Projections o emand,
Supply a d Marketing Patterns
Potatoes in 1990

([2

for U. .
Stephen 0. Sparks

 

LIBRARY

Michigan State
University

 

MSU RETURNING MATERIALS:
Place in book drop to
, LIBRARJES remove this checkout from
"w your Y‘ECOY‘d. FINES wil]
be charged if book is
returned after the date
stamped be1ow.

 

 

 

PROJECTIONS OF DEMAND, SUPPLY AND MARKETING
PATTERNS FOR U.S. POTATOES IN 1980

by

Stephen 0. Sparks

 

Research Paper Prepared as Partial Fulfillment
of the Requirements for the Degree of

MASTER OF SCIENCE

Department of Agricultural Economics

Michigan State University

1973

L'BRAR‘ES
MICHIGAN SIAYE UNIVEQS‘TY
EP.SY LANSING. MCH 48624-1048

 

 

AC KNOWLEDG EMENTS

I would like to take this opportunity to express
my appreciation to my major professor, Dr. Kelly Harrison.
His criticisms and recommendations for improvements of
this research paper and guidance throughout my graduate
career were tremendously valuable. I would like to thank
the remaining members of my committee, Dr. Lester
Manderscheid, Dr. Marvin Hayenga, and Dr. John Allen, for
their recommendations for improvements of this research
paper and guidance during my Master's program. I would
also like to thank Dr. Lloyd Teigan for his helpful sug-
gestions in the early stages of this study.

I am most grateful for the generous financial and
professional support afforded me by the Departmrnt of
Agricultural Economics at Michigan State University, under
the direction of Dr. Dale Hathaway and Dr. Harold Riley.

I would like to express my appreciation to Mrs.
Laura Flanders for her assistance with the data processing
and programming aspects of this research report. Also, the
efforts of Mrs. Patti Stiffler and Mrs. Enid Maitland in
the preparation of earlier drafts is appreciated.

Finally, a special note of thanks is due to my
wife, Linda, who, among other things, typed the first

ii.

draft of this report and constantly reminded me of the

warm weather and sunshine in California.

iii.

TABLE OF CONTENTS

Page
ACKNOWLEDGEMENTS . . . . . . . . . . . . ii
LIST OF TABLES . . . . . . . . . . . . vi
CHAPTER
I. POTATO CONSUMPTION, PRODUCTION, AND
MARKETING . . . . . . . . . . . . 5
Introduction . . . . . . . . . . 5
Potato Consumption . . . . . . . . 5
Potato Production . . . . . . . . 8
Shifts in Potato Production . . . . . 11
Location of Processors . .. . . . . . 13
Marketing Channels . . . . . . . . l6
Movements of Raw Potatoes in the United
States. . . . . . . . . . . . 23
The Eastern Region . . . . . . .. 26
The Southern Region . . . . . . . 30
The Midwestern Region . . . . . . 35
The Western Region . . . . . . . 38
II. U. S. DEMAND-SUPPLY PROJECTIONS TO 1980 . . 42
Introduction . . . . . . . . . . 42
The Model Structure . . . . . . . . 43
The Acreage Equation . . . . . . . 43
The Yield Equation . . . . . . . 47
The Marketing Margin Equation . . . . 48
The Retail Price Equation . . . . . 50
The Identity Equations . . . . . . S3
Estimation Results . . . . . . . . 53
The Acreage Equation . . . . . . . 53
The Yield Equation . . . . . . 56
The Marketing Margin Equation . . . . 57
The Retail Price Equation . . . . . 59
The Projection Procedure . . . . . . 63
Results and Interpretation . . . . . 67
Potato Consumption and Utilization . . . 73
Procedure for Projecting Potato Consurption
and Utilization . . . . . . . . 82
Results and Interpretation . . . . . 83

iv.

CHAPTER Page

IV. U. S. PRODUCTION PROJECTIONS BY STATE

AND SEASON TO 1980 . . . . . . . . 89
Introduction . . . . . . . . . 89

Procedure . . . . . . . . . . 89

Results . . . . . . . . . . . 91
Interpretation of Results . . . . . 103

V. SUMMARY AND CONCLUSIONS . . . . . . 123
BIBLIOGRAPHY . . . . . . . . . . . . . 131

Table

10.

11.

12.

LIST OF TABLES

Per Capita Consumption of Fresh Potatoes and
Processed Potato Products, 1950, 1960,
1.970. O O 0 O O O O I O O O 0

Total Potato Production by State for 1970 .

Potato Processing Plant Locations in the
United States, 1972. . . . . .

National Listing of the Forty—One Cities .

Flows of Raw Potatoes from Major Producing
States by Months by Carlot to the
Eastern Region in 1971. . . . . . .

Flows of Raw Potatoes from Major Producing
States by Months by Carlot to the
Southern Region in 1971. . . . . .

Flows of Fresh Potatoes from Major Producing
States by Months to the Midwestern Region
in 1971

Flows of Fresh Potatoes from Major Producing
States by Months to the Western Region
in 1971. . . . . . . . . . . .

Demand-Supply Projections to 1980 . . .

‘Per Capita Consumption of Fresh and Pro~

cessed Potato Products in the United
States, 1950, 1960 and 1970 . . . . .

Potato Utilization, 1950, 1960 and 1970. .
Per capita Consumption of Fresh and Pro-
cessed Potato Products in 1970 and

Projected Per Capita Consumption of Fresh
and Processed Potato Products in 1980 .

vi.

10

14

25

27

32

36

4O

68

74

75

85

Table

13.

14.

15.

16.

17.

18.

19.

20.

Potato Utilization in 1970 and Projected
Potato Utilization in 1980 . . . . .

Total Potato Production in 1970 and Pro-
jected Total Potato Production in 1980
by Major Producing State . . . . .

Winter Potato Production in 1970 and Pro-
jected Winter Potato Production in 1980
in Florida and California . . . . .

Early Spring Potato Production in 1970
and Projected Early Spring Potato Pro-
duction in 1980 in Florida . . . . .

Late Spring Potato Production in 1970 and
Projected Late Spring Potato Production
in 1980 by Major Producing States . .

Early Summer Potato Production in 1970 and
Projected Early Summer Potato Production
in 1980 in Texas, Virginia, and
California . . . . .',. .

Late Summer Potato Production in 1970 and
Projected Late Summer Potato Production
in 1980 by Major Producing States. .

Fall Potato Production in 1970 and Fall

Potato Production in 1980 by Major
Producing States . . . . . . . .

vii.

Page

86

93

96

96

97

98

98

CHAPTER I

INTRODUCTION

Potatoes, like an increasing number of fruits and
vegetables, have gone through an evolutionary process which
has changed the form in which the product is marketed.
Since 1950 there has been a dramatic increase in the volume
of potatoes marketed in processed form. The resulting
changes in marketing patterns have had a tremendous impact

on potato production and marketing arrangements.

Purpose

The study undertakes one part of a research pro-
ject which is seeking to determine those changes in the
potato marketing systems which would increase total revenue
‘in the Michigan potato industry. This research report lays
the foundation for the entire research project by examining
past, present and projected future relationships in the
total United States potato industry. The information in
this study provides the background necessary for a full
understanding of the important market relationships in the
potato industry. This complete understanding is necessary
if the impact of policy proposals is to be adequately

analyzed.

Objectives

 

To provide the background necessary for a thorough

understanding of the potato industry, several objectives

must be met.

1.

The objectives of this research report are:
Identify changes in potato consumption patterns
over time, discuss the factors which may have
led to these shifts and project these consump-
tion patterns to 1980.

Identify major potato producing states in the
United States, identify the shifts in location
and seasonability of potato production, and
discuss the factors which may have led to
these shifts.

Identify the location of processors of potato
products and discuss the possible rationale
behind their location pattern.

Identify and describe the movements of raw
potatoes in the United States.

Identify and describe the marketing channels
for potatoes.

Project United States demand-supply relation—
ships to 1980.

Identify the amount and shift of potato utili-
zation patterns over time and project these

utilization patterns to 1980.

8. Project the production of potatoes by state
and season in 1980 and discuss the factors

which may lead to these projected shifts.

Procedure

 

To identify and examine past and present relation-
ships in the potato industry, secondary data are collected
for a sample period of twenty years from various U. S.
Department of Agriculture, university, and potato industry
publications. ,The data dealing with potato consumption,
production and utilization are assimilated so the shifts
which have occurred during the past two decades could be
readily identified. A11 collected data are analyzed
using U. S. Department of Agriculture, university and
industry publications and knowledge of the market.

The projection of United States demand—supply
relationship makes use of the same secondary data sources
for a sample period of 1949—1970. A model of the demand-
supply relationship is developed using statistical and
economic theory. This recursive model has four behavioral
equations and two identities. The structural relationship
of the behavioral equations in the model is estimated by
the ordinary least squares estimation procedure. The
estimated regression coefficients of the variables in the
behavioral equations are used in conjunction with the value

of projected predetermined variables, the value of

variables determined within the recursive model and the
identity equations to determine projected demand-supply
relationships.

The projections of consumption, utilization and
production by state and season in 1980 also make use of
these sources of secondary data. The sample period for
the projections of consumption and utilization is 1950-
1970, while the sample period used to project potato pro-
duction by state and season is 1955-1972. The projections
for consumption, utilization and production are determined

by extrapolation of the trends of the time series data.

CHAPTER II

POTATO CONSUMPTION, PRODUCTION, AND
MARKETING

Introduction

 

The evolutionary process which has changed the
form in which the potato product marketed has had a
major impact on potato consumption, production and
marketing patterns. This chapter will identify these
shifts and discuss the factors which may have led to
these shifts. Also, the movements of raw potatoes will

be analyzed.

Potato Consumption

 

The evolution in potato processing was a major
factor in the reversal of the potato consumption pattern.
Through the thirties and forties per capita potato con—
sumption decreased while total potato consumption remained
constant due to population growth. The majority of the
potatoes consumed were fresh, although there was a small
amount of potatoes consumed as chips, starch and flour.

In the late forties and through the fifties the potato
processing evolution began to take place. Processors
found that potatoes could be marketed successfully in

processed forms. Processors found demand increasing for

frozen potato products (primarily french fries) and dehy—
drated potato products.

The acceptance of processed potato products had a
major impact on the potato utilization patterns in the
sixties. Where the fifties were the beginning period
for the potato processing industry, the sixties were the
period of rapid growth. The three major forms of pro-

cessed potatoes grew substantially during the sixties.

TABLE 1.--Per Capita Consumption of Fresh Potatoes and
Processed Potato Products, 1950, 1960, 1970.

.__..__ .—-—4-———.——.——.—.__._—‘. .. .. -.
_--...'—-- ~....~-— .—_-. .__ - .-_-___._

 

 

ggdaPrgi:::ed FrGSh Eggiissed
1950 106.3 100.0 6.3
1960 108.4 84.7 23.7
1970 118.6 59.6 59.0

 

Source: American Potato Yearbook, 1951, 1961, and 1971.

 

Table 1 shows the reversal of the downward trend
in total per capita potato consumption during the fifties.
This reversal of the fifties is due to the increase in con-
sumption of processed potato products. From 1950 to 1960
the increase in per capita consumption of processed potato
products was larger than the decrease in per capita con-
sumption of fresh potato products. Most of this increase

occurred in the late fifties when consumers became aware

6

of the processed product. This set the stage for the
sixties, when the consumption of processed potato products
increased by 35.3 pounds per person. During this same
time span, the annual per capita consumption of fresh
potatoes decreased by 25.1 pounds. Thus, there was an
increase in the per capita consumption of potatoes.

This rapid rise in the consumption of processed
potato products was the result of a combination of rising
incomes and changing tastes. Consumer incomes had been
rising since the depression. Prior to the fifties, fresh
potatoes were an inferior good because as consumer incomes
increased the consumption of fresh potatoes decreased.
This was a major reason for the decreasing per capita
consumption of potatoes during this period. However,
during the fifties, as per capita incomes continued to
rise, the consumer demanded more leisure time. This led
to more eating out and use of convenience foods. Also,
the rise in consumer involvement in activities outside
the home led to the rapid growth in fast food outlets,
restaurants, and other institutions. These outlets found
that it was much more economical to purchase potato pro-
ducts in processed form than to purchase fresh potatoes
and perform whatever operation was necessary themselves.
Rising consumer incomes also produced greater demand for
convenience foods which require less preparation time in

the kitchen. With a rising number of housewives holding

jobs, processors of dehydrated potatoes and frozen potato
products had another segment of the market which they could
supply. The increase in consumption of snack foods was
another taste factor of particular importance to potato
chippers. These changes which deve10ped in the fifties

had a major impact on the market for fresh potatoes. Per
capita consumption might have continued to decline through
the fifties and sixties without the occurrence of all these
factors simultaneously, but the decrease probably would

not have been as rapid as it was in the late sixties.

These factors have combined to speed the decline in per
capita consumption of fresh potatoes and hasten the increase

in per capita consumption of processed potato products.-

Potato Production

 

Potatoes have shifted from a secondary crOp, with
the amount of acres planted a function of the number of
acres allocated to the major crOp, to a highly specialized,
primary crop in many areas of the United States. This
evolution has led to a shifting of potato production to
areas of the United States which are more suited to large
scale production techniques.1

Although potatoes are produced commercially in all

fifty states, there are nineteen states which accounted

 

lWill M. Simmons, "An Economic Study of the U. S.
Potato Industry," Agricultural Economics Report No. 6
(Washington, D.C.: U. S. Department of Agriculture, March,
1962) p.6.

for 94.1 percent of total potato production in 1970 (See
Table 2). Idaho is the major potato producer. Idaho
produces more than twice as many potatoes as Maine, the
second largest producer, and Idaho produced 23.0 percent
of the total U. S. potato production in 1970. The third
largest producing state, Washington, is rapidly closing in
on Maine, which has remained relatively stable over the
last 15 years. The three states located in the Northwestern
corner of the United States, Idaho, Washington, and Oregon,
rank first, third, and seventh respectively in production
and produced 38.0 percent of the total United States crop
in 1970. If California is included in the grouping, these
Western states produced 47.2 percent of the total U. S.
potato production. Although its potato production has
trended downward, California is a major producer of pota-
toes in all producing seasons except early spring.

The major producing states in the Midwest are
North Dakota, Minnesota, Wisconsin and Michigan. The
majority of potatoes produced in North Dakota and Minnesota
are produced in the Red River Valley. These four Mid-
western states accounted for 16.7 percent of the total
United States production in 1970. Production in these
states has been on the rise, although not as rapidly as the
Northwestern states.

Besides Maine, New York and Pennsylvania are the

only major producers of fall potatoes in the Eastern region

10

TABLE 2.--Total Potato Production By State for 1970.

 

 

----_—--..__..-. __.-__..-- _._ -_.__..—_.-_.._
. —-—_-_ _ -__. -‘ ._.... .—

———_. “—~ .__._ _ ._
.‘_‘_ ___._.._, ._..——. <.—- .- .— - - -_.___-

Production Percent of
State (1000 cwt.) Rank Total production

 

Alabama 2,127 18 0.7
Arizona 2,712 17 0.8
California 29,760 4 9.2
Colorado 12,916 10 4.0
Florida . 5,936 13 1.8
Idaho 74,660 1 23.0
Maine 35,699 2 11.0
Michigan 10,550 11 3.2
Minnesota 13,380 8 4.1
New Jersey 3,380 16 1.0
New York 16,977 6 5.2
North Carolina 2,093 19 0.6
North Dakota 17,550 5 5.4
Oregon 15,229 7 4.7
Pennsylvania 8,280 12 ~2.5
Texas 4,593 14 1.4
Washington 33,590 3 10.3
Wisconsin 13,038 9 4.0
Virginia 3,990 15 1.2
All Other States 18,340 5.9
Total Production 324,801 100.0

 

Source: U.S.D.A., Statistical Reporting Service, Crop
Reporting Board, Crop Production, 1972.

 

ll

of the United States. New York was sixth and Pennsylvania
was twelvth in total potato production in 1970, but

potato production has been trending downward in both states.
The other major potato producing states in the Eastern and
Southern regions of the United States are major producers
of seasonal potato crops. New Jersey is a major producer
of late summer potatoes. North Carolina and Alabama are
major suppliers of late spring potatoes. Texas and Florida
produce early spring potatoes, while Florida is a major
producer of winter potatoes. Potato production in the
areas which produce early season crops is relatively small

and has been trending downward in New Jersey and Florida.

Shifts in Potato Production

 

The majority of the potato production was located
near the major centers of population early in the 29th
century for a variety of reasons. The technology had not
yet been developed which would retard the spoilage of the
fresh product, so production had to occur near major
markets. Because transportation costs were high relative
to value of potatoes, production was not feasible in more
distant areas. The adaptability of potatoes to most all
soils is another factor which led to potato production near
major markets. The fact that there wasn't a highly
developed processing industry also led to this concentra—

tion. With most major markets located in the Eastern

12

region of the United States, the majority of potato pro-
duction was located in the Eastern region.2 Many of the
early potato chippers, operating before the evolution in
potato processing during the fifties, located near the
centers of population in the East which encouraged potato
production.

As transportation improved and the potato pro-
cessing industry developed, production moved into the more
efficient producing areas. These producing areas were
well adapted to the large scale, highly mechanized pro-
duction techniques (such as mechanical harvesting) which
are necessary for efficient production of potatoes. These
new areas of production were well adapted to irrigation,
which has been an important factor in increasing yields
of potatoes.3 The increasing value of land in producing
states such as California, New York, Florida, New Jersey
and Pennsylvania, and the competition for acreage by other
crops in Michigan, California, Florida and North Carolina,
has led to concentration of production in areas where land
values are relatively low and there are few competitive
crops. Storage has also played an important part in the
shift of production westward. As processing became more

prevalent, potatoes became storable for the entire year.

 

21bidol pp06-7o

3Ibid., pp.6-7

13

Also, as consumption of processed potato products increased,
potato processors established their plants in the Northwest
near the source of low cost raw materials.4 This had a
multiplier effect in the expansion of potato production on
the major Western potato producing states. Storage has
also had an impact on the fresh market. Fresh potatoes are
now available from the fall crop almost all year long
because of improved storage. This has reduced the need

for the different seasonal potato crops. The relation of
the better storage and higher land values in major pro-
ducing states of seasonal crops has led to the downward

trend of production in these states.

