I

71-2098
KREBS, Jr. Edward Hugo, 1944SIMULATED PRICE AND SUPPLY CONTROL PROGRAMS
FOR THE MICHIGAN NAVY BEAN INDUSTRY.
Michigan State University, Ph.D., 1970
Economics, agricultural

University Microfilms, A XEROXC om pany, A nn Arbor, Michigan

THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED

SIMULATED PRICE AND SUPPLY CONTROL PROGRAMS
FOR THE MICHIGAN NAVY BEAN INDUSTRY
By
C
Edward H^ JKrebs
;

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Agricultural Economics
1970

PLEASE NOTE:
Some pages have small
and indistinct type
Filmed as received*
University Microfilms

ABSTRACT
SIMULATED PRICE AND SUPPLY CONTROL PROGRAMS
FOR THE MICHIGAN NAVY BEAN INDUSTRY
By
Edward H. Krebs
Michigan produces over 99 percent: of the navy beans
grown in the United States.

Navy beans are one of the most

important agricultural products grown in Michigan and have
a wide domestic and export market.

Because this crop has

had wide yield, price, and income fluctuations in the past,
a government price support program has been in effect in
the industry for the last 30 years to help reduce the fluc­
tuations in income and to increase producer income.
The objectives of this study are;

(1) to estimate

the basic supply and demand relationships existent in the
Michigan navy bean industry for the period from 1951 - 1967
and

(2) to evaluate whether selected price and supply con­

trol programs would be desirable to employ in the Michigan
navy bean industry, either individually or when combined
with a government price support program.

Such information

might be particularly useful if the government were to

Edward H . Krebs
reduce its expenditures for agricultural support programs
to any extent.
To fulfill the first objective, an economic model
consisting of a navy bean acreage planted equation and a
three equation demand structure

(domestic demand, exports,

and price of small white beans equations) were developed.
The first equation was estimated using ordinary least
squares and the demand structure was estimated using three
stage least squares.

A high navy bean planted acreage was

found to be associated with a high navy bean price to pro­
ducers the previous year.

In the demand structure, both

domestic consumption and exports moved in the opposite
direction of the price of navy b e a n s , though both have a
low responsiveness to change in navy bean p r i c e .
The estimated supply-demand relationships were the
basis of a recursive simulation model used to examine dif­
ferent price and supply control programs.
examined with the simulation model include:
in the government price support level,
market situation,
program,

The programs
(1) variations

(2) an unsupported

(3) a private price support - storage

(4) a marketing quota,

and (6) a two price system.

(5) an acreage control,

The program simulations were

run for the period 1953 - 1967 and were intended to answer
the question:

What would have happened during this period

had the programs under study actually been in operation?
The results of the variations in the government price
support level simulations showed that this program did

Edward H. Krebs
increase producer gross income from navy beans over what:
would have occurred in an unsupported situation.

The lower

the price support level was set, the less effective the
government program and the lower the government expenditures
A private price support - storage program would have been
able to decrease price fluctuations in some crop years,
but would not have substantially improved producer income
from navy beans.

A marketing quota and an acreage control

would have been able to increase producer gross income from
navy beans, but would also have decreased available supplies
Under a two price system, producer gross income from navy
beans would have been increased more in the fifteen years
if the domestic price had been set at higher levels.
The general conclusions obtained from the examination
of the different control programs was that such programs
would have increased producer income from navy beans above
what it would have been without controls.

The government

price support program at the actual price support levels
was superior in this regard

(judging from the price levels

considered for the private programs).

At some point all the

programs ran into constraints that limited their ability to
increase producer income or price beyond a certain point.
Not all the programs examined could have been used
in the actual industry because of current legal structure
and/or organizational deficiencies.

The different industry

Edward H. Krebs
groups and the public must decide if any of the different
programs examined should be tested in the actual market
or if they should be studied in further detail.

ACKNOWLEDGMENTS
The author wishes to express his appreciation to
Dr. Marvin Hayenga, chairman of the Thesis Committee, for
his advice and counsel during the graduate program and the
development of this study.
Appreciation is also extended to members of the
Thesis Committee, Dr. Lester Manderschied and Dr. James
Shaffer, for their helpful suggestions.
The author also wishes to thank Dr. L. L. Boger and
Dr. Dale E. Hathaway for the financial assistance that the
Department of Agricultural Economics provided for the
study and for the entire graduate program.
And last, but not least, the author expresses his
gratitude to his wife, Ruth, for her patience, understand­
ing, and sacrifice throughout his graduate program.

ii

TABLE OF CONTENTS
LIST OF T A B L E S .....................................

Page
V

LIST OF F I G U R E S ...................................... vi
CHAPTER
I.

INTRODUCTION .................................

1

5
Objectives of the S t u d y ....................
Previous Research
..........................
8
Outline of the S t u d y ........................... 10
II.

GENERAL ORGANIZATIONAL SETTING OF THE
MICHIGAN NAVY BEAN I N D U S T R Y .................. 11
Early Development of the I n d u s t r y ........... 11
Early Cooperative Movement ..................
13
Cooperative Elevators Today
...............
15
..................
15
Cooperative Pooling Today
Government Price Support Program ...........
16
Markets for Navy B e a n s .........................18
Navy Bean Shippers and C a n n e r s ................ 20
The Michigan Bean Commission and the
Michigan Bean Shippers Association . . . .
22
S u m m a r y ........................................ 24

III.

FORMULATION OF THE ECONOMETRIC MODEL . . . .

25

S u p p l y .......................................... 25
D e m a n d .......................................... 28
Domestic Demand
......................
29
Demand for E x p o r t s ...................... 32
Price of Small White B e a n s .............. 33
Government Role in Navy Beans
. . . .
34
Model Estimation Procedure
..............37
IV.

ESTIMATED SUPPLY AND DEMAND RELATIONSHIPS
IN THE MICHIGAN NAVY BEAN INDUSTRY, 1951-67.

41

Estimated Results
..........................
41
Navy Bean Acreage P l a n t e d ..............45
Navy Bean Domestic D e m a n d ..............47
Navy Bean Export D e m a n d ................ 49
Small White Bean P r i c e .................. 50
iii

Page
Data Required for More Detailed Analysis . . 51
Summary and Conclusions......................... 52
V.

EVALUATION OF PRICE AND SUPPLY CONTROL
P R O G R A M S ........................................ 54
Basic Dynamic F r a m e w o r k ...................... 54
60
Validation of the M o d e l .............
Program Testing and E v a l u a t i o n ................ 67
Variation of the Government Price
Support Level
......................
68
Private Price Support-Storage
P r o g r a m ................. ............73
Marketing Quota or Control ...........
78
Acreage Control Program
...........
82
Two Price System (Price Discimina................................. 86
tion)
S u m m a r y ................................. 91

VI.

IMPLICATIONS FOR DIFFERENT INDUSTRY GROUPS .

97

P r o d u c e r s ...................................... 98
Consumers and T a x p a y e r s ..................... 102
Farm Input G r o u p s ............................ 105
Shippers . . . . .
107
P r o c e s s o r s ..................................... 109
Required Institutional Setting ............. 110
VII.

S U M M A R Y ....................................... 117

LIST OF R E F E R E N C E S .................................. 131
A P P E N D I C E S ........................................... 135
A.
B.

The Ontario Navy Bean I n d u s t r y .......... 135
Data Used in the Study but not
Presented in the T e x t ................. 142

LIST OF TABLES
Table
1-1
II-l
11-2

IV-1
V—1

V-2

V-3

V-4
V-5
V-6
A-l
B—1

Page
U.S. and Michigan Production of Dry Beans
and Navy Beans, 1951-67 ......................

2

Government Support Price, Takeover and Ex­
penditures for Michigan Navy Beans, 1951-67 .

19

U.S. Exports of Navy Beans and Navy Bean
Exports as a Percentage of U.S. Navy Bean
Production, 1951-67 ..........................

21

Estimates of Structural Equations in
the M o d e l ........................................ 42
Actual and Simulated Results of the Navy
Bean Government Price Support Program,
1953-67 ........................................

62

Average Annual Results for Actual,
Simulated Actual, and Simulated Reduced
Navy Bean Price Support Levels, 1953-67 . . .

70

Simulated Results for Ten Minimum Price
Support Levels Operated by a Private
Storage Agency, 1953-67 ......................

76

Simulated Results for Ten Marketing Quota
Levels for the Period 1953-67 ...............

81

Simulated Results for Nine Selected Acreage
Control Levels for the Period 1953-67 . . . .

84

Results of the Two Price System Simulations
for the Period 1953-67

89

Ontario Production and Exports of Navy
Beans, 1958-67

137

Data Used in Study but not Presented in
the T e x t ....................................... 142

v

LIST OF FIGURES
Figure

Page

1-1

Dry Bean Acreage (Thousand Acres) by
Counties, 1 9 6 7 ..................................3

III-l

Operation of Government Price Support
P r o g r a m ....................................... 35

V—1

Information-Feedback Design Used in the
Simulation Model ............................. 55

vi

CHAPTER I
INTRODUCTION
Michigan produces approximately one third of the
dry beans grown in the United States

(Table 1-1).

12,391 farms in Michigan grew dry b e a n s .^

In 1964,

Dry beans pro­

duced in Michigan had a value of $45.7 million in 1967,
which was 13.5 percent of the value of all field crops
grown in the state that year.

2

In terms of cash receipts

from marketings, the dry bean industry was the s t a t e 1s
sixth largest agricultural enterprise in 1964.
By far the largest proportion of dry beans grown in
Michigan are navy beans

(Table 1-1).

In terms of physical

output, generally over 90 percent of dry bean production
in Michigan has been navy b e a n s .

Michigan now produces

over 99 percent of the navy beans grown in the United States.
Production within Michigan is concentrated in the "Thumb"
region of the state

(Figure 1-1).

Because of the concen­

tration of navy bean production in Michigan and the nature
^U.S. Bureau of the Census, 1964 United States Census
of Agruculture (Washington, D.C.:
Government Printing
Office, 1$64) Vol. I, part 13, p. 13.
2
Michigan Department of Agriculture, Michigan Agri­
cultural Statistics (Lansing, 1968).
3
U.S. Bureau of the Census, 1964 United States Census
of Agriculture, pp. 13-18.
1

2

TABLE 1-1
U.S. AND MICHIGAN PRODUCTION OF DRY BEANS AND
NAVY BEANS, 1951-67
(Production in 1000 cwt.)
Crop
Year
(Sept. 1)
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967

Sources:

U.S. Production3
Dry Beans
15,828
14,917
16,498
16,939
16 ,672
17,234
15,670
19,287
18,505
17,411
19,672
17 ,942
19,982
17 ,375
16,347
19,962
15,472

Navy Beans
4 ,072
3,412
3,601
3,158
4,428
5,020
3,358
5,042
6,069
5,845
6,755
6,725
7,599
6 ,801
5 ,480
7 ,289
4,787

Michigan Production*5
Dry Beans

Navy Beans

4 ,022
3,474
3,750
3,295
4 ,536
5,389
3,508
5,226
6,413
6 ,248
7 ,358
7 ,392
8,585
7,601
6 ,175
8 ,026
5,382

3 ,782
3,202
3,428
3 ,000
4,343
4 ,905
3,272
4 ,949
6,006
5,811
6 ,714
6,696
7,565
6,771
5,460
7,289
4,787

aU.S. Department of Agriculture, Statistical
Reporting Service, Crop Reporting Board, Crop
Production (Various Annual Summaries).
bMichigan Department of Agriculture, Michigan
Agricultural Statistics, (Lansing, Various
Issues).

3

3

OTStMO

r* * m rb

FIGURE 1-1
Dry Beans Acreage
(Thous. Acres)
By Counties,
1967

87.0
14.0
m om tC A u i

23.0162.0

62.0

a tT A m

Source t Michigan Depart
ment of Agriculture,
Michigan Agricultural
Statistics

17.0
L/rrmtf*

15.0
auM n

mm

87.0

21.0

of -the navy bean marketing problems, this study will focus
on the navy bean industry.
The navy bean industry has exhibited substantial
farm price, production, and farm income fluctuations in
the past.

Weather, yield, and acreage changes for this

weather-sensitive crop provide only part of the answer
for these fluctuations.

The minimal amounts of market

information available to some industry participants may
be a contributing factor.

There appears to be substantial

risk and uncertainty for all participants in the dry bean
industry because of the lack of market information.
A governmental price support program, which has been
in operation in the commodity for approximately thirty
years, has helped reduce fluctuations from year to year.
This program has resulted in substantial government ex­
penditures in the industry.

However, pressures being

exerted on the federal government by certain segments of
the U.S. society may force reallocation of federal funds
away from agricultural programs such as this one.

While

the price support program probably will not be eliminated,
the price support level and associated government expend­
itures may be reduced.

Alternative industry or government

programs might be considered by some industry participants
as a supplement to or replacement of the current or modi­
fied government price support program.

5
Objectives of the Study
In the last few years, there has been increasing
interest in examining various government or industry pro­
grams to determine effects of program shifts or adjustments.
Until the late 19 20's the traditional method of coordination
of production and marketing in U.S. agriculture was an open
market system with no price or production controls.

Under

this system wide fluctuations in production and prices
have occurred due to vagaries of weather, uncertain price
expectations, etc.

To alleviate the wide fluctuations and

general low level of income to the growers of these com­
modities, various programs have evolved which modify the
market.

These include bargaining associations, coopera­

tive marketing organizations, price support programs,
marketing orders, and production or marketing control pro­
grams .

Another program which has been used extensively

in Canada and Europe is the marketing board.

In 1966 the

U.S. National Commission on Food Marketing recommended
that Congress should provide the passing of similar en­
abling legislation for U.S. agricultural industries that
wish to create marketing boards.
Many different objectives have been suggested which
these supplementary institutions are intended to achieve.
The most important of these include:
quate consumer supplies,

(1) to insure ade­

(2) to improve the absolute level

of grower income and the returns to grower resources,

6
(3) to improve the stability of grower income and

(4) to

increase the market power of growers.
To fulfill these objectives, various programs have
been used.

A list of these includes:

(1) regulating

quantity flow within and between seasons,
supply or marketings,
ing and promotion,
trade practices,

(2) controlling

(3) regulating quality,

(5) research,

(4) advertis­

(6) prohibiting unfair

(7) minimum price guarantees for agri­

cultural commodities and

(8) providing market information

for commodity participants.
This study tests the hypothesis that certain price
and supply control programs other than those now existing
in the Michigan navy bean industry could have advantages,
either as replacements of or supplements to existing pro­
grams, for some interest groups.

While it will not be

possible to evaluate the effects of each program mentioned
above, some major supply and price control measures which
can be employed will be studied and their relative degree
of success or failure in serving particular interests will
be analyzed.

The programs which can incorporate one or

more of these measures can then be partially evaluated.
The programs to be examined are:
control of quantities marketed,

(1) acreage control,
(3) two price system,

non-governmental storage program minimum price, and
government price support program.

(2)
(4)

(5)

The market without

supply or demand controls will also be examined.

Some of

7
-these have already been tried in -the Industry while o-thers
have not.

They are, however, ones which economic theory

suggests might produce beneficial results for some in­
dustry participants.

New enabling legislation would have

to be passed before some of these measures could be used
in various industry programs.
To measure the advantages and test the validity of
the hypothesis for the various programs, several criteria
which appear applicable to one or more of the interest
groups concerned with industry performance will be used:
(1) level of producer income,
income,

(2) stability of producer

(3) stability of price,

(4) level of price,

(5)

amount of government expenditures, and (6) per capita
supplies of navy beans.

Since changes in some of the

criteria can be simultaneously "good" for one group and
"bad" in the view of another group, the desirability of
the programs is evaluated from several different, poten­
tially competing, viewpoints.

It is quite possible that

no one measure or program would make everyone better off.
However, the quantitative evaluation of these programs
should provide each interested group with some reasonable
estimates of potential benefits and costs of each program
which might be promoted by and adopted in the industry.
In summary, the general objectives of this study are:

8
1)

To estimate the basic supply and demand relation­
ships existent in the Michigan navy bean industry
in the past 17 years.

2)

To evaluate within the above supply and demand
context, the expected consequences of alterna­
tive methods of regulating or structuring the
market for navy b e a n s .

Previous Research
The amount of published economic research related
to the navy bean industry is limited.

Two studies examined

the effects of the government price support program on
the navy bean industry.

In 1955, Hathaway, doing a study

of the Michigan dry bean industry, found that the price
support program was an important factor in maintaining
incomes for bean producers, in some post-war years.

In

1967, Vandenborre, doing a general study of the important
dry bean varieties grown in the United States, concluded
that the price support program has substantially aided
5
navy bean producers in recent y e a r s .
4

Dale E. Hathaway, The Effects of the Price Support
Program on the Dry Bean Industry in Michigan (Michigan
Agricultural Experiment Station Technical Bulletin 250,
East Lansing, April 1955), p. 3.
^Roger J . Vandenborre, An Econometric Investigation
of the Impact of Governmental Support Programs on the
Production and Disappearance of Important Varieties of Dry
Edible Beans" (Giannxni Foundation Research T e p o r t No. 294,
California Agricultural Experiment Station, Berkeley,
December 1967), p. 85.

9
Both of these studies involved estimating econo­
metric models which were then used in running simula­
tions to evaluate the price support program against an
unsupported market.

However, neither study examined

situations other than the government price support pro­
gram or the unsupported market.

Further, neither study

differentiated commercial demand into domestic and export
segments.

With exports taking a large proportion of

production in recent years, a model explicitly incorporat­
ing exports in it should increase our understanding of
the forces operating in the industry.
A third study on the navy bean industry was pub­
lished by Hayenga in 1968.

His emphasis was on describing

the structure and related problems of the navy bean mar­
keting system.

He found that four shippers handled

approximately 95 percent of Michigan bean production.®
He also found that the industry was concentrated at the
7
canner level.
Some of these and other industry partici­
pants lack adequate market information which results in
increased risk and uncertainty for them.

This study will

be examining several programs which would necessarily
involve structural changes in the industry, and
g
Marvin L. Hayenga, Structure and Problems of the
Navy Bean Marketing System (Michigan State University
Agricultural Economics Report 91, East Lansing, April
1968), p. 9.
7Ibid., p. 17.

10
corresponding changes in the information available to
some industry participants.
Outline of the Study
Initially, the historical development of the
Michigan navy bean industry programs and organizations
will be described.

The annual supply and demand rela­

tionships which have recently existed in the Michigan
navy bean industry will then be modeled.

After econo-

metrically estimating the parameters of this model, the
implications of the estimates will be discussed.

These

estimated relationships will then provide an initial
basis for the supply and demand structure which will be
incorporated into a simulation model.

By modifying the

estimated relationships to conform with expected behavior
under alternative programs, the simulation model can be
used to evaluate the impact which particular production
or marketing programs would have had during 1953-1967.
Consequently, the likely impact of such programs can be
evaluated without resort to potentially "expensive"
actual implementation within the industry.

Using the

results of the simulation study, a descriptive analysis
of the effects of different programs will then be pre­
sented.

CHAPTER II
GENERAL ORGANIZATIONAL SETTING OF
THE MICHIGAN NAVY BEAN INDUSTRY
The success of any new industry program is tem­
pered by how compatible it is with existing organizations
and programs.

More importantly, the actual acceptance

and implementation of a new program is dependent on how
it is received by different industry participants.
Therefore, one should consider where the industry is at
present before he will be able to adequately judge any
new programs.
Early Development of the Industry^
Beans, as a field crop in Michigan, became import­
ant enough by 1884 to take a place in that year's state
census.

By 1900 Michigan was the leading producer of

dry beans in the United States.

Since that time, Michigan

has continued as the largest producer of dry beans in
this country.

The favorable soil and climatic conditions

1The historical facts in this section are summar­
ized from Wilbur O. Hedrick, Marketing Michigan Beans
(Michigan Agricultural Experiment Station Special Bulletin 217, East Lansing, Nov. 1931).

11

12
for beans in the "Thumb" region of the state is a major
factor leading to this situation.
While Michigan was gaining prominence in the grow­
ing of dry beans, production and consumption decisions
were based upon a price determined by supply and demand
conditions.

There was no attempt to establish minimum

price guarantees or to control supplies through pooling
arrangements by growers.
However, because of poor transportation conditions,
the price the grower received was determined by bargaining
between him and a local elevator operator.

