herefore, based on the cum ulative results, the initial transactions costs null
hypothesis will be rejected.

5.4.6 Social Closeness
Social closeness betw een the ten an t and landlord was m easured in several
different ways. Being essentially new ground to investigate, there was little if any
precedent to follow.

R espondents w ere asked to evaluate the interpersonal

relationship at the outset of the agreem ent.

Later, each was asked to indicate

participation in social activities. Closeness was also examined with respect to the
degree of shared or joint involvement in production and m anagem ent decision
making.

O th er indicators such as com m ents on attitude and satisfaction were

exam ined as proxies for closeness. Only one general hypothesis was form ulated from

137
the m odel in C h ap ter III. It was expected that this null hypothesis would be rejected:

(3)

H 0: Social closeness m akes no difference in the choice of contract type.

As with oth er variables studied above, the responses of landlords and tenants
showed som e variation with respect to the im portance placed on closeness m easures.
N evertheless, the variables representing social closeness provided strong statistical
evidence for the rejection of the null hypothesis on the p art of both tenants and
landlords. T he variables and corresponding statistics discussed below can be found
in Tables IV ,26 and IV,29.
F o r the landlords, social interaction (SO C IA LIZIN G .02) responses constituted
the strongest association with th e choice of the share agreem ent followed closely by
social closeness at the outset o f the agreem ent (SOCIA L C LO SEN ESS initial .03).
T he probability of the choice o f a share agreem ent by landlords was also associated
with the evaluation of fairness of the agreem ent (FA IR N ESS .07). The involvement
in joint decision making also contributed to the choice of the share agreem ent
(JO IN T .10).
T enants on the other hand w ere not m otivated by social interaction p er se but
showed th at interaction in decision making (JO IN T .03), knowledge of the landlord
(K N O W L E D G E .06), and willingness of the landlord to help (W ILLING .01) all
contributed to the probability o f the choice of the share agreem ent. These responses
provided the grounds for rejection of the final hypothesis:

138

(4)

Hc: The managerial resources of the landlord and the disposition to provide
such make no difference in the choice of contract.

D r a on the physical condition of the farm also ap p eared to correspond to the
probability of the choice of the share agreem ent and the rejection of the null
hypothesis. It appears th at th e landlord’s efforts in m aintaining the farm in good
condition (U PK E E P .04) may lead th e tenants to believe th at the landlord may have
som ething to contribute in th e choice of th e share agreem ent. O th er variables that
show evidence of landlord and ten an t confidence in each other are the strength of
the oral agreem ent and the anticipation th at the lease will continue for som e time in
the future. F o r the tenant, th e length of existence of the lease (L E N G T H .01) is
statistically the m ost significant variable in the equation.

5.4.7 Further study of social closeness
Though the evidence above shows strong support for the inclusion of social
closeness variables in a model of contract choice, num erous im provem ents could be
m ade to construct questions to gather this data. Scaling of choices for respondents
would be one area to consider in particular. Q uestion tw enty-three provides a good
exam ple of this. The choices are strictly "yes" or "no" to a question that probes the
landlord’s attitude tow ard the tenant.

R arely was a "no" answer given.

It is the

opinion of the author th at an unhappy landlord would simply find a new ten an t with
whom a m ore satisfactory relationship could prevail. However, if given a few m ore
choices such as "excellent", "above average", "average", and "fair" (for exam ple), the

139
responding landlord may be b etter able to reveal m arginal dissatisfaction with a
ten an t and correspondingly the effect on contract choice.
By contrast question forty-one provided too many choices.

The responses

w ere very polarized. Possibly one middle choice would be sufficient in this case.
Q uestions twenty-four, twenty-five and thirty-two provided som e useful data
b u t did not lend them selves to providing d ata for the logit regression analysis.
A n o th er phenom enon th at was anticipated and borne out in the responses to these
questions is th a t there is a tendency for some close family m em bers to turn away
from financial involvement, possibly to avoid conflict. This tendency runs contrary
to the hypothesis proposed and, depending on its strength, may result in conflicting
findings.
A n o th er pool of potentially useful data that was not tapped with the approach
used in this study, is the responses of landowners and farm ers formerly involved in
leasing th at have forsaken the practice for w hatever reason. It would seem that the
continuum of responses would thus be m ore com plete.

5.5 Sum man
T he results presented in this chapter suggest th at the choice of the share
agreem ent may be m otivated by a num ber of different factors. D epending on the
im portance of the relevant factors to the parties involved, the choice ultimately
depends on which of the factors dom inates the decision. Risk, transactions costs, and
social closeness all ap p ear to play a role. T here is strong evidence that interpersonal

140
relationships (described and evaluated as social closeness) m ust not be ignored in a
study of farm land leasing in Michigan. H owever, the author would caution th at based
on the evidence of the success of the two equations, landlord and tenant, in
predicting the choice of the share agreem ent, that accurate prediction of the cash
agreem ent was much m ore successful (see Tables IV,25 and IV,28). It may be that
we have found som e necessary but not sufficient conditions for the choice of the
share agreem ent.
In this study we exam ined the choice of contract. T he implication is th at the
landlord and ten an t will choose an agreem ent that best suits their needs eith er for
individual or joint maximization of profits. T he traditional research has focused on
the idea (predicted from the traditional m odel) th at lower productivity and hence
lower profits would result from the choice of the share agreem ent.

M uch of the

resulting d ata suggests th at this is not the case. It stands to reason that researchers
have been looking in the w rong direction. An investigation into the potential for
increased profits as a result o f share tenancy would be in order.
This study provides evidence that the traditional models of farm land leasing
are incom plete. T he m otivation of both ten an t and landlord is im pacted by social
closeness, a factor not included in traditional modeling of contract choice. F urther
exam ination of the contribution of social factors in profitability is recom m ended.

CHAPTER VI

CONCLUSIONS

6.1 Review
This study has g athered inform ation to address the paradox of the crop share
lease. L et us review briefly th at paradox and the surrounding issues. It has been
dem onstrated by num erous authors that, based on the tenets of econom ic theory, the
share lease is inefficient and will not result in ap propriate allocation of resources.
T he argum ent stem s from a conflict faced by the tenant. U n d er the share agreem ent
the ten an t receives only a portion of the output and consequently, when equating
m arginal costs and returns, will choose a level of production below that which would
be chosen by the landlord or a ten an t with a cash lease agreem ent, resulting in an
inefficient allocation of resources.

This seeming conflict has been the focus of

num erous attem pts to provided explanations as to why the share agreem ent would
persist if the landow ner is attem pting to maximize profits. And, though many articles
have been w ritten, very little em pirical d ata has been gathered to substantiate the
claims and explanations. H ence, the purpose of this study was to form ulate a model
o f leasing behavior, generate testable hypotheses with respect to the motivations for
choice of contract, and then gather data and test the hypotheses.
In form ulating a m odel of contract choice in leasing, effort was m ade to solicit

141

142
reasons for choice of contract from landlords and tenants. The issues of risk sharing,
ease of handling, tradition, and fairness w ere m ost prevalent in the responses. Risk
sharing and transactions costs associated with adm inistering the agreem ent had been
addressed in the literature and incorporated into existing models.

H owever, the

issues of fairness and tradition p rom pted fu rther inquiry. In addition, interest had
previously been generated with respect to interpersonal relationships and the effects
on contracting and econom ic choices, both form al and informal, by a new literature
(R obison 1987).