Location of Processors

 

Potato producers are located in every state. The
majority of the potato processors are located in the
major potato producing states (Table 3). As expected,
Idaho has the largest number of processors. Approximately
40 percent of all potato freezers are located in Idaho,
while over half of all dehydrators are located in the state.
Idaho also is the location of the largest number of starch
and flour processors, approximately 40 percent. Maine and
Washington are tied for the second largest number of
potato freezers, with seven freezers located in each state.

The ten starch and flour processors located in Maine are

 

41bid., pp.6-7.

14

TABLE 3.--Potato Processing Plant Locations in the
United States, 1972.

 

Dehy— Starch
State Free— dra- Chip— Can- and Total
zers tors pers ners Flour
.Alabama -- -- 1 1 -- 2
Arizona —- -- 3 -- -- 3
California 1 -- 13 12 l 27
Colorado -- —- 7 1 -— 8
Florida -— -- 5 1 -- 6
Idaho l9 l8 4 -- 12 53
Maine 7 l 4 f— 10 22
Michigan 1 l 15 4 —- 21
Minnesota 2 2 5 2 2 13
New Jersey 2 -- ,l 3 -- 6
New Yerk 2 2 l3 5 1 23
North Carolina —- - 3 3 l 6
North Dakota 1 4 1 —- 2 8
Oregon 4 l 3 -- -- 8
Pennsylvania 2] -- 25 3 -- 30
Texas -- -— 18 2 -- 20
hshington 7 4 5 1 -- 17
Wisconsin 1 —- 8 7 -- 16
Virginia -- -~ 6 4 -- 10
Total 49 33 140 49 30

 

Source: U.S. Department of Agriculture, Potato Marketing
Report. (September, 1972); G. H. Sullivan, The
Potato Industry, Purdue Research Bulletin No.862,
(1968).

 

 

15

33 percent of the total number of U. S. starch and flour
processors.

The Northwestern corner of the United States is
the location for a majority of the processors of frozen
potato products and processors of dehydrated potatoes. If
the one potato freezing plant in California, which is
located near the Oregon border is counted in the North-
western total, 31 of the 49 potato freezers, or 60 percent,
are located in this area.

An even larger percentage of potato dehydrating
plants are located in the Northwest, approximately 66
percent. The large amount of relatively cheap labor and
the production of a good, high yielding processing potato
have induced processors to locate in this area.

Potato chippers and potato canners show a somewhat
different geographical dispersion than potato freezers,
potato dehydrators and potato starch and flour processors.
The majority of the potato chipping plants are located in
the potato producing states with large population centers.
Although potato chippers might like to make use of potatoes
with high total solids content and reduce costs by using
relatively cheap labor by locating in the Northwest,
transportation costs make this prohibitive.5 Potato

chippers must locate near the final consumers to reduce

 

5G. H. Sullivan, "The Potato Industry," Purdue

Research Bulletin No. 862 (1968) p. 16.

16

distribution costs as much as possible. This is illus-
trated by the fact that states with ten or more potato
chippers, California, New York, Michigan, Pennsylvania, and
Texas, all have or are located near large population
centers. Canners of potatoes are located throughout the
United States wherever canners of vegetable products are
located. The majority of the potato canners are located

in California.

Marketing Channels

 

The shift to a larger percentage of the potato
crop being processed has had an impact on the marketing
channels available to the primary producer. Prior to the
fifties, the producer had one major outlet for his pota-
toes, the fresh market. Small amounts of potatoes were
going to potato chippers, while cull potatoes were processed
into starch and flour. A majority of these processors
were located in the Eastern region of the United States.
With the evolution of the potato processing industry,
the spectrum of possible marketing channels for potato
producers was significantly broadened, as shown in Chart 1.
This is especially true in the Northwestern portion of the
United States where a large number of the various types
of potato processors have located.

The discussion of the marketing channel begins with

the primary producer. Potatoes produced for processing

l7

 

mmmzomzoo Til

 

 

.mawcemfiu oefluoxuez opened we upmzu 30H

 

 

mumpsuomwscmz
woom

 

   

 

mummdflno Q

 

 

 

 

 

Ammnoum
poom

use

:80 r

lsuflumcHe

mumaflmumm

 

 

 

 

 

mumammmaosz

 

 

 

 

 

mHOmmmooum

weapoum cmmoum

 

 

muoumupmnmo

a--.» “some

 

 

 

 

 

MMOmmOUOHm

sonmum
one uson

mmMUDaomm

 

 

 

 

 

 

 

umxommmm if

 

 

 

 

fl

 

 

 

when:
HmfluumspoH

 

mummmflzm

 

Mmdszm

 

 

 

\

l8

purposes are taken directly from the field to the processing
plant. The grower performs no other function except that '
of the producer. But, potatoes which go to the fresh

market may be another story. Many of the smaller growers
will send potatoes to shippers directly from the field.
Shippers consolidate the production of these smaller growers.
Shippers are usually located in or very near the production
areas. Shippers perform the sorting function, segregating
potatoes by size and quality. The next member of the
channel, repackers, receive these sorted potatoes.

Repackers are primarily located in or near major consuming
markets. Repackers package potatoes for sale at retail

food stores and package potatoes for use by institutional
food establishments. From the repackers, potatoes may

go to wholesalers or directly to retailers.

In reality it is doubtful that any fresh potatoes
are actually handled by all the various members of the
market channel. Although the number of handlers increases
as distance to the market increases, as growers have become
larger it has become more feasible for producers to per-
form other functions in the marketing channel. An
increasing amount of growers of potatoes are taking over
the function of the shipper. These growers are shipping
sorted potatoes to repackers. Most of the larger growers
are even by-passing the repacker, packaging the potatoes

themselves and shipping the packaged potatoes to wholesalers

19

or directly to retailers. Some of the larger shippers have
also taken on the packaging function. The shipment of
fresh potatoes has been enhanced by the rise in the use of
contracts by retailers, primarily large supermarket chains.
The contract may be desirable from the primary producers
standpoint because he knows exactly what price he will
receive for his crop before it is planted. The contract
may be desirable from the retailers standpoint because he
knows exactly what the potatoes will cost him and because
he has a specified supply. The contract may be desirable
for both parties because the retailer and the producer can
achieve a working relationship which will be to the advan-
tage of both parties. At the present time a small
percentage of the potatoes going to the fresh market are
contracted, but because of the advantages the amount con—
tracted is increasing.

Dehydrators of potatoes make greater use of con-
tracts than do retailers of fresh potatoes. Evidence
suggests that dehydrators contract directly with the
grower for about half of the raw potatoes needed for pro-
cessing. Although dehydrators like to process potatoes
with the high total solids content, when there are not
enough of these potatoes available or their price is too

high, cull potatoes will be processed.6 If cull potatoes

 

61bid., p. 17.

20

are used, they are supplied by growers and shippers of
fresh potatoes. After the potatoes have been processed,

a large portion of the product goes to wholesalers or
retailers. This movement may be facilitated by a food
broker. The food broker acts on behalf of many different
companies in the same role played by a salesman. A com-
pany will use a food broker if it isn't large enough to
warrant its own sales force. Dehydrators have another
option available. They can ship the processed potato pro-
duct to food manufacturers. These food manufacturers use
the dehydrated potatoes in other food products. For
example, dehydrated potatoes may be used by the makers of
frozen dinners or they may be processed into products
which are similar to potato chips. The final processed
product from the food manufacturers will also go to either
wholesalers or retailers. Food brokers could also be
involved in these movements.

Frozen potato product processors make even greater
use of contracts for raw potatoes than do dehydrators.
Somewhat over 50'percent of all potatoes processed into
frozen products are obtained via the contract. The pri—
mary processed frozen product is french fries. The best
french fries come from U. S. number one, two inch minimum

potatoes which are high in total solids.7 The demand for

 

7Paul W. Wilkes, Procurement Manager for Ora-Ida
Foods, Inc. A personal interview at Greenfield, Michigan,
September, 1971.

21

a high quality potato product accounts for the large amount
of contracting. Other frozen potato products which are
processed from potatoes not meeting the standards for french
fries are tater tots, hash browns, and other special pro-
ducts. After processing, frozen potato products can go to
wholesalers or retailers of food manufacturers. Movements
of frozen potato products from potato processors and food
manufacturers to wholesalers or retailers many times
involve food brokers.

Of the three major processors of potato products,
potato chippers contract with the grower for the raw pota-
toes the least, somewhat less than 50 percent. Unlike
frozen potato processors and potato dehydrators, potato
chippers are located in almost all of the major potato pr0*
ducing areas throughout the United States. In many of
these areas, the potato chipper is the only alternative
to the fresh market. Also, many potato chippers are
relatively small and do not use large quantities of raw
potatoes. Thus, the potato chippers often can get the
quality of potato necessary to make the best potato chips
without the large use of contracts. Potato chippers will
be forced to increase contracting in the future because of
the decline in the production in many producing areas.
Potato chippers ship the processed product to wholesalers
and retailers. Again, food brokers may facilitate this

movement.

22

The processors of potatoes for flour and starch
deal primarily with cull potatoes. Flour and starch pro-
cessors have decreased in importance with the rapid
evolution of the other potato processors, the production
of more high quality potatoes, and the ability to store
potatoes longer. The supply of potatoes for starch and
flour processors are the excess supply, cull potatoes
of growers and cull potatoes of fresh potato shippers.
Potato flour and starch is shipped to industrial users.

The final two members of the market channel are
wholesalers and retailers. Wholesalers link potato pro-
cessors and fresh potato distributors with retailers.
There are many types of wholesalers in existence today.
Independent wholesalers serve as a warehouse and a dis-
tributor for small retail food stores without a warehouse
system. As these small retail food stores disappear,
independent wholesalers will become less important. Another
type is the cooperative wholesaler. This wholesaler ser-
vices a number of independent retail food stores which
are members of the cooperative organization. Many of the
large retail food store chains have integrated backward in
the market channel and perform the wholesaling function.

Most potato products go through retailers to reach
the final consumer. Consumers can obtain potato products
from two types of retailers, institutional and food stores.

Institutional retailers prepare and sell potato products

23

in the ready to eat form. Institutional retailers are
establishments such as restaurants, hotels, hospitals,
cafeterias, cafes, military and other government agencies,
caterers, and fast food outlets. Retail food stores may
sell potato products in the ready to eat form, such as
potato chips, but the potato products are not prepared

by the retail food stores.

Movements of Raw Potatoes in
the United States

 

 

To analyze the flow of potatoes, the United States
is broken into four major consumption regions, the Eastern
region, the Southern region, the Midwestern region, and
the Western region. These regions are developed directly
from the flow data available in the Fresh Fruit and
Vegetable Unloads for 41 Cities reports. These reports
give the flow of fresh potatoes,in rail and truck unloads,
from the state of production to the city of consumption.
Regional consumption data is tabulated by consolidating
the carlot or truckload shipments from a particular potato
producing state to all the reporting cities within that
region. 80, flows to the four regions are determined by
the cities within that region which are included within
the unloads reports. These unload reports are not all—
inclusive. Many large cities, such as Jacksonville,
Florida; Las Vegas, Nevada; and Lansing, Michigan; are not

included. The consumption of potatoes from these areas

24

is omitted. Table 4 lists the reporting cities within
each particular region.

Before the flows of potatoes to a particular region
are discussed, some of the limitations of the data used to
determine these flows should be pointed out. First, the
unload reports deal with raw potatoes. The data may not'
represent the actual flow of potatoes from one producing
state to a consumption region. A potato producing state
which has a large processing industry may not bc totally
represented by these reports. Second, there is no way of
telling if the potatoes supplied are for fresh consumption
entirely or if some of the unloads are processed. For
example, unloads of potatoes in Detroit may go to the
Frito-Lay plant which in turn distributes the processed
potato product to the markets which the Detroit plant
serves. The third limitation to the use of the data deals
with the size of a carlot. For potatoes, the unloads
report assumes that one carlot equals 50,000 pounds. But,
the actual shipping weight per carlot for several states
is heavier than 50,000 pounds. So, an exact comparison of
carlot shipments to a particular consumption region from
different producing states can not be made. Because of
this variability of carlot weights, the data is left in
unloads per consumption region. If the data were con—

verted, the figures in many cases would be understatements

25

 

.Hnma .mmfleflo atouumm he mmmczcu

.coflmflbwa mflflmuemm> Use uflSpm .ooflapmm mswpexx

m. 4.-....~ wrx
(IA .
J.

’4
1

umu¢> tam begun xmomu .gmcmum mBoz uexufiz
umamncou .epuuamcqnmd we usaeuhmmmo .m .D "condom

 

A. .mmmz .9603. 6ch
.cmms .mfiuuamm

A.mo .ncwaxmo .ocHV
.wo .oowwocmum new

.ao .mnwo mama “Ham

. .HO a UGMHPHOQ

.mu .mmammc4.mou

.OHOU .Hm>cmo

.oz .mwaog .um

. Suez . wflaonwmocdfia

. mum...» . wmxsméfiflz
.hm .mHHH>mHSQH

.02 .muflo mmmcmm

.ch .mflHOQMCMflch

.az .ufiouuoe
.o .pcmfim>eau
.o .flmfinofio

HHH .ommoaao

.xoe toflcouq4 new
.afluo .muflo «earmaxo
.mq .mcwwHHO 3mz
.ccee .oflaflbxmmz
.mfim .flsaaz

.ccmH .mflxmemz
.xoe .coumsom
.xme .ruuos .um
.xme .mmdfimo

.U.m .mflnsnaou
.maa .Emamcaeuwm

.mo .muaaapa

.o.o .coumcflnmmz
.H.m .ooqufiboum
.mm .auanmpufla

.m. .maaaammaaaau
A.m.z .xumzmz .och
c» .z .038. 3%

.w.z .onwmam

.mmmz .couwom

.ez .muosauamm

.m.z rmcmnaa

 

conmm cumumma

:ofimmm chmpmwzpflz

coflmmm cumnunom

soflmmm cumummm

 

.mmflufiu accumuuom ecu mo mcfluwflq Hmcoflmmm us.v mqmde

26

or overstatements of the actual amount of potatoes shipped
from a producing state to a consumption region. Finally,
the Fresh Fruit and Vegetable Unloads reports only count
the number of unloads in 41 cities in the United States.
Even though these 41 cities are the major consuming cities
in the United States, there are a number of important
cities excluded which could effect the conclusions about
supplies by a particular potato producing state to a par-
ticular consumption region. For example, if unloads in
Richmond, Virginia, or the Norfolk—Hampton—Newport News,
Virginia, area were reported, the potato producing states
of Virginia and North Carolina would have a greater impact
on the Eastern consumption region. There are other cities
excluded which could increase the amount of potatoes
supplied by a particular state to a particular consumption

region.

The Eastern Region

 

Keeping the limitations of the data in mind, the
flows of raw potatoes, from the major producing states to
the four major consumption regions can be analyzed. The
first consumption region to be analyzed is the Eastern
region. The 1971 total and monthly flows of raw potatoes
to the Eastern region is shown in Table 5. Maine is the
largest supplier of raw potatoes with more than twice as

many unloads as any other major potato producing state.

27

 

.23 .838 Emummm fi 38?: m33m8> new page smears. £83m 2&2 8st

 

 

.coflmflbflo magnummm>.pcm pflsum .mofl>han dawumxumz nwenmcou .muduasoflwmm mo unmeunmmmo .m .D "meadow
mfio.am memm nmem mmom NNHN ommm vomm vmmm nNOm mcwm mNHm omem nemm Hence
Amuwuwaw
.mmxmB .mwOwoaaez
.mueaaareaam
.comouo .cmmflnoezv
mwo ow Hm mm mom mma mm mm mm Hague
mum mm moa oma mm mm Gamecomflz
mwoa moH mma Hmm mam mom ma. NH ma Ha wmmuom zez
0mm How omv mam cowmcarmma
oom mOH Hm mafiaoumo sumo:
one MHH mam oom em” mfimcfim
owe me me mm .2 2 .2 am S S S mafiflsmfima
$3 8m 8: N3 massed».
momm ma 0H Hm mm 5mm mHvH whoa ma weSMOMHHmu
mmwm mom Ham vow mmm mow on an «ma mma mmm Hue mam xuow 3wz
vnmv «me New wmm HH mm mme mom Hnm mom mom 0mm oampH
nmmaa maoa maoa wem an mm mmoa voma mmmm mmma mmma hem mafia:
H38. 08 >02 ”Bo flaw age has was was “ma um: new new 8.3m

 

.4\ma ea seamen cueummu map on

28

Besides being the second largest producer of potatoes,
Maine has a geographical advantage in the Eastern region.
Maine is the largest or very close to the largest unloader
of raw potatoes in all the reporting Eastern cities, except
Buffalo, which is located nearer to a major potato producing
region in New York. Producing primarily fall potatoes,
Maine does not supply the Eastern region the year round.
Unloads from Maine begin to occur in September and October.
The major impact on the market of the fall crOp is from
November through June.

The second largest supplier of potatoes to the
Eastern region is New York. New York, like Maine, enjoys
a geographical advantage in the Eastern region. New York
supplies all the reporting cities except Philadelphia and
Providence, with New York City being the largest market
outlet. The seasonal supply pattern of New York is some-
what different from the seasonal supply pattern of Maine.
New York is a major producer of late summer potatoes as.
well as fall potatoes. So, New York unloads a large number
of raw potatoes in August, September and October, before
the fall crop has had much impact on the market. The
supplies of New York potatoes begin to decrease in February
and continue to decline until June.

Unloads of raw potatoes from Idaho are an inter-
regional flow of potatoes or a flow of potatoes into

the Eastern region from the Western region. Although

29

the number of unloads of raw potatoes from Idaho is smaller
than the number of unloads from New York, the actual amount
of raw potatoes supplied by Idaho may be greater than the
actual amount supplied by New York, if the actual carlot
weights were measured.

Idaho supplies all of the reporting Eastern cities,
but the majority of the unloads are in New York City and
Philadelphia. The seasonal unload pattern of Idaho res-
embles the seasonal unload pattern of Maine. Unloads of
the fall crop begin in September and October. The number
of unloads increases to a peak in March and then declines.