With the

limited number of elevators in any one locality, growers
were often at a disadvantage in getting a price that
broader supply and demand conditions may have warranted.
As transportation improved, the market area served by
elevators expanded and the options open to the growers
increased.

Some elevators started providing storage at

reduced rates and more credit.

Besides buying and sell­

ing beans, the elevators also cleaned and graded them.
The actual marketing or selling of the beans to
canners and retailers was done by bean shippers.

In

1930 there were three or four shippers keeping a regular
daily buying market in beans, with twenty or more shippers
handling beans at some time during the year.

A few of

the shippers operated their own elevators while others
bought most of their beans from independent elevators.

13
Some of the independent elevators also sold direct ly
2
to canners and grocers.
Early Cooperative Movement

3

Cooperative elevators were established in the
Michigan bean industry after World War I.

In 1920 these

cooperative elevators associated themselves into a central
exchange

(Michigan Elevator Exchange) which acted as

their terminal sales agency.

At first these new coopera­

tive marketing organizations added no new features to
bean marketing, simply participating in the system like
any other elevator.

The producer members were issued

stock shares in the cooperatives and received patronage
dividends.
In 1931 the Federal Farm Board added impetus to
cooperative marketing by establishing a pool plan.

Under

this plan, farmers joined local cooperative associations
and pooled their beans.
2

After the marketing season they

At this time a sizable proportion of navy beans
went to the dry market. Although no data is available
as to the breakdown today, it is estimated by people in
the industry that the dry market now takes about ten per­
cent of production.
3
Thxs section was summarized from Hedrick, Market­
ing Michigan Beans, and Wilbur O. Hedrick, A Decade of
Michigan Cooperative Elevators (Michigan Agricultural
Experiment Station Special Bulletin 291, East Lansing,
May 19 38).
4
The Michigan Elevator Exchange is one of the
four large shippers still operating in the industry.

14
were to receive a pool price which reflected the benefits
of joining the association.

Growers also had the option

of selling their beans to the pool at the current market
price if they so desired.

Growers could be advanced

some funds on their beans if they delivered all their
beans to the p o o l .

The local cooperative associations

used the Michigan Elevator Exchange as their sales agency.
However, because of an industry practice requiring
payments upon delivery and the consequent high capital
requirement, the project failed during the depression.
Besides these two factors, the fact that the Farm Board
died a political death after three years also contributed
to the failure.

After the extinction of the Farm Board

and the collapse of the overall pool, several of the
local pool associations remained active.
The early cooperative movement had several lasting
effects on the industry.

The first was that local coop­

erative elevators and the Michigan Elevator Exchange
became a permanent fixture in the industry.

This has

increased the power of growers to some extent within the
marketing channel.

Second, while the pooling arrange­

ment was a failure, it marked the first attempt of the
government and the growers to cooperate in trying to
raise farm income for Michigan beans.

It probably left

growers with the lasting impression that government action
was one way to increase their income.

A third effect was

15
that/ while not always strong, an interest in cooperative
marketing pools has remained in the industry.
Cooperative Elevators Today
There have been cooperative elevators in the Michi­
gan dry bean industry for more than fifty years.

Until

1969, the cooperative elevators operated with the Michigan
Elevator Exchange as their marketing agency on a voluntary
basis:

that is, they could market their beans through

the Elevator Exchange if they so chose, but were not
required to do so.

Beginning with the 1969 crop year,

the agreement between the Exchange and the local coopera­
tives was drawn up so that the local participating coop­
eratives are required to market all their beans through
the Exchange.

The local cooperative elevators handle

approximately 25 percent of the total crop.

The Exchange

handles approximately 30-35 percent of the crop, picking
up the additional amount from non-cooperative elevators.
Cooperative Pooling Today
Since the collapse of the overall pool in the 1930's,
there have been continuing attempts by various coopera­
tive marketing associations to organize pools.

The latest

association, Michigan Bean Growers Marketing Cooperative,
was organized in March of 1965.

Its major goal was to

concentrate a major part of the bean supply in the control
of a grower marketing organization.

By doing this it was

16
hoped -that growers could increase their market power and
influence prices in their direction.

This association

set a goal of getting 75 percent of navy bean acreage
under membership before the pool would become operative.
Up to this time, the association has not been able to
attract the necessary acreage thought necessary to imple­
ment its program.
In 1968, however, the association had a local
cooperative elevator operate a smaller pool as a stop
gap measure until 75 percent of navy bean acreage could
be attracted into the larger p o o l .

The smaller pool

legally had no connection with the association and associ­
ation members were not required to join it.

While the

small pool did not raise prices, industry people suggest
that it might have kept fluctuations within the market­
ing system at a low level for the year it was in operation.
The association was relatively inactive during the 1969
crop year, but it plans a new membership drive for the
1970 crop.
Government Price Support Program
After the demise of the Farm Board, the federal
government moved toward price supports for agricultural
commodities.

However, there was no program intended

specifically to support the price of dry beans prior to

17
World War II.

5

Purchases of dry beans for rellev feed­

ing, school lunches, and other consumption programs did
affect market prices, but there were no indications that
this was their specific objective.
The first definite support program for dry beans
came as a result of World War II.

In 1941 a price sup­

port program was implemented to expand acreage of white
varieties of dry beans.6
purchases from dealers.

The program was operated through
The main objective of this pro­

gram was to increase acreage by stabilizing prices.
After the war, the first support program was ter­
minated.

It was replaced by a support program operated

through direct non-recourse loan and purchase agreements
with growers.

This latter type of support is still in

operation today.

The objectives of this program have

been to stabilize the industry and raise producer income
while maintaining at least a minimum acreage of beans.^
This program has helped increase farmer income over what
would have resulted under an unsupported market system.
It has also increased stability over what it would have
been under an unsupported market system.
The magnitude of government takeover of Michigan
navy beans and government expenditures for this takeover
SDale E. Tathaway, The Effects of the Price Support
Program on the Dry Bean Industry in Michigan, p. 13.
^Ibid., p. 14.
^Ibid., p. 15.

18
shows the dependence of this Michigan industry on the
price support program for the period 1951-67.
ment has had takeover in 14 of the 17 years

The govern­

(Table II-l).

In this 17 year period approximately $87 million has been
spent by the government in taking over an average 13.5
percent of Michigan navy bean production.

This dependence

on the price support program has been persistent even
though the support price level was adjusted downward
during this time period.
The government takeover of navy beans from this
program has partially been disposed of through non-market
channels such as the school lunch program and Public
Law 480 donations.

There has also been some disposal

back into normal market ghannels, both domestic and ex ­
port/ but these statistics are not available.

Therefore,

the total effect of government actions on the commercial
market can not be fully evaluated.
Markets for Navy Beans
The outlets for navy beans have changed during
the last thirty years.

Thirty years ago a large percent­

age of the navy beans sold to consumers were in the dry
form.

Today, it is estimated that 90 percent of navy

bean production goes to the domestic consumer in canned
form.

While navy beans compete with great northern beans

in the dry market, great northerns are not effective

19
TABLE II-l
GOVERNMENT SUPPORT PRICE, TAKEOVER AND EXPENDITURES
FOR MICHIGAN NAVY BEANS, 1951-67a
A
YEAR

1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
Total

Support
Price
$
7.94
8.75
8.80
8.41
7.43
7.38
7.29
7.17
6.43
6.46
7.15
7.15
7.15
7.15
6 .90
6.90
6.90

B
Amount
Government
Takeover
1000 cwt
1130
260
377
0
623
1747
31
9
177
1838
1611
964
1012
601
0
1676
0
12,056

C
Government
Expend!ture s®

8,972,200
2,275,000
3,317,600
0
4,628,890
12,892,860
225,990
64,530
1,138,110
11,873,480
11,518,650
6,892,600
7,235,800
4,297,150
0
11,564,400
0
86,897,260

Sources: For years 1951-54 and 1957-67 a per­
sonal communication with the U.S. Department of Agriculture,
Agricultural Stabilization and Conservation Service, East
Lansing, Michigan.
For years 1955-56, U.S. Department of
Agriculture, Federal Extension Service, Statistical Summary:
Dry Edible Beans and Dry Field P e a s , {October, 1964), p. 31.
u
Government expenditures are equal to the support
price per hundred-weight multiplied by takeover per hundred­
weight.

competitors in the canned market.

This leaves small

white beans, grown in California, as the main competitor
in the white bean canned market.
Exports add another dimension to the navy bean
market.

Navy beans have become increasingly dependent

on the export market

(Table II*-2) .

The United Kingdom

is the primary export market where Michigan navy beans
compete primarily with Ontario navy b e a n s .
Navy Bean Shippers and Canners
There are two types of marketing firms which
handle navy beans between the time they leave the pro­
ducer and the time they reach the wholesaler or retailer.
They are the bean shippers and bean canners.
Producers generally deliver their navy beans to
firms commonly called navy bean shippers.

These shippers

operate grain elevators which provide storage, drying,
and transportation loading facilities.

The shippers buy

beans from producers and then sell them to processors or
canners.

The canners can be domestic or foreign.

The largest bean shippers are corporate or coopera
tive elevator chains which control a number of elevators.
The four largest are the Michigan Bean Company, the
Michigan Elevator Exchange, Frutchey Bean Company, and
the Wallace and Morley Company.

It is estimated that

TABLE II-2
U.S. EXPORTS OF NAVY BEANS AND NAVY BEAN EXPORTS AS A
PERCENTAGE OF U.S. NAVY BEAN PRODUCTION, 1951-1967
(Navy Bean Exports in 1000 cwt.)

Crop Year
Starting
Sept. 1

Navy Bean
Exports3

Exports as
a % of U.S.
Production15

Crop Year
(Sept. 1)

Navy
Bean
Exports

Exports as
a % of U.S.
Production

1951

701

17

1960

640

11

1952

250

7

1961

902

13

1953

716

20

1962

1,704

25

1954

52

2

1963

1,938

26

1955

674

15

1964

1,550

23

1956

1,209

24

1965

1,075

20

1957

8

0

1966

2,044

28

1958

1,029

20

1967

762

16

1959

1,433

24

SO\*L3TC6S *

cl

For years 1951-53, Roger J. Vandenborre, An Econometric Investigation
of the Impact of Governmental Support Programs on the Production and Disappear­
ance of Important Varieties of Dry Edible Beans (Giannini Foundation Research
Report No. 2i?4, California Agricultural Experiment Station, Berkeley, Dec. 1967).
For years 1951-67, U.S. Department of Agriculture, Agricultural Statistics
(Washington, D. C.: Government Printing Office, Various Issues).
^Production statistics taken from:
Crop Production (Various Annual Summaries).

U.S. Department of Agriculture,

22
these four shippers handle 90-95 percent of the navy
Q
beans produced in Michigan.
Domestically, navy bean canners are the primary
buyers of navy beans from these shippers.

Canners also

are the most important outlet for navy beans destined
for consumption in the U.S.

There are twelve major navy

bean canners which buy dry beans and process them into
canned products.

Two of these, Stokely Van-Camp, Inc.

and Campbell Soup Company reportedly sell nearly one
half of the canned beans sold in the United States.

9

The navy bean canners are generally large multi-product
companies which have canned navy beans as part of their
product line.
The Michigan Bean Commission and the Michigan Bean
Shippers Association
The Michigan Bean Commission was legislatively
commissioned to begin January 1, 1966.^°

The commission

consists of six grower representatives, two shipperhandier representatives, and one bean canner representa­
tive, with ex officio members from the Michigan
0
This discussion is taken from Marvin L. Hayenga,
Structure and Problems of the Navy Bean Marketing System,

pTT

9Ibid., p. 17.
10Michigan Department of Agriculture, Bean Commis­
sion Law {Pamphlet put out by the Commission, Act No.
ll4, Public Acts of 1965, Lansing).

i

23
Department: of Agriculture and the Michigan State Uni­
versity College of Agriculture.
The commission may levy and collect assessment
(currently 2 cents per hundredweight)

from all partici­

pating dry bean growers in Michigan.

According to the

Bean Commission Law, the Commission may allocate its
funds into various bean advertising, research, informa­
tion or market development activities.

The Commission

maintains an office in Lansing and has been active in
promoting navy bean products and supporting a small
amount of research related to navy bean production
practices, etc.
A marketing organization that has been in exist­
ence in the industry since 1892 is the Michigan Bean
Shippers A s s o c i a t i o n . ^

Its main objective is to ad­

vance the general interests of those engaged in the
growing, handling and processing of Michigan dry edible
beans.
The main function of this association is to es­
tablish trade rules for industry shippers, to acquire
and disseminate pertinent information among its members,
and to assist in establishing and maintaining uniform
grades for Michigan dry edible b e a n s .

The establishing

^ M i c h i g a n Bean Shippers Association, Constitution
and By-Laws of the Michigan Bean Shippers Association,
(Publication put out by the Association, Saginaw, I3TT5) .

24
and maintaining of grades is done in conjunction with
the Michigan Department of Agriculture and the United
States Department of Agriculture.

The Association main­

tains an office in Saginaw.
Summary
, Michigan has been prominent in the growing of
beans since the turn of the century.

During this time

period, local elevator cooperatives, grower marketing
pools, and a cooperative elevator exchange were started.
Many of the local cooperative elevators and the elevator
exchange are still operating.

Other supporting market

organizations in the industry are the Michigan Bean
Commission and the Michigan Bean Shippers Association.
The industry has also become highly concentrated at the
shipper and canner level.
While all of these have played an important part
in the industry,

the dominant feature in the industry

for the last thirty years has been a government price
support program which has set a minimum price in the
industry.

This has given the industry participants some

protection against production and price fluctuations.
The producer and market organizations in the industry have
become dependent on this program.

Thus, the way the in­

dustry is now structured can in part be considered a re­
sult of this program, and a definite factor affecting the
likely acceptance of any new or changed programs in the
industry.

CHAPTER III
FORMULATION OF THE ECONOMETRIC MODEL
Annual navy bean production is primarily deter­
mined by the acreage planted and subsequent weather
during the single growing and harvest season each year.
Once production is determined, it plus carryover from
the previous year define the U.S. supply for the current
year.

Since supply is fixed after harvest, demand con­

ditions primarily determine the price at which the cur­
rent year's supply will be sold.
The primary factors affecting annual navy bean
supply and demand will be examined and formulated into
a recursive econometric model.

The available data and

appropriate estimation procedure are subsequently con­
sidered.

supply
Determining the acreage planted to navy beans is
the first step in determining production and, consequently,
supply of this crop.

Navy beans are produced in Michigan

as a cash crop; therefore, it is hypothesized growers
respond readily to economic factors likely to change
their income from navy b e a n s .
25

Navy bean price is an

26
important element affecting profitability and net income
of navy bean producers.

Thus, one would expect that

navy bean producers respond to a higher expected price
by increasing their planted acreage.
The specific way in which growers form their
price expectations is probably complex and highly vari­
able among growers.

In this study, it is assumed that

an 8-month average price (September to April) during
the previous marketing season is the primary basis on
which growers base their price expectations for next
year and their corresponding planting decisions.

Since

much of the price variability in the industry arises
from abandonment of acreage and yield variability, the
grower has little more than last year's experience as
to what prices might be in the current year, although
USDA planting intentions report may plan a minor role.^
A second element that can affect grower acreage
decisions is the price of competing crops that can be
grown on the same land.

Corn is the principle competi­

tor of navy beans in Michigan.

It is hypothesized that

an increase in the average price of corn for the months
October to April will lead to a decrease in the acreage
planted to navy beans in the current year.
^The announced government support price might also
affect acreage planting decisions for some commodities.
However, when equations were run making acreage planted
a function of the announced support price, no statisti­
cally significant relationships were found.

27
A third variable assumed to be associated with
acreage planted is the acreage planted the previous year.
It is expected that acreage planted in the current year
is positively related to acreage planted in the previous
year; that is, if acreage planted was high in the pre­
vious year, it will be high in the current year.

While

this variable is primarily a predictor of acreage planted,
several possible causal explanations can help explain
this type of producer behavior.

Producers could continue

to grow the same crops because of habit or specialized
know-how.

They might also have crop rotations which

they want to maintain or investments in specialized
machinery which they want to recover.
The following equation represents the important
variables assumed to influence planted navy bean acreage:
APNB = f(PNB
PC
APNB
)
t-1'
t-1'
t-1
where:

APNB is acreage of navy beans planted in Michigan
in thousands of acres.
PNB is average price (September to April) for
choice handpicked navy beans in dollars per 2
hundredweight paid to producers in Michigan.
PC is average price (October to April) for corn
in dollars per bushel paid to Michigan producers.

2

This price is a simple unweighted average of the
quotation on the fifteenth of each month; 1951-56, Michi
gan Bean Company; 1957-67, Michigan Elevator Exchange.

28
Along with the acreage planted, two other factors
affect production of navy beans.

One of these is the

acreage abandoned or not harvested.

Producers might

abandon acreage because of a low price for the product.
This would show up as a negative relationship between
acreage abandoned and price.

The simple correlation of

these two variables, however, was highly positive.

This

suggests that acreage abandoned was the cause rather than
the effect of a low price.

The abandoned acreage is,

therefore, assumed to be determined by weather factors
which can not be predicted.

Consequently, an equation

for acreage abandoned was not estimated.
The final factor affecting navy bean production
is yield.

Attempts were made to construct and estimate

a navy bean yield function using acreage planted, expected
price, a time variable for improved varieties and cultural
practices, and a weather index.

The results of the esti-

mation were totally disappointing.

Weather is undoubtedly

the most important variable influencing yield in any
given year, but no adequate weather index is available.
Therefore, no yield function is estimated.
Demand
Commercial demand for navy beans consists of domes­
tic and export demand.

Since these two demand components

are affected by different factors, they are considered
separately in this section.

The demand for small white

29

beans, a competing bean, and government demand, created
by the price support program, are also discussed.
Domestic Demand
Sales of a commodity are determined by several
factors.

Typically, as price of the commodity goes up,

the quantity sold goes down.

Population and consumer

income also are factors affecting the sales of a commod­
ity.

The quantity sold generally increases as popula-

tion and consumer income increase.

Another determinant

which helps to establish the quantity sold is the tastes
and preferences of the individuals consuming the commod­
ity.

The price of related commodites, both substitutes

and complements, make up still another determinant affect­
ing the level of demand for a commodity.
The factors assumed to influence the domestic de­
mand for navy beans at the local elevator level are:

1)

the price of navy beans paid to producers, 2) population
of the U.S., and 3) price of small white beans.

The

price paid to producers probably affects the price and
profitability at each subsequent level in the marketing
process, and the consequent quantity which would be

3

Increases in quantity sold brought about by a
population increase are generally independent of the
absolute, or starting, population level:
a two percent
rise in population will bring about an approximate two
percent rise in sales.

30
purchased at the elevator level.

4

U.S. population and

per capita income trends have been very highly correlated.
Thus, U.S. population was used in the demand structure to
account for both per capita income and population effects.5
The price of small white beans was included because small
white beans are a competitor of navy beans in certain
areas of the country.6

A lower small white bean price

would lead to a shift to using small white beans by can­
ners, and correspondingly less navy bean consumption at
the current market price.
4

The price used in the demand structure is the same
eight-month unweighted average price used in the acreage
planted equation.
The simple correlation between the 8
and 12-month average prices was .98. However, because of
wide price fluctuations during the last four months of
the crop year, caused by the selling of the remnants of
the current year's production and expectations of the
next year's production, the 8-month price gives a more
accurate representation of price received for the current
year's bean crop.
5Both population and income were not used because
of the high correlation between the two variables which
would result in multicollinearity. Multicollinearity
results in less efficient estimators.
6Great northern beans, a larger type of white beans,
were included in Hathaway's economic model; Dale E. Hatha­
way, The Effects of the Price Support Program on the Dry
Bean Industry in Michigan. However, when included in the
economic model of this study, no significant relationship
between great northern and navy beans was found. A possible
explanation for great northern and navy beans not being
related to any significant extent at this point in time
is that great northern beans compete almost exclusively
in the dry bean retail trade, which now is very small,
while navy beans are used mainly in canned bean products
which are now the primary intermediate uses of dry beans
in the United States.

31

A fourth variable included in the domestic demand
equation was a dummy variable.