T he share lease seem ed to offer an ap p ro p riate forum for

exam ination of this issue. T hus the concept of social closeness was incorporated as
a variable in the model.
Close social relationships w ere hypothesized to have the effect of reducing
som e of the costs associated with risk and transactions and of increasing the M VP of
the inputs in production including the te n a n t’s labor.

Furtherm ore, if the self

interests of the contracting parties could becom e subordinate to the interests of the
unit, the conflict would be resolved. Individuals with strong family ties, friendships,
or loyalty might be expected to act in the interest of the whole. H ence, the issue of
social closeness took a m ajor p art in the study.

6.2 Empirical findings
Collection of data was accom plished by conducting a survey using a mail
questionnaire to a sam ple o f farm ers and landowners in M ichigan p rep ared by
M ichigan Agricultural Statistics Service. The resulting data w ere coded and analyzed

143
using a logit m odel and the com puter program , The Statistical Package for the Social
Sciences.
C aution m ust be exercised in regional generalization of the implications. The
sam ple taken only from M ichigan landlords and tenants, and there ap p ears to be
substantial differences in the incidence of share leasing com pared to the findings of
another recent study in the mid-west.

The M ichigan ratio of share crop leasing

agreem ents to cash rent agreem ents was roughly one to three, som ew hat less
frequent than the N ebraska-South D akota study which found an almost even split in
South D akota and the reverse o r three share leases to one cash rent in N ebraska
(Johnson et al 1990). F urtherm ore, parcel size and num ber of parcels p er ten an t
differed significantly am ong these states.
It was dem onstrated th at the three main factors hypothesized as m otivators
for the choice of the share agreem ent -risk , transaction costs, and social closeness,
—all played a role. T here w ere differences with respect to landlords and tenants on
som e o f the m otivating factors.

6.2.1 Findings on risk
T he respondent provided inform ation on his or her own attitudes tow ard risk
by the response to the self evaluation question (question 40). The resulting d ata was
then used as a variable in the regression equations for landlords and tenants which
exam ined the correlation of this solicited risk variable with the choice of contract.
As discussed previously, the choice of contract carried implications for exposure to

144
risk which are quite different for landlords and tenants.
The cash agreem ent provides relative income security to the landlord, leaving
the ten an t to b ear production, price, and ultim ate income risk due to stochastic
fluctuations. By contrast, the share agreem ent involves the landlord in bearing risk
but only partially relieves th e tenant. H ence, it was hypothesized that risk averse
landlords and tenants would p refer the cash and share agreem ents respectively.
The findings revealed that a statistically significant correlation was evident
betw een risk preferences and choice of contract for landlords, but not for tenants.
This might suggest that the landlords have g reater income dependence on the
outcom e of th e lease than the tenants. O ne might further speculate th a t the tenants,
having m ore leases on average, are m ore diversified and thus control exposure to risk
through oth er means.

It may also be that other considerations dom inated the

te n a n t’s choice sufficiently to overshadow the influence of risk.
Previous studies focused the exam ination of risk on a slightly different
approach.

C rops were identified as having inherently g reater or lessor risk of

successful production and harvest for a give area. It was then hypothesized that the
riskier crop would be chosen by tenants with share agreem ents w here the exposure
to risk was lower. Conflicting results ensued.
This study suggests that consideration of landlords and tenants separately when
examining risk may provide a solution to the above conflicts.

F urtherm ore, the

author has earlier provided considerable discussion on the various aspects of risk and
how this variable might affect the choice of contract. In particular, the gains from

145
favorable social interaction and social closeness betw een th e landlord and tenant
would have the effect of reducing or elim inating the effects of risk in many cases.

6.2.2 Findings on social closeness
Landlords had a g reater likelihood of entering into share agreem ents with
tenants whom they encountered on a social basis.

This variable did not play as

significant a role in the ten an t’s choice of share contract. Given that the landlord is
placing not only valuable assets but also anticipated incom e in the care and
m anagem ent skills of the tenant, it m akes sense th at the landlord would seek an
individual in whom confidence could be placed.

146

Table VI. 1

Selected social closeness and interaction
corresponding statistics for landlords and tenants.

variables

and

T enants

Landlords
Variable*

(5 value

Significance

P value

Significance

Socializing

2.31

.02

.02

.98

Joint

1.55

.10

1.70

.03

Social Closeness
(Initial)

2.84

.03
1.49

.18

Social Closeness
(C urrent)
Fairness

2.22

.07

1.06

.22

O ral

.07

.96

1.92

.06

2.04

.01

Willing

Source: 1992 M SU Survey of L andlords and T enants
* See T able IV.28 for description of variables.

C om parison of the variables and corresponding statistics in Table V I.1 reveals
that tenants and landlords have substantially different needs and views with respect
to social aspects of contract choice. F o r the landlord it appears th at the interest
tends to focus on factors thay create trust and loyalty such as socializing, social
closeness, and perceived fairness that may substitute for the foregone security of the
cash agreem ent. O n the o th er hand, the tenants ap p ear to be m ore interested in
w hat the landlord has to offer in interactive agreem ent. F or example, tenants w ere
much m ore likely to engage in a share agreem ent with landlords dem onstrating both

147
resources and knowledge as well as willingness or disposition to participate in
m anagem ent decisions. By contrast, the socializing relationship was not statistically
significant for tenants. This may reflect the natu re of discovery of contracting parties.
T he ten an t looks for land, hoping for an am icable relationship and willing to en ter
into a share agreem ent if the landlord ap p ears to have the resources and willingness
to participate. T he landlord looks for th e "right" tenant, putting high value on w hat
he knows about the ten an t through social channels.
F urther,

in terpretation

of th e

num erous

com m ents

sections

on

the

questionnaire leads the au th o r to the conclusion that frequently the share agreem ent
was chosen by the landlord to provide optim al assistance and incentive to a family
m em ber leasing the farm. A "we’re in this together attitude" often cam e through
indicating a great interest in fairness and a sharing of both hardship and bounty.
O th er motivations included such things as the landlord’s desire to rem ain involved,
retention of identity with the farm and farming, desire to help the ten an t "get
started", and tradition corresponded with the choice of the share agreem ent. F or the
landlord, the share agreem ent seems to m ean m ore than simply renting out the land
for a source of income.

6.2.3 Findings on transaction costs
Theoretical effects of transaction costs on the choice of contract w ere
supported by this study.

T he share agreem ent necessarily requires increased

interaction, monitoring, record keeping relative to costs shared, and evaluation and

148
division of the crop. T he variables representing the type of crop and the frequency
o f visits by the landlord to th e land, w ere found to be statistically significant in the
landlord regression equation, dem onstrating th at these concerns influence the
landlord’s choice of contract.
O n the p art of the ten an t, the variables representing the num ber of parcels
and formality of the agreem ent (O R A L ) indicate that the tenant also takes into
consideration th e costs associated with carrying out various transactions in choosing
the type of agreem ent.

T he com m ents en tered by tenants further establish the

dim ension of this aspect of contract choice. O n several occasions individuals took the
tim e to write lengthy explanations about how the share agreem ent would not be
feasible w here the ten an t is farm ing num erous parcels of leased land. In such cases
the burden of record keeping, tim e required in joint decision making and even the
aspect of storing crops separately until weighing and m easuring could be com pleted
was com pletely im practical.

Frequently a ten an t would write in response to the

question o f w hether (s)he would be willing to change to a share agreem ent that scales
w ere not available to m easure the crop. This was seen as an insurm ountable barrier.

