II B R.AR.Y

OF THL

UNIVERSITY
Of ILLINOIS

NOTICE: Return or renew all Library Materials! The Minimum Fee for
each Lost Book is $50.00.

The person charging this material is responsible for
its return to the library from which it was withdrawn
on or before the Latest Date stamped below.

Theft, mutilation, and underlining of books are reasons for discipli-
nary action and may result in dismissal from the University.
To renew call Telephone Center, 333-8400

UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN

DEC 2 0

?no4

L161— O-1096

The pricing and utilization of

LEGUME AND
GRASS SEEDS

By
WILLIAM HERR

and
G. L. JORDAN

Bulletin 582

UNIVERSITY OF ILLINOIS • AGRICULTURAL EXPERIMENT STATION

CONTENTS

PAGE

Seed-Producing Regions 5

Historical Analysis of Production, Acreage, Yield, and Prices 10

Factors Influencing the Annual Production of Forage Seeds 17

Seasonal Price Variations 19

Price Analysis of Six Forage Seed Crops 22

Demand factors affecting seed prices 22

Variations in demand 24

Supply factors affecting seed prices 25

Price analysis of five legume seeds 28

Price analysis of timothy seed 36

Geographical Price Areas 36

How geographical areas affect price analyses 40

Future studies desirable 42

Marketing Margins 42

Trends since 1 926 42

Relative variability of total margins 46

Size of total margin 47

Relationship of retail prices to prices received by farmers 47

Other forces affecting margin 48

Summary 48

ACKNOWLEDGMENT

The authors wish to acknowledge the cooperation received from G. C. Elder,
in charge of the Seed Section, Division of Field Crop Statistics of the Bureau of
Agricultural Economics; and from J. A. Swing and W. H. Ebling, agricultural
statisticians for Illinois and Wisconsin, who made suggestions and gave aid during
various phases of the project. Through Dr. Ebling's connection with the Field
Seed Institute of North America, some of the research done by this organization
was inspected and was helpful. Professor J. A. Jackobs, Department of Agronomy,
University of Illinois, made constructive suggestions which were incorporated into
the manuscript.

This report is based on a study financed in part by funds provided
by the Agricultural Marketing Act of 1946 (RMA, Title II)
under a contract with the Illinois Agricultural Experiment Station.

Urbana, Illinois December, 1954

Publications in the Bulletin series report the results of investigations made
or sponsored by the Experiment Station

THE PRICING AND UTILIZATION
OF LEGUME AND GRASS SEEDS

By WILLIAM HERB and G. L. JORDAN*

IN CHOOSING THE GRASS AND LEGUME SEEDS to be in-
cluded in this study, the first question was: "Which forage crops
are best adapted to the northeastern quarter of the United States —
or that area which includes Illinois?"

As can be seen from Fig. 1, there are five major areas of adaptation
for forage crops in the United States, with Area 1 extending over the
entire northeastern section. Following is a list of the grasses and
legumes which are adapted to this area and commonly grown there:2

I Grasses

Bluegrass, Canada Foxtail, Meadow

Kentucky Oatgrass, Tall

Bromegrass, Smooth Orchard grass

Canary grass, Reed Redtop

Fescue, Chewings Ryegrass, Common
Meadow Perennial

Red Sudan grass

I Tall Timothy

Legumes

Alfalfa Clover, Alsike

Birdsfoot trefoil Crimson

Lespedeza, Korean Ladino

Common Red

Sericea Sweet

Common white

To narrow the list still further, it was necessary to consider the
relative importance of the different seeds. This, of course, can be com-
puted on a number of different bases. It can, for example, be based on
the acres harvested for seed or the acres that the forage seed produced
will plant. The purpose that the seed will be used for also helps to
determine its importance. If we were primarily interested in soil con-
servation we would choose a different measure than if we were inter-
ested in land utilization.

1 WILLIAM HERR, formerly Assistant in Agricultural Economics, and G. L.
JORDAN, Professor of Agricultural Economics.

2U. S. Department of Agriculture Yearbook for 1948. pp. 743-752.