This variable is zero for

the years 1951-57 and one for the years 1958-67.

It was

included after observation of the data showed a large
increase in domestic demand after 1957.

While no apparent

reason for this phenomena has been forthcoming, two possible
explanations are offered.

One, there could have been a

change in the way data was reported, or two, there could
have been a change in taste occurring for the commodity,
although this seems highly unlikely in such a short time
period.

The effect of this variable is to allow a shift

in the constant term in the domestic demand equation
after 1957.
The following equation represents the variables
assumed to influence domestic demand for navy b e a n s.
DDNB = f (PNB, PSW, USPOP, DUMMY)
where:

DDNB is domestic demand for navy beans in 1000
cwt., 12-month {September 1 - August 31).
PNB is average price (September to April) for
choice handpicked navy beans in dollars per
hundredweight paid to producers in Michigan.
PSW is 8-month average price for small white
beans received by growers in dollars per cwt.
(September 1 - April 31) .
USPOP is average U.S. population
people) during the year.

(millions of

DUMMY is a variable which is zero for years
1951-57 and one for years 1958-67.

32
Demand for Exports
A large proportion of United States navy bean ex­
ports go to the United Kingdom.

The other regular exporter

of navy beans to the United Kingdom is Canada.
of Canada's exports go to this market.

Almost all

Since the commodity

being exported by these two countries is the same, exports
from both countries are added together in the export
equation.

For this study, United States navy bean exports

are assumed to be determined within the model while Cana­
dian exports are assumed to be primarily determined by
forces other than the variables included in the export
demand equation.
The demand for exports should be influenced by
determinants similar to those influencing domestic demand.
The price of navy beans paid to Michigan farmers is the
price assumed to influence the quantity demanded.

How­

ever, a difference between domestic and export demand
becomes apparent when the determinants influencing the
level of demand in the export equation are considered.
Since the United Kingdom is the primary market for both
United States and Canadian exports, United Kingdom popu­
lation is assumed to influence the level of demand for
navy beans.

Because this variable has been closely

correlated with time, its estimated coefficient will
probably pick up influence of other variables also

33
related to time but not included in the equation.

There­

fore, the clear interpretation of this coefficient may
be difficult.
The equation to be estimated for exports i s :
EXNB - f (PNB, UKPOP)
where:

PNB was previously defined.
EXNB are United States exports (USEX) and Cana­
dian exports (CANEX) for the crop year in 1000
(U.S.) cwt. EXNB equals USEX plus CANEX.
UKPOP is average United Kingdom population
(millions of people) during the year.

Price of Small White Beans
The price of small whites should be influenced
by the quantity of small white beans available.

Other

factors influencing price are demand shifters such as
the price of competitive beans and population or income.7
The factors that are assumed to influence the
price of small white beans are:
small white beans,

(1) the production of

(2) the price of navy beans, and (3)

United States population.
The equation to be estimated i s :
PSW = f (PRSW, PNB, USPOP)
where:

PSW, PNB and USPOP were previously defined.

7Small white beans compete with navy beans for use
in canned products like pork and beans.

34
PRSW is production of small white beans in 1000
cwt.
Government Role in Navy Beans
The demand structure for navy beans is not complete
without considering the role of the government price sup­
port program.

The government operates a price support

program under which it sets a minimum price and theoreti­
cally takes sufficient quantity off the market to insure
that the market price does not fall below the support
p
price.
Whether a price support operation will be effec­
tive in any given year depends on the total supply of the
commodity available in that year and the willingness of
enough producers to allow the government to take owner­
ship of the beans if the market price does not exceed
or is not expected to exceed the support price.
A simplified example may make the operation of
this program clearer.
curve is assumed

In this example, a linear demand

(ABD in Figure 1).

Supply is also

fixed after the crop is harvested for a particular crop
year.
If the net government support price is P^:

as

shown in the diagram, the demand curve facing the
g

The quoted support price is a gross price.
To
obtain a price equivalent to the market price or net
price received by growers, a handling charge of approxi­
mately $.95 must be subtracted from the given support
price.

35
producer is ABC as compared to ABD without the price
support.

The demand curve is completely elastic for

quantities greater than Q 1 -

If supply for a given year

is greater than Q^, the price does not fall below P^,
the price support level, because the government pur­
chases the quantity needed to maintain the market price
at P^.

For example, if supply for a given year is Q2 ,

without a price support program, the market price would
be

and the entire quantity Q2 would go into the

commercial market.

However, with the price support

program, the government purchases a quantity necessary
to maintain price at the support level.
of supply Q2 , quantity Q 2 ment, while quantity

In the case

is bought by the govern­

is sold in the commercial
FIGURE III-l

Operation of Government Price Support Program
Example
P

P0
P1
P2

0
Q

36
market.

If supply is less than Q^, the price support

is not effective for that time period and no government
takeover occurs.

For example, if supply after harvest

is Qq , market price is P q , which is above the support
level of P^.

Therefore, the entire quantity goes into

the commercial market.
In actual practice, the market price sometimes
falls a small amount below the support price because of
the way the government mechanism works.

In large crop

years, if market price approaches the net support price,
producers can put all or part of their production tinder
loan to the government.
price.

They receive the net support

If the market price goes above the support price,

they can sell their beans in the commercial market and
repay the loan to the government.

If the market price

does not rise above the support level by a certain date
(set by the CCC each year), the producer can let the
government take over the quantity of beans under laon,
as payment for the loan.

Uncertainty as to what market

price will do may cause enough producers to retain owner­
ship so that the market price may fall below the support
price due to insufficient takeover in some years.
Given supply, commercial demand, and the price
support level for navy beans, the amount of government
takeover in any given year can be determined.

The price

support program, by setting a floor or market price,

37
creates an elastic demand for the product at this price
and government takeover is a residual amount.

Therefore,

an equation for government takeover need not be estimated.

Two identities complete the demand structure.

The first identity fixes the supply at the beginning of
a crop year; the second identity in effect subtracts
the government price support pruchases from total supply
to give total commercial consumption, thus closing the
demand s truc ture.
The identities are:
SUPNB - PRNB + BINV
SUPNB = DDNB + USEX + GDNB
where:

DDNB and USEX were previously defined.
PBNB = production of navy beans in U.S. in
1000 cwt.
BINV — beginning inventory
navy beans in 1000 cwt.

(September 1) for

SUPNB = U.S. supply of navy beans in 1000 cwt.
GDNB — government navy bean takeover in 1000
cwt.
Model Estimation Procedure
Any model development involves certain arbitrary
judgments as to variables chosen and estimation procedures
9
The estimation of the demand structure is done
under the assumption that government disposal of navy
beans has not entered back into the commercial market.
This probably leads to an underestimation of domestic
commercial disappearance since some of the government
takeover has gotten back into the commercial market in
some short crop years.

used.

Data limitations and cost of estimation are prime

considerations in these judgments.

Only recently have

navy bean data been separated from all dry bean data.
Because of this, only the years 1951-67 are used in the
estimation procedure.

The use of two different estimation

procedures were used due to the differing structures of
the supply and demand relationships within the industry.
The behavioral model can be summarized into the following
stochastic model.
Supply Structure:
APNB* - A 0 + A 1PNBt_1 + A 2 PCt_1 + A 3APNBfc_1 + V**
Demand Structure:
DDNB* = B1 q -f B 11 PNB* + B 12 PSW* B 13 USPOP + B 14 Dummy + U**
EXNB* « B20 + B 21 PNB* + B22 UKPOP + U2 **
PSW* » B3(> + B 31 PNB* + B 32 USPOP + B 33 PRSW + U 3**
SUPNB - GDNB « DDNB* + EXNB* - CANEX
*Endogenous variables
**V and U are the disturbance terms for the equations.
The acreage planted equation has only one endogenous
variable.

Ordinary least squares was used to estimate it.

A linear functional relationship between variables was
also used.

Ordinary least squares gives the best, linear,

unbiased, consistent estimated coefficients of the equa­
tion under the following assumptions:

1) disturbance

terms have a zero mean, 2) variance is equal over all

39
observations

(homoscedasticity), 3) there is zero cor­

relation between disturbance terms for any length of
time between them, and 4) all variables other them the
dependent variable are p r e d e t e r m i n e d . ^
As shown in the stochastic presentation, the de­
mand structure consists of three equations and an identity.
There are four endogenous variables in the structure.

In

reality, the values of these endogenous variables are
determined simultaneously

(within the same period of time).

It has been shown that ordinary least squares estimation
of the demand structure would result in biased estimates
of the parameters.'*'1

If the equations in the structure

are just-identified or over-identified, the equations
can be estimated using simultaneous estimation proce­
dures.

Since the equations in the demand structure are

identified or over-identified according to the rank and
order conditions, they were estimated using three stage
least squares

(3SLS).

Three stage least squares gives

consistent and asymptotically efficient estimates when
it is assumed that the demand structure exhibits the
following properties:

1) the disturbance terms have a

^■°A simultaneous estimation procedure could also
have been used with the results being the same as the
ordinary least squares results.
Since ordinary least
squares estimation procedure is the least expensive, it
was used.
^ J . Johnston, Econometric Methods
McGraw-Hill Book Company, l9(>3) , pi 233.

(New York:

zero mean, 2) the variance of disturbance terms is con­
stant over all time periods# 3) the disturbance terms are
independent over all time periods, and 4) the exogenous
variables are nonstochastic.

CHAPTER IV
ESTIMATED SUPPLY AND DEMAND RELATIONSHIPS IN THE
MICHIGAN NAVY BEAN INDUSTRY, 1951-67
The economic forces at work in the Michigan navy
bean market during 1951-67 can provide insight into why
the market has behaved as it has, and be useful in in­
terpreting future market behavior.

The statistical

estimates of the navy bean industry supply and demand
model provide a picture of the major influences on market
behavior.

In analyzing individual equations, an evalua­

tion of important results obtained and their implications
are given.

Some shortcomings and weaknesses of the re­

sults are also considered and some implications for
specific industry programs are suggested.
Estimated ResultB
The results of the ordinary least squares estima­
tion of acreage planted and the 3SLS estimation of the
demand structure are given in Table IV-1.

The variables

shown in this table were defined in Chapter III.
procedure used to calculate the R

2

The

(the percentage varia­

tion in the normalized endogenous variable explained by

41

TABLE IV-1
ESTIMATES OF STRUCTURAL EQUATIONS IN THE MODEL

Equation

Normalized
Endogenous
Variable

1

APNB

Coefficients a (standard errors of
coefficients) and explanatory variablesb
317.5776
(114.8333)

0.5112APNB . +
(0.1307) t_A

+

R2
(Durb.-Wat. Stat.)

20.4132PNB .
(7.1419)
.89
(1.71)

-181.0371PC ,
(51.3822)
2

DDNB

80.1858NB
(822.9839)

-548.0385
(4098.8389)

+

13.6156PSW
(468.6216)

+24.5519USPOP + 696.2777DUMMY
(28.0938)
(1358.3249)
3

EXNB

-15332.2373
(4268.6814)

4

PSW

-2.3352
(3.5183)

63.7641PNB
(159.1106)
+

0.7357PNB
(0.2190)

.75
(2.48)
.55
+ 323.5717UKPOP
(2.41)
(77.3603)
0.0034PRSW
(0.0011)

+0.0514USPOP
(0.0199)

^ o t all the coefficients are significant to four decimal places.
^Definitions of variables are given in Chapter III.

.88
(2.58)

43
the other variables in the equation)

for each equation

in the demand structure is the same procedure that is
used in the ordinary least squares estimation method,
which was used for the acreage equation.

Consequently

these values are not strictly valid when 3SLS is the
estimation procedure.

However, they do give an approximate

measure of the fit of the arbitrarily designated dependent
variable to the other variables in the equation.
Durbin-Watson statistics, used to test for serial
correlation of residuals, are also presented for each
equation, but again these in the demand structure are
not strictly valid for a simultaneous system of equations.
They can, however, provide a quasi-index of serial cor­
relation.

If the demand structure was a system of inde­

pendent equations, all the Durbin-Watson statistics are
within the acceptable range.
While it is difficult to evaluate the performance
of a model with any single criterion, there are several
criteria that can give a fairly reliable indication of
the model's performance.

These criteria include:

(1)

the variation in the normalized endogenous variable ex2

plained by the other variables in the equation— R , (2 )
the statistical significance of the model's coefficients,
and (3) the sign and magnitude of the model coefficients
when compared to the economist's expectations
real world and theoretical considerations).

(based on

44
Using the first criterion, the model has fairly
good results.

The R

equation was .89.

2

value for the acreage planted

The R

2

values in the simultaneously

estimated demand structure ranged from a low of .55 to
a high of .8 8 .

Export demand was the endogenous variable

least well explained by its equation.

This result could

have been anticipated from the formulation of the equa­
tions.

When an equation estimates the exports of two

countries, the likelihood of omitting important “local"
variables affecting exports is greater.
The second criterion, the significance of the co­
efficients, can be evaluated by

calculating the ratio

the coefficient to its standard

error {both shown in the

table).

of

While some variables may not be statistically

significant when compared to an

arbitrarily chosen

significance level, they may be

included to make the

model compatible with its further u ses.

This criterion

does, however, provide a guide to the retention or re­
jection of variables not otherwise crucial in the remainder
of this study.

The above criteria and the conformity of

the statistical results with those expected on the basis
of our theoretical understanding of market behavior are
discussed in detail in what follows.*
*The analysis of estimated results of the demand
structure with more than one jointly dependent variable
in each equation needs clarification.
This analysis can
be done on the equations as they were presented in Chapter
III.
It can also be done by making each jointly dependent
variable a function of only exogenous variables (reduced

45
Navy Bean Acreage Planted

2

The explanatory value of the variables expected
to be related to acreage planted decisions of farmers
was quite high.

The calculated R

2

for equation one was

estimated at .89.
The signs of the coefficients are reasonable, al­
though the magnitudes may be less than some economists
might expect.

An increase in navy bean price in the pre­

ceding year has been associated with an increase in
planted navy bean acreage.

An increase in last year's

navy bean price by.one dollar has typically resulted in
a 20,413 acre increase in the current year's acreage
planted; that is, a one percent change in the previous
year's navy bean price has been associated with a .28
percent change in acreage planted in the current year in
3
the same direction.
These results indicate the acreage
planted is relatively unresponsive to different price
levels.

In a similar earlier study, Vandenborre found

an elasticity of dry bean acreage planted with respect to
4
last year's dry bean price of .39.
form equations). The analysis will be done on the equa­
tions as they were presented in this Chapter III; that is,
only direct effects will be considered.
2

In the calculations in the descriptions all other
explanatory variables are held constant when the first
derivative is taken.
3
The calculation of elasticities in this study are
done at the mean for the variables used.
4
Roger J. Vandenborre, An Econometric Investigation

46
Corn, as a competitive cash crop, would be expec­
ted to infringe on navy bean acreage if corn became
relatively more profitable, or vice versa.

A ten cent

drop in last year's corn price increased the current
year's planted acreage of navy beans by 18,103 acres.
The elasticity of acreage planted with respect to corn
price the previous year was estimated at -.44.
borre estimated this elasticity at -.34.

5

Vanden-

This result

indicates that growers have substituted the planting of
corn for navy beans, if the relative price position the
previous year warrants such a move.
The acreage planted the previous year has been
positively related to acreage planted in the current
year.

This would seem to indicate that growers' de­

cisions are influenced by their previous decisions or
by the continuing influence of the same factors which
prompted their decision last year.

If a grower planted

a large acreage the previous year , he will plant a large
acreage in the current year.
The implication of the estimated parameters is
that wide price fluctuations in the industry will only
lead to small percentage responses in acreage planted.
of the Impact of Governmental Support Programs on the
Production and Disappearance of Important Varieties of
Dry Edible B e a n s , p . 33.
5 Ibid.

47
The size of -the constant term is quite large, represent­
ing over half the acreage planted in any of the years.
This could be a result of the government price support
program causing a stable expected price, which could
have led to a fairly large "base acreage" in the time
period under study.
Navy Bean Domestic Demand
The percentage of variation in domestic demand
2

(R ) explained by the equation was 75 percent in equation
two.

Signs of the coefficients were consistent with ex­

pectations.

Navy bean price has been inversely related

to domestic sales of navy beans; that is, as price of
navy beans moved in one direction, domestic sales moved
in the opposite direction.

The price elasticity of

demand with respect to navy bean price has been -.14.
Retail price generally is quite stable, perhaps because
beans only contribute a small proportion of the total
cost of the retail canned beans.

This may be part of the

cause of the inelastic demand situation at the elevator
level.
Domestic demand and the price of small white beans
move in the same direction.'

A one dollar increase in the

small white bean price has been associated with a 13,615
hundredweight increase in domestic consumption of navy
beans.

The cross-elasticity of demand has been .03.

Thus, the small white bean price has very little influence

48
on the domestic demand for navy b e ans.

Since small white

bean production is so small in proportion to total navy
bean production, this result was to be expected.
United States population changes have had a
positive effect on domestic demand.

A one million in­

crease in population has been associated with a 24,551
hundredweight increase in domestic demand.

In percentage

terms, a one percent change in U.S. population has been
associated with a 1.14 percent change in domestic demand.
Over the estimated period, domestic demand was growing
at a rate slightly larger than population was growing.
Since there is a high correlation between population,
income and time, the population variable could have
picked up some of the effects of income increases or
other variables related to time.
The dummy variable has a coefficient of 696,
which means the constant term is increased by that many
1000 hundredweight after 1957.

This was the result ex­

pected of this variable after observation of the raw
data.

However, no clear-cut explanation of this ob­

served phenomena has been found.
In summary, the domestic demand for navy beans
is quite inelastic with demand growing only slightly
faster than population.

This, coupled with the low

influence of small white bean price on navy bean domestic
demand, suggest that substantial increases in price would
result from small decreases in supply.

49
Navy Bean Export Demand
Approximately 55 percent of the variation in U.S.
and Canadian exports was explained by the other variables
in the equations.

The signs of the coefficients are

consistent with expectations.
expanded when price decreased.

Exports have typically
The price elasticity of

export demand has been -.36, which makes it approximately
two and one-half times as elastic as domestic demand.

It

was hypothesized that exports are more price elastic
than domestic demand.

However/ even though this has

been observed/ the fact that export demand was inelastic
(price elasticity less than an absolute value of - 1 ) is
somewhat surprising, although the elasticity for two
countries

(as in this equation) would be less elastic

than one country.

The competitiveness of international

markets and the general availability of substitute pro­
ducts usually make export demand price elastic (price
elasticity greater than an absolute value of -1).

How­

ever, since this particular export market involves only
a few countries (United Kingdom and a few other western
European countries), an inelastic export demand might be
expected.

The export market stimulates more countries

into the market at low prices.

If the price support

level had not been as high in the past, other importing
countries might have been prompted to enter the market at
lower prices, creating a more elastic export demand curve.

50
There has been a positive relationship between
navy bean exports and the United Kingdom's population,
which is consistent with expectations.

However, the

magnitude of the coefficient is extremely large, which
suggests that this variable has picked up the influence
of other variables omitted from the equation, but associ­
ated with time (U.K. population size also has been highly
correlated with time).

Therefore, by itself, the U.K.

population coefficient does not fully explain the growth
in navy bean exports over the estimated time period.
Part of the magnitude of this variable might be the in­
fluence of export market development activities which
have increased in recent years but which cannot be
adequately quantified.
Small White Bean Price
The percentage of variation in small white bean
2
price (R ) explained by the equation was 8 8 percent.
The signs of the coefficients are consistent with ex­
pectations.

There has been a negative relationship be­

tween small white bean production and price.
in small white bean production of

100,000

An increase

hundredweight

has been associated with a decrease of $.34 in small
white bean price.

The price flexibility of demand was

-.23; that is, a one percent change in production was
associated with a .23 percent change in price.

51
Small white bean price and navy bean price were
positively related in this equation.

A dollar increase

in navy bean price has been associated with a $.73 in­
crease in small white bean price.

Since navy beans are

usually shipped to small white bean production areas,
one would expect that small white bean price would be
strongly related to navy bean price.

In percentage

terms, a one percent change in navy bean price has been
associated with a .52 percent increase in small white
bean price.

A large percentage of the variation in

small white bean price is explained by variation in
navy bean price.
U.S. population has also been positively related
to the price of small white beans.