6.2.4 Summary of findings
It may not be possible to com pletely sep arate the effects of social closeness
and transaction costs and risk. N evertheless, in each case for the landlord the null
hypothesis was rejected, indicating that the findings w ere sufficient to consider that
risk, transaction costs and social closeness m ust be taken into consideration when

149
examining the choice of contract. Especially the influence of the social closeness
variables on the choice of contract should not be ignored.

6.3 Implications
Probably the m ost im portant finding of this study is the support for the social
closeness factors in contract choice.

Contrary to neoclassical m icroeconom ic

assum ptions, there is evidence that interpersonal relationships play a role in
establishing the characteristics of an econom ic agreem ent. A similar conclusion was
reached by M arcelo Siles (1993) in a study on lending behavior.

T he evidence

indicates th at traditional econom ic models do not adequately account for the
im portance of social relationships.

This can lead to incorrect conclusions, farm

m anagem ent advice, and policy.

6.3.1 Implications for landlords
T he implications of neoclassical theory for the effects of both share leasing
and cash rent agreem ents are som ewhat negative for landlords. In the case of the
share lease, the theory finds an incentive for the tenant to apply less than optim al
variable inputs, resulting in depletion of soil fertility.

U n d er the cash agreem ent,

exploitation may also occur either through mining of the soil, failure to perform
conservation and erosion prevention, or neglecting m aintenance.

However, the

presence of social closeness would tend to provide at least partial counter incentives
for both o f these kinds of problem s. As one farm er put it, "We all tend to abuse the

150
land in po o r econom ic times. T he difference is that the good farm er will replenish
and resto re when tim es are better." A ten an t that experiences social closeness to the
landlord would be m ore likely to try to keep the land and assets in good condition.
T he findings here suggest that landlords may be well advised to invest in
closeness. R egardless of the type of contract, the tenant that considers the landlord
to be a friend and a source of help and advice from time to time, will be much less
likely to exploit that landlord.

Savings should be expected in transaction costs,

especially supervision, which would be greatly reduced w here m utual understanding
and trust prevails. It is reasonable to hypothesize that the M VP of inputs may be
enhanced by investm ent in social closeness and that enhanced M VP contributes to
profitability.

6.3.2 Implications for tenants
It may be that the ten an t does not have as much to gain from investing in
social closeness in the sam e way as the landlord. As noted, the coefficients on the
SO C IA LIZ IN G variable for landlords and tenants were quite different as were the
significance levels observed. O n the other hand, the W ILLING variable (indicating
the te n a n t’s perceived willingness of the landlord to contribute to the agreem ent)
played a m ajor role in the te n a n t’s choice of contract. This propensity to provide
assistance may be sought by prospective tenants.
Som e possible advantages cited earlier include the potential for cost savings
through advice, contacts with m arkets and suppliers, and understanding of

151
peculiarities of the land.

These are consistent with higher factor m arginal value

products for the inputs supplied by and paid for by the tenant.

In addition som e

(socially close) landlords are willing to contribute a couple of days work, the use of
storage areas, and possibly some im plem ents or tools.

D epending on the

arrangem ent this could be a valuable resource, contributing to higher income for the
tenant.
Willingness of the landlord to accom m odate the needs of a tenant suffering
from crop failure might also be of considerable value to th e tenant. A lthough the
incidence of adjustm ent was rath er rare in the study, further investigation into
preconditions for adjustm ents not formally p art of the contract could be w arranted.

63.3 Implications for farm management and extension personnel
T he advantages and disadvantages associated with cash and share tenancy
should be fully understood in o rder to provide accurate inform ation to inquiring
landlords and tenants. The im portance of m aintaining or prom oting a positive social
relationship betw een landlords and tenants should be considered. M isunderstanding
leads to conflict and possibly lower productivity.
O ne practice that has been frequently encouraged in the extension literature
is the sharing of production inputs. This author has observed that the sharing of
inputs does tend to achieve, at least in part, a m ore equitable allocation of resources.
H owever, an o th er role seem s to exist.

Given the traditional rigidity of the share

splits, (fifty-fifty, one-third-tw o-thirds, etc.) the choice and num ber of shared inputs

152
tends to fine tune the agreem ent. This was evident in the N ebraska South D akota
study as well as in conversation with farm ers and farm m anagers. F u rth er research
in this area might provide links to help bridge the gap betw een theory and practice.
Land owned by governm ent or institutions is often m ore vulnerable to
exploitation. T he im portance of establishing a farm m anager as acting landlord with
genuine interest in the land and the ten an t may have beneficial results.

6.3.4 Public policy implications
Traditional aversion to the share agreem ent, evident in the academ ic
literature, is not w arranted.

The fact that an increasingly large am ount of the

nation’s landbase is farm ed by tenants, many of which are share tenants, cannot be
interpreted to m ean that th ere is a deterioration in the efficiency of food production.
Public policy to discourage share agreem ents seems unw arranted.
In a nation reluctant to officially adopt any m oral guidelines, it is doubtful that
public policy can be form ulated to encourage social closeness. It may be possible
how ever to focus on the m erits of cooperation, voluntary contributions of all kinds,
and community spirit building events, activities, and projects.

To the extent that

individuals feel a closeness, or a sense of belonging, and a trust of fellow m em bers
of com munity or society, the benefits associated with closeness may em erge (Schmid
1987).
O n a b ro ad er scale, th e author envisions that this line of investigation could
have beneficial effects on international relationships. Few er confrontations might

153
occur if negotiating parties could find com m on ground for choosing trad e agreem ents
based on established com m on interests and expectations of improving bonds betw een
nations. T h ere may be reason to believe that individuals with close social bonds
dem onstrate willingness to m ake choices in the interest of the unit rath er than the
individual.

6.3.5 Theoretical implications
T he inconsistency of theory and practice is partially resolved by adding social
closeness variables to the theory. T he persistance of the leasing agreem ent may not
be conclusively resolved in this study. New lines of inquiry are suggested.

M uch

m ore research m ust be done to provide support for these findings. N evertheless, the
m ain areas of potential gain through social closeness ap p ear to be in the cost savings
associated with transaction costs and risk. T he underlying implication is that socially
close individuals with an interest in the w elfare of the other can help prevent, avoid,
or elim inate altogether some of these costs.

6.4 Recommendations for further research
R eplication of this study in an o th er location would provide much in the way
of furth er support for the findings. It is recom m ended that considerable effort should
be expended to secure a m ore com plete sam ple in an o th er state or region.
T he model developed here focused on factors influencing contract choice.
Learning from this m odel, we can now ask about the supply of social closeness. A

154
m ore general m odel linking several steps might b etter illum inate the complexity of
the issues being exam ined. T he following functions might be considered to further
develop the mondel:

social closeness = /(investment)
MVP = /(social closeness)
profit = /(MVP)

This m odel would lend itself to an econom ic evaluation of th e returns to investm ent
in closeness rath e r than limiting the inquiry to how a given supply of closeness (social
capital) influences the contract choice as in the m odel used in this study.
Closeness could affect M VP in several ways. Closeness affects productivity
and ultimately profitability via its interaction with risk and transaction costs. It was
n oted earlier th a t only one aspect of risk and contract choice was exam ined in this
study. O f particular interest would be work in the area of examining the substitution
of closeness for alternative ways of dealing with risk. It occurs to the au th o r th at
individuals enjoying social closeness would reduce the probability of an unfortunate
event occurring, as well as spread the risk of unavoidable events. W ould this m ake
a tenan t or landlord less risk averse? How would alternate forms of contracting
influence risk attitudes and choice of technology and enterprises?
Similarly there are many areas w here there ap p ears to be a relationship
betw een closeness and transaction costs. F u rth er study on this issue, especially that
which would prom ote understanding of actual expected dollar savings due to

155
closeness, would be very useful.
O ne im plied assum ption that could and should receive additional attention and
study is the idea th a t th e share agreem ent is chosen because it enhances profitability.
N o attem p t was m ade in this study to m easure and evaluate profitability specifically.
T he author anticipates th at creation of a m eans of m easuring the gains from socially
close relationship would be a rich addition to this area of study.
The au th o r had anticipated analyzing and com paring changes in contract
choice in cases w here the landlord and tenant w ere initially strangers and becam e
close friends. This com parison was not m ade because of lack of sufficient data on
individuals th at w ere strangers at the outset of the agreem ent and changed from one
kind of agreem ent to another.
gathering a larger sample.