There are several states which supply the Eastern
region with seasonal production. The major unloader of
seasonal potatoes is California. Although the major sup-
ply of raw potatoes by California is during only three
months, June, July and August, California is still the
fourth largest supplier in the Eastern region. California
is the major supplier in these months to all the reporting
Eastern cities, except Baltimore and Washington. The
largest amount of California potatoes are unloaded in New
York City and Boston. Virginia is another of these
seasonal suppliers of the Eastern region in the summer
months, with the majority of the unloads occurring in July.
Virginia, like California, is taking advantage of the small

number of unloads through these months by the largest three

suppliers of this region. Unloads of raw potatoes from

30

Virginia exceed unloads from California in Washington and
Baltimore because of the geographical advantage. Florida
unloads raw potatoes from the winter and early spring

crops in the Eastern region from March through June, with
peak amount of unloads in May. The majority of the Florida
unloads are split between Baltimore, Washington, Philadelphia
and New York City. Washington and New Jersey supply the
Eastern region with late summer potatoes through August,
September and October, just before the fall crop is
marketed. Washington primarily supplies New York City,
Boston and Philadelphia, while New Jersey supplies New York
City, Philadelphia and Washington. 'Wisconsin and Pennsyl-
vania are small suppliers of fall potatoes. Wisconsin ships
raw potatoes to New York City and Pittsburg. Almost all
Pennsylvania potatoes are unloaded in Pittsburg. Pittsburg
also receives unloads of potatoes from Michigan from

August through November.

The Southern Region

 

The Southern region of the United States receives
the smallest total number of unloads of raw potatoes of any
region. This is due to the smaller population in the
Southern reporting cities when compared to the population
in reporting cities in the other regions of the United
States. Another reason for the smaller number of unloads

of raw potatoes in the Southern region is the composition

31

of the diet of the people in this region. Southern people
do not include potatoes in their diet as much as people in
other regions of the United States.8 Finally, there is no
major producer of fall potatoes in the Southern region.

All the producing states within the region produce smaller,
seasonal crops.

The major supplier of raw potatoes in the Southern
region, Colorado, as shown in Table 6, primarily supplies
the Texas and Oklahoma markets. The largest unloads of
raw potatoes from Colorado occurs in Dallas and Oklahoma
City. This emphasizes the second reason given for the
smaller number of unloads in the Southern region. The
Texas and Oklahoma markets have consumption patterns more
like those found in the Southwestern portion of the United
States, where potatoes are more important part of the
population's diet, than the Southeastern region. Colorado
also has a geographical advantage in these markets over
other major fall potato producing states. Since Colorado
produces fall potatoes and late summer potatoes, the
seasonal supply pattern is similar to that of New York.
Unloads of late summer potatoes begin to occur in August
and September. October through May are the months in which
the largest number of unloads of raw potatoes from Colorado

OCCUI‘ .

 

8"Homemakers Preferences and Buying Practices for
Selected Potato, Rice and Wheat Products," Marketing Research
Report No. 939, (October, 1971) p. 27.

 

.sems .mcssso auansaom
cw memoaca manmummw> tam uflsum pmmum .nocmum m3mz umxumz .COHmH>HQ manmummm>
vcm ufinum .mofl>umm mcflumxumz quSmQOU .munussoflumd we pawaunmmmo .w .2 "mousom

 

 

32

mmosmm Hmma mNmH mnma mmma mama Hmma vnmm nwmm nmom comm omwa onam Hence

sflhAOHflquh-n
.mdfifié%@h8m
. HCHmHH> scmmflnofl:

m
\h‘ ‘1 I .\|l(.:.
rename sex .sc»ttts

saw mm ma mm mms mms ems as as ms cm as “ease
smas an ass was owe oma «mm was was am sm sm mamas
came mmm mam cam mm was as mm mam awe as» was mam ceasesoo
mssm ham aam mom mos mas ms. ss em sa sas mas mmm cschOmsz
sass ems sms ass sms mm mm mm mas was was ous show sum
mam am sa on oss ems am as mos masmz
mom am mam mas mss aesnosm
sew mm oms mmm mam os cosmcsamms
amsm Nam asm mmm was so was new mes. mmm muosaa asses
sans amm mmm msm mes ss em as sss mmm qsm mam asommcaax
msmm wam 5mm mes am ss sms «mm mas mms mma 0mm 0am amass
mm» mm mom was mm maoNssa
mmas as an em as was awe ssm mm ms mm as am eschewssmo
woos ms as New omm nos ma memnmsa

 

sauce 08 >02 poo ham was 3% was. sass sea Has has new 33m

 

33

Idaho is the second largest supplier of potatoes
in the Southern region. Idaho ships potatoes to all
reporting cities in the Southern region with no one parti-
cular city unloading a large amount of raw potatoes. The
seasonal unload pattern is primarily the same as the Eastern
region in the Southern region, except for the peak unload
month. The peak number of unloads occur in June in the
Southern region as compared to March is the Eastern region.

The number of unloads of raw potatoes from Wis-
consin and North Dakota are almost equal in the Southern
region. Both states supply the same cities in the Southern
region, with Wisconsin unloading primarily in Atlanta,
Birmingham, Memphis, Nashville, New Orleans and Oklahoma
City. Seasonal unload patterns of these two states differ
slightly. Unloads from the late summer crop in Wisconsin
begin in August. From October to January the number of
unloads is fairly stable. After January they begin to
decline to a low point in June. Unloads of North Dakota
raw potatoes begin in September. The peak supply period
of North Dakota potatoes is From October to A ri .

Minnesota and New York supply the Southern region
primarily with fall potatoes, although New York does sup-
ply this region with a small amount of late summer potatoes.
The unload pattern of Minnesota potatoes closely follows
the unload pattern of North Dakota potatoes, with the major

market for Minnesota potatoes being Dallas. New York raw

34

potatoes are unloaded primarily in three Southern cities,
Atlanta, Miami and Columbia. New York has a relatively
stable number of unloads from September through December,
a peak number in March, and then a decline to June.

There are a number of seasonal and small suppliers
of the Southern region. California, as in the Eastern
. region, is one of the major seasonal suppliers. The major
thrust of the California supply is again in June, and July
and August. California supplies all reporting Southern
cities except Memphis, with the largest number of unloads
occurring in Houston. Alabama is another of these seasonal
suppliers. The majority of Alabama potatoes are unloaded
in Atlanta and Birmingham. Unloads of Texas raw potatoes
occur not only in June, July and August, but also in May,
September, October and November. The unloads of Texas
potatoes outside the June, July and August period occur
primarily in the four reporting Texas cities. Arizona
ships raw potatoes from the Late Spring crop to the Southern
region mainly in June and July. The major market for
Arizona raw potatoes is Dallas. Florida again primarily
unloads potatoes from the winter and early spring crops in
March, April and May. The major market for Florida pota-
toes is Atlanta. Washington supplies the Southern region
with raw potatoes from August through October. All

reporting Southern cities unload Washington raw potatoes.

35

Maine is a small supplier of fall potatoes. The majority
of the Maine raw potatoes are unloaded in Atlanta and Miami,

with Maine being the major supplier in the latter city.

The Midwestern Region

 

Idaho is the major supplier of the Midwestern
region, as shown in Table 7. Unloads of Idaho potatoes
occur in all reporting cities in the Midwestern region.
Idaho is the major supplier of potatoes in Cincinnati,
Cleveland, Indianapolis and Louisville, and very close to
the major supplier in Chicago. The largest number of
unloads of Idaho raw potatoes occur in Chicago, Cleveland,
Detroit, and Indianapolis. Although Idaho unloads potatoes
in the Midwestern region in all months, the seasonal unload
pattern is not drastically different from the Eastern
region or the Southern region. From a low in August, the
number of unloads increases as the fall crop is harvested
and marketed. The three peak unload months are March,
April, and May.

The second largest supplier of raw.potatoes in the
Midwestern region is Wisconsin. Being centrally located in
this region gives Wisconsin a geographical advantage in
many reporting Midwestern cities. Chicago is a goOd example
of this. The unloads of Wisconsin potatoes in Chicago
exceed all other supplying states. Wisconsin is also the

major supplier of raw potatoes in Louisville and Milwaukee

 

.Hnma .mmfluflu
cswpmmzpwz aw memoHcD msnmuwwo> paw pflssm nwmsm .nusmnm mzmz uwxumz .COHmH>HQ
msnmuwmm> paw uwsum .mow>n0m mceumxsmz umEsmcoo .eusuasowumd mo ucmEuHmmwo .m.D "mousom

 

 

36

emw.mm mmnm comm mmmm mHmN wflmm vam momv moot wmom mwmm mmnm mmom Hmuoe
AOUmHOaOO stO%
302 reflce>awmssgm
.mscflmufl> .mcflmeumo

5.82 £8an

 

mafia mm om em moa wHN vmm mm mm mm mm mm Hm Hmnuo
momm va Ham new wmm Hmm HH OOH Ham mmm mow owe mam CHmCOomwm
msm mms cam as was mass.
mmm mm mom mag ea , ma NH Ha coumCHnmmE
Heme mmm mmm mvm moa mw mum Mmm mew oom vow muoxmm anhoz
mmmv whv ovm Hum mmm mmm am am hmm mum mow vmm mmm MUOmmfifiHm
whee mmw mom mom mam how hm em mum mmv mom mmm CMWHQUHE
hHm om ma om cam mNN MHH wm hm OQHME
mnmn 0mm mmw mew mwm mma mmm mmm mmm mom Nam can, mmn OJdUH
avma mmm mmm mum me ma MUHHOHM
mHmm mmm vmwa Gama mm MACHOMflHMU
mam mam mNo .mm MQONfiHm
mmw Hm mmH mmm mm memnmad
sauce owe >02 poo one >s5h muse we: and an: new can mpmum

 

37 ’

and is a close second to Idaho in Cincinnati. Because of

the production of late summer potatoes, unloads of Wis-
consin potatoes occur earlier in the year than unloads of

a fall crop state like Idaho or Maine. The number of

unloads of raw potatoes from Wisconsin begins to increase

in August to a peak level in September. From October through
March, the number of unloads is fairly stable. The number

of unloads then decreases after March.

The number of unloads from the three remaining
potato producing states in the Midwestern region are almost
equal. These three states, Michigan, Minnesota, and North
Dakota, are all within the Midwestern region so these are
intra-regional flows, as are flows from Wisconsin. North
Dakota and Minnesota are large suppliers of raw potatoes
in Chicago, Cincinnati, Kansas City, Louisville, Minnea-
polis, and St. Louis. Almost all potatoes produced in
Michigan are unloaded in Detroit, although there are unloads
in Louisville, Cincinnati and Cleveland. 'The seasonal
unload pattern in Minnesota and North Dakota is similar to
the seasonal unload pattern in Michigan, which is similar
to other states which produce late summer and fall pota-
toes, such as New York or Wisconsin.

The largest seasonal supplier of raw potatoes in
the Midwestern region is California. California supplies
raw potatoes from May through August, with unloads occur-

ring in all reporting Midwestern cities. Unloads of

38

California potatoes are largest in Chicago, Cleveland,
Detroit and Minneapolis. Alabama also supplies the Mid-
western region during the May through August time period.
The major markets for Alabama potatoes are Cincinnati and
Louisville. Unloads of Arizona potatoes occur in the
Midwestern region from May to July. Arizona primarily
supplies Chicago, Cleveland, Detroit and Minneapolis.

Texas is a supplier of early summer potatoes to the Midwest.
The majority of the unloads of Texas potatoes are in
Chicago, Cleveland, and St. Louis. Unloads of raw potatoes
from Florida occur from March through June. Florida ships
raw potatoes to all reporting Midwestern cities. Detroit
is the largest market for Florida potatoes. Unloads of
Washington potatoes occur primarily from August to October.
Washington supplies most of the reporting cities, with
Chicago being the largest market. A small number of unloads
of the fall potato crops from Maine, Pennsylvania, and

New York occur in Cleveland.

 

The Western Region
The Western region of the United States has a
characteristic which distinguishes it from all other regions.
All flows of raw potatoes are intra-regional. In other
1
words, all major suppliers of raw potatoes are located

within the region. There are two major factors which explain

this occurrance. First, most of the reporting cities in

39

the Western region are located along the Pacific Ocean.

The transport costs to ship raw potatoes into these Western
areas from the other producing states outside the Western
region make it impossible to compete with potatoes pro-
duced in the Western region. Second, the potato producing
states in the Western region are ranked first, third,
fourth, seventh, tenth and seventeenth in potato production.
The Western potato growers produce more than enough to meet
the demand within the region.

The largest unloader of raw potatoes in the Western
region is California, as shown in Table 8. This is appa-
rently due to the large urban consuming centers in
California. In the other three regions California is
primarily a seasonal supplier of raw potatoes. Even in
the Western region, the largest number of unloads of
California potatoes occur in June and July. But, in the
Western region California is a major supplier of raw pota-
toes the year round. With California producing potatoes
in five of the six growing seasons, there is a steady flow
of raw potatoes fer the entire year.‘ The majority of
unloads in the early and late months of the year occur in
San Francisco and Los Angeles, the latter being the largest
unloader of California potatoes.

Oregon is the second largest supplier of the Western
region. Oregon, gains some competitive advantage over

Idaho because of closer proximity to the major Western

4O

 

.Hnma smefl HO cumummE CH memoHcp oaeouooeb mam uflnum gmwsm .gommum msmz umxsmz

 

 

.COsmH>so msnmumqm>_©cm pflnwm .wow>wwm mcfipoxumz Heesmcou sensuanofisma mo pcmspumamo .m .D "mossom
Hmm.mm vsmm omnm mmmm Hmwm mnum Hmmm mnsm Nmmm Hmmm ovum amen Nomm Hence
smmxme .mpHMOHm
. Sousa fisozv
ham cs mm as ms om mms mm sm segue
ammm oem hem mmw men ham Hm mm mas omm Hmm Hos mmm coumceemmz
onmw «me mmv Ham ems mew me Hom mew 0mm mom Hmm mew comono
shoe mmm has hem mm mas me nmm one mom wmv mmm mmm osmcH
ommva mom «mm mom moss mama mmmm veom mmoa mom mnoa Hmoa whoa mflcuomssmu
mmn Hm mm mum mam He aneuwsm
oomm Ham mmm mmm Haw mmm ma vom Nam mav mmm Ham ocmuoHoo
sang 08 >02 80 seam was 3.2., was E: has .82 has can 33m

 

.ssms cw cosmmm cumummz
map Op mnucoz ha wmumum mCHUDUOHm Hoflmz EOHM mmOpmuOm nmmum HO m304h-l.o aqnfls

41

markets. The majority of the unloads of raw potatoes
from Oregon go to three cities, Los Angeles, Portland,
and San Francisco. The seasonal unload pattern of Oregon,
a producer of fall potatoes, resembles that of Idaho.

The third and fourth major supplier of potatoes
in the Western region are Idaho and Washington. Idaho is
a supplier of potatoes in all reporting Western cities,
as is Washington. The number of unloads of Idaho potatoes
are largest in Los Angeles, Salt Lake City and San
Francisco. Washington is an important supplier of raw
potatoes in Los Angeles and San Francisco during August,
September and October. But, the largest number of unloads
of Washington potatoes occur in Seattle. Washington sup-
plies Seattle all twelve months. The seasonal unload
pattern in Idaho in the Western region is similar to the
seasonal unload pattern in the other consumption regions.
The seasonal unload pattern in Washington is similar to
the seasonal unload pattern of other states which produce
late summer potatoes and fall potatoes.

Arizona is the one true seasonal supplier in the
Western region. Arizona unloads raw potatoes from May to
July. Almost all unloads of Arizona potatoes in the
Western region are in Denver. All unloads of Colorado

potatoes occur in Denver.

CHAPTER III

U. S. DEMAND-SUPPLY PROJECTIONS
IN 1980

Introduction

 

A recursive system is used to project the supply and
demand for potatoes for all uses at the farm level through-
out the United States. The four equations in the recursive
system incorporate major behavioral variables which influ-
ence the supply and demand for potatoes. The first of these
equations, the acreage equation, states that the yearly
acreage of potatoes is a function of the farm price of
potatoes in the previous period, the acreage of potatoes in
the previous period and the farm price of wheat in the
previous period. The average yield of potatoes is stated
as a function of time. The average marketing margin of
potatoes is stated as a function of marketing labor costs.
The final equation, the average retail price of potatoes is
a function of marketing margin, total production, per capita
disposable income and time. Two identities are used in
the model to tie the recursive system together.

Each of the equations in the system is estimated

separately via the ordinary least square procedure of

42

43

estimation.1

The ordinary least squares estimation pro-
cedure yields best linear unbiased estimates. The equations
are estimated using time series data from 1949-1970.

Also, the shifts which have occurred in the fifties
and sixties in potato consumption and utilization are
examined in greater detail. Specifically, shifts in the
consumption of the various processed potato products and
shifts in the utilization of raw potatoes are examined. Pro-
jections of consumption and utilization are made by extra-

polating trends of consumption and utilization time series

data from 1950-1970.

The Model Structure

 

The Acreage Equation

 

The acreage equation explaining variations in the
yearly acreage of potatoes in the United States for the
current period. The acreage equation is:

At = alDl+ lelFPt-l + bZDlAt-l + b3Dlwt-l + a2D2+
b4D2FPt_l + bSDZAt-l + e where:

At = Annual acreage in the current period (1,000 acres).

 

lA simultaneous equation model was developed in the
early stages of the study making use of both two stage least
squares estimation procedures. Comparison of results from
the simultaneous equation model and the recursive system
lead to rejection of the simultaneous equation model because
the results were significantly not different.

44

FP = Average price of potatoes received by farmers
in the previous period (dollars per cwt).

A = Annual potato acreage in the previous period
t-l

(1,000 acres).

Wt-l = Average price of wheat received by farmers in
the previous period (dollars per cwt).

D1 = Dummy variable, where D1 = 0 if 1949—1953 and
D1 = 1 if 1954-1970.

D2 = Dummy variable, where D2 = 0 if 1949—1953 and
D2 = 1 if 1954-1970.

e = Statistical error term.

a1, a2 and b1 to b5 = Regression coefficients to be
estimated.

The acreage equation states that the total potato
acreage (At) is a function of three variables prior to
1954, the farm price of potatoes received in the previous
year (Pt-l)’ the total potato acreage in the previous year
(At-l)’ and price of wheat in the previous year (Wt_1), and
two variables after 1954, the farm price of potatoes in the
previous year,and the total potato acreage in the previous
year.

The inclusion of farm prices in the previous period
reflects the assumption that producers expect prices to be
the same in the current year as they were in the previous
year. It is expected that if farm prices for potatoes are
high in the previous year producers will increase potato
acreage and that both b1 and b4 will be positive.