A million person .

increase in population has been associated with an in­
crease of $.05 in small white bean price.
Data Required for More Detailed Analysis
The price used in the model was an unweighted
8 -month

average.

If sales volume at each price could

be obtained, an improved weighted average price could
be used in the demand structure, probably resulting in
improved R

2

and significance of the variables.

The magnitude of the coefficient for U.K popu­
lation probably reflected the effect of some ommitted
variables in the export equation.

A closer study of the

export market, if and when data become available, might

52
show the need for more than one export equation and the
addition of a market development activity variable to
the existing equation.

Improved data and additional

equations would help reduce part of the present unex­
plained variation in exports.
A more accurate breakdown of where production
goes would also help improve the estimated results.
Some of the unexplained fluctuations in domestic and
export sales could then be explained, thus, improving
2
the R and significance of the variables.
Summary and Conelusions
The model and the estimated results seem reason­
able when evaluated in the context of signs of coeffi­
cients and the correlation coefficients.

However, the

significance and magnitudes of some variables suggest
that further improvements in the model could be made
when data becomes available.
In terms of usefulness, the estimated model pro­
vides a quantitative picture of recent demand and supply
relationships in the navy bean industry.

This informa­

tion can be useful to persons contemplating public policy
changes in the industry.

Further, these equations can

be used for prediction purposes.

Given the acreage

planted the previous year, the price of navy beans and
of corn for the previous year, an estimated acreage

53
planted for the current year can be obtained.

Using

estimates of production, an estimated price, domestic
demand, and export demand can be obtained.

These es­

timates can help different industry participants plan
their market activity for the current year.

Finally,

further research needed in the industry and the data
needed to carry out the research are pointed out.

CHAPTER V
EVALUATION OF PRICE AND SUPPLY CONTROL PROGRAMS
The previously estimated equations were used in
developing a simulation model which simulates the supplydemand behavior in the navy bean industry.

This dynamic

computer model provides a means of experimentally evalu­
ating .alternative price and supply control programs which
might be instituted in the industry without exposing
the industry to the actual consequences of each program.
Basically, the simulation model can be used to answer
the question:

What would have resulted in the navy bean

industry if the government or private control programs
under study had been in operation during the period
1953-67?

This answer should also provide some insight

into the likely relative future benefits of these pro­
grams if they would be implemented in the industry.
Basic Dynamic Framework
The design of the simulation model was based on
an information-feedback loop involving the price and
production of navy and small white beans

(See Figure V - l ) .

The previous year's navy bean price was fed into the navy
54

FIGURE V-l

INFORMATION-FEEDBACK DESIGN USED IN THE SIMULATION MODEL
STEFS:

Variables determined
outside of simulation
model but entering
into its process:

Facts fed to computer:
Initial* prices of navy
and small white beans
and previous year's
acreage planted for
navy beans

Variables deter­
mined outside
of simulation
model but enter­
ing into its
process:

SUPPLY
STRUCTURE

Corn price

Barley and sorghum
gross revenue

7

Competing bean ___
price index

Current year
navy bean
acreage____
f.

Calculate current
year small white
! production______

Calculate current year navy
bean production

AM

Navy bean yield
% of acreage planted
harvested in current
year
in
in

Calculate simultaneously for current year:
1Domestic
consumption
of navy
beans
Navy bean
price

Exported
quantity
of navy
beans

Government
expendi­
tures

Small white
price

to Supply Structure to determine the next
year's production levels for navy and small
white beans.
*1952 for navy beans, 1951-52 for small white beans

(U.S. population,
_ U.K. population,
~Canadian exports,
DUMMY variable)

DEMAND
STRUCTURE

56
bean acreage planted equation.

Using the actual pre­

vious year's corn price, the planted acreage was pre­
dicted.

This estimate was multiplied times the actual

proportion of dry bean harvested in Michigan and actual
navy bean yield for that year to obtain navy bean pro­
duction.
For small white beans, the small white price of
the previous two years was used, along with the previous
year's barley and sorghum gross revenue and competing
bean prices, to determine the production of small white
beans.

These two production figures were then put into

the simultaneous demand structure.

Along with the actual

values for the variables determined outside the model
(U.S. and U.K. population and Canadian exports), these
variables were used to determine price of navy b e a n s ,
price of small white beans, and domestic consumption,
exports, and government purchases of navy b e a n s .

The

generated prices were then fed back into the production
equations to start the process again for the next year.
Each year all the variables were stored, printed
out, and used in subsequent analysis and evaluation.
For example, these variables were used to calculate pro­
ducer gross income from navy beans and domestic per capita
consumption of navy beans, two measures of program per­
formance which might be considered important by some
industry members.

The value of each of the generated

57
variables for evaluation purposes varies for different
industry groups.

All industry groups would be inter­

ested in the price of the product since it affects
everyone's welfare.

Farm input suppliers would be

interested in acreage planted because this would probably
be closely related to the volume of purchased inputs.
Canners would be interested in domestic consumption since
this is a measure of their aggregate volume.

Shippers

would be interested in the production level, which deter­
mines their volume of business.

Consumers would be

interested in per capita consumption and the price level
at the elevator level, which might affect their retail
prices somewhat.

The g^joss income from navy beams rela­

tive to the number of acres planted would be of most
interest to producers.
However, the reader must be cautioned in examin­
ing this income figure.

It is only for gross income

from navy beans and not the navy bean producers' total
gross income.

The study is too limited to exaunine the

implications of all possible switches in cropping
mix.

What can be said is that if producers' gross in­

come from navy beans rises on fewer acres, one would
expect that the navy bean producers' net income from
beans and total gross income would also rise.

So, navy

bean producers' total gross income for different

58
industry programs should not be compared without con­
sidering the associated navy bean acreage under each
program.
The equation used for estimating

small white

bean production was taken from a previous study.^

The

influence of the price of small white beans the pre­
vious two y e a r s , an index of competing bean pri c e s,
and average gross income of barley and sorghum per acre
to California producers were consistent with expectations and significantly different from zero.

2

The estimated results for this equation along
with its correlation coefficient, standard errors, and
Durbin Watson statistic are:
PRSW - 2,739.6010 - 729.3139 log GIBS.
(157.0015)

,

+ 647.6832 log PSW. .
(106.1815)
t_J+ 119.0698 log PSW fc_ 2 - 406.6399 log CBPI t _ 1
R2 =

.86

D-W statistic = 2.31

where PRSW and PSW were previously defined. GIBS
is average gross income of barley and sorghum to
California producers. CBPI is average price of
competing beans (to small white beans) per cwt.
paid to producers.

^■Roger J. Vandenborre, An Econometric Investigation
of the Impact of Governmental Support Programs on the
Production and Disappearance of Important Varieties of
Dry Edible Bea n s , p. 30.
2

fornia.

Small white beans are grown exclusively in Cali­

59
The five behavioral equations used in the simu­
lation model are shown below:
APNB* - 317.5776 + 0.5112APNB<t_ 1 + 20.4132 PNBt _x
-181.0371 PC t _ 1
PRSW*

= 2,739.6010 - 729.3139 log GIBS^^^
+ 647.6832 log PSW t _ 1 + 119.0698

log PSW fc_ 2

- 406.6399 log CBPI t - 1
DDNB* =-548.0395 - 80.1858PNB* + 13.6156 PSW*
+ 24.5519 USPOP + 696.2777DUMMY
USEX*

= - 15,332.2373 - 63.7641 PNB*
+ 323.5717 UKPOP - 1.0 CANEX

PSW* =

-2.3352 + 0.7357PNB* - 0.0034 PRSW
+ 0.0514 USPOP

*
Endogenous variables
The behavioral structure of the simulation model was
these equations except where specific modifications
seemed desirable to improve the realism of the simulated
results.
By simulating industry behavior during 1953-1967,
actual data could be used for the variables assumed to
be determined outside the economic model.

To start

each simulation run, the actual 1952 navy bean price
and planted acreage were used.

The actual 1951 and

1952 small white bean prices were also used.

In succeeding

60
years of each simulation run, the simulated navy bean
price, planted acreage, and small white bean prices
from the previous year were used.
Validation of the Model
After the simulation model was built, it was
tested to see if it functioned realistically; that is,
did it duplicate reality with any degree of accuracy?
To test

this, the actual government price support pro­

gram that operated in the navy bean industry and the
small white bean industry from 1953-1967 was inserted
into the basic simulation model.

In this simulation,

it was assumed that government takeover was not placed
back into the commercial market.

This assumption was

made because a breakdown of government commercial sales
and donations of navy bean takeover was not available.
Since it appeared that most of the government disposal
was via donation, this assumption appeared fairly real­
istic.
In the first runs of the simulation model, the
estimated equations in the demand structure were used
with no restrictions on the top portion of the demand
curve.

The results of these early runs showed that in

the years when there was no government takeover of navy
beans, the simulated navy bean price was substantially
above the actual price observed in those years.

From

studying the actual price and supply values, it was

61
observed -that the average navy bean price for a year
went above $8.50 in only one year from 1953 to 1967,
even though supply was small in several years.

This

suggested that demand was less inelastic at higher
price levels.

At higher price levels, there

are

prob­

ably other goods that consumers substitute for navy
beans•
To make the top portion of the navy bean demand
structure conform more closely to past behavior, the
estimated linear demand structure was modified for navy
bean prices greater than $8.50.

For supply less than

800,000 hundredweight, a maximum price of $10.50 was
set.

Between $8.50 and $10.50, the following equation

was use d :
PNB = 10.9 - 0.0005 SUPNB.
This flatter demand structure better fit the few obser­
vations available in the upper portion of the demand
curve.
With the addition of the government price support
program and the above modification in the demand struc­
ture, the model was then run to simulate what actually
occurred from 1953 to 196 7 in the navy bean industry.
The annual values for some selected performance vari­
ables from this run are presented in Table V - l .

They

include navy bean acreage planted, producer gross income

TABLE V-l
ACTUAL AMD SIMULATED RESULTS OF THE NAVY BEAN GOVERNMENT
PRICE SUPPORT PROGRAM, 1953-1967a

Navy Bean
Domestic
Consumption
1000 cwt.

YEAR

Actual

2508
3106
3131
2064
3319
4004
4459
3367
4242
4054
4649
4640
4405
3567
4025

1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
Average
Difference

Simu­
lated
2884
2836
3120
3191
3055
4072
4181
4265
4250
4325
4375
4455
4364
4590
4466

303

Navy Bean
Exports
1000 cwt.

Actual Simu­
lated
716
52
674
1209
8
1029
1433
640
902
1704
1938
1550
1075
2044
762

153
0
702
711
220
1063
1086
1109
1295
1222
1527
1348
332
1416
388

362

Navy Bean
Government Takeover
1000 cwt.

Government Expendi­
tures for Navy Bean
Takeover (Million $)

Actual

Simulated

Actual

Simulated

377
0
623
1747
31
9
177
1838
1611
967
1012
601
0
1676
0

592
0
403
1132
0
254
1047
712
1322
1515
2000
637
0
1070
0

3.3
0.0
4.6
12.9
.2
.1
1.1
11.9
11.5
6.9
7.2
4.2
0.0
11.6
0.0

5.2
0.0
2.9
8.4
0.0
1.8
6.7
4.6
9.4
10.8
14.3
4.6
0.0
7.4
0.0

414

aThe sources for the actual data are given in Appendix B.

2. 7

TABLE V-l (continued)

YEAR

Navy Bean
Planted
Acreage
aooo A)
Actual Simu­
lated

1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967

350
441
489
477
482
514
511
497
508
516
519
556
600
577
493

Average
Difference

371
417
476
490
483
559
538
521
520
545
543
529
517
560
500

24

Navy Bean
Production
1000 cwt.
Actual Simulated

3601
3158
4428
6020
3358
5042
6069
5845
6755
6725
7599
6801
5480
7289
4787

3629
2336
4225
5034
3275
5388
6314
6086
6367
7062
7903
6440
4696
7075
4354

244

Small White
Production
1000 cwt.
Actual Simulated

560
731
884
771
759
800
943
618
438
542
607
514
578
670
470

576
678
752
717
657
846
669
618
430
506
575
576
435
576
458

58

Navy Bean
Price
$/cwt.
Actual Simu­
lated

7.83
9.37
6.48
6.17
7.62
6.44
5.46
5.46
6.00
6.07
6.15
6.61
8.35
6.01
8.35

7.85
9.48
6.48
6.43
9.26
6.22
5.48
5.51
6.20
6.20
6.20
6.20
8.55
5.95
8.50

.23

Small white
Price
$/cwt•
Actual

9.81
10.97
8.26
7.71
9.05
9.23
7.48
8.46
10.78
9.02
9.36
10.95
12.35
9.12
12.95

Simu­
lated

9.73
10.72
8.40
8.64
11.09
8.36
8.57
8.92
10.33
10.12
9.99
10.14
12.50
10.21
13.27

.70

Producer gross
income from
navy beans
(million $)
Actual Simulated

28.2
29.7
28.7
31.0
25.6
32.5
33.1
31.9
40.5
40.8
46.7
45.0
45.8
43.8
40.0

28.5
26.9
27,4
32.4
30.3
33.5
34.6
33.5
42.6
43.8
49.0
39.9
40.2
42.1
41.2

2.4

64
from navy beans, government expenditures on navy bean
takeover, government takeover of navy beans, navy bean
domestic consumption, navy bean exports, and government
takeover of small white beans.

3

The comparable actual

results are also given in the same table.
Comparison of the actual results and simulated
results indicated that the simulation model performed
fairly well in simulating values for the selected vari­
ables .

The average values for these comparisons per­

formed better than the year to year results
values are shown in Table V-2).

(the average

This was to be expected

because the few variables in each estimated equation did
not completely explain the variation of the dependent
variables.

The largest difference arose in the alloca­

tion of supply between domestic and export markets with
domestic consumption higher and exports lower than what
actually occurred.
After duplicating the actual situation, the next
step in validating the model was extending the estimated
relationships beyond observed price levels so other pro­
grams potentially involving these price levels could be
evaluated.

The extension of the model in the unobserved

price range was accomplished by simulating an unsup­
ported market.
3

The first simulations of this situation

While the government price support program was in
effect for small white beans from 1953-1967, almost no
government takeover occurred. The simulation had no
small white bean government takeover.

65
were based on the assumption that the estimated relation­
ships between variables were valid over the entire navy
bean price range.

However, this assumption may be ques­

tioned because there were no price observations below
the government price support level.

But, the unsup­

ported market and some of the other simulated programs
might touch these lower price levels.

Consequently, the

validity of this assumption was carefully examined for
both the demand and acreage response equations during
the unsupported market simulation.
To improve the realism of the unsupported market
simulation, it was assumed that navy bean price would not
fall below $2.00 per hundredweight.

As price falls,

new demands for navy beans would probably show up.

For

example, there is a price where navy beans could be
used as livestock feed.

There also can be a speculative

storage demand where beans would be bought and stored
at a low price in the hope that price would go up in the
future and they could then sell at a profit.

The $2.00

floor was chosen because it was consistent with obser­
vations of the lower limits of average navy bean prices
prior to introduction of the government price support
program.
The acreage planted equation j was assumed to have
the same slope below the price support level that it
had above that price level for all the simulations.

An

66
analysis of costs and returns for navy beans showed
that navy bean price would have had to average approxi­
mately $4.57 per hundredweight to be at a breakeven
profit level with corn for a large number of producers.

4

Producers expect navy bean price fluctuations and fur­
ther , have specialized technical knowledge and equip­
ment.

They also are influenced by fairly fixed crop

rotations and habit.

Thus, one might reasonably expect

that producers would continue to grow some beans even
though price might fall to very low levels in one year
(so long as the average price received over a longer
period of time still made them competitive with
4

In a study of cost and returns to cash crops
(Larry J. Connor, Costs and Returns for Major Cash Crops
in Southern Michigan: Michigan State University Agricultural Economics Report No. 87, East Lansing, Nov •
1967; pp. 8 & 16), it was estimated the cash cost for
growing field beans was approximately $14 less than for
growing corn on the loam-clay soil found in the "Thumb."
From 1953-1967, average price of corn in Michigan was
$1.14 per bushel and the yield was 56.8 bushels per acre.
This gives an average gross income of $64.75 per acre.
To obtain an income figure comparable with navy bean
gross income, the $14 additional cash cost per acre for
corn was subtracted from this gross income figure.
The
value obtained was $50.75. Given that the average navy
bean yield was 11.1 hundredweight per acre for 1953-196 7,
an average navy bean price of $4.57 would have been the
breakeven point between corn and navy beans.
The data to
compute the average values was taken from: U.S. Depart­
ment of Agriculture, Agricultural Prices (Washington, D .C.:
Government Printing Office, various issues) for corn
prices; Michigan Department of Agriculture, Michigan
Agricultural Statistics (various issues) for corn yield;
and personal communication with U.S. Department of Agri­
culture, Agricultural Stabilization and Conservation
Service, East Lansing, Michigan for navy bean yield.

67
alternative crops).

If the average navy bean price for

the 15 year period were to fall to $4.57 for any of the
simulations, a re-examination of the acreage equation
should be made.

However, it was felt that a slope

similar to that obtained in recent years was a reason­
able approximation of the actual average response which
producers would have had to lower price level in this
study.
Based on the comparisons of the actual and simu­
lated results of the actual situation, it appeared that
the simulation model came reasonably close to duplicat­
ing likely industry behavior.

The modifications made

to the estimated linear equations appeared to be reason­
able approximations of industry behavior under low price
situations which have not been observed recently.

Con­

sequently, it was decided that the simulation model
could be used to realistically test price and supply
control programs which might be considered for the industry.

5

Program Testing and Evaluation
The different programs to be examined included:
(1) variations in the government price support levels,
5The simulation program used in the study is
stored in the Department of Agricultural Economics,
Michigan State University, East Lansing.

68
including an unsupported market situation,

(2) private

(non-governmental) price support-storage program,
marketing quota or control,
(5) a two price system.6

(3)

(4) acreage control, and

These programs are not the

only ones that might have relevance for the navy bean
industry.

However, they are programs which have been

considered or utilized in other commodity situations
plagued by highly variable production and price levels.
Variation of the Government Price Support Level
One factor normally considered by policy-makers
when they are discussing a price support program is
government expenditures.

With the decreasing relative

importance of agriculture in the total economy, it has
become harder to get large amounts of government expendi*>
tures for a price support program and higher price
support levels.

At the same time other groups in the

country are placing demands for increased government
expenditures in the urban areas of the country.

Demands

are also being placed for government expenditures to
fight pollution wherever it is found.

Part of the funds

now going to agriculture may be shifted into other
6With the exception of the variation in the govern­
ment price support levels, the programs were not used
for small white beans in the simulations. Small white
bean price and supply were not controlled or supported.
In the variation of the government price support levels,
the small white support price was reduced the same per­
centage as the navy bean support price.

69
programs.

This could mean a reduction in price support

levels or elimination of some price support programs.
If this were to occur, what would be the likely conse­
quences for the navy bean industry?

These simulations

were designed to give some insight into such a situation
by answering the question:

What would have happened

in the navy bean industry from 1953 to 1967 if the yearly
price support level had been reduced for navy beans?
By showing what impact the program would have had in
the past, consequences in the future can also be in­
ferred if the general pattern of industry behavior and
exogenous variable variation continues in the future.
Further, the basic model structure could subsequently
be used to test new or different programs under expected
future conditions.
Six different 15 year simulation runs were made
(Table V - 2 ) .

In each run the actual price support levels

for navy and small white beans for each of the 15 years
were adjusted downward by a predetermined percentage.
For example, in the first simulation, the actual annual
price support level in each of the 15 years was reduced
by 10 percent.
ported situation

The last simulation shown is the unsup­
(no government price support program).

With a 10 percent reduction in the yearly price
support level, average navy bean planted acreage and
corresponding production levels would have decreased

TABLE V -2
AVERAGE ANNUAL RESULTS FOR ACTUAL, SIMULATED ACTUAL, AND
SIMULATED REDUCED NAVY BEAN PRICE SUPPORT LEVELS, 1953-67

Government Price
Support Level in
Effect for the
15 Year Period

AVERAGE YEARLY
Navy Bean
Planted
Acreage

Production
of
Navy Beans

Production
of Small
White Beans

Domestic
Consumption
of Navy Beans

Navy Bean
Exports

1000 A.

1000 cwt.