This might be rem edied in further research by

A PP E N D IX I

T H E Q U E S T IO N N A IR E

156

1992
Michigan State University
Farmland Leasing Survey

Farm land leasing Is an Important part of today’s production agriculture. Vet, It Is often
difficult for ten an ts an d landlords to gain a clear understanding of leasing practices
within their locality and the state. By com pleting this questionnaire, you will b e helping
to com pile lease market Information for 1992, which w e will b e happy to sh are with
you.

This survey is being sent to a random sam ple of both tenants and landlords. S o m e
qu estio n s m ay not apply to you, but please respond a s completely a s possible. Your
answ ers will b e kept confidential and u sed In compiling total and average resp o n ses.

Even though you m ay have more than one lease, plea se answ er the

GENERAL INFORMATION

questions (or ju st ONE lea se agreement- either your MOST IMPORTANT
t . A s a land ow ner or farm operator In Michigan In 1992, how m any acres of farmland,
if any, d o you:
v
Total
Estim ated
A cres
Market Value
a. o w n ? .................................................................................
b. lease to others?

OR MOST TYPICAL lease.
5.

In what county Is this leased land located?
P lease Indicate the predom inant soil t y p e _____________________________

$__ /acre

.........................................................

$__ /acre

c. lease from o t h e r s ? .......................................................

$__ /acre

d. Total that you farm y o u rse lf? .....................................

$__ /acre

6. How m any years have you leased this land?

______________

7. For this agreem ent, (check o n e for each question)

2. P lease list a n d categorize a ll your leases. E ach parcel or contract should be

a. you a r e ? ........................te n a n t....... ....................................

co n sid ered a s a se p a ra te lease.

b. the lease Is? ..................
S hare or

L ease

N um ber

Major C rops

N um ber

of A cres P ercent Field C rops

o ra l...... .......................................

c. the lease Is renew ed? annually

landlord___
w ritten___

tor a multi-year p e rio d ___

C ash Rent

P ercent H ay/Pasture

fClrcle One!

1.

.

% ___________ % ............ S

C

2.

.

% ___________ % ............S

C

3.

.

% ___________ % ............S

C

4.

.

% ___________ % ............ S

C

8. P lease check the o n e category that best d escribes your relationship with the other
party at the lime of entering Into this agreem ent:
Very clo se friend or family
Friendly acquaintance (e.g. neighbor)

5.

.

% ___________ % ............ S

C

_______

6.

.

% ___________ % ............ S

C

_______

7.

.

% ___________ % ............ S

C

Individual previously unknown to you (stranger)
Familiar Institution (a com pany, government agency, etc. with which
you h ad previously dealt or established a relationship.)
Unfamiliar Institution (a com pany, governm ent agency, etc. with which

3.

you h ad never dealt previously.)

W hat kind of contract d o you p refe r- sh a re or c a s h rent?.

9. P lease check the category that best d escribes the physical condition of the farm at

P lease give the m ain reason(s) for your choice__________

th e time you entered Into this agreem ent:
Excellent

4. How m any of your lease agreem ents are: a. written?^

b. oral?

A verage

Run down

4

3
The following p a g e s are divided Into two sections, one for crop-share lease

13.

ag reem en ts an d the other for cash lease agreements. P ie e a e a n s w e r th e

landlord's sh are? (com plete all that apply)

Of any C R O P PRODUCTION INPUT co sts that are shared, what Is the

q u e s tio n s O N L Y In t h e s e c tio n th a t b e s t r e p r e s e n ts th e l e a s e y o u

Landlord's
sh are

h a v e c h o s e n a s m o s t ty p ic a l o r m o s t Im p o rtan t. The questions are
d esig n ed lo b e answ ered by either landlords or tenants.

Landlord's
share

a se e d ....................................

%

b. fertilizer................................

%

c. l i m e ....................................

%

d. herbicide and/or Insecticide

%
%

Following that, there are se p ara te questions for land lords an d tenants. T hen a few

e. co st to apply chem icals .

%

f. Irrigation e n e r g y .................

sum m ary q uestions for all respondents. P lease re sp o n d to all questions that apply to

g. h a rv e stin g ..........................

%

h. tra n sp o rta tio n ....................

%

you.

I. d r y i n g .................................

%

I. other (soecifv)

%

C RO P-SH ARE LEASE SECTION
14.

Since you b eg a n leasing this land, has:

10. Is the lease you will be describing a C R O P SH ARE lease?
a the lease ch an g e d from c ash to share
Y e s . K "Yes' g o to Q uestion 11.

N o . H “No" g o to Q uestion 16.

Yes

No

___

___

or has,
Increased

11. P lease Indicate the nature of the farmland, y o u r e s t i m a t e of value (for farming
p u rp o ses), m ajor crop(s) a n d corresponding sh a re, (com plete all that apply)

Relevant
A creage Is

Estimated
Market Value of
th e Land

T enant’s S hare
of Output

farmland:
a.

dryland . . . .

a c re s . . . $________ . . ________________
(per acre)

b. the landlord's percent

sh a re of Inputs? . . . ...

c. the num ber of sh a re d

Inputs ch an g ed ? . . . . . .............. . .

d. the ten an t's c ro p sh a re chan g ed ?
Major
C rop(s)

Irrigated . . .

a c re s . . . $_______
(per acre)

. . ________________

12a. The a d d e d rent am ounts to $_

...................... .....................

%

%

S u p p o se the other parly In this agreem ent expressed an Interest In changing the

lease from cro p-share to c ash rent.

________________

Is there a c a s h paym ent In addition lo th e sh a re rent?
Y e s If ■Yes" g o to Q uestion 1 2 a

..... ........

________________

Would this b e acceptable to you?

12.

No
C hange

e. P lease explain the reason(s) for any ch an g e noted.______________

15.
b.

D ecreased

N o . If *No' g o to Q uestion 13.
per/acre or a total of $.

1 5 a P lease explain yourreason(s)_

YES _

NO

vo

t

6

CASH LEASE SECTION

18b. If an adjustm ent of this type h as been m ade any time within the p ast 5
years, please briefly describe the circum stances an d the am ount ot the
adjustment.

16. Is th e le a se you will b e describing a CASH lease agreem ent?
Yes. If "Yes' g o to Q uestion 17*
No. It "No' g o to Q uestion 21.
19.

Since you beg an leasing this land, has:
Yes

17. W hat w ere/are the 1991 or 1992 per acre c a s h rents, m ajor c ro p s an d y o u r
e s t i m a t e of th e 1992 per acre market value (for farming purposes), of this le a s e d
la n d ?

a land ow nership ch anged? ................................. ........
b. there b een a different te n a n t? ....................................
c. the lease changed from share to ca sh rent? . ___

Estim ated

d. P lease explain the reason(s) for any ch an g es noted..