Potato acreage in the previous period is included

in the equation as a proxy for variables which cannot be

45

readily quantified, but are important determinates of the
amount of potato acreage. Reflected through this variable
is the crop rotation practice of potato producers, the
amount of fixed assets used in potato production, and pro-
ducer management habits. With the high level of fixed
capital investment required to produce potatoes, especially
since 1954, a producer is not likely to decrease or increase
production a great deal. The producer usually will plant
sufficient acreage to effectively utilize this specialized
equipment and will hesitate to buy extra equipment. This
probably is a major reason for the relatively small varia—
bility in the number of acres planted since 1954.

The farm price of wheat in the previous period is
also included in the acreage equation prior to 1954. The
productien of potatoes in the many different areas of the
United States makes it difficult to determine a universal
crop or crops which compete with potatoes for available
acreage. Although wheat is not grown in all major pro-
ducing regions, it is the only other major crop which com-
petes with potatoes for acreage across the entire United
States. Again it is assumed that producers expect farm
prices to remain the same in the current year as they were
in the previous year. Using these prices, producers made
decisions about the allocation of acreage. The price of

wheat was a significant variable prior to 1954 because

46

producers were less specialized and could more easily shift
acreage from potato production to wheat production. After
1954, with the shift to mechanical potato harvesters, pro-
ducers were no longer as flexible to change crops because

of the high fixed capital investment and the resulting push
toward specialization. Thus, the price of wheat was no
longer a significant influence on the amount of potato acres
planted. The value of the parameter bl is expected to be
negative because a relatively high price of wheat would
stimulate more wheat acreage and less potato acreage.

The two dummy variables, D and D2,are included to

1
account for an unexplained shift in the potato producers
supply behavior which apparently occurred in the mid-
fifties. The primary cause for this shift was the intro-
duction of mechanical potato harvesting. As mentioned

.
above, the high fixed capital investment requirement, due
to the introduction of the mechanical harvester and other
capital intensive production techniques, combined with the
increase specialization, probably reduced acreage variability.
Thus, a possible shift in the functional relationship of
the variables is allowed for by including both slope and
intercept dummy variable shifters in the equation, selecting
1954 as the shifting point because after that year the

variability in the actual amount of potato acreage was

reduced.

47

The Yield Equation

 

The yield equation explains variations in the
average yearly yield of potatoes in the United States for

the current period. The yield equation is:

Yt = a1 + blT + e

where:

Yt = Average yield of potatoes in the current period
(lbs per acre).

T = Time

e = Statistical error term.

a1 and bl = Regression coefficients to be estimated.
The average yield of potatoes (Yt) is stated as a

function of time (T). The independent variable time is
used as a proxy for the amount of variable inputs used per
acre which can not be readily quantified. The amount of
the various inputs used in the production of potatoes will
be a great influence on the yield obtained. As the prices
of inputs change, the mix of inputs used will also change.
The mix of inputs will directly affect yield. Lack of data
over the time period specified (1949-1970) is the major
obstacle to using this variable. The major producing states
do not have time series data on this variable, although
some is available for specific years. There would be a
problem even if a time series of the amount of variable
inputs used per acre were-available for a major potato pro-

ducing state. The amounts of inputs used vary from state

48

to state and season to season. So, data from one state for
one season would not be sufficient to reflect the amount of

inputs used in all other states and seasons.2

The Marketing Margin Equation

 

The marketing margin equation explains variations
in the average yearly marketing margin for potatoes in the
United States for the current period. The marketing margin
equation is:

MMt = a1 + blMCt + e

This formulation of the marketing margin equation
was found to have serial correlation.in the statistical
error term when it was estimated. The adjusted marketing
margin equation is3:

(MMt - DMMt-l) = a2 + b2(MCt - FMCt—l) + e
this equation can be transformed to:

MM = a + b MCt - b pMC

t 2 2

where:

+ VIM + e
p: t_

2 l

t-l

 

2The index of prices paid for were inputs by farmers
and the expenditures of inputs in Idaho were tried in the
yield equations as a proxy for the amount of inputs used. The
coefficients for both variables were not significantly
different from zero. '

3The procedure used here was suggested by Dr. Lester
V. Manderschied. The procedure is a variation of the
Hildreth-Ln method of solving the serial correlation problem.

49

MMt : The average marketing margin for potatoes in the
current period (cents per pound).

MMt_1 = The average marketing margin for potatoes in
the previous period (cents per pound).
MCt = The average marketing labor costs in the current

period (dollars per hour).

MCt__l = The average marketing labor cost in the pre-
vious period (dollars per hour).

e = Statistical error term.

p = A constant used to adjust the equation for
serial correlation.

a1, a2, and b , b2 = Regression coefficients to be

estimate .

The average marketing margin for potatoes in the
current period (MMt) is a function of the average marketing
labor cost in the current period (MCt)' Marketing labor
cost is an average of wages paid for labor necessary to move
the product through the market channel from the producer to
the final consumer. If marketing labor costs could be
eliminated, the differential between farm and retail price
would be close to zero. Although this is unrealistic, it
does lead to a prediction about the sign of b2. A rela-
tively high marketing labor cost will cause a higher
marketing margin due to retailers desire to incorporate
this increased cost into the retail price. So, it is
expected that b2 will be positive.

Marketing margin in the previous period (MMt_l),

marketing labor cost in the previous period (MCt_l), and p

50

are included in the equation to adjust for serial corre-
lation present in the first equation formulated. The value
of p is calculated using the Durbin—Watson statistic (d),
which shows the presence of serial correlation in the first
form of the equation. The formula used to calculate the
value of p is d = 2(l-p). The use of the lagged variables
and 0 change the form of the independent and dependent

variables which reduces the serial correlation problem.

The Retail Price Equation

 

The retail price equation explains variation in the
average yearly retail price of fresh potatoes in the United
States for the current period. The retail price euqation is:

RPt = athblMMthPIthTb4e
which can be transformed to the linear form for estimation
by taking the logarithm of each side, thus:

log RP = log a + bl log Qt + b log MM +

t l 2 t
+ +
b3 log PIt . b4log T log e
where:
PRt = The average retail price of fresh potatoes in
the Current period (dollars per cwt).
Qt = The total production of potatoes for all uses

in the current period (1,000 cwt).

MMt = The average marketing margin for potatoes in
the current period (cents per pound).

PIt = Per capita annual disposable income ($1,000).
T = Time

e = Statistical error term.

51

a1 and b to b4 = Regression coefficients to be

estimated.

The retail price of potatoes (RPt) in the current
period is a function of four variables, the total production
of potatoes in the current period (Qt)' the marketing margin
for potatoes in the current period (MMt), per capita dis-
posable income in the current period (PIt) and time (T).

The total production of potatoes reflects the influence of
the supply of potatoes on the retail price of potatoes.
Assuming a normal, downward leping demand curve for pota-
toes, the relationship between retail price and the quantity
supplied can be determined. A relatively small supply of
potatoes will yield a higher retail price. A relatively
large supply of potatoes will yield a lower retail price of
potatoes. Although the degree of variability in the retail
price change due to a change in the quantity of potatoes
supplied depends on the price elasticity of demand, it can
be assumed that retail price and quantity supplied vary
inversely. This indicated that the value of the coefficient
b1 should be negative.

The marketing margin is an important variable in
the retail price equation because retailers include the costs
which determine this margin in the retail price. With col-
lective bargaining being of little significance for
potatoes, prices are determined by the interaction of

demand and supply, and marketing margin can be assumed to be

52

independent of farm price. The desire of retailers to
include all costs which determine marketing margin, not just
marketing labor costs, make this a significant factor in
determining the level of the retail price. As the marketing
margin increases, retailers will increase the retail price.
So it is expected that the regression coefficient for the
marketing margin (b2) should be positive.

The per capita disposable income will influence the
demand for potatoes, which will be a direct influence on
the retail price of potatoes. As disposable income
increases, the demand for potatoes will shift. With a fixed
supply of potatoes, changes in the demand for potatoes will
also change the retail price of potatoes. The direction
of the change in demand depends on whether consumers con-
sider potatoes are an inferior commodity, that is,
increases in disposable income cause a decrease in demand
for the commodity, or if potatoes are a normal commodity,
that is, increases in disposable income cause an increase
in demand for the commodity. Population also influences
the demand for potatoes, but because of the high correla-
tion between disposable income and population (.98) it was
not included in the equation as a separate variable.
Instead, the per capita disposable income variable was
used to incorporate the influence of both disposable

income and population.

53

As in the yield equation, there are variables which
affect the demand for potatoes which can not be readily
quantified. Time is specified in the equation as a proxy
for these variables which gradually change over the sample
period. The most important of these variables is consumer
tastes and preferences. The consumption tastes and pre-
ferences of the consumers will have a direct effect on the
commodities demanded and the quantity of each commodity
demanded. As these tastes and preferences change, so does

the demand.

The Identity Equations

 

The identities are two definitional equations
which are necessary when the projection procedure is used.
The identity equations are:

= A X Y

Qt t t

FP = RP - MM
t t t

where all variables are the same as defined above.

Estimation Results

 

The Acreage Equation

 

An ordinary least square estimation procedure was
used to estimate the regression coefficients for the acreage
equation from the 1949-1970 time series data. The results

are:4

 

4The figures in parentheses are the standard errors
of the coefficients.

54

3,.
II

t 420.32 + 299.96 DlFPt_l

(53.93) (.15)

307.98 DZWt—l + 56.41 DZFPt~I + .58 DZAt-l

(109.12) (28.12) (.22)
R2 = .795
d = 2.436

The results indicate that the variables included in
the above equation explain approximately 79 percent of the
potato acreage variation during the 1949—1970 time period.
All signs of the regression coefficients are consistent
with the expected sign and using the t~test, all regression
coefficients are significantly different from zero at the
5 percent significance level. There is no serial cor-
relation present in the residuals.

The regression coefficients for the farm price of
potatoes prior to 1954 indicates that if the farm price in
the previous period increased by ten cents per hundredweight,
the potato acreage in the present period would increase by
approximately 30,000 acres. After 1954, the same ten cent
increase in farm price in the previous period would induce

producers to increase potato acreage only 8640 acres.

 

5R2 is the coefficient of determination which indi-
cates the percentage of variation in the dependent variable
explained by the independent variables in the equation.

6d is the calculated Durbin-Watson statistic which
is used to test for serial correlation.

55

These results indicate that the number of acres put into
production of potatoes has become much more inflexible to
price changes probably due to the shift to mechanized
potato production, especially mechanized harvesting.

The regression coefficients for potato acreage in
the previous period indicate that potato acreage in the
present period increased as potato acreage in the previous
period increased. Prior to 1954, an increase in potato
acreage in the previous period of 1000 acres increased potato
acreage in the present period by 910 acres. But, after 1954,
the same 1000 acre increase in the previous period increased
potato acreage in the present period by 580 acres. This
decrease in the variability of potato acreage in the present
period reflects structural changes in the three major
variables for which acreage in the previous period is a
proxy. The amount of fixed assets used in potato production
increased as producers shifted to mechanization. Producers
had a greater incentive to maintain sufficient acreage to
effectively utilize these highly specialized fixed assets.
This led to increased specialization which had reper—
cussions on rotation practices. The producers were forced
to abandon the rotation practices used before the intro-
duction of the specialized equipment and adopt new ones
which allowed them to obtain maximum utilization of this
equipment. The increased amount of fixed assets also

changed producer crop management habits. The grower was

56

forced to become a specialized potato producer. Diversifi-
cation became a thing of the past.

The farm price of wheat in the previous period
also influenced potato acreage prior to 1954. The regres-
sion coefficient for the farm price of wheat indicates that
a ten cent increase in the price of wheat in the previous
period decreased potato acreage in the present period by
30,798 acres. The negative regression coefficient for this
variable indicates that wheat was a competing crop prior to
1954. In early formulations of the acreage equation, the
farm price of wheat was included in the equation after 1954,
but was found not to be significantly different from zero.
Again, the influence of the shift to more mechanized potato
production had a great deal to do with the reduction of the
significance of wheat prices. The increase in mechaniza-
tion has forced producers to be more specialized and to
increase the size of potato farms to obtain economics of
scale. Thus, wheat prices in the previous period after
1954 were not significant in determining potato acreage in

the present period:

The Yield Equation

 

The results of the yield equation are:

Yt = 137.48 + 4.06 T
(.20)

R2 = .95

d = 1.79

57

The results obtained indicate that the independent
variable time explains approximately 95 percent of the yield
variations during the 1949-1970 time period. This high
coefficient of determination value indicates that the
actual yield of potatoes during the time period closely
resembles this linear approximation. The shifting of potato
production from areas of the country with lower annual
yields to areas of the United States which are better suited
to produce potatoes and thus obtain larger yields, as dis-
cussed in Chapter II, is one important factor which has
caused yields to rise over the sample time period. Also,
yields in each of the potato production areas have increased
for a number of other reasons. Some of these reasons are
production of higher yielding varieties, use of better seed,
wider use of irrigation, more use of fertilizers, better
crop management, use of advanced production technology and
'use of pesticides and herbicides. The regression coeffi-
cient is judged to be significantly different from zero at
the 5 percent significance level and there is no serial

correlation in the’residuals.

The Marketing Margin Equation

 

The results for the marketing margin equation are:

(MMt — .51 MMt_l) = .03 + t iiiMCt - .51 Mct_1)

or using the transformed version of this equation:

58

MMt = .03 + 7:22)Mct — t:%2;.51 MCt_l) + .51 MMt_l
R2 = .80
d = 1.79
5 = .517

The independent variable, the adjusted first dif-
ference of the marketing labor cost, explains approximately
80 percent of the variation in the adjusted first differ—
ence of the marketing margin. Before the equation was
adjusted for serial correlation, market labor costs
explained approximately 93 percent of the variation in the
marketing margin. The coefficient bl is positive, as pre-
dicted above, and significantly different from zero at the
5 percent significance level. The original equation yielded
a value of .98 for the Durbin-Watson statistic which indi-
cated that serial correlation was present. After adjusting
the equation for the serial correlation, a Durbin-Watson
statistic of 1.79 was obtained which shows that-the serial
correlation problem was reduced by the adjustment procedure
described above.

It can be shown that the negative value of the

marketing labor cost in the previous period will almost be

completely offset by the positive value of the marketing

 

7With a value of the Durbin-Watson statistic in

the original equation of .98, p is formulated from the
equation .98 = 2(l-p).

59

margin in the previous period. Thus, for explanation pur-
poses, changes in the marketing margin the present period
are dependent on changes in the marketing labor cost in the
present period. For example, in period t, assuming no
change in the previous period, a ten cent increase in mar-
keting labor cost will increase the marketing margin by

21 cents. In period t + 1, assuming marketing labor cost
stabilizes at this new, higher level, marketing labor cost
in the previous period has increased by ten cents while
marketing margin has increased by 21 cents and after inser-
ting these values into the marketing margin equation and
doing the arithmetic, the marketing margin in period t + 1
is unchanged from that for period t. The increase in mar-
keting margin due to an increase in marketing labor cost
reflects the desire of retailers to cover most costs asso-
ciated with getting the product to market in the retail

price.

The Retail Price Equation

The results for the retail price equation are:
_ - .45 1.41 .27 -.29
RPt — 3.35Qt MMt PIt T

or in the linear logarithmic form:

log RPt = log 3.35 - .45 log Qt + 1.41 log MMt +
(.24) (.31)

.27 log PIt -.29 log T
(.21) (.08)

R2 = .85

d = 2.04

60

The value of the coefficient of determination
indicates that the four independent variables explain approxi-
mately 85 percent of the variation in the logarithm of
retail price. An R2 value of 85 for the retail price equa-
tion is quite good, especially when it is realized that the
data used is the average of retail price data collected in
selected cities across the United States on a non-
standardized product. All regression coefficients are con—
sistent with expected results; Three out of four independent
variables, total production, marketing margin, and time,
are significantly different from zero at the 5 percent
significance level. The fourth independent variable, per
capita disposable income is not significantly different
from zero at the 5 percent significance level. There are
two reasons why the variable is retained even though it
isn't significant at the 5 percent level. The presence of
multicollinearity may have biased the standard error of the
coefficient upward.8 If the standard error of the coef-
ficient is biased, the regression coefficient for per capita
disposable income may be actually significant at the 5

percent significance level. Also, per capita disposable

 

8The use of exponential form of the retail price
equation reduced the correlation between the time variable
and the per capita disposable income variable from .96
for the arithmetic equation to .89. This reduction in
correlation reduces the severity of the multicollinearity
problem, although it is still present.

61

income is an important economic variable. Thus, it should
be retained in the equation. There is no serial correlation
in the residuals.

The estimated regression coefficient for the total
production of potatoes for all uses indicates that as potato
production increases, retail price decreases. But, the
magnitude of the regression coefficient seems smaller than
would be normally expected for a staple food product. The
price flexibility coefficient, which can be taken directly
from the estimation results, is --.45.9 If all cross flexi-
bilities and cross elasticities are assumed to be zero, the
price elasticity equals one over the price flexibility. So,
the corresponding price elasticity is —2.22. This result
implies that the demand for potatoes is very elastic. An
elastic demand indicates that a large change in the quantity
supplied has a very small effect on retail prices. Most
prior studies indicate that the price elasticity is approxi-
mately —.3, where farm price is a function of total
production. Although the calculated elasticity may be too
high because of a slight simultaneous equation bias, there
is a plausible explanation for this apparent inconsistency.
It should be noted that this study uses retail prices for
fresh potatoes instead of farm prices for potatoes. At the

farm level, changes in production of potatoes will cause

 

Price flexibility = is; x

= b
1

"UIIO

62

large fluctuations in the price of potatoes. But these
variations in farm price need not carry over to the retail
price. The shift toward increased processing of potatoes
has been a major factor which has pushed the price elasti-
cities upward and reduced the large variability of retail
price. Processors can buy potatoes when farm prices are
low and absorb much of this production increase because of
the ability to store the processed product for relatively
long lengths of time. Also, if farm prices are depressed,
potato farmers may find it economically feasible to feed
potatoes to livestock or keep them as seed for next year's
crop. So, instead of a large increase in the amount of
potatoes supplied to the fresh market, which would have
occurred before the processing revolution, processors can
absorb a majority of this increased production and reduce
the size of the increase in the amount of potatoes sup—
plied to the fresh market.

The regression coefficient for the marketing margin
variable indicates that the retail price of potatoes
increased as the marketing margin increased. This is con-
sistent with the result predicted above and is based on the
assumption that retailers incorporate any increase in the
costs which determine the marketing margin into the retail
price.