1000 cwt.

1000 cwt.

1000 cwt.

497.0

5409.0

652.6

3666.3

1038.6

Actual Support Level
(ASL)
($7.24)

499.5

5391.1

584.8

3856.4

829.8

90% of ASL

($6.52)

480.0

5176.4

559.7

3888.9

822.7

80% of ASL

($5.79)

468.1

5046.2

543.6

3899.1

842.1

70% of ASL

($5,07)

460.7

4965.4

531.6

3894.1

853.7

60% of ASL

($4.34)

463.4

4890.7

518.6

3890.2

846.1

50% of ASL

($3.62)

448.1

4844.0

506.2

3882.4

832.6

Unsupported Market
($0.00)

446.2

4819.7

496.8

3959.7

860.0

(Average Support.
Price in $/cwt.)
Actual®

($7.24)

Simulated Results:

aThe sources of the actual data are given in Appendix B.
bThis is the quoted support price. To get a price equivalent to the market
price, handling charges (assumed to be 95 cents in this study) must be subtracted.

1

TABLE V-2 (Continued)

Government Price
Support Level in
Effect for the
15 Year Period
(Average Gross
Price in $/cwt.)

Actual

($7.24)

Average Yearly
Producer
Gross
Income from
Navy Beans
Million $

Government Government
Takeover Expenditures
of
for
Navy Beans Navy Beans
1000 cwt.

Million $

Government
Takeover
of Small
White Beans
1000 cwt.

Domestic
Per Capita
Consumption
of Navy
Beans*
lbs.

Average
Weighted
Price to
Producers
from
Navy Beans'*
$/cwt.

36.1

704.2

5.0

0.0

2.07

6.67

Simulated Results:
Actual Support Level
(ASL)

($7.24)

36.0

705.1

5.0

0.0

2.18

6.68

90% of ASL

($6.52)

31.7

464.8

3.0

0.0

2.19

6.12

80% of ASL

($5.79)

29.3

305.0

1.7

0.0

2.20

5.80

70% of ASL

($5.07)

27.6

217.6

1.1

0.0

2.20

5.55

60% of ASL

($4.34

25.7

161.1

.7

0.0

2.20

5.26

50% of ASL

($3.62)

24.4

129.0

.4

0.0

2.19

5.03

23.1

0.0

0.0

0.0

2.21

4.79

Unsupported Market
($0.00)

Average yearly domestic per capita consumption of navy beans equals total navy bean domestic
consumption for the 15 year period divided by the average U.S. population for the same period divided
by 15.
Ttevy bean average price equals total producer gross income from navy beans for the 15 year
period divided by the total navy bean production for the same period.

72
slightly.

Average yearly producer gross income from navy

beans would have dropped more than 10 percent due to the
inelastic demand curve for navy beans.

Government expend­

itures would have dropped 40 percent because of decreased
government takeover resulting from the reduced price sup7
port level.
Average yearly domestic consumption would
have increased slightly because of the lower price in
some years.

The weighted average navy bean price re­

ceived by the producer would have dropped $.56.
With further reductions in the support level,
the movement of the selected variables would, in most
cases, have been farther in the same direction.

As

the support price level was dropped, less navy bean
acreage was planted, resulting in less production.
Average annual producer gross income from navy beans
and average price per unit sold would have fallen as
the price support level was dropped, with the lowest
producer gross income from navy beans and lowest average
price coming in the simulation where the price support
program was eliminated.

Domestic consumption would

have also been highest when the government program was
eliminated.

If the navy bean price support in each

year had been 50 percent lower in each of the 15 years,
7
Mo government small white bean takeover would
have occurred under any of the price support levels
tested.

*

73
government expenditures would have averaged only
O
$400,000.
Without a price support program, govern­
ment expenditure would have been zero.
Private Price Support-storage Program
If a government price support program can be
successful in increasing average annual producer gross
income from navy beans, the possibility exists that a
private program could also have the same effect if a
government program were not available.

A private indus­

try price maintenance and storage program might be a
potential substitute for a government program in the
future.

Some measure of its effects should be of value

to industry decision-makers.
The private price support-storage program, in the
simulation model, was similar to the government program.
The storage agency would have purchased a large enough
quantity in large crop years so that the market price
would not have fallen below the stated support price.
It would have been different in that the quantity taken
over by the storage agency would have been put back into
the market when supply and demand conditions allowed the
minimum price to be maintained.

Under the government

Q
Government expenditures include the cost of beans
and not administrative costs. Therefore, total govern­
ment costs would be greater than this. To the extent
that some commercial sales took place, expenditures would
be reduced correspondingly.

74
program, an attempt was made to dispose of the stored
quantity in non-market channels.

However, the biggest

difference would have been that the private agency would
not have had any outside funds to pay for the beans
taken over by it.
In order to pay the producers for the amount
taken over, the private agency would have had to borrow
money.

The agency would have had to pay assumed inter­

est charges of 6 percent per y e a r .

The agency would

also have had to pay storage charges of $.72 per hundred­
weight per year.

After the agency sold its stored beans,

it would have had to pay back the money borrowed and
redistributed any extra funds back to producers.

If

the minimum price was set too h i g h , the private agency
would not have been able to sell all the beans it took
over.

Its alternatives would have been to either

lower the price support level or make producers pay for
the costs of the program; which, in effect, lowers the
support price received by producers.
What minimum price level could the agency have
maintained over the period?

To answer this question,

ten simulations were run, each with a different minimum
price.

The minimum price levels were selected over

the relevant price range for navy b e a n s .

The important

variables of interest are the annual ending storage,
the 15 year ending*storage and the agency financial

75
position.

The producer gross income from navy beans is

misleading if the agency was not able to sell all the
beans it had taken over.

Part of it was money borrowed

by the agency to pay the producer for the beans that
had been taken over.

A more meaningful gross income

figure can be obtained by adjusting the average annual
gross income producers would have received for their
navy beans by one-fifteenth of the ending agency finan­
cial position.

It is not claimed that the higher mimi-

mum price levels could have been maintained.

They are

shown to give the reader some insight into the inventory
buildup and increasing debt position of the agency at
higher minimum price level resulting from the inability
of the agency to sell its beans.
The higher the minimum price was set for the 15
year period, the more years storage would have taken
place (Table V-3).

At a $2.50 minimum, storage would

have occurred in 5 of the 15 years with average annual
ending inventory of 114,600 hundredweight.

At a $7.00

minimum, storage would have occurred in 14 out of the
15 years, with an average annual inventory of 5,028,900
hundredweight.

Average annual ending inventory would

have increased the higher the minimum price was set.
The average price per hundredweight received by farmers,
a measure of profitability, would not have been increased
substantially over the unsupported situation (see

1

I

TMI1 V-3
omulatib

Mininaa lupport
rrloa for tba
IS laar farlod

woulto roa m manat race canon u v su o m u n n
* ravMt nowei u n n , 1053-1007

Avaraga taarly
aary Oaan
Planted
Acreage

Snail
Mayy Baao
Chita
Production Production

1000 A.

1000 a>t.

02.50

443.0

4707.2

500.7

Cl.SO

445.1

4004.0

507.2

S3.SO

441.0

4771.5

St.SB

440.7

04.50

Drastic
Cary
Com tnptian MSB
of
Osports
Navy Mans

Donastic
Par capita
Canaanptisn
Maryof
baaoab

Soling
bgancy
Otoraga

■itebsr of
Tatra
Mary Mans
Etorad

Mary laana
Stood at
aad of tba
IS Tatra

financial
Position
of Otoraga
bganoy at aad
of IS yoara6

1000 cwt.

Million f

Avaraga
naightbd
Orica to
Otodaoatb
fra
b am Baaaa
0

1000 cut. Million 0

Iba.

3)<i!.4

050.7

” •'<23.0)

2.22

114.4

5

0

4.3

5.03

ji:o.7

075.0

M ‘St24.7)

2.22

152.)

5

0

3.4

5.14

505.1

1104.0

074.7

»-»(23.7)

2.20

204.3

0

0

-.4

4.07

4050.0

514.4

37J5.2

024.0

11,J(25.3)

2.21

542.0

0

0

0.1

5.22

450.3

4043.1

510.0

3722.7

040.4

” -»C25.0)

2.21

000.0

10

0

-12.0

5.14

05.00

440.1

4002.4

534.0

3110.0

1044.4

” •‘(34.1)

2.22

1300.3

10

0

-22.7

5.23

05.50

472.3

5100.0

550.4

3722.4

1044.0

” •*(14.4)

2.21

2070.0

12

2150.0

-50.1

5.20

00.00

404.4

5270.0

540.)

3H02.5

1034.7

3J,1(27.5)

2.20

3140.2

12

5205.0

-02.0

5.21

04.H

404.0

5352.5

570.7

31)40.0

1027.4

U , *(27.0)

2.10

3544.0

14

4010.0

-117.0

5.10

07.00

510.3

5532.0

500.3

3U3S.2

1004.0

)*-*(27.4)

2.14

5020.0

14

10444.0

-102.1

4.05

1000 art.

10)0 cwt.

frodacar Cross
Insaaa r r a
aary baaoa

1000 art.

*Tba mint la ptrsnthasaa la the gross iacans fra navy beans raoaivad by producers aftar adjoatiag for U w fiaasclal position of tba atoroga agency at tba aad of tka period.
b

a m lT yM r parted dlrtdndby*!)0**1** 00“ B^ ,U0° o t
p0,tti0" 01

b#“ * •*'“ u to t*1 * " * , t i c e«ne«nptlan of nary baaoa for tba parlod divided by tba avaraga 0.0.

for tba

naocy la agoal to the lneeaa raoaivad fra aaUlng of atorad baaoa alaaa dabt for unpaid borrooad capital, raald lauraat, and — pfM atoraga

********
n*oa paid to prodaeara fra aayy baaoa equals total prodaoar groaa laooaa fra nary baaoa (adjaatad for aadlag agaacy flaaaelal position) for tba IS yoar
parlod dWldad by tba total navy baao pndaetioo for tba tan parlod.
-v~
r—

77
Table V-2 for unsupported situation) for any of the mini­
mum price levels.
A problem would have occurred for any minimum
price above $5.00 because of the increased storage at
high minimum prices.

At minimum prices above this level,

there would have been storage occurring in most of the
15 years, with some beans still being stored at the end
of the period.

When a portion of the current year's

production would have been stored in most of the 15
year period, very few stored beans would have been sold.
This means that very little money would have flowed
back into the storage agency and it would not have been
able to pay its interest charges, storage costs or capi­
tal debt.

It would have had to borrow additional money

to meet these expenses and also administrative costs,
if the minimum price level was to be maintained.
The breakeven point in terms of the agency being
addle to sell all its beans and cover storage and interQ

est charges would have been at a minimum price of $4.00.
Above a $5.00 minimum, the agency would not have been
able to sell all its takeover by the end of the 15 years.
9
The $4.00 minimum's ending financial position would
have been better than the $3.50 minimum's financial position
because with a higher minimum price in some large crop
years, more beans would have been stored.
When the beans
were sold in following years the agency had more beans to
sell at a higher price (in the elastic portion of the de­
mand structure), but not enough of a quantity to have put
price back on to the inelastic portion of the demand
structure.

78

The higher -the minimum price level went above a $5.00
minimum, the larger the storage would have been at the
end of the 15 years.
Compared to the actual government price support
program, the private price support-storage program would
have been substantially lower in producer gross income
and average price from navy beans.

However, the program

would have offered a benefit of less price variation
within each season by establishing certainty as to the
minimum price possible.

The program would have offered

more price and income stability but not substantial in­
creases in either price or income compared to the govern­
ment program.
Marketing Quota or Control
If one wishes to increase producer gross income
and reduce price stability, another program exists which
could have maintained a minimum price in the industry.
The program would have involved controlling the amount
of beans marketed so that the market price would not
have gone below the minimum price level.

This could

have been accomplished by issuing marketing certificates
for the beans that could be marketed, based on a quota
for each producer.

If a producer did not receive a

certificate for a certain portion of his production,
he could not have marketed that portion of his production.
When supply and demand conditions warranted, the producer

79
would have been given certificates which would have
allowed the marketing of some stored production.

A

program of this type would have required an industry
or government agency that had a monopoly on the selling
of beans.
Under this type of program, the market price would
have been paid only for the production allowed to be
marketed.

The producer would have retained ownership

of the rest, and would have paid commercial storage
rates in this simulation.

He would have retained owner­

ship and received no payment for the unmarketed produc­
tion in some years.

In simulating his production b e ­

havior, it was assumed that he reacted to this situation
as if he had received a lower price and sold his total
supply.

He then planned his next year's production

taking this price (as a proxy for the effect of the un­
marketed production)

into account.

To estimate the results of this program in the
navy bean industry, ten 15 year simulations were run
at different relevant minimum price levels

(Table V -4).

The main difference between this program and the private
price support-storage program ^ s

in planted acreage.

Under the private price support-minimum price program,
the producer would have received at least the minimum
price for all his production and, therefore, would have
made acreage decisions based upon the market price.

80
Under the marketing quota, the producer would have only
received the minimum price for the marketed portion of
production.

Given the assumption stated in the last

paragraph, he, therefore, planned next year's produc­
tion

(in some years) with a price lower than the minimum

stated level.

This accounted for the lower acreage

planted.
Production would generally have been less for the
15 years if the minimum price for the period was increased.
With a higher minimum price level, the producer would
have been able to market a smaller quantity of beans.
Therefore, he would have produced fewer beans.

For ex­

ample, if the minimum price level was set at $4.00 for
the 15 y e a r s , average annual production would have been
4,720,000

hundredweight, while at a $6.00 minimum,

average annual production would have been 4,599,000
hundredweight.

Because of the reduced production, the

higher price level could have been maintained with less
storage than occurred under the private price supportstorage program.

At the same time, the increase in

price would have more than offset the decrease in pro­
duction, thereby, increasing producer gross income from
navy beans.

In fact, at a $7.00 minimum price, the m a r ­

keting quota would have given an average yearly producer
gross income from the navy beans he sold of $34.2 million,

TABLE V-4

SIMULATED RESULTS FXJR TEE MARKETING QUOTA LEVELS FOB THE PERIOD 1953-1967
Navy Bean
Price Main­
tained by
Use of tbe
Marketing
Quota

Averaoe Yearly
Doaestic Navy Bean Producer Doaestic Unaarketed Unaar- Busbar
Navy Bean Navy Bean
Snail
Constaption
In- per capita Navy Beans ketad of years
Planted Production White
Exports Gross
Acreage
conn froa Consumption at end of Navy marketing
Production
of
Navy Beans
Navy Beans
of
crop year Beans
Quota
Navy Beans
at end of Operated
15 year
period
1000 A.

1000 cwt.

1000 cwt.

1000 cwt.

1000 cwt. Million $

lbs.

1000 cwt. 1000 cwt.

Average
Weighted
Price to
Producers
froa
.
Navy Beans
$

$2.50

440.4

4752.0

503.1

3918.9

833.1

24.2

2.21

121.3

0

5

5.05

$3.00

440.4

4752.5

512.8

3910.1

842.4

25.2

2.21

141.6

0

5

5.24

$3.50

439.0

4746.3

523.3

841.8

26.2

2.20

200.1

0

6

5.46

$4.00

437.0

4720.3

538.2

3887.5

832.8

27.5

2.19

251.7

0

7

5.77

$4.50

436.6

4705.7

552.1

3872.9

832.7

28.8

2.19

286.0

0

7

6.06

$5.00

436.2

4691.8

564.0

e0

836.0

29.9

2.17

330.7

0

8

6.31

$5.50

432.9

4650.5

577.9

3834.7

815.8

31.1

2.16

406.1

0

8

6.61

$6.00

428.3

4599.6

591.4

3820.2

779.4

32.1

2.16

488.3

0

8

6.91

$6.50

428.5

4600.5

604.5

3801.4

799.1

33.5

2.15

584.6

0

9

7.21

$7.00

424.0

4555.8

614.7

Jf

781.2

34.2

2.13

666.5

0

10

7.41

“Average yearly domestic per capita coostaption of navy beana equals total domestic consuaption of navy beana
for tbe 15 year period divided
by the average U.S. population for the sans period divided by15.
^Average weighted price
fox tbe 15 year period divided

paid to producers froa navy beana equals total producer grossincomefroa navy beans
by tbe total navy bean production for tbe sane period.

82
which was only $1.8 million less than what happened
under the actual government price support program.
At higher minimum price levels for the period,
average yearly domestic consumption and exports would
have decreased over what would have occurred under the
private price support-storage program with the same
minimum price level.

On the other hand, average storage

would have been less because of the reduced production.
In none of the simulations would there have been stor­
age remaining at the end of the 15 year period with this
program.
From viewing the results of the marketing quota
simulations, some people might suggest that price could
have been raised considerably higher than the highest
experimented level of $7.00.

While the model might

suggest this, at higher prices other areas of the country
might start to grow beans.

There probably also would be

substitutions of other beans or other food products for
navy beans that would not have taken place at lower
prices, especially over a prolonged period like 15
years.

Therefore, there are constraints that would

keep navy bean prices from reaching very high levels
over an extended period of time.
Acreage Control Program
Under the programs considered up to this point,
the main emphasis has been on the controlling of price

83
by con-trolling navy bean market supply within different
years.

Alternatively, the major production input, land,

needed to produce navy beans might be controlled.

To­

bacco is just one example of where the land input is
restricted.
An acreage restriction could be the prime control
device in operating such a program for navy b e a n s .

Nine

acreage maximums were arbitrarily set to test this pro­
gram.

The model was run nine times for the 15 year

period.

Each run was made with a different maximum in

effect.

The results are given in Table V-5.
If the acreage maximum h a d been set at 300,000

acres per year, that amount would have been planted
each year.

Average annual gross income from navy beans

would have been $29.2 million.

However, production and,

therefore, domestic consumption and exports would have
been drastically reduced.

If the acreage maximum had

been set higher, average yearly production, producer
gross income from navy beans and consumption would have
increased until the 400,000 acre limit was reached.
Beyond this point, average yearly producer gross income

The possibility exists that by use of more fer­
tilizer, improved seeds, etc., part of the effect of the
acreage control could have been reduced through increased
yields.
If this had occurred, the decrease in production
at the different control levels would not have been as
great as the simulations suggest.

I
TABLE V-5

SIMULATED RESULTS FOR MIKE SELECTED ACREAGE CONTROL LEVELS FOR THE PERIOD 1953-1967
Average Yearly
Size of the acre­
age planted
Doaestic
Navy Bean Navy Bean
Snail
Navy Bean Producer
Donestic
Average
Years
restriction for
White
Conauaption Exports
Gross
Per Capita Weighted . Acreage
Production
Planted
the 15 year
Acreage
Production
of
Incoaw
Conauaption Price to
Restriction
period (per
Navy
Beans
fron
Producers
Level
o£
a
year)
Navy Beans Navy Beans
froa
.
was
Navy Beans Reached
1000 cwt.
1000A.
1000 cwt.
1000 cwt.
lbs.
1000 cwt. Million $
$
300,000 Acres

300.0

3166.0

684.3

3133.4

32.7

29.2

1.77

9.23

15

325,000 Acres

325.0

3429.8

671.6

3367.9

97.4

30.7

1.88

8.94

15

550,000 Acres

346.3

3692.6

660.5

3509.9

221.6

32.2

1.96

8.71

14

375,000 Acres

367.8

3933.5

643.2

3594.2

388.6

32.9

2.00

8.37

13

400,000 Acres

389.3

4170.0

629.4

3667.4

544.3

33.7

2.05

8.07

13

425,000 Acres

407.7

4386.8

591.1

3754.4

676.1

31.3

2.10

7.14

12

450,000 Acres

422.5

4556.8

545.4

3859.6

742.7

27.5

2.16

6.04

10

475,000 Acres

433.6

4679.2

527.1

3920.2

805.8

26.6

2.19

5.69

5

500,000 Acres

438.9

4734.2

516.5

3922.0

859.3

25.9

2.19

5.47

5

aAverage yearly doaestic per capita consunption of navy beana equals total domestic consunption of navy beana
for the15 year period divided by the average U.S. population for the sane period divided by 15.
V

for the

Average weighted price paid to producers froa navy beans equals total producer gross incomefroanavy beans
15 year period
dividedby the total navy bean production for the saae period.