Relevant

Market Value of

Major

Per Acre

A creage is

the Land

Crop(s)

C a sh Rent

Farm land:
& dryland

acres . . . $_________

..

(per acre)

20.

S u p p o se the other party In this agreem ent expressed an Interest In changing the

lease from c ash rent to crop-share.

Would this be acceptable to you?
b. Irrigated . . .

(per acre)

18.

YES

NO

ac re s . . . $

Are there lease provisions that vary th e am ount of c a s h rent d u e to c h a n g e s In

yields or prices?

Y es. If "Yes" g o to Question 1 8 a

No. H ‘No* g o to Q uestion 19.

1 8 a Is rent adjusted for ch a n g e s In: (check alt that apply)
Higher than normal yields
Lower than normal yields
Higher than normal cro p prices
Lower than normal cro p prices
Other reasons. P lease explain____________________________

2 0 a Please explain your reason(s)______________________________________________

O

I

8
25.

SPECIFIC LANDLORD INFORMATION

S u p p o se o n ce again you h ad a parcel of land that you could not farm yourself.

You are p resented with a variety of potential tenants. This time you are very uncenaln
about the ability of each one. They may or may not b e g o o d farm ers but you just don't
IF Y O U L E A S E T O O T H E R S , please answ er Q uestions 22 through 27. If not, g o

know an d have no way of getting m ore Information. What kind of lease would you

to Q uestio n 28.

prefer with ea c h of the following kinds of tenants? (Please check one for each te n a n t)

'

21. S ecu rin g acce p ta b le ten an ts Is: (circle one)
Prospective tenant:
1
Quite

2
S om ew hat

Difficult

4

Generally

Very

Easy

C ash

no

L ease

agreem ent

a Very clo se friend or f a m ily ..........

3
Difficult

Share
L ease

b. Frelndly a c q u a in ta n c e ..................
c. S tr a n g e r ...........................................

Easy

d. B usiness or Institution..................
e. Firm or Individual with whom you

22. Did y ou ask for references w hen you engaged
Y e s _____ No

th e p resen t tenant?

Os

23. D o y o u regard your tenant as:

2 6 a W ere you, or are you currently actively en g ag ed In farming yourself?

Yes
Trustworthy an d conscientious

had a serious d isa g re e m e n t.............

.............................................. ...........
Y e s . If ‘y es' give num ber of years

Timely In perform ance of th e farming o p e r a tio n ...............................
M akes g o o d ch o ices of cro p s to grow an d Inputs to use.
U p to d a te an d know ledgeable

.

No

. ____

.............................................. ...........

24. S u p p o s e you had a parcel of land that you could not farm yourself. You are

26b. Did you o n ce farm the land about which you are responding In this

p re se n te d with a variety of potential tenants. You conclude that e a c h Is of equal ability,

questionnaire?

skills a n d resources. What kind of lease would y o u prefer with ea c h of the following

Y es

kinds of ten ants? (Please ch ech one for ea c h kind of tenant.)

P ro sp ectiv e tenant:

S hare

C ash

no

L ease

L ease

agreem ent

No

26c. Do you visit the land frequently? (At least o n ce every couple of m onths)

Y es

No

a Very clo se friend or f a m ily ....................... ............. . . .

_____

b. Friendly a c q u a in ta n c e ............................................

...

_____

c. S t r a n g e r ........................................................ ............. . . .

_____

27.

d. B u sin ess or In stltu llo n ............................................ . . .

_____

llvesiock or other on-farm use?

Do you retain any of your sh are of the production from the farm for your own

e. Firm o r Individual with whom you
h a d a serio u s d isa g re e m e n t.......................................

...

_____

Y es

No

1—1

i

10
32.

SPECIFIC TENANT INFORMATION

S upposing you w ere Interested In farming m ore land and you were presented with

the following possible landlords. What kind of contract would you prefer? (Please

IF YOU LEASE FROM OTHERS, please answer Questions 28 through 33. If not
g o to Q uestion 34.

check on e for each kind of landlord.)
I would refuse

k

28. How did y ou typically first learn that the land you lease was available to rent?

Prospective landlord:

(check one)

a C lose friend or family

Share

C ash

to enter Into

Lease

Lease

any agreem ent

............................ ....................... ............. ................ ..........

b. A c q u a in ta n c e ........................................... .......... ............. ............. ................ ..........

From a relative.

c. S t r a n g e r ..................................................... ............. ............. ............. ................ ..........
d. B usiness or In stitu tio n......................................... ............. ............. ................ ..........

From a neighbor o r o ther Individual

e. Firm or Individual with whom you

From landow ner directly.
V

had a serious disagreem ent

From n ew sp aper o r other m edia ad.

............... ............. ............. ............. ................

..........

From other so u rc e (explain)____________________________
3 2 a P lease give the mein reaso n s for your preference of contract._______________
29. At the time of your original agreem ent, w ere you
aw are of com petition from others? . ___Y e s

i— i

---------------------------------------------------------------------------------

. ____ No

30. W hen you renew your lease, are you
usually In com petition with others? . .

Y e s . ___ No

31. P lease evaluate th e landlord’s ability an d contribution to your agreem ent.
31 a The landlord provides land only.

Y e s . If ■yes' g o to question 32.
N o . If "no" g o to question 31b.

The land lord Is:
31b. The landlord has:

Wen Qualified

Willing to help

(circle one)

(circle one)

1) Knowledge of farm ing

Yes No

Yes No

2) Knowledge of m arkets an d marketing

Y es N o

Ye3 No

3) R eso u rces (a ss e ts an d skills)

Yes N o

Yes No

33.

Would a ch an g e In the type of lease change the way the land Is farmed? P lease

explain.

Os

11

12
36.

How would you classify your current relationship with the other parly In this lease?

GENERAL RENTAL MARKET AND
Very close
Friendly

RESPONDENT INFORMATION

Strictly b u siness
Distant

Questions 34 through 47 are lor ALL RESPONDENTS.
34.

P lease list any a s s e ts (other than the land) u se d In c ro p production, storage or

transportation that a re supplied (without additional coqt) by th e landlord.

39.

Over the past 5 years or since this lease agreem ent began (If less than 5 years),

h a s there b een any investment In perm anent Improvements?
Y e s , If "Yes' p lease go to questions 39a an d 39b.
N o , If ’No‘ please go to question 40.

35. W ho Is responsible for care and m aintenance c o s ts
Landlord
a L and

.............................................................

3 9 a How w as the co st of the Improvement s h a re d _____% landlord

of:

% . . . _____ %
% . . . _____ %

c. O ther fixtures &equipm ent

% . . . _____ %

39b. P lease Indlcale the kind of Improvement (I.e. tiling, Irrigation system etc.)

36. W hich category below b e st describes how m uch th e other party In this agreem ent
c a re s ab o u t the as se ts , Including upkeep en d m aintenance w here applicable.

40.

Som e Investm ents have high expected return but also high variability from o n e

year to another (som etim es great gain, but som etim es great losses).
a

a great deal
Indicate your preference for Income-rtsk trade-ott. (Circle a num ber below):

b .____ som ew hat
c .____ very little

Higher average Income

d .____ not at all

Lower Average Incom e

Higher risk

Lower Risk

37. How long d o you expect this lease to continue?
1
o n e y ear

two to five years

^

u>
Tenant

b. Buildings .......................................................
......................

% tenant.

m ore th an five years

2

3

4

5

6

7

8

9

10

13

14
45. Your ag e Is (check one)

39. Your sex Is?

41. P te a se ch eck th e category that b est describ es the deg ree of interaction of landlord
and ten an t In production a n d m anagem ent decisions.