Increases in per capita disposable income increased

the retail price of potatoes during the sample period as

63

indicated by the positive regression coefficient. This
indicates that as per capita disposable income rises, the
demand for potatoes rises to induce the increase in the
retail price of potatoes. It cannot be determined whether
potatoes are a normal or inferior good because the per
capita disposable income variable measures the effects of
disposable income and population. The estimated regression
coefficient for the time variable indicates that the retail
price of potatoes decreased every succeeding year during
the sample period due to gradually changing unquantified
variables such as tastes and preferences. It should be
pointed out that the multicollinearity problem does not
affect the estimated regression coefficients for these two
variables. Multicollinearity will not bias the estimated
regression coefficients either upward or downward, but it

may bias the standard'errors of the coefficients.10

The Projection Procedure

 

There are three variables which are not determined
within the model and must be predicted before any calcula-
tions can occur. Of the three predetermined variables, time,
per capita disposable income, and marketing labor cost, only

two need to be estimated by use of a statistical technique.

 

10Ronald J. Wonnacott and Thomas H. Wonnacott,
Econometrics. (John Wiley and Sons, Inc., 1970) p. 59.

 

64

The extrapolation of linear trends is used to estimate the
per capita disposable income and marketing labor cost
variables. The linear trend equation is of the form

Y = a + bT + e where Y is the estimated predetermined
variable and T is time. The coefficient of determination
was quite good for both variables, being 96 percent for per
capita disposable income and 93 percent for marketing labor
cost. These linear trends are extrapolated into future to
obtain values of per capita disposable income and marketing
labor cost.

Before the actual projection procedure is discussed,
two important assumptions should be made explicit. It is
known that linear increases or decreases will not continue
at a constant rate to infinity. At some point in time, the
rate of increase or decrease will begin to slow. This occurs
because the value of the variable is approaching the asymp-
tote, the upper limit of increase or lower limit of
decrease for that variable. It is assumed in this model
that during the prediction period the rate of increase or
decrease will not slow or that the asymptote will not have
been reached. An example of this would be the slowing of
the increase in the yield of potatoes due to environmental
factors. One of these factors could be a shortage of water
for irrigation which would definitely reduce yields. It
is also assumed that the structure of the estimated equa-

tions of the model remains constant over the projection

65

period. It is a distinct possibility, especially in the
rapidly changing potato industry, that sometime during the
next ten years the structural relationship within an
equation will change. A good example of this was the shift
to mechanical harvesters which occurred in the mid—fifties.
This changed the structural relationship within the acreage
equation. If a change does occur during the seventies, it
is not known what this change will be or when it will
occur. So, for projection purposes, it is assumed that the
estimated structure will continue to 1980.

For projection purposes, the specification of the
marketing margin equation is changed slightly. The equation
is modified by adding p et, where et is the residual value
in the last period (1970 in this case). This last modifi-
cation is necessary to assure the proper unbiased
projections.ll This allows the serial correlation present
between the last error term and all succeeding error terms
to be built into the projection. If this term was not
added to the equation, the resultant projection of the mar-
keting margin would be biased. The p value must be squared
for the second time period, cubed for the third time period,
and so on to account for the decreasing correlation between
the error term in the last observed time period and all

succeeding periods. With c): .51 and et= .04 in 1970, the

 

11 J. Johnston, Econometric Methods. (McGraw-Hill,
1963) p. 187.

 

66

adjustment term becomes very close to zero after only two
projections.

To explain the projection procedure, the year 1971
will be used as an example. To estimate the total potato
acreage, the value for total potato acreage in 1970 and
the farm price in 1970 are inserted into the acreage equa—
tion (the first three variables are not used because D1= 0).
These values times their respective regression coefficients
plus the constant term determines the total potato acreage
in 1971. The projected yield for 1971 is the value of the
predetermined variable time in 1971 multiplied by its
regression coefficient plus the constant term. The mar-
keting margin in 1971 can also be easily projected. The
market margin in 1970 is known, the marketing labor cost in
1970 is known, and the marketing labor cost in 1971 is a
projected predetermined variable. All other constants and
regression coefficients needed to project marketing margin
are known. Simple arithmatic calculations determine the
marketing margin in 1971. At this point,-the first identity
is introduced. Total production in 1971 is the product of
the projected total acreage in 1971 and the projected yield
in 1971. The logarithm of the calculated total production
for 1971 plus the logarithm of the projected marketing margin
for 1971 are inserted into the retail price equation with
the logarithms of the projected predetermined variables, per

capita disposable income and time. From this equation, the

67

logarithm of the estimated retail price in 1971 is deter-
mined. Taking the anti-logarithm of this value yields the
projected retail price in 1972. To determine estimated farm
price in 1971, the second identity is introduced. The
projected marketing margin in 1971 is subtracted from the
projected retail price in 1971 to determine the projected
farm price in 1971. The projected farm price in 1971 is
inserted into the acreage equation and the system is
started again. This system is repeated until 1980. Use of
this procedure produces projections of potato acreage, ave-
rage yields, marketing margins, total potato production,

retail prices and farm prices.

Results and Interpretation

 

Using the projection procedure described above,
results are obtained for the projection period, 1971 to
1980, as shown in Table 9. Examining these projections
leads to the conclusion that the predictions obtained from
the recursive model are extremely stable.

As expected, the total acreage projection during
the projection period is very stable, varying only 12,000
acres throughout the entire period. The structure of the
acreage equation leads to the resulting stable acreage pro-
jection. With acreage being dependent on farm price in the
Previous period and with the projected slow, steady increase

Of farm prices, potato acreage also increases at a slow and

 

68

 

.uocusm one we pousmEoo "mousom
mm.m aa.m m.a www.mmm m.aem mmaa omma
mm.m mm.m V m.a oee.mmm m.mom amaa mama
NN.N oa.m ~.a aaa.aam ~.mmm emea mama
NN.~ am.m o.a ema.aam a.mmm mmaa aama
am.m mo.m m.e mmm.amm a.amm mmaa mama
om.m mm.m a.e mma.mmm o.aam amaa mama
om.m aa.m m.e moa.mmm o.mvm maaa aama
ma.m ae.m v.0 oma.mam m.mmm aaaa mama
aa.~ ma.m m.e omm.mmm m.amm aaaa mama
ma.~ am.m a.m aom.mmm m.omm mmwa aama
a.na

a.pso awe a.uso Hem Hem mucmov A.pso oooav amnom Amanda

madaaoev mamaaoev aamumz aoaaoseoum “ma.mgav oooav
moaum game moaum aamumm mcaamxumz amuoe eawaw mmmmuua new»

 

69

steady rate. The one decrease in acreage in 1972 is due
to the decline in farm price for potatoes between 1970

and 1971. Although potato acreage will probably vary much
more than projected here, the average acreage will pro-
bably continue to be around the 1,450,000 level, as it

has been for the last fifteen years. So, the projected
potato acreage values are not out of line with past
experience and expectations of the future.

The yield of potatoes is projected to continue to
increase at a constant rate of approximately 4.1 pounds
per acre per year throughout the projection period. By
1980, the yield is projected to be 267.3 pounds per acre,
as compared to 228.0 pounds per acre in 1970. This is an
increase of 39.3 pounds per acre during the ten year period.
Continued shifts to areas of the UniuaiStates which are
better suited for potato production, are more specialized
in potato production, use more irrigation, use better seed
and make greater use of fertilizers will be the primary
contributors to these increases in yields.

Total production of potatoes is also projected to
increase during the seventies. With total production being
formulated in the model as the product of the projected
total potato acreage, which is increasing in all but one
year throughout the projection period, and the projected
yield, which is increasing throughout the projection period,

total production must increase in at least nine out of the

70

ten years. The only possible decrease in projected total
production could occur in 1972, but the increase in yield
more than offsets the decrease in acreage, so projected
production increased. In reality, it could be expected
that total production of potatoes would vary upward and
downward, depending on the variation of total potato
acreage and the existence of random shocks. If potato
acreage does vary around this 1,450,000 acre figure, total
potato production will also vary. These increases and
decreases in production will probably continue to trend
upward over the projection period, as they have over the
last fifteen years. The projection of total production
does reflect this upward trend of actual production, even
though the large increases and large decreases are not
reflected.

The predicted marketing margin increases throughout
the prediction period because marketing labor cost increases
linearly throughout the prediction period. This is the
same result which would be expected of the actual marketing
margin.

The predicted value of the retail price is a func—
tion of the predicted marketing margin, the predicted
total production, time and per capita disposable income.
All four of these independent variables are increasing

throughout the projection period. The results indicate

71

-that the influence of the positive variables outweigh the
influence of the negative variables. Retail price is
predicted to increase from $8.21 per cwt. in 1971 to $9.71
per cwt. in 1980. Actual retail prices are also expected
to increase during the seventies, but the increases are
not expected to be a constant rate as predicted by the
model. Actual retail price will probably fluctuate more
during the projection period due to variations in total
production and random factors.

The final variable, farm price, like total potato
production, is calculated by use of an identity. Variations
in the farm price are dependent on variations in the
retail price and the marketing margin. The projected
values of the farm price indicate that retail price is
increasing slightly faster than marketing margin. The
slow, steady increase in farm price seems to be unrealis-
tic. With an inelastic demand curve at the farm level,
prices should fluctuate by greater amounts than projected
by the recursive model. If total production does vary
during the projection period, as is expected because of
random factors, farm price will also fluctuate substan-
tially. Although it is almost certain that these
fluctuations in the farm price will occur, the projection
of the farm price should be a good indicator of the trend
in actual farm prices if the industry behavioral structure

does not change significantly over the next decade.

72

The projections of the model can be compared to
actual data for the first two years of the projection period,
1971 and 1972. In both 1971 and 1972, total acreage
decreased by much larger amounts than projected by the model.
The reduction in total acreage in 1971 was due to a slight
drop in the farm price in 1970. The increase in actual
yield in 1971, which is very close to the yield projection,
did not increase enough to offset the decrease in acreage,
thus a decrease in actual total production. This decrease
is contrary to the increase projected for total production.
The actual marketing margin and actual retail price are
slightly higher than the projections due to higher rise
in marketing labor costs in 1971 than projected and the
reduction in total production. The actual farm price is
at a lower level in 1971 than predicted farm price because
of the large carryover of stocks from the large potato
production in 1970 which had a depressing effect on 1971
farm prices.

In 1972, the difference between projections and
actual values is larger than in 1971 due to the influence
of a random shock. The weather in 1972 was extremely bad
during the harvesting season for the fall potato crop.

This weather greatly reduced the total number of harvested
acres (especially in the Midwest and the East). Although
yield again increased, total production decreased. This

differs from the projected increase in total production

73

because of this random shock. Although both actual
marketing margin and actual retail price are slightly
higher than the projected values for these variables,
actual farm price increased to a much higher level than
projected by the model. This large increase in the actual
farm price is due to the random shock of bad weather,
which reduced total production in 1972 and the inelastic

farm demand.

Potato Consumption and Utilization

Over the past two decades there has been a shift
in potato utilization. In 1970, approximately half the
potatoes produced for human consumption were shipped to
potato processors, which can be contrasted with the five
percent shipped to processors in 1950. This indicates the
magnitude of the rapid evolution of the potato processing
industry which occurred during the late fifties and
through the sixties. The shifts in potato utilization can
be examined by looking at shifts in per capita consumption
or by looking at the usage of the potatoes produced.
Tables 10 and 11 will facilitate this examination. Table
10 shows the shift in per capita consumption from 1950 to
1970, while Table 11 shows the shift in potato utilization
from 1950 to 1970.

Total per capita consumption of potatoes has been

on the rise since 1950. This rise in consumption is the

74

TABLE 10.--Per Capita Consumption of Fresh and Processed
Potato Products in the United States, 1950,
1960 and 1970.

v-W—_ fl*-.-. _...-.,
. ——--—-~ --..——- m.‘- w“-

—-—_-l.._...--....- -__..
-.—.-‘—.._-...- .c.-.—

 

 

1950 1960 1970

(pounds) (pounds) (pounds)
Total Potato
Consumption 106.3 108.4 118.6
Fresh 100.0 84.7 59.6
Processed 6.3 23.7 59.0
Frozen French
Fries .3 5.7 24.5
Other Frozen
Products 0.0 .8 3.3
Chips 4.5 11.7 17.7
Dehydrated 1.2 5.0 12.8
Canned and
Flour .3 .5 .7
Source: American Potato Yearbook, 1951, 1961 and 1971.

 

75

TABLE 11.--Potato Utilization, 1950, 1960 and 1970.

 

__ ___,._ _,.__ .___._.. ____. -__.._,_
——————-o _, _.__. .—

 

 

__..
__.——_ .— _-_—_..__.—..__ - .—

 

1950 1960 1970

(1000 cwt.) (1000 cwt.) (1000 cwt.)
Total Production 259,112 257,104 324,801
Table Stock 151,863 149,376 129,242
Total Processed 31,318 59,442 136,137
Chips 13,720 21,310 35,861
Dehydrated 1,849 10,104 26,053
Frozen French
Fries 720 13,373 54,478
Other Frozen
Products 0 1,669 7,381
Starch, Flour
and Canned 15,529 12,986 12,364
Seed, Feed,
Shrinkage
and Loss 45,588 48,885 60,209

 

Source: American Potato Yearbook, 1951, l961,and 1971.

 

76

result of interaction of the rapid evolution of the pro-
cessing industry and the rapid change in consumer tastes.
Prior to 1950, with primarily fresh potatoes being con-
sumed, per capita potato consumption had decreased for
forty years. Without the rapid evolution of the processing
industry, per capita potato consumption would probably
still be declining. This is shown by the fact that per
capita fresh consumption has continued to decline through-
out the fifties and sixties. Clearly the decline would
not have been as rapid were processed potato products not
available to substitute for fresh potatoes. Fresh potato
per capita consumption decreased by 15.3 pounds per person
between 1950 and 1960, when the great influx of processed
potato products began to reach the market. During the
sixties, the decade of very rapid increases in consumption
of processed potato products, fresh potato consumption
dropped to 59.6 pounds per person, a decrease of 25.1
pounds per person. This is a decline of 30 percent during
the ten year period between 1960 and 1970. During the
twenty year period between 1950 and 1970, per capita con-
sumption of processed products grew rapidly. Consumption
of processed products increased by 17.4 pounds per person
in the fifties and by 35.3 pounds per person in the sixties.
The decreases in per capita consumption of fresh potatoes

was more than offset by the large increases in per capita

77

consumption of processed potato products, so total potato
consumption increased during these two decades.

The per capita consumption of each of the various
types of the processed potato products can be examined to
determine which of the processed products were most res-
ponsible for the increase in consumption of potatoes.

Per capita consumption of frozen potato products was the
largest of all processed potato products in 1970, although
it was the smallest in 1950. Frozen potato products
experienced the largest increase in per capita consumption
during the sixties, when per capita consumption increased
from 6.5 pounds per person in 1960 to 27.8 pounds per
person in 1970. Frozen french fries account for approxi-
mately 90 percent of this consumption. Per capita
consumption of frozen french fries increased by 18.8
pounds per person during the sixties. Per capita consump-
tion of other frozen potato products, (hash browns, tater
tots, etc.) has also been on the rise. Per capita con-
sumption of these frozen potato products rose from .8
pounds per person’in 1960 to 3.3 pounds per person in 1970.
The increases in the percapita consumption of frozen
potato products, primarily frozen french fries, has been
the major factor in the rapid increase in the per capita
consumption of processed potato products.

Increases in the per capita consumption of dehydra-

ted potatoes in 1950 was 1.2 pounds per person. By 1960,

78

per capita consumption had risen to 5.0 pounds per person,
an increase of 3.6 pounds per person, and by 1970 per
capita consumption had risen to 12.8 pounds per person.
This is an increase of one and one—half times the level of
consumption in 1960.

Per capita consumption of potato chips has also
increased during the past two decades. During the fifties
per capita consumption of potato chips increased by 7.2
pounds. It was during this decade that technological
advancements were made in the chipping industry which
allowed chippers to meet the increasing demand for proces-
sed products.12 An increase in per capita consumption of
6 pounds per person was experienced during the sixties.
Although per capita consumption of frozen potato products
is the largest, per capita consumption of potato chips is
still the second largest of all processed products. Per
capita consumption of canned potatoes and potato flour
has also been increasing, but a per capita consumption of
only .7 pounds per person in 1970 indicates that these are
not important processed products for human consumption.

The shifts in potato utilization also can be
examined by looking at the use of the potatoes produced.
The amount of potatoes going to the fresh market has dec-

reased over the past two decades. The decrease between

 

12Sullivan, loc. cit.

79

1950 and 1960 was approximately 2.5 million cwt., which
parallels the decrease in total production of 2 million
cwt. The largest decrease in the amount of potatoes going
to the fresh market occurred during the sixties. During
this ten year period the quantity decreased by 20 million
cwt. This large decrease reflects the shifting of sup-
plies of raw potatoes from the fresh market to processors.
Supplies of potatoes to processors have increased
greatly during the fifties and the sixties to meet the
increasing consumer demand. Supplies of potatoes to chip-
pers has increased by two and one-half times during these
two decades. Chippers increased the amount of potatoes
processed during the fifties by 7.6 million cwt. The
increase during the following decade was almost twice that
at 14.5 million cwt. There appears to be an inconsistency
between the consumption data and usage data. The consump-
tion of potato chips during the fifties and sixties
increased by equal amounts, while increases in supplies
of potatoes to potato chippers in the sixties nearly
doubled the increases in supplies during the fifties. This
apparent inconsistency is explained by the technological
advancement which occurred during the fifties. The new
technology allowed potato chippers to produce the same
amount of potato chips with a smaller amount of raw potatoes.

So, supplies to potato chippers during the fifties did

80

not have to increase by 14 million cwt. as they did in the
sixties to meet the increased demand.

The quantity of potatoes supplied to potato dehyd-
rators has also increased rapidly during the fifties and
the sixties. The quantity supplied increased from 1.8
million cwt. in 1950 to 10.1 million cwt. in 1960, an
increase of 8.3 million cwt. Dehydrators were supplied
with 26.1 million cwt. of potatoes in 1970, an increase of
16 million cwt. over the 1960 level. The increase in
supply to dehydrators during the sixties was twice as large
as the increase in supply to dehydrators during the fifties.