85
from navy beans would have fallen in the 15 year period
while the acreage maximum was reached less often.

If a

500,000 acre maximum had been in operation, it would
have been effective in only one-third of the years.
The results of these simulations suggest that if
acreage planted had been restricted too much, producer
gross income from navy beans would have been reduced
because the elastic portion of the demand curve was
reached and quantity demanded declined relatively faster
than price increases at extremely high prices

(although

the net income from navy beans could be higher).

If

acreage had not been restricted enough, producer gross
income from navy beans also would have decreased because
the inelastic portion of the demand curve was reached
and price decreased relatively faster than quantity in­
creased.

Thus, the intermediate restriction zone seems

to be where the highest producer gross income from navy
beans would have been.
This program was the only one examined that did
not entirely base planting decisions on the competitive .
position between crops when the planting decision was
made.

Consequently, the question of what happens to

total producer gross income arises.

In some years,

producers would not have been able to raise as many
beans as the competitive situation suggested they wanted
to grow.

The model was not designed to explicitly ^

86
handle this question.

When comparing this program to

the other programs, the reader should remember that
producers would have had an opportunity to plant some
of the acreage diverted from navy beans in other crops.
Therefore, the producers' gross income from navy beans
is not necessarily equivalent to the navy bean producers'
gross income.
Two Price System

(Price Discrimination)

Another type of supply management program can
control the quantity of navy beans going into different
markets.^1

The idea behind this type program is to get

a higher price for the commodity in the market that is
less responsive to price changes.

By reducing the quan­

tity supplied in the market which is less responsive to
price changes, and increasing supply in the other mar­
ket, producer gross income can be increased.

This form

of price discrimination can take place if the total
market for the commodity can be legally and effectively
divided into submarkets which exhibit different respon­
siveness to price.

Price discrimination is used quite

frequently in both agricultural and non-agricultural
products.

The selling of milk for fluid uses at a higher

■^A discussion of price discrinimation is given in
Richard H. Leftwich, The Price System and Resource Allocation (New Yorks
Holt” Rinehart and Winston, l£f>6) . d d .
197-200.

87
price than milk for manufactured uses is an example of
price discrimination in agriculture.
The proposed separation of markets under this
program in the navy bean industry would have been be­
tween domestic and export markets.

A higher price would

have been charged in the domestic market because domestic
consumption is less responsive to price changes than are
exports.

Therefore, a higher price could be charged

in the domestic market with a corresponding small de­
crease in consumption.

The quantity of beans not sold

in the restricted domestic market at the higher price
would have been sold in the export market where it would
have less effect on the world price.
To maintain a separation of the domestic and ex­
port markets, governmental action would have been re­
quired.

The tariff on dry beans has averaged out to

approximately $1.50 above the export or world price.

12

If the difference had become greater than this, imports
would have driven the domestic price down.

If a larger

difference in price had been desired, either a larger
tarriff or a quota restriction on imports into the
United States would have been needed.

12

U.S. Tariff Commission, Tariff Schedules of the
United States Annotated (1969) (T. C. Publication 272,
Washington D.C.: Government Printing Office, 196 8),
p. 41.

88
In the first simulation of a two price system,
no restrictions were placed on production.

Instead,

the simultaneous demand structure was adjusted so that
the domestic price would have been $1.50 above the ex­
port price.

This price setup is compatible with the

existing U.S. tariff structure on dry beans.

Under

this two price system, domestic marketings would have
been controlled so that domestic price would have stayed
$1.50 above the export or world price.

Acreage planted

decisions were assumed to be based on the average of
the two prices, weighted by the quantity sold in both
markets.
The results of the simulation were quite similar
to the unsupported market simulation (Tables V-2 and
V-6).

Average yearly producer gross income from navy

beans would have risen only $200,000 during the 15
years over what it would have been with the unsupported
market situation.

Domestic consumption and exports

would have been virtually unchanged.

Since neither

domestic consumption nor exports are very responsive to
changes in price, these results were not surprising.
The reason for this was that production was so large in
five of the fifteen years that the $1.50 difference
between domestic and export prices could not have been
maintained, with both prices reaching the $2.00 minimum
set by the model.

TABLE V-6

RESULTS Of THE W O PUCE SYSTEM SIMULATIONS FOR THE PERIOD 1953-1967

Average Yearly

Type of Two
Price Systea

$5.50 doaestic
niniaua
$5.75 doaestic
$6.00 doaestic
miniaua
$6.25 doaestic
■iniata
$6.50 doaestic
niniaua
$6.75 doaestic
$7.00 doaestic
niniaua
$7.25 doaestic
■tnt —

Snail
Navy Beans Navy Bean
Navy Bean Producer
Doaestic
Coaaarcial. Average
Doaestic
Inventory Weighted
Production
White
per capita
Acreage
Consuaption Exports
Gross
Xncoae
Consuaption at end of Price to
Production
Planted
of
15 years Producers
Navy Beans
froa
froa
.
Beansa
Navy Beans Navyof
Navy Beans
1000 A.
1000 cut.
1000 cvt.
1000 cvt.
1000 cut. Million $
lbs.
1000 cvt.
$

459.9

4961.2

562.1

3922,8

1038.3

27.3

2.19

0

5.50

460.1

4976.4

567.1

3911.8

1064.6

27.1

2.19

0

5.45

464.5

5015.5

575.5

3900.1

1115.3

27.8

2.18

0

5.55

468.3

5057.2

584.3

3887.6

1169.6

28.6

2.17

0

5.66

472.1

5099.8

593.7

3884.4

1215.4

28.9

2.17

0

5.79

476.0

5141.6

602.8

3869.5

1226.8

29.8

2.16

679

5.90

480.0

5181.9

611.9

3856.3

1233.2

30.6

2.15

1387

6.02

483.3

5223.0

620.8

3840.9

1242.7

31.5

2.15

2092

6.13

486.9

5262.4

629.5

3828.9

1252.2

32.3

2.14

2721

6.26

490.0

5297.9

636.5

3816.5

1253.9

33.1

2.13

3418

447.4

4825.8

504.6

3959.3

866.2

23.3

2.21

0

$7.50 doaestic
$7.75 doaestic
niniaua
$1.50 price dif­
ference
(no niniaua)
for

4.83

yearly doaestic per capita consuaption of navy baana equals total doaestic constaption of navy beans
the 15 year period divided by the average U.S. population for the saae period divided
by 15.

Average weighted price paid to producers froa navy beans equals total producer gross incoas froanavybeans
15 year period
divided by the total navy bean production for tne saae period.

for the

90
In a second simulation of a two price system, the
model was set up with a minimum domestic price for the
fifteen year period.

Domestic navy bean producers would

have been restricted so that beans could not have been
bought for domestic consumption

from producers for less

than the minimum level in any of the 15 years.

The re­

maining supply of navy beans was marketed in the export
market, with price allowed to fall to the $2.00 minimum
set by the model.

This two price system would have

caused a difference greater than $1.50 in price between
the export and domestic markets.

It was assumed that

a quota or higher tariff would have been put into
operation to restrict imports into the U.S. domestic
market.
Ten fifteen year simulations, each with a differ­
ent minimum domestic price, were run (Table V-6).

These

results show that the two price system would have in­
creased producer gross income from navy beans.

However,

it would also have caused domestic consumption to be
reduced and exports to be expanded over what they would
have been without controls.

Producer gross income from

navy beans would have increased more if the minimum
domestic price had been set at a higher level.

An in­

crease in navy bean acreage planted also would have
occurred.

91
However, there is a problem associated with this
two price simulation.

It was assumed some private in­

dividuals or groups would store beans at the $2.00
minimum in the hope they would have been able to sell
them in the export market when prices went up.

At

minimum domestic prices above $6.50, these speculators
would not have been able to sell their b e a n s .

There­

fore, it is doubtful that very much of this type storage
would have taken place at domestic prices above $6.50.
What probably would have happened was producers would
have had to take a reduced price in the domestic market
or store the beans themselves.
Summary
Attention has been focused, in this section, on
simulating several price and supply control programs.
The simulations were designed to give estimates of how
these programs would have performed in the navy bean
industry from 1953 to 1967.

The variables generated

and examined included navy bean planted acreage,
production, domestic consumption, exports, government
takeover, and producer gross income from navy beans.
The first program examined was the lowering of
the government price support level.

The larger the re­

duction, the less government takeover and expenditures
would have occurred.

Producer gross income from navy

beans and the average producer price would have been

92
progressively less as the government price support
level was reduced.

However, even a reduction to 50

percent of the actual price support level for 1953-67
would have kept the average producers' price received
above the price they would have received in an unsup­
ported market.
A private price support-storage program, similar
to the government program, was then examined.

This

program would not have raised average price received
by producers above $5.23 for any of the minimum price
levels tested.

The breakeven point (that is, having

no agency storage or debt at the end of the 15 years)
would have been at a minimum support price of $4.00.
Third, a marketing quota or control was examined.
This program would have been able to increase the pro­
ducer gross income from navy beans and the average
producer price.

The amount of the increase was depend­

ent on the degree of marketing restrictions.

As mar­

ketings were restricted, less domestic consumption and
exports would have taken place.
An acreage control was the fourth program tested.
The acreage control would have reduced production of
navy beans more, the greater the restriction for the 15
year period.

The highest average price obtained in the

simulations was where acreage was restricted to only
300,000 acres per year.

However, the largest average

93

annual producer gross income from navy beans would have
been obtained at a maximum acreage planted restriction
of 400,000 acres per year over the entire 15 year period.
The last program simulated was a two price sys­
tem.

In one simulation, the domestic price was set to

be a maximum of $1.50 above the export or world price.
This simulation produced an average price and producer
gross income from navy beans similar to the unsupported
market simulation.

In the other two price simulations,

a minimum domestic price was set, with export price
allowed to fall below the domestic minimum.

These simu­

lations showed a larger producer gross income from navy
beans and also a larger average price as the domestic
minimum was raised.

Above a price of $6.50 for the

domestic minimum, not all of the 15 years' production
could have been sold by the end of the period.

This

might have caused problems in maintaining the domestic
minimum at these high levels.
Comparison of producer gross income from navy
beans and the acreage planted together, give an indica­
tion of what total navy bean producer gross income
would have been under each program.

In their best

performance, the acreage control and the marketing quota
would have given a higher level of producer gross income
from navy beans when compared with other private pro­
grams.

This would have been accomplished with less

94

planted acreage.

These two programs' average annual

producer gross income from navy beans would also have
come within several million dollars of the actual
average annual producer gross income from navy b e a ns.
They would have accomplished this with less planted
acreage.

If this diverted acreage had been planted

to other crops, these programs would have been in the
same or in a slightly higher range in terms of average
annual total producer gross income, when compared to
the actual government price support program.
A final qualification of the simulation results
needs to be mentioned.

It is possible that the re­

sponse of exports to lower prices might have been
greater than the export demand curve indicated.

If

this had been the case, the results just examined would
have been slightly different.

Average price would not

have fallen to the bottom of the price range as often.
There most likely would have been a small upward shift
in average price because of increased navy bean exports
for all programs at lower price levels.

However, the

private minimum price-storage program and the two price
system probably would have improved their position
relative to the other programs.

These two programs were

constrained by not being able to sell beans at low
prices, thus, accumulating large inventories.

If more

beans could have been exported at lower prices, these

95

two programs might have performed better than the
current simulation results suggest.
The summary of the simulation results would not
be complete without discussing the relevance of the
results for the industry in the future.
used to predict the future?

Can they be

What can be said is that

if the trend of variables used in the model but deter­
mined outside it (corn price, Canadian exports, popu­
lation) continue to move in the same way they have in
the past, the relative results obtained from the dif­
ferent programs in the future most likely would be quite
similar.

For example, one could reasonably expect a

private price support-storage program would bring some
stability by evening out supply from year to year.
However, it probably would not increase price or in­
come to any large extent over what it would be with an
unsupported market.
However, there are some indications that the
variables determined outside the system are changing
or will change differently from how they moved in the
past.

Feed grain prices have been moving downward.

If this continues, more planted navy bean acreage will
most likely occur at each expected price level.

This

might make an acreage control look relatively more
attractive in the future than it looked for 1953-1967.
In the export market, Canada is now operating a two

96

price system with a lower export price compared to their
domestic price.

Canadian exports may increase and place

pressure on our export market.

This change might make a

two price system for the U.S. navy bean industry look
better relative to how it looked for 1953-1967, particu­
larly if export market development continues and leads
to a more elastic export demand curve.

CHAPTER VI
IMPLICATIONS FOR DIFFERENT INDUSTRY GROUPS
To a large extent this study has been approached
from the producer viewpoint.

This has been particularly

true in building and applying the models.

The main rea­

son for this emphasis has been that the data available
for the industry has been kept primarily by public
agencies which have concentrated on recording economic
activity at the farm or producer level.

A simulation

model constructed using this type of data generates
values that are primarily concerned with the effects
at the producer level.

This level of activity certainly

affects and reflects related behavior farther along in
the marketing process.

To build a model that directly

takes into account the behavior of other groups in­
volved in the navy bean production-consumption system,
more information and data concerning these activities
will first have to become available to the researcher.
However, some implications of the price and supply
control programs for most industry participants can be
examined through this simulation model.

The navy bean

industry can be divided into many different groups with
97

98
many different: concerns.

A partial listing of the

groups that would be affected by the use of the pro­
grams are:
payers,

(1) producers,

(2) consumers,

(4) farm input group,

processors.

(3) tax­

(5) shippers, and (6)

The industry's outlook on its economic

well-being is discussed within the context of these dif­
ferent industry groups.

Since many individuals find

themselves in more than one of these groups, they may
have several, possibly conflicting, views of any one
program.
Producers
It is assumed that producers' primary goal is a
high level of economic well-being.

The price, gross in­

come, and acreage planted to navy beans are partial ways
of measuring this well-being.

It is also assumed that

producers have a high regard for individual freedom of
action.

A recent study of Michigan farmers found that

they do not want their actions restricted.1
Past attempts to raise producer income in agri­
culture have caused problems because most proposals have
conflicted with producers' individual freedom of action.
This problem still exists with the control programs
1Dale E. Hathaway, et a l . Michigan Farmers in
the Mid Sixties (Michigan State University Agricultural
Experiment Station Research Report 54, East Lansing,
1965), p. 63.

99
examined in this study.

The evaluation of the programs

from the producers1 viewpoint can be looked at as a con­
flict between a desire for high price and income and a
desire for individual freedom of action.

The private

price and supply control programs would replace the in­
dividual's freedom of action with the more restrictive
group freedom of action.
One fact that stands out from the producer view­
point is that the government price support program has
been quite successful in attaining the goal of higher
average navy bean price and year to year price stability,
when compared to the unsupported situation.

At the same

time very little of the producers' freedom of action has
been sacrificed, since the government support program
has been merely another marketing option available to
farmers.

However, if price support levels had been

dropped as little as 10 percent, some of the other pro­
grams might have provided higher average price and gross
income from navy beans.

The acreage control program

might have given equally high producer gross income if
the acreage diverted from navy beans could have been
planted to other crops.
The non-government programs would have involved
more restriction on producer freedom of action than the
government price support program.

However, if no govern­

ment price support program existed, the producers might
have found that one or more of the other programs offered

100

enough increase in average price and gross income from
navy beans to balance off the decrease in their in­
dividual freedom of action.

Running a government price

support program without restriction on acreage or supply
probably is feasible only for minor crops, except where
very low support levels are involved.

Governmental

price support programs for major crops (corn, wheat and
cotton) usually have restrictions on producers' actions.
Without some type of supply restriction, the money the
government has spent in the past on major crop programs
rapidly reached politically dangerous levels.

To make

the major crop price support programs work effectively,
but at lower costs, supply or acreage restrictions are
now used.
Increased average price and gross income from
navy beans over what would have resulted with an un­
supported market could have been achieved with an acreage
control, marketing control or quota, or two price system.
All three of these programs would have involved a loss
of some producer freedom of action:

one in the size of

the navy bean enterprise, and the other two in when,
how much, and where beans could be marketed.

Under a

private price support program there would have been
gains over an unsupported situation, but less than other
programs would have provided.

However, the producers'

freedom of action would have been affected less than

101

under the other non-governmental programs.

If exports

had increased more at lower prices than the navy bean
export demand equation suggests, the performance of the
private storage program would have been improved.
Because of conflicting producer g o a l s , no clear
cut program preference can be deduced from the simula­
tion r u n s .

The results do give producers information

on what might have happened if the different programs
had been used in the navy bean industry from 1953 to
1967.

This information provides a basis from which the

producers, acting individually and as a group, can better
decide the tradeoffs from different courses of action
they might consider in the future.
Even though no data is available on the distri­
bution of producers by size, it is obvious that in any
program based on production, the benefits are directly
proportional to the size of each farmer's production.
Therefore, larger producers with higher production re­
ceive more of the benefits.

Since the income benefits

of a program may be capitalized into the value of the
land where navy beans can be grown, the benefits of the
prograpi have accrued and usually benefit the original
landowner most.

New producers, who pay a higher price

for the land because of the program, are dependent on
the program continuing.

Ending of a program could be

financially disastrous for producers who have purchased
land at program-inflated prices.

This potential drop

102

in current: income and land values could give added
impetus for finding different programs to substitute
for or supplement a reduced government price support
program.
Consumers and Taxpayers
It is difficult to separate the consumer and the
taxpayer.

Most individuals belong to both groups, but

a higher proportion of a low income p e r s o n 1s expendi­
tures go for food, less for taxes; thus, low income
consumers are more concerned with food prices than
taxes.

Nevertheless, there are conflicts between the

values individuals hold as consumers and the values
they hold as taxpayers.

Consumers as a group value

having a high quality and sufficient quantity of food
available to them at the lowest possible price.

Tax­

payers as a group have supported policies that help
provide a more equitable distribution of income between
different sectors of the economy.

However, the attain­

ment of equitable distribution or the desire to attain
it is not completely dominant.

Taxpayers do place a

restraint on the amount of government expenditure that
can go for this purpose.

They want some equality, but

they want it obtained at the lowest possible expense.
The navy bean policy conflict can be illustrated
by comparing the simulated results of the actual

103
government price support program for navy beans from
1953 to 1967 and the simulated results of the unsup­
ported market.

Using the simulation, we find that

average annual producer income from navy beans at the
actual price support level was $36.1 million (Table V-l).
This was obtained with the help of an average yearly
expenditure of $5.0 million by the government for price
supports.

According to the simulation results, without

the price support program, average yearly producer in­
come from navy beans would have been $23.1 million,
but with no expenditures by the government.
With these income figures as a basis, the income
transfer that resulted from the government price support
program can be partially examined.

The average yearly

increase in gross income going to navy bean producers
because of the program was $13.0 million ($886 per farm).

2

Taxpayers paid $5.0 million ($341 per farm) of this
total in the form of expenditures for government take­
over.

Consumers paid the remaining $8.0 million ($545
3
per farm) in the form of higher prices.
What the price
------- 5------------The per farm figure is obtained by dividing the
gross income figure by 14,665. This figure is the average
number of Michigan dry bean farms as reported in 1954,
1958 and 1964: U.S. Bureau of the Census, 1964 United
States Census of Agriculture, p. 13.
Dr. C. Bedford, Department of Food Science,
Michigan State University estimates a one pound can of
prepared beans has approximately four ounces of CH.P.
dry beans in it. Therefore, a $1.00 increase in the
price paid to producers will increase the price of a
one pound can of prepared beans $.0025 (other things
constant). Since the difference between the actual

104

support program did was place part of the income trans­
fer burden on the taxpayer and part of it on the con­
sumer.

An alternative to this program could have been

a direct payment of the $13.0 million to producers with­
out use of a price support program.

Under this type of

program, the taxpayer would have carried the entire
burden of income transfer.

Another alternative could

have been the use of non-government price and supply
control programs like those that are examined in this
study.

This would have shifted the entire burden of

income transfer to the consumer.

If consumers used

navy beans in the same proportion that they paid taxes,
the burden would have been the same for all individuals.
However, beans are consumed in large quantities by low
income individuals who pay proportionately less taxes.
In terms of these low income consumers, the non-govern­
mental programs would be the least desirable.