L ess than 25 years

Male

25 to 34 years

Female

M ost prod u ction an d m anagem ent decisions are m ade jointly.

35 to 44 years

Only th e m o st im portant production an d m anagem ent decisions are m ad e Jointly.

45 to 54 years

Rarely are production an d m anagem ent decisions m ade Jointly
D ecisions are m a d e solely by th e tenant with so m e advice from the landlord

55 to 64 years
65 or m ore years

Production a n d m anagem ent decision are m ade solely by the tenant with n o Input
from th e landlord.

i
46. Your residence Is:

42. D o y ou Interact o n a social b a sis outside of the farming agreem ent? (P lease check
any that apply)

a

M em ber of the sa m e club, church, or other organization

com m ents, p lease provide them below.

Work (off farm) at th e sa m e location

43. From th e standpoint of fairness to you, how would you classify your leasing

arrangamen t(s)? (circle one)
1

2

3

4

5

Very b a d

P oor

A dequate

G ood

Excellent

44. O n av erag e, n e t Incom e from cro p an d livestock production or farm land rental
contributes w hat p ercen tag e of your total household Incom e? (check one)
L ess th an 30%
50% to 80%
More th an 80%

County

S tate

47. W e thank you for completing this questionnaire. If you have any additional

Enjoy leisure activities together

30% to 49%

_________

on

A PP E N D IX II

STA TISTICA L D A T A
FR O M T H E LO G ISTIC R EG R E SSIO N S

165

166
In this appendix the output of the regression runs is
provided with explanation of the statistics.
Landlord regression:
Choice of contract variable Q10 dependent.
LOGISTIC REGRESSION Q10 WITH CROP CONT Q42 DEC Q40 Q8 ORA Q43
Q36 Q 3 7 EXP LAN Q 9 .
Number of selected cases:

97

Number rejected because of missing data:

22

Number of cases included in the analysis: 75
Dependent Variable..

Q10

Beginning Block Number
-2 Log Likelihood

0.

Initial Log Likelihood Function

94.030419

* Constant is included in the model.
Two goodness

of

fit

tests

are performed on the

regression

model. The first compares the goodness of fit to a model using
only

a constant

second

then

in

the

examines

regression

the

(a perfect

goodness

of

fit

model).

with

all

The
the

variables.
Log likelihood (versus perfect model)
To test the null hypothesis that the observed likelihood
does not differ
model

that

likelihood

from

fits

1 (the value of the likelihood for a

perfectly),

we

use

the

value

of

(-2LL). Under the null hypothesis that

fits perfectly, -2LL has a chi-square distribution

-2
the

log
model

with

N-p

degrees of freedom, where N is the number of cases and p is
the

number

of

observations

parameters

and

13

estimated.

parameters

Since

estimated,

there
the

are

75

degrees

of

freedom are 62. Consulting the chi-square table the value -2LL
of

94.03

and

62

degrees

of

freedom

corresponds

to

a

167
significance

level

off

the charts

or

greater than the

.01

level of significance. Since the observed significance level
is large, we do not reject the hypothesis that the model fits.
Variable(s) Entered on Step Number
CROP

type of crop grown

CONT

number of contract or parcels of land

Q42

degree of interaction socially

DEC

participation in decision making

Q40

self risk evaluation

Q8
ORA

relationship to other party in the lease

Q43

evaluation of fairness

Q36

formality of the agreement
other parties demonstrated care for ,
assets

Q37
EXP

continuation of the lease
years of experience

LAN

years of experience on this land

Q9

condition of the farm at beginning of
lease

Estimation terminated at iteration number 6 because
Log Likelihood decreased by less than .01 percent.

Goodness of fit with all variables
A second test of goodness of fit of the model examines
the improvement as variables are added to the regression. The
high

significance

values

for

lines

one

and

four

further

substantiate the findings of the first test above. In line two
a comparison is made between the model with only the constant
(-2LL value 94.03 above) and the complete model (48.72 in line
1). The difference is called the model chi-square (45.31). The
degrees of freedom is also the difference between the values
for the two models. The low significance value means that we
reject the hypothesis that the coefficients for the variables
are all equal to 0.

168

Chi-Square

df Significance

-2 Log Likelihood

48.720

61

.8718

Model Chi-Square
Improvement
Goodness of Fit

45.310
45.310
42.588

13
13
61

.0000
.0000
.9649

Classification Table for Q10
Predicted
share

cash

Percent Correct

c

s
Observed
share

s

1

15

|

9

1

62.50%

cash

c

|

7

1

44

|

86.27%

Overall

78.67%

Variables in the Equation
Variable
CROP
C0NT
Q42
DEC
Q40
Q8
ORA
043
Q36
Q37
EXP
LAN
09
Constant

-1.4453
.4879
2.3125
1.5514
-1.7011
2.8485
.0737
2.2290
.2136
1.8838
-1.5220
4.0931
-1.7160
-2.0079

S.E.

Wald

.8620
1.0256
.9971
.9302
.9178
1.3744
1.3982
1.2179
.7714
1.7017
2.0253
2.4676
.8222
3.1300

2.8113
.2263
5.3786
2.7814
3.4352
4.2953
.0028
3.3500
.0767
1.2255
.5648
2.7514
4.3554
.4115

Sig

R

Exp(B)

.0936 -.0929
.2357
.6343
.0000 1.6289
.0204
.1896 10.0992
.0954
.0912 4.7180
.0638 -.1235
.1825
.0382
.1562 17.2620
.9580
.0000 1.0764
.0672
.1198 9.2909
.7819
.0000 1.2381
.2683
.0000 6.5782
.4523
.0000
.2183
.0972
.0894 59.9245
.0369 -.1583
.1798
.5212

169

LOGISTIC REGRESSION Q10 WITH CROP CONT Q42 DEC Q40 KNO HEL Q38
ORA Q43 Q 3 7 .
Dependent Variable..

Q10

Beginning Block Number

0.

-2 Log Likelihood

type of lease chosen
Initial Log Likelihood Function

119.19557

* Constant is included in the model.
Again we

fail

to

reject

the

hypothesis

that

the

model

is

significantly different from the perfect model.
) Entered on Step Number
type of crop grown
CROP
CONT

number of contracts or parcels of land

Q42

degree of interaction socially

DEC

participation in decision making

Q40
KNO
HEL

self risk evaluation
tenant perception of landlord's knowledge
tenant perception of 11 willingness to
help

Q38

current relationship

ORA

formality of the agreement

Q43

evaluation of fairness

Q37

continuation of the lease

Estimation terminated at iteration number 5 because
Log Likelihood decreased by less than .01 percent.
Chi-Square

df Significance

170
-2 Log Likelihood

70 .835

106

.9966

Model Chi-Square

11

.0000

Improvement

48 .361
48 .361

11

.0000

Goodness of Fit

91 .993

106

.8319

Testing and outcome are consistent with the landlord equation
above.
Classification Table for Q10
Predicted
share

cash

Percent Correct

Observed
share

13

cash

11

54.17%

86

91.49%

Overall

83.90%

Variables in the Equation - -- Variable

B

CROP
CONT
Q42
DEC
Q40
KNO
HEL
Q38
ORA
Q43
Q37
Constant

.8238
-.2712
.0219
1.7024
-.1388
1.9115
2.0443
1.4897
1.9208
1.0622
3.5294
-11.0547

S.E.