The largest increases experienced during the
fifties and the sixties occurred in supplies of potatoes
to potato freezers. During the fifties supplies to pro-
cessors of all frozen potato products increased from 720
thousand cwt. to 15 million cwt. During the sixties, the
increase in supplies was even more dramatic. Supplies of
potatoes to potato freezers in 1970 were four times
larger than in 1960. This is an increase of 46.8 million
cwt. Approximately 90 percent of all supplies of potatoes
shipped to processors of frozen potato products were pro—
cessed into frozen french fries, just as 90 percent of all
frozen potato products consumed were frozen french fries.
Supplies of potatoes to potato freezers which were pro-

cessed into frozen french fries increased by 12.7 million

81

cwt.during the fifties and 41.0 million cwt. during the
sixties. Supplies of potatoes processed into other frozen
potato products have also increased during these two
decades. The increase between 1950 and 1970 was 5.8 mil-
lion cwt.

Supplies of potatoes to flour and starch proces-
sors have varied greatly during the fifties and sixties
because most cull and excess supply potatoes are shipped
to these processors. The general trend of supplies to
flour and starch processors has been downward. Supplies
of potatoes to potato canners, also primarily cull pota-
toes, increased during the fifties and sixties. In total,
supplies to these processors decreased by 2.5 million cwt.
during the fifties and decreased by 600 thousand cwt.
during the sixties. The amount of cull potatoes increases
as total production increases which accounts for the
smaller decrease in supplies to these processors during the
sixties.

The total amount of potatoes used for feed and seed
and not marketed because of shrinkage and loss increased
over the past two decades. The amount of potatoes used as
seed decreased during the fifties by 3 million cwt.,
primarily because of improvements in potato seed during the
fifties. With the increase in total production during
the sixties, the amount of potatoes used for seed increased

by 1.7 million cwt. The quantity of potatoes used as feed

82

increased by 5 million cwt. during the fifties and 2
million cwt. during the sixties. The quantity of potatoes
not marketed due to shrinkage and loss also increased
during the fifties, by 1.3 million cwt. and the sixties by
7.8 million cwt. Potatoes used for feed and seed, which
are primarily cull potatoes, and potatoes not marketed
because of shrinkage and loss will tend to increase as
production increases. This accounts for the increases in
the amount of potatoes going to these uses in the sixties.

Procedure for Projecting Potato
Consumption and Utilization

 

 

To project the per capita consumption of fresh and
processed potatoes, linear trends were extrapolated to 1980.
The equation to determine the trend is of the familiar form,
Y = a + bT + e, where Y is the per capita consumption of
fresh or processed potatoes and T is time. A twenty year
sample period, 1950 - 1970, was used to determine the
trends. The coefficient of determination for the fresh
equation is 78 percent, which is acceptable, while the
coefficient of determination for the processed equation is
95 percent, which is quite good. The per capita consump—
tion of each of the various types of processed potato
products in 1980 is also determined by the extrapolation
of the linear trends. A11 coefficients of determination

are close to 85 percent. This same procedure can be used

83

to determine the supply of potatoes to the fresh market

and to each of the various types of processors. The pro-
jected quantity of potatoes used as feed and seed and not
marketed because of shrinkage and loss is the difference
between the projected total production and projected fresh
and processed potato usage. The projected total production

is determined by the demand-supply model discussed above.

Results and Interpretation

 

Before the results of the projections are dis-
cussed, a basic assumption of this type of projection pro-
cedure and implications of that assumption should be
pointed out. It is assumed that during the projection
period there will be no basic reversal of current consumer
tastes and preferences. Use of linear trends for projec-
tion purposes assumes that if consumer tastes and pre-
ferences changed during the sample period, they will change
in the same direction during the projection period. In
reality, there may be forces in the market which could
increase or decrease the tendency toward consumption of
processed potatoes. One of these forces may be the
consumerism movement. People may become disenchanted with
potato processors and return to the consumption of fresh
potatoes. Living patterns could drastically change during
the projection period which may change the consumption

patterns of consumers. Cooking innovations may be which

84

would reduce the preparation time of fresh potatoes may be
marketed. The microwave oven may be just such an innovation.
This oven would allow the consumer to prepare a baked
potato in a very short amount of time, which may have a
substantial affect on sales of fresh potatoes. Potato
processing costs could change substantially during the
projection period which would change the price of the
processed potato products at the retail level. This change
in prices could cause consumers to adjust consumption
patterns. Changes in retailing practices could also affect
consumer behavior. For example, the move toward standardi-
zation could have a tremendous impact on the consumption

of potato chips. Retail food stores may stop stocking
bagged potato chips which would drastically reduce the con-
sumption of this processed potato product. The move of
retail food storestxnstocking only high volume items could
also change consumption patterns. Unit pricing and changes
in merchandising practices could affect the consumers de-
cision of which potato product to purchase.

Using the procedure described above yields the
results given in Tables 12 and 13. As expected, per capita
consumption of fresh potatoes is projected to decrease
during the seventies and per capita consumption of pro-
cessed potatoes is projected to increase, with a net
result of an increase in total potato per capita consump-

tion of potatoes of 8.3 pounds per person.

85

TABLE 12.--Per Capita Consumption of Fresh and Processed
Potato Products in 1970 and Projected Per
Capita Consumption of Fresh and Processed
Potato Products in 1980.

_-_____.,‘_- ,._
___. ____-.__... ___._ a-.. .

 

1970 1980

(pounds) (pounds)
Total Potato
Consumption 118.6 126.9
Fresh 59.6 47.9
Processed 59.0 79.0
Frozen French
Fries 24.5 34.6
Other Frozen
Products 3.3 4.0
Chips 17.7 23.0
Dehydrated 12.8 16.6
Canned and
Flour .7 1.0

 

Source: American Potato Yearbook, 1971. Projections

 

Calculated by the Author.

86

TABLE l3.--Potato Utilization in 1970 and Projected
Potato Utilization in 1980.

 

 

1970 1980

(1000 th.) (1000 cwt.)
Total Production 324,801 389,780
Table Stock 129,242 112,697
Total Processed 136,137 191,363
Chips 35,861 53,655
Dehydrated 26,053 38,443
Frozen French
Fries 54,478 77,289
Other Frozen 7,381 . 8,588
Products
Starch, Flour
and Canned 12,364 13,388
Seed, Feed,
Shrinkage,
and Loss 60,209 85,720

 

 

Source: American Potato Yearbook, 1971. Projections
Calculated by the Author.

 

87

The per capita consumption of fresh potatoes is
projected to decrease by 11.7 pounds per person. Per
capita consumption of processed potato products during the
seventies is projected to increase by 20 pounds per person,
as compared to an increase of 35.3 pounds per person
during the sixties. The projected increase in the per
capita consumption of processed potato products during the
seventies is determined by the projected increases in per
capita consumption of the different types of processed
potato products. The per capita consumption of potato
chips is projected to increase from 17.7 pounds per person
in 1970 to 23 pounds per person in 1980, and increase
of 5.3 pounds per person. Per capita dehydrated potato
consumption is projected to increase by 3.8 pounds per
person during the seventies. Per capita consumption of
frozen potato products is projected to increase by the
largest amount of any processed potato product. It is
projected that per capita consumption of all frozen potato
products will be 38.4 pounds per person in 1980, an
increase of 10.6 pounds per person. Again frozen french
fries will be approximately 90 percent of the per capita
consumption of frozen potato products at 35.6 pounds per
person. Consumption of canned potatoes and potato flour
is projected to increase by .3 pounds per person during
the seventies.

These projections can also be examined from the

utilization side. Supplies to the fresh market are pro-

88

jected to continue to decrease during the seventies.

Total supplies to the fresh markets in 1970 were 129.2 mil-
lion cwt. and are projected to decrease to 112.7 million
cwt. by 1980. A11 supplies to processors of potato pro-
cessors of potato products are projected to increase
during the seventies. Supplies to potato chippers are
projected to increase by 12.3 million cwt. in 1980 from
35.9 million cwt. in 1970. This is an increase of 17.8
million cwt; Supplies to dehydrators are projected to
increase by 12.3 million cwt. during the seventies. Sup-
plies to potato freezers are projected to be 45 percent

of all potatoes supplied to processors in 1980. This is

a projected increase of 24 million cwt. during the
seventies. Supplies of potatoes to processors of potato
flour and starch and potato canners are also projected to
increase, reversing a downward trend of the fifties and
sixties. Potatoes used for feed and seed and potatoes not
marketed because of shrinkage and loss are also expected
to increase during the seventies. These increases are due
to the increased amount of cull potatoes produced as

total production increases.

CHAPTER IV

U. S. PRODUCTION PROJECTIONS BY
STATE AND SEASON IN 1980

Introduction

 

There has been a gradual shifting of potato pro-
duction over the past two decades. These shifts were
identified and possible factors leading to these shifts
were discussed in Chapter II. Potato production by state
and season is projected in this chapter to indicate pos—
sible future increases and decreases in potato production.
The projections are determined by extrapolation of trends
of time series data from 1955-1970. Also, factors which

may lead to these projected shifts are discussed.

Procedure

 

The extrapolation of the trend of production in
the nineteen major potato producing states to 1980 is the
procedure used to'project the production of potatoes in
1980. The extrapolation of trends was used above to pro-
ject the value of various variables to 1980. Besides the
linear trend, the logarithmic type of nonlinear trend is
used to predict potato production in the nineteen major

producing states. The equation for the linear trend,

89

9O

><
II

a + bT, is transformed into its exponential form,

Y = aTb, where Y is the quantity of potatoes produced in
a particular state and T is time. To calculate estimated
‘production values by ordinary least squares, the equation
is transformed into a linear form by taking the logarithm
of both sides, log Y = log a + b log T. The trend line
obtained from this equation is a straight line if graphed
on logarithmic scaled paper and curvilinear if graphed on
arithmetic scaled paper. The logarithmic trend line has
either a positive or negative slope throughout the 1955
to 1980 time range like the linear trend line, but unlike
the linear trend line, the amount of increase or decrease
is not constant.

The procedure used to project the value of the
dependent variable is the same procedure used in Chapter
III when dealing with trends. There is only a slight
adjustment necessary when the logarithmic equation is used.
To project the logarithm of the value of the dependent
variable, the logarithm of the independent variable time
must be taken. After calculating the projection, the anti-
logarithm must be taken to transform this projected value
from logarithms to natural numbers. Except for these two
small adjustments, the projection procedure is the same
as used above.

The criteria for the selection of the type of

trend to be used was a combination of the comparisons of

91

the relative sizes of the coefficients of determination

for the various trend equations for a particular state and
a visual and graphic extrapolation of the trend of the
observed data. The coefficient of determination, which
measures the goodness of fit of the estimated values to the
observed or actual values, is the first of the two criteria
for selection. The trend equation which yields the highest
coefficient of determination or the trend equation which
best fits the observed data is selected. The second cri-
terion for selection compares the projected trend with a
visual and graphic extrapolation of the trend of the
observed data. This criterion is used to make sure that
the estimated trend yields reasonable projection results.
Even though an equation has a high coefficient of determina-
tion, it could be rejected if the projected values are

not reasonable..13

Results
Before the results are examined, a major assump-
tion of this method of projection should be discussed.
Trends are based on historical data. Projections based on

historical data assume that what has happened in the past

—__¥

l3Various polynomial trend equations were compared
with the linear and logarithmic trend equations used. In
some cases the coefficient of determination was larger with
polynomial trend equations. But, in all cases, the pre-
dicted values were not within reasonable expectations.

92

will continue to happen in the future. This is not always
the case. The assumption disregards any environmental fac-
tors which may slow yield increases and any shortages which
may occur in land availability. The push of environmen-
talists to outlaw certain types of agricultural chemicals
may affect yields during the seventies. If replacements
for these fertilizers are not readily available, increases
in yields could be slowed. The lack of adequate amounts

of irrigation water is also a very important environmental
factor. Most major potato producing states use irrigation.
If the amount.of water used for irrigation is reduced,
yields could be drastically effected. The problem of ade-
quate amounts of irrigation water unless new sources are
discovered, will be most important during the late seven-
ties and early eighties in most major potato producing
states. Land availability could also become a problem
during the seventies. If land is not available to be con—
verted into potato production, projected increases in
potato production in some states may not be feasible.

The results of the trends of total potato produc—
tion by major producing states are shown in Table 14. The
1980 projected production in each state is compared to the
actual production of potatoes in each state in 1970. Of
the nineteen major potato producing states, eleven are
predicted to increase total potato production. The largest

projected increase in potato production will occur in

93

TABLE 14.--Total Potato Production in 1970 and Projected
Total Potato Production in 1980 by Major
Producing States.

 

 

__ ._ l...— -___..——a- __.__. . .
_b -——.—...—.-l-~- ~_—-..-_. - _..‘—_ __ ._.-_ . .~_._ - -

 

State 1970 Rank 1980 Rank
(1000 cwt.) (1000 cwt.)
Alabama 2,127 18 2,742 17
Arizona 2,712 17 3,568 16
California 29,760 4 28,000 4
Colorado 12,916 10 11,620 11
Florida 5,836 13 5,180 14
Idaho 74,550 1 99,155 1
Maine 35,700 2 35,759 3
Michigan 10,550 11 11,854 10
Minnesota 13,390 8 18,550 7
New Jersey 3,380 16 2,742 18
New York 16,977 6 14,076 9
North Carolina 2,093 19 1,341 19
North Dakota 17,550 5 21,143 5
Oregon 14,660 7 16,641 8
Pennsylvania 8,280 12 6,732 13
Texas 4,593 14 6,956 12
Washington 33,590 3 62,431 2
Wisconsin 13,028 9 19,770 6
Virginia 3,990 15 4,173 15
Total 306,461 372,449

 

Source: U.S. Department of Agriculture, Statistical Repor-
ting Service, Crop Reporting Board. Crop Production
(1972). Projections calculated by the author.

 

94

Washington, where projected 1980 production will double
that of 1970. The projections also indicate that Idaho
production will continue to increase rapidly. Potato pro-
duction in Idaho is projected to increase by 25 million
cwt. or 33 percent by 1980. The three most important
potato producing states in the Midwest, Wisconsin, North
Dakota and Minnesota, are also projected to increase total
potato production substantially. Wisconsin is projected
to increase production by approximately 6.5 million cwt.,
while Minnesota production is projected to expand by 5.2
million cwt. North Dakota production is projected to
increase from 17.5 million cwt. to 21.1 million cwt., an
increase of 3.6 million cwt. Increases in total potato
production of over a million cwt. are projected to occur
in Texas, Michigan, and Oregon. Potato production is pro-
jected to increase by 2.4 million cwt. or 52 percent in
Texas, 1.4 million cwt. in Oregon and 1.3 million cwt. or
11 percent in Michigan.

The remaining eight major potato producing states
are projected to decrease production by 1980, with the
exception of one state, Maine. Maine is the only state
where total production in 1980 is projected to remain at
its 1970 level of 35.7 million cwt. The largest projected
decrease in production will occur in New York, a decrease
of almost 3 million cwt. This is a reduction from 17

million cwt. in 1970 to 14 million cwt. in 1980.

95

California, Colorado, and Pennsylvania are also projected
to experience reductions in potato production of over one
million cwt. Potato production is projected to decrease
by 800 thousand cwt. in Florida and 700 thousand cwt. in
North Carolina._

Extrapolation of seasonal potato production trends
by major producing states for each season are shown in
Tables 15 through 20. Most states producing winter and
spring potatoes are projected to reduce production between
1970 and 1980. Both major producing states of winter
potatoes, California and Florida, are projected to decrease
production by 1980. Florida production is projected to
decline by 500 thousand cwt., while California produc-
tion is projected to decline by 650 thousand cwt. This
is a total decrease in production of winter potatoes of
1.2 million cwt. or 33 percent. Production of early
spring potatoes is also projected to decrease over the ten
year time span. Florida, the only state which produces
a large amount of early spring potatoes, is projected to
decrease productiOn by 230 thousand cwt. or 50 percent.
Three of the four states producing late spring potatoes,
California, North Carolina, and Alabama, are projected to
decrease production of late spring potatoes, with Califor-
nia experiencing the largest decline. The prodrction of
early spring potatoes in California is projected to decline

by 4.1 million cwt. In percentage terms this is a

96

TABLE 15.--Winter Potato Production in 1970 and Projected
Winter Potato Production in 1980 in Florida
and California.

 

-—--_ -, ”~-_—_.-- m .—.-—

 

State 1970 1980
(1000 cwt.) (1000 cwt.)

California 1955 1290

Florida 1627 1106

Total 3582 2396

 

Source: U. S. Department of Agriculture, Statistical Re-
porting Service, Crop Reporting Board, Crop
Production (1972). Projections calculated by
the author.

 

 

TABLE 16.--Early Spring Potato Production in 1970 and
Projected Early Spring Potato Production in
1980 in Florida.

W*—.._._. ..—-— __.__.~,-_ m
M. _ .__._. _._ _.__. .._..__..__-_-_. __... _ .________..,..___. .--.~~__.___._a

 

State 1970 1980
(1000 cwt.) (1000 cwt.)

Florida 4309 4074

Total 4309 4074

 

Source: U. S. Department of Agriculture, Statistical
Reporting Service, Crop Reporting Board, Crop
Production (1972). Projections calculated by
the author.

 

97

TABLE 17.—- Late Spring Potato Production in 1970 and
Projected Late Spring Potato Production in
1980 by Major Producing State.

 

 

State 1970 Rank 1980 Rank
(1000 cwt.) (1000 cwt.)
Alabama 1,027. 4 864 4
Arizona 2,712 2 3,568 2
California 14,516 1 10,457 1
North Carolina 1,747 3 ' 1,052 3
Total 20,005 15,941

 

Source: U. S. Department of Agriculture, Statistical Repor-
ting Service, Crop Reporting Board, Crop Production

 

(1972). Projections calculated by the author.

TABLE 18.--Early Summer Potato Production in 1970 and
Projected Early Summer Potato Production in
1980 in Texas, Virginia, and California.

 

 

State 1970 Rank 1980 Rank
(1000 cwt.) (1000 cwt.)
California 1,650 3 507 3
Texas 3,569 2 5,634 ‘ 1
Virginia [3,918 1 4,173 2
Total 9,137 10,314

 

Source: U. S. Department of Agriculture, Statistical
Reporting Service, Crop Reporting Board, Crop
Production, (1972). Projections calculated
by the author.

 

n

TABLE l9.-—Late Summer Potato Production in 1970 and
Projected Late Summer Potato Production in
1980 by Major Producing States.