These

low income consumers might be for either a program of
government support of preferably no support at all in
order to keep bean prices low.
The consumers and taxpayers can not view the
programs to be used in the navy bean industry in isolation.
average weighted price and the unsupported market's
average weighted price was $1.88(see Table V - 2 ) , the
government price support program increased the price of
beans an average of $.0047 for a one pound can over
what it would have been with an unsupported market.

105
The programs must: be considered as part of an overall
farm and welfare program or policy.

Therefore, no clear

consensus as to how the different navy bean programs
might fit into the overall policy can be predicted.
The implementation of any of the non-governmental pro­
grams would mostly like come from a compromise between
overall government farm and welfare policy.

The simu­

lations can be helpful in giving each group information
which consumers and taxpayers can use in deciding which
program(s) is most desirable or acceptable to them.
Farm Input Group
It seems reasonable to assume that the farm input
industry serving the navy bean industry wishes to maxi­
mize its profit.

The input group is interested in

having a large and, if possible, a growing demand for
the products and services it sells.

This demand will

remain large if acreage goes up and producers receive
a high price and income for the products they produce.
If producers receive higher prices and income and use
more purchased inputs to accommodate a growing demand
for their products, the farm input industry's interests
would not be in conflict with the interests of producers.
However, the price and supply control programs
attempt to increase the price and income producers re­
ceive by regulating and restricting supply rather than
by demand expansion programs.

The acreage control and

106

marketing control or quota would have reduced the navy
bean acreage over the period under study.

Looking at

this one fact alone, the input group might oppose imple­
mentation of these programs.
A broader view on the problem might change the
situation.

Even if acreage of one commodity is reduced,

this does not necessarily mean that the demand for pur­
chased inputs by producers is also reduced.

The acre­

age taken out of production of one commodity probably
will end up in the production of another commodity.
Further, the navy bean income per acre may increase,
leading to increased demand for inputs.

The picture

is further confused when the simulation without controls
is examined.

If no control had been used, price and

income received by producers would have been lower.
In response to this, producers would have planted
fewer acres to navy b e a n s .

In these last two cases,

the farm input group might be either neutral or for
price and supply control, depending on their restrictive
nature and the expected transfer effect to other com­
modities .
The preceding discussion shows that the use .of
price and supply control programs have effects on the
producer input group as well as other participants in
the industry.

Depending on the programs used and the

level at which they are used, the demand for the

107

products and services supplied by the farm input group
could either increase or decrease.

The study does not

provide sufficient information to determine the exact
tradeoffs between acreage planted and producer gross
income as they affect the demand for purchased inputs.
To provide this information accurately, studies on a
multi-commodity basis would have to be made.
Shippers
Like the farm input group, the shippers are
assumed to be guided by the goal of profit maximiza­
tion.

If the shippers operate with a certain margin

of profit per unit of beans handled, total profit is
dependent on the total number of units handled.

Ship­

pers, therefore, are interested in handling a large
volume of beans in order to insure a large profit.
The larger the volume handled, the lower the profit
margin on each unit handled has to be in order to obtain
the same total profit.

Shippers probably are interested

in programs that increase the volume of beans they
handle.

They are also interested in having a certain

minimum quantity available for shipping each year.
The government price support program resulted in
an average yearly production of 5,409,000 hundredweight.
The simulation without controls resulted in an average
yearly production of 4,819,700 hundredweight.

The

108
government program has had the effect of increasing the
number of beans that the shippers handled over what
they would have handled without the program.

If the

other price and supply control programs had been sub­
stituted for the government price support program,
production of navy beans would have fallen below what
it was with the government program in operation

(with

the exception of the private price support-storage pro­
gram at very high minimum price levels).
The price stability provided by some of the
control programs probably would be considered a posi­
tive factor by most shippers.

The price support pro­

grams, the marketing control or quota, and the two
price system restrict the price movement so that ship­
pers could purchase beans for resale with some assur­
ance that price would not fall below a certain price.
The storage feature of the marketing control or quota
and the price support program also should be considered
a positive factor by the shippers.

Stored supplies

insure supplies for export in short crop yea r s , which
makes it easier to develop and hold export markets.
The programs that require storage of beans would also
help the shippers1 business through increased use of
their elevator facilities.
Consequently, shippers would receive some bene­
fits from the use of the private price and supply

control programs.

However, private control programs

imply control over trading, so some possible duplica­
tion of salesmen in various competing shippers' offices
might be eliminated.

Also, the volume that shippers

would handle would probably decrease with the use of
any of the private price and supply control programs
over what it would be with a government price support
program that is maintained at price support levels
equivalent with past ones.

The handling and storage

of government takeover has been done by licensed
elevators.

Since almost all elevators are licensed,

and producers have the option to take beans bound for
government storage to any licensed elevator, reduction
in volume stored would be spread proportionately over
all elevators.
One final comment on the effect of the private
price and supply control programs on the shippers is
needed.

One of the two largest shippers is a coopera­

tive run on a non-profit basis by producers.

Since it

is run for the benefit of producers and not for a pro­
fit like the corporate shippers in the industry, its
view of the programs may be close to that taken by the
producers.
Processors
Processors' profit, as it concerns navy beans,
is dependent on the volume of beans it processes and

110

sells.

Generally, the more beans the processors handle,

the greater their profit will be.

Navy beans are only

one of many products handled by most processors
marily bean canners).

(pri­

The data generated by the simula­

tions , consequently, does not give a clear picture of
the effects of the programs on the profit or operation
of the processors in the area of navy beans.

The simu­

lations with the two control programs and the two price
system suggest that domestic consumption of navy beans
would decrease slightly if these programs are imple­
mented.

However, this does not indicate what would

happen to total processors' profit, since consumers
may increase consumption of another product within
their product line.
A more stable price and supply would help can­
ners in their planning, processing, promotion, and
merchandising practices.

All the supply and price con­

trol programs offer more of this stability than would
an unsupported market.
Required Institutional Setting
The use of the private price and supply control
programs would all have required producer group action
and the loss of some individual freedom of action.

The

group action could involve regulating market supply,
seating minimum prices, storing beans, and controlling
acreage planted.

These activities would require an

Ill
appropriate institutional setting.

Let us examine some

institutional forms that could provide the legal and
economic framework necessary for successful operation
of the programs.
In theory, all the programs could have operated
in the navy bean industry under existing cooperative
legislation.

However, cooperative programs which have

tried to regulate price and supply have had very little
success in increasing producer income.

Many producers

have tried to get the benefits from these cooperative
programs without joining the cooperative and paying
the associated costs.

The name most commonly used for

this is the "free rider" problem.

The chief weakness

of these cooperative programs has been in the lack of
power to compel all producers or handlers to act in
the manner required to make a certain control program
work.
The marketing order, an option open to many U.S.
commodity groups, has used compulsory powers.

It was

authorized by the Agricultural Marketing Agreement Act
of 1937.

Under marketing order legislation, an industry

group may operate certain supply and price control pro­
grams if it is accepted by a two-thirds majority of
producers.
4

4

However, marketing orders are not authorized

The two-thirds majority can be by number of pro­
ducers or by quantity produced.
For some citrus crops
the required majority is higher than two-thirds.

112

for navy beans at this time.

A marketing order may be

issued only for certain products, chiefly milk and some
fresh fruits and vegetables.

Marketing orders for a

commodity may have one or more of the provisions dis­
cussed below.

Before a particular order is issued, it

must be voted on by the producers in the area covered
by the order.

Once an order is issued by the Secretary

of Agriculture, it is mandatory that all handlers af­
fected by the order comply with the provisions.

Mar­

keting orders for some fruits and vegetables must also
be approved by the handlers of the product.

A committee

recommended by the industry and appointed by the Secre­
tary of Agriculture makes recommendations of courses
of action to take.

This committee can be made up of

both producers and handlers.5
Market orders for milk generally have required
handlers to pay for milk on the basis of their use of
the milk, with different prices for each use.

Milk

marketing orders specify minimum prices for each use
class.

Orders that have been issued for other allowed

commodities have not involved price control directly.
Instead, limitations have been placed on shipments, the
kinds of containers used have been regulated, and
5

U.S. Department of Agriculture, Agricultural
Stabilization and Conservation Service, Compilation of
Statutes Relating to Soil Conservation et a l . (Agri­
cultural Handbook NoT 242, January 1, 1963), pp. 275-91

113
research and development have promoted.

Prices have

been raised mainly

through limitations on shipments or

allocation between

markets.

Powers given under the

order can be used to limit:
crop that may be sold,

(1) the total amount of a

(2) the grade, size or quality

of the product that may be sold,

(3) the amount that

may be sold in any

time period, and (4) the amount that

may be sold in any

market.

The effect of the marketing order as a price and
income raising tool has been open to question because
orders directly regulate the handler and not the pro­
ducer.
tion.

This has meant no direct controls of produc­
Marketing orders' major contribution has been

short run stability in the markets where the marketing
order applies.
The marketing board, an institution that has
some features similar to the marketing order, has been
used to help improve the results flowing from various
agricultural products in the United Kingdom, Canada,
and Australia.

A Guelph bulletin defines it as:

"A

producer-controlled compulsory, horizontal organization
sanctioned by government... to

perform

specific mar­

keting operations... in the interest of the producer of
the commodity concerned."6

The marketing board can

Department of Agricultural Economics, University
of Guelph, A Comparative Study of Agricultural Marketing
Legislation"in Canada, Australia, Unitea Kingdom and the
United States (Department Publication No. A. E./64-65/11,
Buelph, Ontario, Nov. 1964), pp. 3 & 4.

114

directly regulate the producer if it so desires.
of this power must be approved by producers.

Use

This

gives this institution the power to affect price for
a commodity either through use of the power or threat
of its use.
The main difference between the U.S. marketing
order and the U.K. style marketing board is the poten­
tially greater power and producer control inherent in
the marketing board.

The marketing board is run by

a board composed of producers, who are elected by and
directly responsible to the producers.

This gives

the wide powers of administration to the producers.
The marketing order committee can only make recommenda­
tions to the Secretary of Agriculture.

The Secretary

is the only one who can make modifications in the
order and he is not directly responsible to the pro­
ducers .

With an order the producers do not have the

powers of administration.
In conclusion, there has been no institution
legally available in the navy bean industry that could
have effectively used or can now use the non-governmental
price and supply control programs examined in this study.

7

Cooperatives are not considered capable of using
the programs because of the "free rider” problem associ­
ated with them because they do not have a mandatory com­
pliance provision.

115
However, there are Institutions that can use one or
more of the programs considered, but they are not now
legally available.

These are the marketing order and

the marketing board.

P

The marketing order affects

individual freedom of action less than the marketing
board.

It probably would be less effective in using

some of the control programs than the marketing board.
For the marketing order to be made available
extension of the Marketing Agreement Act to include
navy beans as one of the applicable commodities would
be required.

In the case of tha marketing board en­

tirely new enabling legislation giving powers not pre­
viously given to producers would have to be passed.
The "Mondale Bill," which has been in committee in the
U.S. Senate for the past several years, would extend
the marketing order to all agricultural commodities.
The bill would also provide for the use of producer
marketing quotas.

This would make the marketing

order have more powers similar to the U.K. style mar­
keting board.
The simulation results suggest the private pro­
grams could improve producer gross income from navy

0

A discussion of the use of a marketing board in
the Ontario Navy bean industry is given in Appendix A.

116
beans in the future if the government price support
level is reduced.

However, this would require the

passing of legislation making the existence of a
marketing order, a marketing board, or some other
similar institution possible in the navy bean industry.

CHAPTER VII
SUMMARY
Michigan produces over 99 percent of the navy
beans grown in the United States.

Navy beans are one

of the most important agricultural products grown in
Michigan and have a wide domestic and export market.
In the domestic market, they are primarily purchased
by canners and processed into canned beans in tomato
sauce, bean soups, etc.

The exports go chiefly to the

United Kingdom where they compete with navy beans from
Ontario.

Because navy beans have had wide yield, price,

and income fluctuations in the past, a government price
support program has been in effect in the industry for
the last 30 years to help reduce the fluctuations in
income and to increase producer income.

This govern­

ment program has involved a large expenditure of
government funds for purchasing part of the production
in large crop years to support the price of navy b e ans.
The industry has relied heavily on this government pro­
gram during these years.
Government expenditures on agricultural programs
may be cut in the future.

If this should occur, the
117

118
industry may have to operate with a reduced government
price support level or no government support program
at all.

This study is designed to look at the effects

of variations in the government price support level
and of some non-governmental price and supply control
programs that could be used individually or in combin­
ation with the current support program.

More specifi­

cally, the objectives of the study are:

(1) to estimate

the basic supply and demand relationships existent in
the Michigan navy bean industry for the period from
1953-67, and (2) to evaluate whether selected price
and supply control programs would be desirable to employ
in the Michigan navy bean industry.
To examine various price and supply control pro­
grams one must first examine the supply and demand re­
lationships that exist in the navy bean industry.

It

is within this context that these programs would have
to operate if they were implemented.

The first step

in the study was to examine the industry's historical
and present position in terms of marketing organizations
and programs.

The second was to design and estimate an

economic model of the industry using price and supply
data for the years 1951-67.

The economic model was

composed of a recursive supply and demand structure.
The supply structure included a navy bean acreage
planted equation, which, when combined with actual yield.

119
predicted the production for a particular year.

The

acreage planted equation was estimated by ordinary
least squares.

A large navy bean planted acreage was

usually associated with a large acreage planted to
navy beans the previous year, a high navy bean producer
price during the previous year, and a low price of corn
received by producers the previous year.
The demand structure included three equations:
domestic demand, export demand, and the price of small
white beans

(a competitor of navy bean s ) .

Three stage

least squares was the estimation procedure used for
these three equations.

Both domestic consumption and

exports moved in the opposite direction of the price of
navy beans.

They were both price inelastic, with price

elasticity of domestic demand being -.14 and price
elasticity of export demand being -.36.

Domestic con­

sumption was found to move in the same direction as U.S.
population, and exports in the same direction as U.K.
population.

The price of small white beans was found

to move in the same direction as the price of navy
beans with a $1.00 change in navy bean price associated
with a $.74 change in small white bean price.
The estimated supply-demand relationships are
the basis of a recursive simulation model.

The basic

information-feedback loop used was price and quantity
data for navy beans and small white beans.

The question

120

to be answered by using the model was:

What would the

effects of using a certain control device have been on
the navy bean industry for the period from 1953 to
1967?

The model was first validated by using a govern­

ment price support program in the model and seeing if
it could duplicate what actually happened in the industry
in that fifteen year period.

Actual values for the

variables determined outside the economic model and
actual government price support levels were used in
the validation.

The yearly values for variables ex­

amined for this validation included price of navy beans#
production of navy beans# production of small white
beans# navy bean producer income# government expendi­
tures for navy bean takeover, government takeover of
navy beans# and government takeover of small white
beans.

Before the validation was completed# the top

portion of the demand curve was modified to make the
simulated results conform more closely with the actual
results.
After the model was validated# the following
price or supply control systems were simulated:

(1)

variations in the government price support level#

(2)

an unsupported market situation (no price or supply
control), (3) private support-storage program#
marketing quota#
price system.

(4) a

(5) an acreage control, and (6) a two

121
In the simulation of these situations or programs,
the actual values for the variables determined outside
the economic model during the fifteen years were used.
The variables examined for effects were the variables
generated within the m o d e l .

These included navy bean

acreage planted, navy bean production, navy bean pro­
ducer income, etc.

Various runs of the model with each

device in it were made.

Each run had the controlled

variable set at a different level.

For example, the

private minimum price level was set at 10 different
levels and a fifteen year run of the model was made
for each minimum price level.
The use of the computer to evaluate economic
questions, as this study has done, allows a fairly
quick and easy manipulation of a complex system of
variables.

It provides valuable information to in­

terested individuals and decision makers who have an
interest in the navy bean industry, without the costs
of actual implementation of undesirable programs.
However, the reader should remember that the results
are simulated and are presented only to help improve
both the actions that might be taken in the industry
and the understanding of the current industry behavior.
The first limitation of the computer simulation
is that it is only as good as the equations that are
its basis.

These equations perform relatively well

122
when they are used over the range of values encountered
for the variables in the actual operation of the indus­
try.

At variable levels not often reached, the rela­

tionships between variables as expressed in the esti­
mated equations may no longer be relevant ones.

If a

certain portion of the variable's range has not been
observed in recent y e ars, as in the case of the low
price range in navy beans, assumptions as to the be­
havior of the variables must be made.
The second limitation concerns the programs ex­
amined and the levels at which the controllable
variables
run.

are

set

at

before

the simulations are

The results are dependent on the programs chosen

for examination.

It may be, however, that the best

program from the total industry viewpoint and sub­
industry viewpoint was not included.

The most important

shortcoming in this area may be the researcher's crea­
tivity or ability to find feasible alternatives to
simulate.

The results probably would also be different

if a different time period had been used.

Further, the

equations may not be valid under all programs.

For ex­

ample, greater uncertainty may change parameters in the
supply equations.

However, the basic model remains

available to test other alternatives that may look
promising.

123
The simulation technique is not used to give the
course of action that would maximize or minimize as in
linear programming.

Each group within the industry has

goals it pursues, but different industry goals may con­
flict; that is, what one group wants maximized another
might want minimized.

If there is to be optimization

or any close approximation thereof, it must be done in
some type of collective action procedure where competing
industry groups or government policy makers examine the
tradeoffs involved in using one or more of the programs.
The final course of action most likely will not be ex­
actly what any one industry group wants, but will be
some compromise between the attainment of competing in­
dustry goals.
The results of varying the government price sup­
port level show that when government price support
levels were lowered a given percentage in each of the
fifteen years and the model run under this condition,
the effects on producer navy bean gross income, govern­
ment takeover of navy beans and government expenditures
for this takeover were significant.

The more the mini­

mum price support level was reduced, the less gross in­
come the producer would have received for his navy bean
production.

At the same time, government takeover and

expenditures would have become less.

If government

price support levels had been only 50 percent of what

124
they actually were for the fifteen year per i o d , govern­
ment yearly average expenditure would have been only
$400,000 as compared to the $5.0 million
that actually occurred.

($341 per farm)

Average yearly producer income

from navy beans would have been $24.4 million as compared
to $36.1 million.
Two conclusions can be drawn from these simula­
tions.

First, the government price support program has

been effective in maintaining producer price above what
it would have been without a price support program.
Second, it would have taken only a slight reduction in
the government price support levels to have significantly
diminished the income producers would have received and
the expenditures of the government.
The private price support program examined was
similar in operation to the government one with two ex­
ceptions .

O n e , the private operation would have had to

put all takeover back on the market when price allowed;
and, two, the private program would have to be paid for
entirely by producers.

The results of the simulations

show that the minimum price could have been set at a
price of $4.00 and not have had any storage or debt
remaining at the end of the 15 year period.

The price

could have been set at $5.00 without any ending storage.
However, the agency would still have been in debt b e ­
cause of accumulated interest and storage charges.

A

125
$5.00 minimum would not have increased average price
paid to producers over the $4.00 level after adjusting
for the ending agency debt.
In order to see the effect of controlling supply
by limiting the amount producers could market and thus
maintaining price at some target level, a marketing
control was added to the model.
have been directly controlled.

Production would not
Instead, producers would

have been allowed to market only a certain maximum
amount of their crop.

Any remaining portion would not

have been allowed to be marketed until doing so would
not have lowered market price below the minimum level.
It was assumed that producers would take the unmarketed
portion of the previous year into account when making
the current year's planting decision, and thus reduce
acreage planted in the current year.

Under this pro­

gram producer gross income from navy beans and average
price would have increased at higher minimum target
price levels.

Because of the decreasing amount of the

crop allowed to be marketed when the minimum target
price level was set at high levels, acreage planted and
production would have decreased as the target price
level was set higher.
Controlling the acreage planted was also tested.
The effect of this acreage control was to decrease
acreage planted, production, and to increase producer

126
Income from navy b e a n s .

The simulation results show

that if acreage planted had been limited to 400,000
acres a year, average yearly gross income from navy
beans would have been $33.7 million.

If the limit had

been either higher or lower, producer gross inooms from
navy beans would have gone down.
The final program tested was a two pries system.
One variation simulated was a price set-up in which
the domestic price would have been kept $1.50 above the
export or world price.