Uald

df

.5539 2.2123
.1658 2.6767
.7680
.0008
.7665 4.9330
.7160
.0376
1.0049 3.6185
.7620 7.1976
1.1012 1.8301
1.0159 3.5749
.8606 1.5235
1.2645 7.7904
3.0799 12.8828

1
1
1
1
1
1
1
1
1
1
1
1

Sig

R

Exp(B)

.0422 2.2792
.1369
.7624
.1018 -.0753
.0000 1.0222
.9772
.1569 5.4872
.0263
.0000
.8462
.8704
.1165 6.7635
.0571
.2088 7.7235
.0073
.0000 4.4356
.1761
.1149 6.8262
.0587
.0000 2.8928
.2171
.2204 34.1050
.0053
.0003

B IB L IO G R A PH Y

A dam s, D ale W., and N orm an Raslc, "Economics of Cost Share Leases in Less
D eveloped Countries", Am erican Journal o f Agricultural Econom ics, 50 935-942
(1968)

A lbrecht, D on E. and John K. Thom as, "Farm T enure: A R etest of Conventional
Knowledge", Rural Sociology, 51 18-30 (1986)

Allen, Franklin, "On the Fixed N ature of Sharecropping Contracts", The Econom ic
Journal 95 30-48 (1985)

Allen, Franklin, "On Share C ontracts and Screening", Bell Journal
o f Econom ics, 13 541-47 (1984)

A lston, Lee J. Sam ar D atta, and Jeffery Nugent, "Tenancy Choice in a Com petitive
Fram ew ork with T ransaction Costs", Journal o f Political Econom y 92 1121-33 (1984)

A pland, Jeffrey, R o b ert Barnes, and Fred Justus, "The Farm Lease: An Analysis of
O w ner-T enant and L andlord P references U nder Risk", Am erican Journal o f
Agricultural Econom ics 66 no. 3 376-384 (1984)

B ardhan, P. IC, and T. N. Srinivasan, "Cropsharing Tenancy in Agriculture: A
Theoretical and Em pirical Analysis", A E R
48-64 (1971)

B ardhan, P ranab IC, "Agricultural D evelopm ent and Land Tenancy in a Peasant
Economy: A T heoretical and Em pirical Analysis", Am erican Journal o f Agricultural
Econom ics 61 48-57 (1979)

B aron, D onald, "Status o f farm land leasing in the U nited States", In Rents and rental
practices in U.S. agriculture", eds. J. P eter D eB raal and G ene W underlich, Farm
171

172
F oundation (1983)

B aron, D onald, "Transaction Costs, Risk Aversion, and Choice of T en u re Revisited",
Agricultural Econom ics Research 32 26-35 (1980)

Berry, Russel L., "Cost Sharing as a M eans of Improving the Share R en t Leases",
Journal o f Farm Econom ics 44 796-807 (1962)

Bliss, C. J. and N. H. Stern, "Palanpur: T he Econom y of an Indian Village", Oxford:
Clarendon Press,
(1982)

Bogue, Allen G., "Foreclosure Tenancy on the N orthern Plains", Agricultural History
39 3-16 (1965)

Boxley, R o b ert F., "Cost-Share Leases Revisited . . . Again", Am erican Journal o f
Agricultural Econom ics 53 529-32 (1971)

Boxley, R o b ert F., "Econom ics of C ropshare, O nce Again: Reply", Am erican Journal
o f Agricultural Econom ics 54 688-89 (1972)

B raverm an, A., and J. E. Stiglitz, "Sharecropping and the Interlinking of A grarian
M arkets", Am erican Econom ic Review 695-715 (1982)

B raverm an, Avishay and Joseph E. Stiglitz, "Cost-Sharing A rrangem ents under
Sharecropping: M oral H azard, Incentive Flexibility, and Risk", Am erican Journal o f
Agricultural Econom ics 68 no.3 642-659 (1986)

Brown, D. J. and J. H. A tkinson, "Cash and Share Renting: An Em pirical T est of the
Link Betw een E n trep ren eu rial Ability and C ontractual Choice", Bell Journal o f
Econom ics 12 296-99 (1985)

Burns, Tom R., Thom as B aum gartner, and Philippe Deville M an, "Decisions, Society:
T he Theory of Actor-System Dynamics for Social Scientists", Soc 810 readings.

173
M ichigan State University

Carlson, John E., and D on A. Dillman, "Influence of kinship arrangem ents on farm er
innovativeness", Rural Sociology 48(2) 183-200 (1983)

Castle, Em ery, "Some A spects of the C rop-Share Lease",
177-179 (1952)

L a n d Econom ics

28

Cheung, Steven N. S., "Private property rights and sharecropping", Journal o f Political
Econom y 76(6) 1107-1122 (1968)

Cheung, Steven N. S., "Theory of Share Tenancy",

Chicago University Press (1969)

Cohen, M. and R. E. Nagel, "An Introduction to Logic and Scientific M ethod",
Chapter X X H arcourt B race, New Y ork (1934)

Currie, J. M., "The Econom ic Theory of Agricultural Land Tenure", Cam bridge
[England]

D atta, Sam ar K., D onald J. O ’H ara, and Jeffrey B. Nugent, "Choice of Agricultural
Tenancy in the Presence o f Transaction Costs", L a n d Econom ics 62 (2) 145-158
(1986)

Dillman, D on A. and John E. Carlson, " Influence of absentee landlords on soil
erosion practices", Journal o f Soil and Water Conservation 37 37-41 (1982)

Dowell, R. S.,
"Risk Diversification and Land T enure in the U nited States
Agriculture", 1890 to 1970 Ph.D. D issertation University o f Chicago
(1977)

D rake, Louis S.,
"Com parative Productivity of Share- and C ash-R ent Systems of
Tenure", Journal o f Farm Econom ics 34 no. 4 535-550 (1952)

Edw ards, William M. and Jess A. Robinson, "Effects of Professional Farm

174
M anagem ent on C rop-S hare Lease Terms", Journal o f the Am erican Society o f Farm
Managers and Rural Appraisers 48 (2) 40-44 (1984)

Ervin, Christine A. and D avid E. Ervin, " Factors affecting the use of soil
conservation practices: H ypotheses, evidence, and policy implications", L and
Econom ics
58(3) 277-292 (1982)

Ervin, David E., "Incentives for practicing soil conservation: T he influence of land
tenure", Am erican Farm land Trust, W ashington, D.C.
(1983)

Ervin, David E., "Conserving Soil: Insights from Socioeconomic Research", eds.
Stephen B. Lovejoy and T ed L. N apier,
Soil Conservation Society of A merica,
Ankey, IA 50021 (1986)

Esw aran, M ukesh and A shok Kotwal, "A Theory of C ontractual Structure in
Agriculture", Am erican Econom ic Review 75 352-67 (1985)

F riedm an, M ilton, "Essays in Positive Economics", The University of Chicago Press
39 (1953)

Gisser, M icha, "Economics of Cost-Share Leases: Comment", Am erican Journal o f
Agricultural Econom ics 51 692-95 (1969)

H allagan, W.,
"Self-Selection by C ontractual Choice and the Theory of
Sharecropping", Bell Journal o f Econom ics, 9 344-354 (1978)

H eady, E arl O., "Econom ics of farm leasing systems", Journal o f Farm Economics
29(93) 659-678 (1947)

H eady, E arl O., "Econom ics of A gricultural Production and R esource Use",
Englewood Cliffs, New Jersey: Prentice-H all, Inc. (1952)

H eady, E arl O.,

"Share Leases and the Interproduct Allocation of Agricultural

175
Resources", Southern E conom ic Journal

18 (3) 362-373 (1952)