 

-———.-—___ __,,___
~~_ - —_~ _.

98

a—_ ._ -.-..-._-_._.._

_....—..—-- ...— _. -.»_. .. .
-..- .— __.__ ...._._..__. .-. —— H- —‘—--

 

State 1970 Rank 1980 Rank
(1000 cwt.) (1000 cwt.)
California 2,135 6 1,914 7
Colorado 3,512 3 3,252 5
Michigan 2,268 5 3,321 4
Minnesota 1,950 8 3,365 3
New Jersey 3,207 4 2,742 6
New York 2,002. 7 1,193 8
Washington 8,580 1 9,031 1
Wisconsin 3,643 2 4,509 2
Total 27,297 29,327

 

Source: U.

 

by the author.

(1972).

S. Department of Agriculture, Statistical
Reporting Service, Crop Reporting Board,
Crop Production,

Projections calculated

99

 

 

 

 

TABLE 20.-- Fall Potato Production in 1970 and Fall
Potato Production in 1980 by Major Producing
State.
State 1970 Rank 1980 Rank
(1000 cwt.) (1000 cwt.)
California 9,504 8 13,832 7
Colorado 9,100 10 8,368 10
Idaho 74,660 1 ' 99,155 1
Maine 35,700 2 35,759 3
Michigan 7,566 12 8,346 11
Minnesota 11,440 7 15,185 6
New York 14,859 6 12,905 9
North Dakota 17,550 4 21,143 4
Oregon 15,229 5 16,641 5
Pennsylvania 8,280 11 6,732 12
Washington 25,010 3 53,400 2
Wisconsin 9,125 9 12,975 8
Total 238,023 304,441

 

Source: U. S. Department of Agriculture, Statistical
Reporting Service, Crop Reporting Board,
Crop Production,(1972). Projections Calculated
by the author.

 

100

projected reduction of 28 percent. The projected total
decline in production of late spring potatoes in both
North Carolina and Alabama is 860 thousand cwt. Arizona
is the only state producing late spring potatoes which

is projected to increase production during the seventies.
Arizona is projected to increase production by 800 thousand
cwt. or 29 percent. The projected increase in production
of late spring potatoes by Arizona is offset by the pro-
jected decrease in production in North Carolina and
Alabama, so total late spring potato production is pro-
jected to decline by 4 million cwt. or 20 percent.

Both summer and fall potato production are
projected to increase during the seventies. Thus, the
projected increase in total potato production will occur
because of projected increases in the production of sum-
mer and fall potato crops. Total production in early
summer is projected to increase because of the rapid
projected rise in early summer potato production in Texas.
Production in Texas is projected to increase by 2.1 mil-
lion cwt. or 60 percent by 1980. The projected small
increase in the production of early summer potatoes in
Virginia will also contribute to the increase in total
early summer potato production. Production is projected
to increase in Virginia by 200 thousand cwt. Production
of early summer potatoes in California is projected to

decline by 1.1 million cwt. The projected production

101

increase in Texas and Virginia outweighs the projected
decrease in California production leading to a projected
increase in total early summer production of 1.2 million
cwt., or 13 percent.

The total production of late summer potatoes is
also projected to increase during the seventies. Of the
eight states producing late summer potatoes, four states
are projected to increase production and four states are
projected to decrease production. The four states
increasing production are Minnesota, Michigan, Wisconsin,
and Washington. The largest increase in projected pro-
duction will occur in Minnesota, where production of late
summer potatoes is projected to increase by 1.4 million
cwt. Late summer Michigan potato production is projected
to increase from 2.2 million cwt. in 1970 to 3.3 million
cwt. in 1980, an increase of 1.1 million cwt. or 33 percent.
Wisconsin is projected to increase production by 900
thousand cwt. while Washington is projected to increase
production 500 thousand cwt. Late summer potato produc-
tion is projected'to decline in New York, New Jersey,
Colorado, and California. None of these projected
decreases exceed one million cwt. New York is projected
to experience the greatest reduction in late summer pro-
duction, a decline of about 40 percent. The second
largest reduction in potato production, 450 thousand cwt.,

is projected to occur in New Jersey. Late summer

 

102

production in Colorado and California is projected to
decline slightly. Total late summer production is pro-
jected to increase by 1 million cwt., primarily due to the
2.5 million cwt. increases in production in Minnesota and
Michigan.

The majority of the projected increase in total
U. S. potato production comes from the projected increase
in fall potato production. Of the twelve states which
produce fall potatoes only three are projected to decrease
production during the seventies and one is projected to
remain stable, that being Maine. The three states which
are projected to decrease production of fall potatoes are
New York, Pennsylvania, and Colorado. Production of fall
potatoes is projected to decline by 2 million cwt.,or 13.5
percent in New York. Pennsylvania and Colorado fall potato
production is projected to decline by 1.5 million cwt.
and 750 thousand cwt. respectively during the seventies.
The remaining eight states which produce fall potatoes
are projected to increase production by an amount which
far exceeds the projected decreases in production. The
largest increase in production is projected to occur in
Washington where production will increase by 100 percent.
Idaho produces primarily fall potatoes, so all of the pro-
jected increase in production will be absorbed in the fall

crop. Major increases in production of fall potatoes are

103

projected to occur in the major Midwestern potato pro-
ducers, Minnesota, North Dakota and Wisconsin. Fall
potato production in California is projected to increase
by 4.3 million cwt., while Oregon fall potato production
is projected to increase by 1.5 million cwt. Fall potato
production in Michigan is projected to increase from 7.5
million cwt. to 8.3 million cwt. or 800 thousand cwt.
Total fall potato production is projected to increase by
66.7 million cwt. or 28 percent during the seventies with
5 million cwt. of this projected increase in fall crop

production occurring in Washington and Idaho.

Interpretation of Results

 

The results of the projections show some definite
shifts in the amount of potatoes that will be produced in
a particular region of the country. Shifts are also
projected to occur in the amount of potatoes produced in
each of the six potato producing seasons. The reasons
for these shifts will be discussed in this section.

There has been and will continue to be a shift
away from potato production in the Northeastern portion of
the United States. Of the four major potato producing
states in this area, Maine, New York, Pennsylvania and New
Jersey, three are projected to experience a decline in
production during the seventies. The only state not

projected to decrease production, Maine, will produce

104

approximately the same amount of potatoes in 1980 as were
produced in 1970. There is a possibility that production
in Maine may decline.

Several of the factors which account for the pro-
jected decline in potato production in New York, New
Jersey and Pennsylvania are common to all three states.
One of the most important factors in all three states, as
well as many other major potato producing states, is the
increasing spread of urbanization. The Northeastern
section of the United States has the highest concentration
of population. Major cities in this area will continue
to grow through the seventies. Asthe cities grow, more
land will be taken out of agricultural production to be
used for residential, commercial and industrial construc-
tion. Couple this with the fact that major potato pro—
ducing areas in these states are located directly in the
path of this increasing urbanization, as shown on Map l,
and the stage is set for the acreage reductions.

Increasing urbanization leads to another factor
which will reduce‘potato acreage in these three states,
that of increasing land values. As the urban areas spread,
land that was once considered only for agricultural pur-
poses becomes a possible area for further urbanization.
With speculation that the land will soon be part of an

urban area, land values increase rapidly. The value of

105

 

American Potato Yearbook,

1972.

Source

 

Map l.--The Location of the Major Potato Producing Regions

in the United States

1972.

in

106

the land becomes so high that it is no longer economically
feasible to grow potatoes in the area.

A major reason for the large population in New York,
New Jersey, and Pennsylvania is the high degree of indus-
trialization in those states. People are needed to work
in the industrial plants, and large concentrations of
people leads to cities and urbanization. As the industries
in these states expand and multiply, more peOple will be
needed. The pressure to expand urban areas will be
increased by the needs of these industries. Industriali-
zation leads to urbanization which leads to higher land
values and the reduction of agricultural land.

Urbanization is a major problem in New York,along
with other factors. In New York, all late summer potatoes
are produced on Long Island. It is obvious that this area
will be hit hard by the spread of urbanization from New
York City and the increasing land value brought by this
urbanization. But, even Northern New York and Western
New York, where the state's fall potato crop is produced,
will feel the pressures of urbanization. Cities such as
Buffalo, Rochester, Syracuse, Utica, and Albany will be
expanding during the seventies right into potato producing
areas. Potato production in New York is also in competi—
tion with many other crops. As the acreage in high valued,
competing crops increases, the land available for potato

production will probably decline.

107

Pennsylvania faces the same type urbanization expan-
sion and high land values that New York faces, but the
urbanization and high land values will have more effect
on Pennsylvania production for two reasons. Pennsylvania
is more industrialized than Western and Northern New York
thus, the push toward urbanization will be greater in
Pennsylvania. Secondly, Pennsylvania is almost completely
mountainous. There are only selected areas in which any
crops can be grown. When land is lost to expanding urban
areas there is no new land to put into production and few
competing crops for which potatoes would be suitable as a
substitute crop. So, urbanization around Pennsylvania
cities such as Harrisburg, Lancaster, Philadelphia, Allen-
town and Scranton, located in the major potato producing
areas, ahs a major impact on potato production. Because
of the vast amount of minerals in Pennsylvania there will
always be the possibility that land now used to produce
potatoes may also be lost to the mining industry.

New Jersey is also faced with increasing land values
and urbanization.’ New Jersey, producing primarily late
summer potatoes, faces urbanization along the western bor-
der from cities of Trenton and Camden, a suburb of
Philadelphia. New Jersey is also a highly industrialized
state, which will speed up the spread of urbanization.

High land values along the Atlantic Coast will make it

108

economically unfeasible to produce potatoes in that area.
The area is used by people from metropolitan New York City
and metropolitan Philadelphia for recreational purposes.
New Jersey is also a large producer of vegetable crops
which compete with potatoes for available acreage. Com-
petition from higher valued vegetable crops will also be
a factor in reduced acreage available for potatoes. The
projected decline in production will be moderated somewhat
by the proximity to Philadelphia and New York City markets.
New Jersey late summer potatoes can enter these markets
before the large fall crop is marketed. Late summer pota-
toes from New Jersey fill the gap between earlier season
potatoes and fall potatoes. The availability of markets
will probably slow the rate of decline in potato production.
Maine is the lone state in the Northeastern region
of the United States in which it is not projected to
decrease production. Potato production in Maine has
remained relatively stable at 35 million cwt. since 1955.
Several reasons account for this stable production. Maine
does not have to face the loss of potato acreage to urbani—
zation and the highly increased land values which accompany
urbanization. All land in Maine that is suitable for pro-
duction of potatoes has been in production for many years.
But, there are very few crops which compete for this

acreage, which tends to offset the land availability

109

restriction. The location of seven potato freezers in
Maine has helped to maintain this stable production level.
The close proximity to major Eastern markets has also been
a factor which helped maintain production at the 35 mil-
lion cwt. level. The Maine potato has been able to com—
pete with the Idaho potato on the fresh market because of
lower transportation costs. If the transportation cost
differential declines, this competitive advantage will

be reduced.

There are several factors which tend to reduce
potato production in all of the Northeastern potato pro-
ducing states. The first of these is the transportation
cost differential, mentioned above. If this differential
becomes smaller, the competitive advantage held by the
Eastern states will decrease. This, combined with the
willingness of consumers to pay a higher price for Idaho
potatoes, because of the high quality image, will lead to
a decrease in demand for Northeastern fresh potatoes.
Another factor tending to decrease potato production is
lower yields in Northeastern states. Both the soil and
the climate are not as conducive to potato production in
the Eastern states as they are in Western potato producing
‘states. These smaller yield increases probably will not
compensate for the acreage decreases in these potato
producing states. Thus, the projected total production

will decrease.

110

All major potato producing states in the Midwestern
region of the United States, North Dakota, Minnesota,
Wisconsin, and Michigan, are projected to increase pro-
duction during the seventies. The three fastest increasing
potato producing states, North Dakota, Minnesota and Wis-
consinr have several factors in common which may cause this
increase in production. A major factor which will allow
for the increase in production is the lack of major com-
petition from other crops. A majority of the major pro—
ducing areas are located in the central to northern
portion of these states. There are very few crops which
can be grown in these areas dur to the weather. In North
Dakota and Minnesota potatoes and wheat are the primary
crops which are grown in these areas. In Wisconsin, the
competition for land comes primarily from dairy operations.
This is not a particular problem in Wisconsin because there
is land available for grazing which is not suitable for
potato production. So, land now producing potatoes need
not be taken out of production and used for grazing.
Urbanization and higher land values probably will not be
instrumental in reducing potato production in these states.
There may be some small losses in potato producing acreage
in Southeastern Wisconsin around Milwaukee and in Minne-
sota around Minneapolis-St. Paul. Another important
factor which will push production upward in these states

is the good potato producing climate. The potato

111

producing areas of North Dakota, Minnesota, and Wisconsin
all experience warm days and cool nights during the potato
growing season. This type climate yields a potato with a
higher percentage of solids. A potato with these character-
istics yields better when processed, so demand by processors
is higher. With processing becoming more important, the
demand for these higher yielding potatoes will increase.
This should give potato production in these three states

a boost. This increase in demand will be met by greater
yields and increased acreage by shifting some land into
potato production that is now not being used to produce
potatoes. As demand for these good processing potatoes
increases prices will increase. These increased prices

will cause land to be shifted to potato production. This
shifting of land will at least offset any loss of land due
to urbanization and higher land values.

Although the potato production is projected to
increase in Michigan, the increase is much smaller than the
projected production expansion in North Dakota, Minnesota,
and Wisconsin. There are a number of factors which will
tend to reduce the magnitude of the increase in production
of potatoes. Urbanization and increasing land values will
have a greater effect in Michigan. Potato producing areas
in the eastern portion of the state will be hurt by the

increase in size of Metropolitan Detroit, Flint and the

112

Saginaw Bay area. The high level of industrialization in
these cities will insure their continued growth and spread.
This could limit the number of acres devoted to potato
production in these eastern areas of the state. Michigan
also has a large number of recreational facilities. As
the large masses of population demand more recreational
facilities, the value of potato land near lakes and other
recreation areas will be bid up to a point where it will
no longer be economically feasible to produce potatoes.
Competition from other crops will also help slow the
increase in production of potatoes. Michigan winters are
not as severe as other Midwestern states because of the
lakes surrounding the state. The Great Lakes tend to
moderate temperatures. This more moderate climate allows
for the production of many different competing crops such
as dry beans, feed grains, and tree fruits. The potato
producer in Michigan has viable alternatives in production
which could influence the amount of potatoes produced.

The introduction of new potato producing land in the South-
west corner of Michigan could counteract the acreage
reductions in other parts of Michigan.14 Average yields
in Michigan are expected to continue to rise. The net

result of a possible reduction in acreage in Southeastern

 

Paul Wilkes, Procurement Manager for Ore-Ida Foods,
Inc. A personal interview at Greenfield, Michigan, in
September, 1971.

 

113

Michigan, possible acreage expansion in the Southwestern
part of the state and rising average yields should be a
slight expansion of Michigan potato production.

By 1980 the three major potato producing states in
the Northwest are projected to produce approximately half.
of the total amount of potatoes produced in the United
States as compared to 40 percent in 1970. None of the
major factors which will reduce potato producing acreage
in the East will affect Idaho. There is no threat of
acreage reductions due to urbanization and the rising land
values which accompany urbanization. Also, there are very
few other crops which compete for acreage in Idaho. A
major factor which will cause the rapid rise in potato
production is a continued increase in average yields. The
combination of the volcanic ash type soil, improved
varieties, improved production technology, and irrigation
will cause these continued increases, at least until the
late seventies. The lack of proper amounts of irrigation
water may cause a slowing of these increases in yield in
the late seventies. The climate during the growing season
is also conducive to producing a potato which is high in
total solids. The increasing demand for processed potatoes
and the demand by the large number of processors for this
type potato will put upward pressure on potato production
in Idaho. The establishment of the Idaho potato as the

tOp potato in the fresh market will also be a factor in

114

increased production. As transportation and storage
improve the marketing season for Idaho will expand. The
increased length of the marketing season will cause an
increase in the demand for fresh Idaho potatoes. Also,
there is a possibility that dehydrated potatoes could be
shipped to Japan. Opening of the export market to Idaho
would also exert upward pressure on potato production. All
these factors combined will increase the demand for Idaho
potatoes. Rising demand for Idaho potatoes will push
prices of potatoes in Idaho upward. As prices increase,
land which is now producing other crops will be switched
to potato production and land on which it is not now
economically feasible to produce potatoes will be brought
_into production.

The largest increase in production is projected to
occur in Washington. The increase in production of late
summer potatoes will be small because of the intense com-
petition for land in the Yakima Valley where the entire
late summer crop is grown. The majority of the tree crops
grown in Washington are grown in this valley. Land that
is taken out of potato production and put into the pro-
duction of tree crops is lost for the entire decade. The
relative prices between the competing crops in the valley
will determine the allocation of acreage. Increasing
yields and increasing demand for late summer potatoes will

offset any decrease in potato producing acreage. Yields

115

will increase in Washington as fast as they do in Idaho
because of similar soil conditions and irrigation practices.
Demand for fresh late summer Washington potatoes will con-
tinue to be strong as a carryover between earlier season
potatoes and fall potatoes in Western and Midwestern mar-
kets.- Also, as the amount of potatoes processed increases,
the demand for early processing stocks will increase.
Processing in the major factor which will cause production
of fall potatoes to rise. At the present time, 80 percent
of the Washington potato crop is processed. As the demand
for processed potato products increases and the demand
for fresh potatoes from the fall crop decreases, pro-
cessing will become even more important. The introduction
of the export market will also divert more potatoes to
processing. The high demand for the good processing potato
production in Washington will exert upward pressure on
potato production. The demand for increased potato supplies
will be met by increase yields and increased amounts of
land in potato production. As in Idaho, increased demand
will cause increased prices which will lead to the shifting
of land from production of less valuable competing crops
to the production of potatoes and reclaimation of land on
which it was not previously economically feasible to
produce potatoes.

Although Oregon is projected to increase production,

the increase is not nearly as large as increases in Idaho

 

116

and Washington. There are several factors which tend to
slow increases in production in Oregon. The potato
processing industry is not as well developed in Oregon as
it is in Washington and Idaho. The well developed pro-
cessing industries in Washington and Idaho have much to do
with the projected large increases in production. But in
Oregon, processing will not have such a strong upward
influence on production. Urbanization will also take some
potato producing land out of production. This urbaniza-
tion will come from the city of Portland, a fast growing
Western city. With urbanization comes increased land
values, but in Oregon, increased land values may effect all
potato producing areas, not just the area near Portland.
The demand for land in Oregon for recreational purposes
has increased and will continue to increase during the
seventies.