BULLETIN No. 582

[December,

INTRODUCED PLANTS
NORTHERN TYPE

WINTER ANNUALS EXCEPT
ON IRRIGATED LANDS

Areas of adaptation for grass and legume crops in the United States (from
U. S. Department of Agriculture Yearbook for 1948, page 743). (Fig. 1)

For this report, however, importance was determined on the basis
of money value, as value is the important consideration from both the
standpoint of the seed industry and that of the farmer. Table 1 shows
the average annual value of the seed production for forage crops that
are widely grown through the north central and northeastern states.1

From Table 1 it can be seen that the five legumes — red clover,
alfalfa, lespedeza, alsike clover, and sweet clover — followed by the
two grass seeds, Kentucky bluegrass and timothy, are most important
from the standpoint of money value. Since data on bluegrass seed are
limited, it could not be studied in much detail. This bulletin is there-
fore devoted primarily to a study of the other six seeds, dealing chiefly
with the economics of production, factors influencing prices received
by farmers, seasonal price movements, and marketing margins.

Other seeds which are increasing in importance, as indicated by a
much larger value in 1950 than during the previous 10 years and also
a larger value in 1951 than in 1950, include common ryegrass, crimson
clover, Ladino clover, and tall fescue. Studies of these seeds, while
desirable from the standpoint of their importance, are limited because
of lack of data.

1 Data for annual values, 1940-1951, are given in Table 1 of a mimeographed
supplement to this bulletin, which may be obtained by writing to the Informa-
tion Office, College of Agriculture, Urbana, 111.

1954] PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

Table 1. — Value of United States Field Seed Crops

(Limited to forage crops commonly used for hay and pasture)

Seed crops

Average value
of production,
1940-1949

Value of
production,
1950

(revised)

Value of
production,
1951

(revised)

Red clover

... 29,190

1,000 dollars
45,306

27,385

Alfalfa

24,806

39,243

47,551

Lespedeza

15,131

12,504

15,067

Alsike clover

5,002

4,734

5,074

Sweet clover

3,927

10,041

4,818

Timothy

3,464

6,433

2,841

Common ryegrass

2,475

5,288

5,463

Crimson clover

2,066

6,231

7,519

Redtop

1,998

4,319

1,700

Sudan grass

1,695

2,587

2,687

Ladino clover

1,474

9,706

12,902

White clover

1,408

S,158

3,412

Bromegrass

l,358a

4,605

1,633

Orchard grass

1,098

2,384

1,923

Perennial ryegrass .

518

750

762

Meadow fescue

139

218

109

Kentucky bluegrass

3,916

8,134

4,030

Tall fescue

2,854b

7,738

11,856

Birdsf oot trefoil

193C

516

480

• Eight-year average, 1942-1949.
b Two-year average, 1948-1949.
c Value for 1949 only.

It is to be noted that the value of the six seed crops included in this
study amounted to only one-third of 1 percent of the average U. S.
gross farm income during the 10 years, 1940-1949. Despite this small
fraction, they deserve attention because of the importance of forage
seeds both in the production of feed and the conservation of soil.

SEED-PRODUCING REGIONS1

The leading states during the 10-year period, 1939-1948, in the pro-
duction of the six forage seed crops studied are shown in Fig. 2.2

1 For this section to be most fruitful, the reader should refer to Frank Beck's
discussion of the location of seed-producing areas by decades in The Field Seed
Industry in the United States, chapter 2, University of Wisconsin Press, Madi-
son, 1944. The objective of the present discussion is to bring Dr. Beck's presenta-
tion up-to-date and to indicate any significant changes in the location of
producing regions by analysis of the decade from 1939 to 1948. As seed produc-
tion is highly variable in the short run, it is necessary to compute long-term
averages.

2 The decade 1939-1948 was chosen as it followed directly the decade 1929-
1938, which was the last one studied by Dr. Beck (see footnote 1). Averages
were calculated from the data given in Tables 2 to 7 of the supplement (see
footnote 1, page 4).