The results of this simulation

were similar to the unsupported market simulation.
another two price simulation,

In

domestic price was set

at some level and the export price was allowed to fluc­
tuate freely within the model framework.

With this sys­

tem producer gross income from navy beans and average
price would have been increased more in the fifteen
years if the domestic price had been set at higher
levels.

However, that part of production kept off the

domestic market to maintain domestic price would have
had to be sold in the export market.

At very high

domestic prices, not all of the production alloted to
the export market could have been sold in the export
market under the assumed demand conditions at those low
export prices.
The general conclusion obtained from the examina­
tion of the different control programs is that they would

127
have increased producer income from navy beans above
what it would have been without controls.

The govern­

ment price support program at the actual price support
levels was superior in this regard (at least at the
price levels considered for the private programs).
The effectiveness of the other control programs was
dependent on their ability to limit production.

Those

that did not have this ability were not as successful
at raising gross income from navy beans or average
price.

At some point all the programs ran into con­

strain ts that limited their ability to raise income or
price beyond a certain point.

Each program also would

have side effects on different groups in the industry
that must be considered by those using or giving the
power to use the programs.
The question as to whether or not non-govern­
mental control programs will be used depends on the
priorities and concerns of the public.

Government

expenditures to low income groups go to those that can
organize in such a way as to make their problem a
national political issue.

In the past, agriculture was

able to do this because it had a large percentage of
the low income people in the country.

The government

price support program came out of this type of setting,
increasing income to those farmers who produced signifi­
cant amounts.

In the past 20 or 30 years, however,

128

agriculture has lost many of its poor to the other areas
of the economy.

The concern of the country for low in-

come people, to a large extent, has turned away from
agriculture toward other areas, particularly the urban
ghettos.

If agriculture is to continue to receive pub­

lic help, it may have to be in the form of law changes
that would permit the operation of non-governmental
control programs similar to those examined in this
study.

If that happens, this study may provide some

useful information which can assist in the evaluation
of alternative programs.

In the interim, the econo­

metric model can be used for predictive and planning
purposes.

It can give estimates of the next yeax*s
It can also be us?;d to estimate navy

planted acreage.

bean price, domestic consumption and government take­
over in the current year.
Some final comments on needed research areas
may prove useful to those who may examine the navy bean
industry in the future.

Several areas of researah can

be suggested:
1.

A more complete study of the export market is
needed when data becomes available in order to
better determine what exports will be at dif­
ferent prices.

Each country which imports

beans should be studied and the factors in­
fluencing demand should be established.

©

A study of where and how the government dis­
poses of its takeover and the effects this
disposal has on the price and demand for navy
beans could be of value to both those people
working in the industry and those doing fur­
ther research on likely consequences of in­
dustry policy change, assuming the government
program continues.
Examination of the domestic supply and demand
situation in Canada might be useful to both
Canada and the United States.

Since the two

countries are the chief producers of the navy
beans grown in the world, an examination of
both U.S., Canadian, and world markets must
take the policies and supply of both countries
into account.

The effects of a market-sharing

duopoly operating under different marketing
policies might be simulated by combining a
Canadian model with the model used in this
study, and could provide some meaningful in­
sight into the effects of the recently insti­
tuted Canadian two-price plan, and the beat
competing policy for the United States.
An analysis of the opinions of producers and
other industry groups on the different price
and supply control programs may be needed

130
before any such programs would be feasible
in the industry.

A study of this type might

show what educational or extension programs
need to be undertaken to show the positive
benefits and costs that could be associated
with each program.
5.

A more detailed analysis of each program
could be done.

This could include examina­

tion of the needed administration each pro­
gram would require and the administrative
costs.

The exact legal and organizational

changes needed before implementation could
also be examined more closely.

o

LIST OF REFERENCES

LIST OF REFERENCES
Articles and Bulletins
Connor, Larry J. Costs and Returns for Major Cash Crops
in Southern Michigan. Michigan State University
Agricultural Economics Report No. 87, East Lan­
sing, Nov. 1967.
Department of Agricultural Economics, University of
Guelph. A Comparative Study of Agricultural
Marketing Legislation in Canada, Australia,
United Kingdom and the United States. Department
Publication No. A. E./64-65/11, Gue l p h , Ontario,
Nov. 1964.
Hathaway, Dale E. The Effects of the Price Support Pro­
gram on the Dry Bean Industry in Michigan.
Michi
gan Agricultural Experiment Station Technical
Bulletin 250, East Lansing, April 1955.
Hathaway, Dale E . , e t a l .
Michigan Farmers in the Mid
Sixties. Michigan State University Agricultural
Experiment Station Research Report 54, East
Lansing.
Hayenga, Marvin L.
Structure and Problems of the Navy
Bean Marketing Syatenu
Michigan sfeate University
Agricultural Economics Report 91, East Lansing,
April 1968.
Hedrick, Wilbur O. A Decade of Michigan Cooperative
Elevators. Michigan Agricultural Experiment
Station Special Bulletin 291, East Lansing, May,
1938.
_______ . Marketing Michigan B e a n s . Michigan Agricultural Experiment Station Special Bulletin 217,
East Lansing, Nov. 1931.
Helmberger, Peter G. and H o o s , Sidney.
Cooperative Bar­
gaining in Agriculture. Berkeley, Calif.0: Uni­
versity of California, Division o f “Agricultural
u Sciences, 1965.
131

132
Meinken, Kenneth W. The Demand and Price Structure for
Wheat. U.S. Department of Agriculture Technical
Bulletin No. 1136, Washington, D.C., Nov. 1965.
Perkin, G. F. Marketing Milestones in Ontario 1935-1960.
Toronto: Ontario Department of Agriculture, 1962.
Tyner, Fred H. and Tweeten, Luther G.
"Simulation as a
Method of Appraising Farm Programs." Journal of
Agricultural Economics, 50: 66-81, FeE"I 1968.
Vandenborre, Roger J. An Econometric Investigation of
the Impact of Governmental Support Programs on
the Production and Disappearance of Important
Varieties of Dry Edible Beans. Glannini Founda­
tion Research Report No. 294, California Agricul­
tural Experiment Station, Berkeley, Calif., Dec.
1967.
_______ . "Dynamic Impact Multipliers: A Case Study of
White Dry Edible Beans." Journal of Agricultural
Economics, 50: 311-327, May 19&8.
Wood, Arthur W.
"The Marketing Board Approach to Collec­
tive Bargaining." Journal of Farm Economics,
49: 1367-75, Dec. 19'67.--------------------Statistical Sources
Council of Economic Advisers. Economic Report of the
President. Washington, D.C.: Government Print­
ing Office, 1969.
Michigan Department of Agriculture. Michigan Agricul­
tural Statistics. Lansing, Various Issues.
United Nations.
Statistical Yearbook.
Various Issues.

New York,

U.S. Bureau of the Census.
1964 United States Census
of Agriculture. Washington, D.C.: Government
Printing Office, 1964.
U.S. Department of Agriculture.
Agricultural Prices.
Washington, D.C.: Government Printing office,
Various Issues.
_______ . Agricultural Statistics. Washington, D.C.:
Government Printing Office, Various Issues.

133
, and California Department of Agriculture.
Dry Beans:
California Market Summary. FederalState Market News Service, Sacramento, Calif.,
Various Issues.
________, Federal Extension Service.
Statistical Summary:
Dry Edible Beans and Dry Field P e a s . October 1964.
________, Foreign Agricultural Service.
World Agricul­
tural Production and Trade. Washington, D .C .,
Various Issues.
________, Statistical Reporting Service, Crop Reporting
Board.
Crop Production. Washington, D . C . ,
Various Annual Summaries.
U.S. Tariff Commission.
Tariff Schedules of the United
States Annotated (1969). T.C. Publication 272,
Washington, D . C . : Government Printing Office,
1968.
Other Sources
Fox, Karl A.
Yorks

Intermediate Economics Statistics.
John Wiley & S o n s , Inc., 1968.

New

Huff, H. Bruce.
"Marketing of Canadian Wheat:
An
Economic Analysis with Projections for 1975
and 1980."
Unpublished Ph.D. dissertation,
Michigan State University, 1969.
Johnston, J. Econometric Methods. New York:
Hill Book Company, Inc., 1*963.

McGraw-

Leftwich, Richard H. The Price System and Resource
Allocation. New York:
Holt, Rinehart and Win­
ston , 1966.
Michigan Bean Shippers Association.
Constitution and
By-Laws of the Michigan Bean Shippers Associa­
tion. Saginaw: Publication put out by the
Association, Saginaw, Mich., 1965.
Michigan Department of Agriculture Bean Commission.
Bean Commission L a w . Pamphlet put out by the
Commission on Act No. 114, Public Acts of 1965,
Lansing.

134
Ontario Bean Producers Marketing Board.
The Ontario
Bean Producers1 Marketing P lan. Pamphlet on'
Ontario Regulation 48/66 as amended by O. Reg.
142/66 and 0. Reg. 385/67 under the Farm Pro­
ducts Marketing Act, London, Ontario.
Schmid, A. Allan and Shaffer, James D.
"Marketing in
Social Perspective," Agricultural Market Analy­
sis. Edited by Vernon L. Sorenson.
East Lan­
sing, Michigan: MSU Business Studies, Graduate
School of Business, 1964.
U.S. Department of Agriculture, Agricultural Stabiliza­
tion and Conservation Service.
Compilation of
Statutes Relating to Soil Conservation iat a l .
Agriculture Handbook No. 242, January' 1, 1373.

APPENDICES

APPENDIX A
THE ONTARIO NAVY BEAN INDUSTRY1
In examining an industry and how it might be
changed to better fulfill its ends in a changing en­
vironment, the organization of similar industries can
provide information for comparisons and directions in
which to move the industry's structure.

The Ontario

navy bean industry is examined with this purpose in
mind, and also to provide background about the organi­
zational setting of the major U.S. competitor in the
export market.
There are several similarities between the Michi­
gan and Ontario navy bean industries.

Both produce al­

most all the navy beans produced in their respective
countries.

Second, production is concentrated in a

small area within the state or province.

In Ontario,

six southwestern counties produce virtually all the navy
beans in the province, as does the "Thumb" region in
Michigan.

Third, production in both state and province

1The information on the Ontario bean industry was
obtained through personal interview with Mr. Charles E.
Broadwell, Manager-Marketing Agent for the Ontario Bean
Producers' Marketing Board, July 1969.
135

136
is in small acreages, with the planting per grower
averaging around 50 acres in the United States and 30
acres in Canada.

Total production in both places has

also trended upward in recent years

(Tables 1-1 and

A— 1) .
The breakdown of sales between domestic and ex­
port markets also are similar.

Both countries are be­

coming more dependent on the export market
and II-2).

(Tables 1-2

The United Kingdom is the primary market

for both countries with Canada having a tariff prefer­
ence which gives it a 6% price advantage on its com2

petitors.

However, there are also several differences be ­
tween the two industries, particularly, in how their
institutions are set up.

A Bean Producers' Marketing

Board voted into existence by producers under the Farm
Products Marketing Act has covered navy and yellow eye
beans in Ontario since 1944.

3

Until 1968 the main

function of this Board was to have a negotiation com­
mittee of producers and dealers
mittee decided:
2

(shippers).

This com­

(1) when payment for beans delivered

This tariff preference will be reduced to 4
percent by January 1, 1971.
3
Under this Act, a local marketing board must have
a plan approved by the Farm Products Marketing Board.
The Farm Products Marketing Board has the right to es­
tablish, amend, and revoke plans of local boards for
control and regulation of the marketing of farm products.

137

TABLE A-l
ONTARIO PRODUCTION AND EXPORTS OP NAVY BEANS,
1958-■67

YEAR

PRODUCTION
1000 cwt.

EXPORTS
1000 cwt

EXPORTS AS A
% OF
PRODUCTION

1958

679

34

5.0

1959

633

58

9.1

1960

620

33

5.3

1961

744

127

17.0

1962

838

200

23.8

1963

863

218

25.2

1964

1117

397

35.5

1965

1182

597

48.9

1966

1396

669

47. 8

1967

821

300

36.5

Source:

Personal communication with the Ontario Bean
Producers' Marketing Board; London, Ontario,
May 2, 1969.

138
to a dealer was to be made;

(2) how disputes as to grade,

moisture content, or conditions of beans were to be
settled; and (3) what the maximum charge by any dealer
for grading and picking was to be.

In time, the agree­

ments between the producers and dealers were standard­
ized by contracts.

In 19 52, the producers started a

cooperative (owned by the Board) to market their own
4
beans.
Promotional efforts for beans was another
activity carried out by the Board.
Starting with the 1968 crop year, the Marketing
Board started operation under a new marketing plan.
This plan's main feature is that all pea and yellow eye
beans grown in Ontario must be marketed by or through
the Board.
In actual practice, after harvest starts, pro­
ducers deliver beans to the elevators of their choice,
at which time the elevator operator, as agent of the
Board, pays the producers an initial payment
in advance by the B o ard.)5

(decided

The Board then pays to the

dealer the money paid out by him on behalf of the Board.
Producers may store beans if they desire, but there is
4

The cooperative and the Board have since been
separated.
The cooperative is now treated like any other
dealer by the Board.
5

There are now 23 dealers operating in navy beans
in Ontario.

139
no advantage for a producer to store beans past a speci­
fied date

(March 16 in the first year of operation).

Once delivered to the dealer, the beans are owned
by the Board, which now sets the minimum prices at which
the beans can be sold.

In its first year of operation,

the Board has operated under a two price system (under
the old marketing plan, only the market price existed).
It sells beans in the domestic market at a higher price
than it does on the export market, the domestic market
being protected by an approximate $1.50 per cwt. tariff.6
This means the domestic price can be up to a $1.50 higher
than the export price.
Before the beginning of the marketing year, the
Board negotiates with the dealers to determine what
charges will be paid to the dealer for handling of the
B o a r d 1s b e a n s .

When the dealer sells the b e a n s , he pays

the Board the price set by the Board minus the agreed
upon handling charges.

The Board can change the price

it sets during the market year if it is necessary to
meet competition.
To make the initial payment to the grower, when
his beans are delivered to an elevator, the Board borrows
from a bank.

As the Board sells the beans, it pays off

6The domestic price was adjusted in Western Canada
to account for freight rate differences between the
United States and Canada.
The United States freight rate
being below the Canadian rate.

140
the loan and forms a pool with the rest of the money.
As money accumulates in this pool, it is paid out to
the growers, based on the quantity of beans they de­
livered to the Board.

After the entire crop has been

sold, all the money in the pool should have been, paid out
to the growers.

7

The growers are charged six cents per

cwt. to cover the administrative costs of the Board.
The Board has the right to restrict the quantity
of beans that a grower can market, but this provision
was not used in the first year of operation.

There are

no plans for this power to be used in the near future.
Another provision of the new marketing plan calls
for an advisory committee composed of a chairman and
eight members.

Three of the members are appointed by the

dealers, four are growers appointed by the Board and
three are appointed by the Ontario Food Processor's
Association.

The Chairman is appointed by the Farm

Products Marketing Board.

The function of this com­

mittee is to advise and make recommendations to the
Board, the dealers, or the Ontario Food Processers'
7

During its first year, the Board operated under
an agreement with the dealers that required that no
more than 5 percent of the 196 8 crop would be remaining
in storage at the end of the crop year.
If this agree­
ment continues, the board will not be able to carry pro­
duction over from a large crop year. This type of
agreement is not required by the Board's marketing
plan.

141
Association with respect to the promotion of greater
efficiency in production and marketing, and prevention
and correction of irregularities in marketing of.beans.

APPENDIX B
DATA USED IN STUDY BUT NOT PRESENTED IN THE TEXT

Year

1
PCa

2
PSWa

3
PRSWa

4
PSSWb

5
PERClf=

1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967

1.71
1.44
1.38
1.37
1.12
1.19
1.07
1.06
1.01
.97
.91
.99
1.06
1.09
1.08
1.24
.98

8.04
8.74
9.81
10.97
8.26
7.71
9.05
9.23
7.48
8.46
10.78
9.02
9.36
10.95
12.35
9.12
12.95

736
540
560
731
884
771
759
800
943
618
438
542
607
514
578
670
473

7.89
8.30
8.35
7.81
6.76
6.71
6.92
6.80
6.06
6.09
7.21
7.33
7.33
7.33
7.52
7.52
7.52

96
94
97
84
97
97
92
98
93
99
98
99
98
97
95
99
91

6
Dummya
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1

aVariable and units defined in Chapters III and V.
^Government price support level for small white beans
in dollars per cwt.
cPercentage of Michigan dry bean planted acreage that
was harvested.
Sources:
Col. 1: U.S. Department of Agriculture, Agricultural
Prices, Various Issues.
Col. 2,3, and 4: U.S. Department of Agriculture and
California Department of Agriculture, Dry Beansi
Cali­
fornia Market Summary (Federal-State Market News Service,
Sacramento, Annual Issues).
Col. 5: Michigan Department of Agriculture, Michigan
Agricultural Statistics (Lansing, Various Issues).

142

143
TABLE B—1 (Continued)

Year

7
PNBa

1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967

6 .51
7.43
7. 83
9.39
6.48
6.17
7.62
6 .44
5.46
5.46
6.00
6.07
6.15
6.61
8 .35
6.01
8.35

8
USPOPa
155
158
160
163
166
169
172
175
178
181
184
187
189
192
19 5
197
199

9
UKPOPa
50
51
51
51
51
51
52
52
52
53
53
53
54
54
55
55
55

10
BINVa

11 .
Yield®

12
AHNB®

400
144
490
74
50
150
244
25
30
363
160
32
338
139
384
499
415

1075
1065
1010
810
920
1065
740
985
1265
1185
1350
1315
1480
1250
960
1280
1070

352
301
339
370
472
461
442
502
475
490
497
509
511
541
569
569
447

dYield of navy beans in Michigan in pounds per acr e .
Michigan harvested navy bean acreage.
Obtained by
dividing Michigan production of navy beans by Michigan
yield of navy b e a n s .
Sources:
Col. 7: Records kept by Department of Agricultural
Economics/ Michigan State University.
Col. 8: Council of Economic Advisers, Economic
Report of the President (Washington, D . C . : Government
Printing Office, 1969), p. 251.
Col. 9i United Nations, Statistical Yearbook (New
York, Various Issues).
Col. 10: Personal communication with the U.S.
Department of Agriculture, Agricultural Stabilization
and Conservation Service, Oilseeds and Special Crops
Division, Washington, D.C.
Col. 11: Personal Communication with the Michigan
Crop Reporting Service, Lansing, Michigan.

144

TABLE B-l (Continued)

Year
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967

apnb£

14
GIBSa

15
CBPIa

365
319
350
441
489
477
482
514
511
497
508
516
519
556
600
577
493

49.03
48.50
47.32
48.42
43.49
45.84
42.99
43.65
44.28
50.09
55.69
58.73
57.23
65.58
62.85
66.14
62.44

10.98
11.44
11.15
10.72
9.90
10.17
10.75
11.10
11.73
16.05
13.84
10.87
11.12
12.52
13.41
20.60
17.91

13

16
CANEXa
16
156
516
150
55
49
82
34
58
33
127
200
218
397
579
669
300

17
DDNB?
2497
2556
2924
3130
3032
1970
3538
3999
4126
3570
4370
3751
4848
4405
4290
3653
4066

f
Navy bean planted acreage in thousand acres.
Obtained by dividing percentage of Michigan dry bean
planted acreage that was harvested (Col. 5) into Michi­
gan navy bean acreage harvested (Col. 12).
^Domestic demand of navy beans obtained by adding
U.S. production plus beginning inventory and then sub­
tracting U.S. exports and government takeover.
Sources:
Col. 14: U.S. Department of Agriculture, Agricul­
tural Statistics, Annual Issues.
Col. 15: Roger J. Vandenborre, An Econometric
Investigation of the Impact of Governmental Support Pro­
grams on the Production and Disappearance of Important
Varieties of Dry Edible Beans.
Col. 16: Years 1951-57, U.S. Department of Agri­
culture, Foreign Agriculture Service, World Agricultural
Production and Trade(Various Issues): Years 1958-67,
Personal communication with Ontario Bean Producers
Marketing Board, London, Ontario.