Two of the major potato producing states in the
West, California and Colorado, are projected to decrease
production of potatoes during the seventies. California
produces potatoes in every crop season except early spring.
Potato production is projected to decrease in all seasons
except fall. The majority of winter, lage spring and
early summer production occurs in the San Joaquin Valley.
The production of vegetable crops, citrus fruits and
grapes also takes place in this valley. Although a large

amount of land has recently been reclaimed along the

ll7

western edge of the valley, there will be little of this
done in the near future because of two major reasons.

One, the costs of reclaiming the land are greater than the
revenues which could be received from that land. Two,
there is a shortage of water for irrigation which is needed
to reclaim this land. So there is a large amount of com-
petition from competing crops for the existing acreage.
Also, with better transportation and longer storage of
fall potatoes the demand for fresh early season potatoes
from California will probably decrease. A strong proces-
sing industry would help support the production of
potatoes, but there are no dehydrators and no freezers in
this part of the state. As the demand for these early
season potatoes decreases, land will be converted into the
production of other, higher valued competing crops. There
are several other areas in Central and Southern California
which produce early season potatoes. All the factors just
discussed apply in these areas, plus acreage will be lost
due to urbanization and increased land values. Again land
values are rising’because of rising demand for land for
urbanization and recreational facilities. The increases
in fall potato production will occur in Northern California,
near the Oregon border. There will be continued expansion
of potato production in this area because of a climate and
soil which is similar to the major potato producing states

in the Northwest. At the present time there is a potato

118

freezer in the area and prospects look good for more pro-
cessors to locate in the area as the demand for processed
potato products increases. The increasing number of
processors will cause potato production to increase as the
demand for supplies increases.

A decrease in potato production is also projected
to occur in Colorado. The amount of land suitable for
potato production is limited in Colorado due to the moun-
tains. This limits any expansion of potato producing
acreage into areas not now in production. This is even
more important when it is realized that acreage now in
potato production will be reduced by higher land values
brought on by the increased demand for recreational facili-
ties in the state. Another factor which will tend to
cause this decline in production is the lack of a substan-
tial processing industry in Colorado. With a small pro-
cessing industry, most potatoes produced in Colorado must
be sold in the fresh market. But, the demand for fresh
potatoes is decreasing. The combination of these factors
will probably lead to the reduced production of potatoes
during the seventies.

The two potato producing states in the Southwest,
Arizona and Texas, are both projected to increase produc-
tion by 1980. Both states produce early season crops,
with Arizona producing all late spring potatoes and Texas

producing primarily early summer potatoes. The most

119

important factor which will allow for the predicted
increases in production is the availability of new land not
in production. With irrigation, this arid land is suitable
for potato production. Rising yields plus increased
acreage will cause increased potato production. Since

both states have almost no processing industry, the speed
of the increase in production depends on the demand for
early season fresh potatoes. Arizona will experience a
smaller increase in demand than Texas because better trans-
portation and longer storage of fresh fall potatoes will
reduce the need for fresh late spring potatoes more than
the need for fresh early summer potatoes. But, both seasons
will be hurt by the decrease in consumption of fresh pota-
toes. Also, Texas has a geographical advantage in the
major Texas and Midwestern markets.

Although two of the four major potato producing
states in the Southern region of the United States are pro-
jected to increase production of potatoes during the
seventies, there is a net projected decrease in production
in the region. Florida produces both winter and early
spring potatoes for fresh sale. Florida, along with all
the other major potato producing states in this region, will
experience a decreasing demand for these early season crops
with the decrease in the consumption of fresh potatoes and
the longer storage and better transportation of all pota-

toes for fresh consumption. There will be losses of

120

potato producing areas because of urbanization, increasing
land values and competition of other crops. Losses due

to urbanization will occur in potato producing areas near
Miami Beach and Fort Myers. Increasing land values will
be a factor in all potato producing areas. There will be
a continued high demand for land for recreational purposes.
Florida has a highly developed citrus industry which com-
petes actively for the limited amount of land available
for agricultural production. Acreage which is transferred
to the production of citrus fruits will be lost at least
until 1980.

Although total potato production in Alabama will
increase, the production of late spring potatoes is pro-
jected to decrease. The decrease in late spring potatoes
is due to the decrease in demand for fresh early season
potatoes and losses in acreage due to the Spread of urbani-
zation from Mobile. Decreases in production of late spring
potatoes will be balanced by increases in production of
other seasonal crops in the northern part of the state.
Production of potatoes give producers an alternative to
cotton production in this part of the state. Also, demand
for fresh early summer and fresh late summer potatoes will
not be affected as much as fresh late spring potatoes by
the better transportation and longer storage of fresh

fall potatoes.

121

North Carolina potato productionis projected to
decrease by 750 thousand cwt., the largest amount of any
potato producing state in the Southern region. The
decrease in demand for fresh early spring potatoes will be
an important factor which will depress production, but
competition for land from other crops will be the major
contributing factor in this production decrease. Tobacco
and peanuts are the major competing crops which are well
suited for production in North Carolina. As the produc-
tion of these high valued crops increases, land available
for potato production will decrease.

A very small increase in potato production is
projected to occur in Virginia, a producer of early summer
potatoes. Potato production in Virginia occurs in the
extreme Southwest corner of the state. Although there are
large amounts of tobacco and peanuts grown in Virginia,
this corner of the state is best suited for potato pro-
duction. The increase in production is small because
reductions in acreage due to urbanization tend to offset
any factors which would cause production to increase. Also,
increased land values, due to urbanization and increased
demand for recreational facilities along the Atlantic Ocean,
will reduce the amount of land available for potato pro-
duction. The close proximity to major Eastern markets will
keep demand for fresh early summer potatoes at a high

level. This will influence potato production upward,

122

and along with increased yields, tend to balance any

reductions in acreage.

CHAPTER V

SUMMARY AND CONCLUSIONS

There has been a major shift in potato utilization
due to shifts in potato consumption patterns during the
past two decades. Since the introduction of processed
potato products in the late fifties, there has been a rapid
shifting of potato consumption from fresh potatoes to pro-
cessed potato products. This shift in the consumption
pattern of consumers has been a major factor which reversed
the downward trend in per capita potato consumption.

The increased demand for processed potato products
has affected potato utilization and opened up new marketing
channels to producers. In 1950, almost all potatoes pro-
duced were shipped to the fresh market. In 1970, half
of the potatoes produced were shipped to the fresh market
and half to potato processors. This shows the shift in
potato utilization which has occurred during the fifties
and sixties.

As the demand for processed potato products con-
tinues to increase, the amount of potatoes supplied to
processors will continue to rise through the seventies.

The increasing supply of potatoes to processors has given

123

124

potato producers a viable alternative to the fresh mar-
ket. Potato producers now have a choice of different mar-
keting channels for their products, at least in states
with a high concentration of processing facilities.

Along with the shifts in potato consumption and
utilization, there has been a geographical and seasonal
shift in potato production. Production of winter, early
spring and late spring potatoes has decreased, production
of early summer and late summer potatoes has increased
slightly, and production of fall potatoes has increased
greatly during the past two decades. Improved storage and
transportation methods and rapidly rising processed potato
consumption are apparently the most important factors
stimulating seasonal shifts in production.

Potato production has shifted away from highly
populated states, which last potato producing acreage
because of the spread of urbanization, and states which
grow a number of competing crops. Production has shifted
to states which were better suited for potato production,
states which produce few competing crops and states which
did not face acreage losses due to urbanization. In geo-
graphical terms, potato production has been shifting from
the major potato producing states in the Northeast and
south to the major potato producing states in the Midwest,

Northwest, and, to a smaller degree, the Southwest.

125

The movements of raw potatoes indicates which
major potato producing states are the major suppliers for
each of the four major consumption regions in the United
States. The Eastern region receives the majority of its
supply of raw potatoes from four states, Maine, Idaho,
New York, and California. The smallest number of unloads
of raw potatoes occurs in the Southern region. The major
suppliers of the Southern region are Colorado, Idaho,
North Dakota, and Wisconsin. The Midwestern consumption
region receives a majority of the supply of raw potatoes
from Idaho, Michigan, Wisconsin, Minnesota, North Dakota,
and California. This largest number of unloads occurs in
this region. The final consumption region, the Western
region, unloads raw potatoes primarily from five states,
Colorado, California, Idaho, Oregon, and Washington.

The trends discussed above are projected to con-
tinue through the seventies, with certain effects on
consumers and marketing channels. Consumers will be eating
more fresh and processed potatoes which were produced in
the Northwest and’the Midwest. The projected decrease in
the consumption of fresh potatoes will primarily affect
the demand for local fresh potatoes, not the demand for
Idaho potatoes. As consumer incomes continue to rise
during the seventies more people will be able to afford
the higher priced fresh potatoes (Idaho), which has a qual-

ity image and is preferred to most local potatoes. Thus,

 

126

the share of the fresh market sold by Idaho will increase.
Consumers will be able to consume fresh fall potatoes for
a longer period of time as storage and transportation
improves during the seventies. The larger marketing season
for fresh fall potatoes will decrease the need for fresh
early season potatoes which fill in the gap between the
beginning and ending of the marketing season for fresh fall
potatoes. As consumption of processed potatoes products
increases during the seventies, consumption of potatoes
produced in the Northwest and Midwest will increase. A
growing majority of potato dehydrators and freezers are
located in the Northwest. Thus, as the consumption of
dehydrated and frozen potato products increases, the con-
sumption of potatoes produced in the Northwest will
increase. Also, as these processors search for other
sources of supply beyond the Northwestern region of the
United States, they will tend to locate in the major
potato producing states in the Midwest which have the
ability to produce a good processing potato and have rela-
tively cheap labor available. Thus, as more processors
locate in the Midwest, the consumption of processed
potatoes produced in the Midwest will increase.

A comparison of the projected total production
in 1980 determined by the recursive model and the pro-
jected total production in 1980 determined by the sum of

the projections of the production in the nineteen major

127

potato producing states to 1980 seems to indicate an
inconsistency. The sum of the projected potato production
in the nineteen major potato producing states is 372
million cwt. in 1980. The recursive model predicts pro-
duction to be 389 million cwt. in 1980. This is a dif-
ference of 17 million cwt. between the two predictions.
There are several possible explanations for this differen-
tial. The projected production of 372 million cwt. is the
sum of the projected production in the nineteen major
potato producing states only. This projection does not
include potato production in the other thirty-one states.
Although the nineteen states accounted for 94.5 percent

of the total potato production in 1970 and this percentage
is increasing as potato production declines in the other
thirty-one states, these nineteen states will not account
for all potato production in 1980. Thus, total production
will be higher than 372 million cwt. If these nineteen
states account for 97 percent of the total production in
1980, a reasonable figure given past increases in the
percent of total production accounted for by these states,
the production will be approximately 380 million cwt.
Assuming that the 389 million cwt. of potatoes are demanded
for all uses in 1980, an additional 9 million cwt. of
potatoes must be produced. These potatoes can be produced
by increasing production in one or more of the nineteen

major producing states or increasing production in one or

 

128

more of the states which border the major producing regions.
The most likely bordering states which would increase pro-
duction are Utah, Wyoming, and Montana because these

states have similar soil and climate which is conducive

to producing good processing potatoes. Also, these

states are reasonably close to a high concentration of
processors.

Another possible explanation for this apparent
inconsistency is the realization that the projection of
the amount of potatoes used as feed and seed and not mar-
keted because of shrinkage and loss may be too high. The
amount of potatoes used as feed and seed and the amount
not marketed because of shrinkage and loss during the
seventies is projected to increase by approximately 25
million cwt., with an increase in production of approxi-
mately 65 million cwt. Although this amount will increase
as production increases, this seems to be an unreasonable
increase, especially when compared to the prior decade.
Total production increased by approximately 65 million
cwt. during the sixties, but the amount of potatoes used
for feed and seed and not marketed because of shrinkage
and loss increased by approximately 12.5 million cwt.
Reducing this projected amount by 9 million cwt. would
reduce projected total production to 380 million cwt.,
which is consistent with the sum of the projected total

production from the total United States, assuming that

129

the nineteen major potato producing states produce 97
percent of the total production.

Finally, the projected amount of total pro-
duction from the recursive model may be too large if the
projected amount of potato acreage and/or the projected
yield is too large. The recursive model projects that
total potato acreage will be at approximately the same
level in 1980 as it was in 1970. If potato acreage is
reduced during the seventies because of urbanization,
increasing land values, competing crops and recreation in
specific major potato producing states, total potato pro-
duction would be lower than projected. Also, environmental
factors may show the actual yield increases during the
seventies. Again potato production would be lower than
projected by the recursive model.

In conclusion, the actual size of the differential
between the two predictions is not as large as it seems.
The nineteen major potato producing states will not pro-
duce all the potatoes in 1980. Assuming that the nineteen
major potato producing states produce 97 percent of the
total production in 1980, the total potato production would
be 380 million cwt. There is still a differential of 9
million cwt. between the two predictions. Total potato
acreage will probably decline during the seventies, so the
prediction yielded by the recursive model is probably

too high. Most of this reduction will be absorbed in

130

reduced amounts of potatoes used for feed and seed and not
marketed because of shrinkages and loss. Thus, potato
production in 1980 will probably be close to 380 million
cwt .

The projections used above are based on the assump-
tion that past shifts continue into the future. This may
not actually be the case. There are several factors which
could cause major structural shifts in potato production,
consumption and utilization. Factors such as increased
production losses due to environmental concerns, lack of
proper amounts of irrigation water and reduced amounts of
potato producing acreage could slow the increase in the
quantity of potatoes produced during the seventies. Factors
such as changes in living patterns, the consumerism move—
ment, changes in retailing practices, changes in the
retail price of processed potato products due to changes
in processing costs, and the development of new food pre-
paration products could change the trends in consumption

and utilization of the prior two decades.

BIBLIOGRAPHY

131

10.

132

BIBLIOGRAPHY

American Potato Yearbook. Various Issues. Stock
Plains, New Jersey: 1950-1972.

Croxton, Frederick E., and Crowden, Dudley J.
Applied General Statistics. Englewood Cliffs,
New Jersey: Prentice—Hall, Inc., 1955.

Hayenga, Marvin L.,and Hacklander, Duane. Monthly
Supply-Demand Relationships for Cattle and
Hog . Michigan State University, Agricul-
ture Experiment Station, Article No. 4884.

Bee, Olman. Demand and Price Analysis for Potatoes.
U. S. Department of Agriculture, Technical
Bulletin No. 1380. July, 1967.

Johnston, J. Econometric Methods. New York: McGraw—
Hill, 1963.

 

Krebs, Edward H.; Hayenga, Marvin L.; and Iehker,
John N. Various Price and Supply Control
Programs for Navy Beans : A Simulation
Analysis. Michigan State University, Agri-
culture Economics Report, No. 212. January,
1972. ' '

Kristianslund, Ivar. Analysis of Intraseasonal
Potato Price Movements. Unpubiished Ph.D.
Thesis, Michigan State University, 1971.

 

Mainker, Kenneth W. Factors That Affect Price and
Distribution of New Jersey Potatoes. New
Jersey Agriculture Experiment Station, Bul-
letin No. 786. June, 1957.

National Potato Council, Annual Statistical Report.
Denver, Colorado: 1971.

 

Simmons, Will M. An Economic Study of the U.S.
Potato Industry. U. S. Department of Agri-
culture, Agricultural Economics Report No. 6,
March, 1962.

 

11.

12.

13.

14.

15.

l6.

17.

18.

19.

20.

21.

133

Sullivan, G. H. The Potato Industry. Purdue

 

University, Agriculture Experiment Station,
Research Bulletin No. 862. 1968.

Department of Agriculture. Agricultural
Statistics. Washington: 1962, 1967, and 1972.

 

 

Department of Agriculture, Statistical Reporting
Service, CrOp Reporting Board. Crop
Production. Washington: 1972.

 

Department of Agriculture. Homemakers Pre-
ferences and Buying Practices for Selected
Potato, Rice and Wheat Products. Marketing
Research Report, No. 939. October, 1971.

 

Department of Agriculture, Consumer Marketing
Service, Fresh Fruit and Vegetable Division,
Market News Branch. Fresh Fruit and Vegetable
Unloads in Eastern Cities. Washington: 1971.

 

Department of Agriculture, Consumer Farketing
Service, Fresh Fruit and Vegetable Division,
Market News Branch. Fresh Fruit and Vegetable

Unloads in Midwestern Cities. Washington:
1971.

 

 

Department of Agriculture, Consumer Marketing
Service, Fresh Fruit and Vegetable Division,
Market News Branch. Fresh Fruit and Vegetable
Unloads in Southern Cities. Washington: 1971.

 

Department of Agriculture, Consumer Marketing
Service, Fresh Fruit and Vegetable Division,
Market News Branch. Fresh Fruit and Vegetable
Unloads in Western Cities. WashingTon: 1971.

 

Department of Agriculture, Economics Research
Service. Marketing and Transportation Situa-
tion, Various Issues. Washington: 1947—1971.

 

Department of Agriculture, Economic Research
Service. National Food Situation. Washington:
1972.

 

Department of Agriculture, Statistical Reporting
Service, CrOp Reporting Board. Potatoes and
Sweet Potatoes. Washington: 1966, l97l, 1972.

 

 

22.

23.

240

25.

26.

27.

134

U. S. Department of Agriculture. Potato Marketing

 

Report. September, 1972.

U. S. Department of Agriculture, Economics Research

Service. Vegetable Situation, Various
Issues. Washington: l952-l97l.

 

U. S. Department of Commerce, Social and Economic

Statistics Administration, Bureau of the
Census. Statistical Abstract of the United
States, Various Issues. Washington:
l952—l97l.

 

U. S. Department of Labor, Bureau of Labor Statistics.

Estimated Retail Food Prices by Cities,

 

Various Issues. Washington: 1950—1963.

Wonnacott, Ronald J., and Wonnacott, Thomas H.

Zusman,

Econometrics. New York: John Wiley and Sons,
Inc., 1970.

 

Pinhas. Econometric Analysis of the Market
for California Early Potatoes. Hilgardia
Val. 33, 1962.

 

 

 

A ~ H,
sq
”III
I