H eady, E arl O. and E arl K ehrberg, "Relationships of C rop-share and Cash leasing
Systems to Farm ing Efficiency R esearch", Iowa Stat College A gricultural E xperim ent
Station. Bulletin 386
(1952)

H iebert, L. D., "U ncertainty and Incentive Effects in Share Contracts", Am erican
Journal o f Agricultural Econom ics 60 536-539 (1978)

H offm an, Philip T., "The Econom ic T heory of Sharecropping in Early M odern
France", Journal o f Econom ic History 44 309-19 (1984)

H offm an, Philip T., "Sharecropping and Investm ent in A griculture in Early M odern
France", Journal o f Econom ic History 42 155-62 (1982)

H olstrom , Bengt,
74-91 (1979)

"Moral H azard and Observability", Bell Journal o f Econom ics

10

H urlburt, V. L.,
"Farm rental practices and problem s in the Midwest", N orth
C entral R egional Publication No. 50, Iowa S tate University, Ames. (1954)

Hurwicz, Leonid and L eonard Shapiro, "Incentive Structures Maximizing Residual
G ain U n d er Incom plete Inform ation", Bell Journal o f Econom ics 9 180-91 (1978)

Hayami, Yujiro and Keijiro O tsuka, The Econom ics o f Contract Choice, Oxford
University Press, New Y ork (1993)

Ip, P. C. and C. W. Stahl "Systems of Land T enure, Allocative Efficiency, and
Econom ic D evelopm ent", Am erican Journal o f Agricultural Econom ics
60 19-28
(1978)

Issawi, Charles,
"Farm O u tp u t U n d er Fixed R ents and Share Tenancy",
Econom ics 33 74-77 (1957

Land

176
Janssen, L., "Farm land ten u re and ownership trends in South D akota", Econom ics
N ew sletter No. 199, South D akota State University, Brookings.
(1983)

Janssen, L. and S. Peterson, "Farm land cash rental term s in South D akota",
Econom ics N ew sletter No. 242, South D akota State University, Brookings. (1986a)

Janssen, L. and S. Peterson, "C ropshare rental term s in South D akota", Econom ics
N ew sletter No. 243, South D ak o ta State University, Brookings.
(1986b)

Johnson, B. B., L. Janssen, M. Lundeen, and J. D. Aiken, "Agricultural land leasing
and rental m arket characteristics: Case study of South D akota and Nebraska",
U npublished com pletion report, US D ep artm en t of A griculture, W ashington DC
(1987)

Johnson, B. B., " T he F arm land R ental M arket--A Case Analysis of Selected Corn
Belt A reas. A gricultural Econom ics R ep o rt No. 235, M ichigan State University, E
Lansing.
(1972)

Johnson, D. G ale, "Resource allocation under share contracts", Journal o f Political
Econom y
58(2) 111-123 (1950)

Kim, ICwan S., "Economics of Cropshare, O nce Again: Comment", American Journal
o f Agricultural Econom ics 54 686-88 (1972

K loppenburg, Jack R. Jr. and C harles C. Geisler, "The Agricultural Ladder: A grarian
Ideology and the Changing Structure of U.S. Agriculture", Journal o f Rural Studies
1 59-72 (1985

Looney, J. W., "Distinguishing Leases from o th er Agreements", In Rents and Rental
Practices in US Agriculture, eds. J. P eter D eB raal and G ene W underlich, Farm
Foundation 55-72 (1983)

Lucas, R. E. B., "Sharing, M onitoring, and Incentives: M arshallian M isallocation
R eassessed”, Journal o f Political Econom y
501-521 (1979)

177
M arshall, A., Principles o f Economics, 8th Edition London: M acmillan & Com pany
(1930)
M atoussi, M oham ed S. and Jeffrey B. N ugent, "The New Institutional Economics and
D evelopm ent T heory and A pplication to Tunisia", C h ap ter 4 The Switch to
Sharecropping in M edjez-el-bab eds. M ustapha K. Nabli and Jeffrey B. N ugent 140-57
N orth-H olland (1989)

Mill, John S., Principles o f Political Econom y 4th ed.;

London (1920)

Newberry, D. M. G., "The State of Leasing Theory: Conflicts Among D ifferent
M odels of D ecision M aking U n d er U ncertainty", In Rents and Rental Practices in U.S.
Agriculture eds. J. P ete r D eB raal and G ene W underlich, Farm Foundation (1983)

O tsuka, Keijiro and Yujiro H ayam i, "Theories of Share Tenancy: A Critical Survey",
Econom ic Development and Cultural Change
31-68 (1988)

O tsuka, Keijiro, Hiroyuki Chum a, and Yujiro Flayami, "Land and Labor C ontracts
in A grarian Econom ies: Theories and Facts", Journal o f Econom ic Literature, XXX
1965-2018 (D ecem ber 1992)

P eterson, Scott R., and Larry L. Janssen, "Farm land leasing in South D akota
A gricultural E xperim ent Station R ep o rt No. B 704", South D akota State University,
U.S. D ep artm en t of A griculture.
(1988)

R ao, C. and IT. H anum antha, "Uncertainty: E n trepreneurship and Sharecropping
in India", Journal o f Political Econom y 79 578-595 (1971)

R eid, Joseph D.,
"Sharecropping and agricultural uncertainty", Econom ic
Development and Cultural Change 24(3) 549-576 (1976

Robison, Lindon J.,
"Deductive Implications of Social D istance Models", Staff
P ap er D ep artm en t o f A gricultural Economics, M ichigan State University
(1987)

178
R obison, Lindon J., "Present V alue M odels and T heir A pplication to Investm ent
Analysis", Staff P ap er D ep artm en t of Agricultural Economics, M ichigan State
U niversity
(1988)

Schickele, R., "Effect of ten u re systems on agricultural efficiency", Journal o f Farm
Econom ics 23(10) 185-207. (1941)

Schmid, A. Allan, Properly, Power, and Public Choice, P raeger Publishers, O ne
M adison A venue, New York, NY (1987)

Scott, J. T. Jr., "How to determ ine the cost of land by observing rents", In Rents and
rental practices in U.S. agriculture, eds. J. P eter D eB raal and G ene W underlich,
F arm Foundation, 93-116 (1983)

Scott, John T. Jr., "Leasing R ecom m endations for Less D eveloped C ountries An
Extension of Leasing T heory", Am erican Journal o f Agricultural Economics, C om m ent
52 610-613 (1970)

Siles, M arcelo E., "The R ole of Social Capital in M ichigan C redit M arkets", PhD
dissertation, M ichigan State University (1992)

Smith, A dam , The Wealth o f Nations

Vol. 2 London: Cadell and Davies, (1805)

Spillman, W., "The A gricultural L adder", Am erican Econom ic Review 9 170-9 (1919)

Sutinen, J. G.,
"The R ational Choice of Share Leasing and Im plications for
Efficiency", Am erican Journal o f Agricultural Economics 57 613-621 (1975

Sutinen, J. G.,
"Agricultural Share C ontracts and Risk", In Rents and Rental
Practices in US Agricultural, eds. J. P eter D eB raal and G ene W underlich, Farm
F oundation 1-17 (1983)

Thurow, L ester C., Dangerous Currents: The State o f Economics, R andom H ouse Inc.,
New Y ork (1983)

179

W underlich, G., "Facts of agricultural leasing", In Rents and rental practices in U.S.
agriculture, eds. J. P ete r D eB raal and G ene W underlich F arm Foundation (1983)