BULLETIN No. 582

[December,

1954}

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

T3 v
C >

rt 3

.2, o

h

X rt

'S oo"

c o

O -u

2 w _,
8 -a 6
> *> o

8 BULLETIN No. 582 [December,

Red clover and alsike clover.1 Production of red clover seed has
not shifted significantly from the corn belt, although there has been
some increase westward, especially to Kansas and Nebraska. There is,
however, a definite trend toward the Pacific Northwest and California
in the production of alsike clover seed, as can be seen by comparing
Figs. 2B and 2C. In 1929-1938 this area produced 18 percent of the
total alsike seed, while in 1939-1948 it produced over 46 percent.

Red clover seed is produced in the corn belt not because it yields
any better here than in the West, but because it is harvested at a
different time than most other crops, and thus offers an opportunity
to use labor and machinery that would otherwise be idle. Production
of red clover seed is therefore likely to be important in the corn belt
as long as red clover occupies an important position in the regular
rotation and is grown in relatively pure stands.

The situation can be explained further by analogy with oats pro-
duction in the corn belt. Despite competition from areas farther north
and west, where yields are better, oats are grown extensively through
the corn belt because there is little competition for the resources of
labor and capital during planting and harvesting.

There are several reasons why production of alsike clover seed is
shifting to the West. For one thing, it is generally grown in combina-
tion with other grasses and legumes in the corn belt, and hence is
not as readily adapted for commercial seed production as red clover.
And even when it is grown in relatively pure stands, the seed yield
is not so high in the corn belt as in the Northwest and California.

Another very good reason for the shift in production is the de-
creased demand in the Midwest. The custom of including alsike clover
as a stand insurance in mixtures, particularly on wet spots, is becoming
less common. It cannot compete with higher-yielding hay legumes,
especially since the increased use of lime has made soils better adapted
to alfalfa and red clover. Also, Ladino clover is being substituted for
it in many forage mixtures.

Sweet clover. The production area of sweet clover seed has not
changed significantly from the 1924-1928 period to the decade ending
in 1948. There have been shifts among the leading producing states,
but at present the ranking of the states has returned pretty much to the

1 For the earlier decades, Dr. Beck treated red and alsike clover as one, rec-
ognizing that such a procedure was not ideal. By doing this he concluded that
the two seed crops are grown in the same regions and are subject to the same
demand factors. While it may be true that the areas of consumption are for
practical purposes identical, the areas of seed production differ significantly.

1954} PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 9

1924-1928 average. From 1929 to 1938 the area of production was
concentrated in Minnesota and North Dakota, but recently production
has shifted somewhat south and west (Fig. 2D).

Timothy. The area of timothy seed production has been stable in
the last three decades, with Iowa, Missouri, Minnesota, Illinois, and
Ohio producing the major portion of the crop (Fig. 2E). The only
apparent change has been the continuing downward trend in produc-
tion. Of the major producing states, only Ohio produced more timothy
seed in the most recent decade (1939-1948) than in the previous
decade. Illinois production declined about 55 percent from 1929-1938
to 1939-1948, showing the largest relative decrease. Iowa showed the
largest decrease in absolute amount, producing about 290,000 bushels
less in 1939-1948 than in the previous decade.

Alfalfa. Commercial production of alfalfa seed continues to be
concentrated in the West (Fig. 2F). There have, however, been pro-
nounced changes in the ranking of the states, first because of insect
damage, and second because of a movement toward irrigated regions
throughout the West. Arizona, for example, has climbed up to third
place as a result of increased production in irrigated areas. Other
leading states, according to figures for the 1939-1948 decade, are
Nebraska, Kansas, Oklahoma, and Montana. Idaho, meanwhile, has
dropped from second place to tenth, its production in 1939-1948 being
just about half of what it was in 1929-1938.

California's production in 1939-1948 was about 50 percent more
than it had been in the previous decade. Production in California is
likely to become more important as the transporting of hardy varieties
to that state for seed production is increased and nonhardy varieties
are planted in the corn belt as soil-improvement crops.

Production in Michigan has also risen significantly — from prac-
tically nothing in 1929-1938 to nearly 60,000 bushels in the following
decade. There are two important reasons for this increase: one is the
higher price of alfalfa seed in comparison with competing crops —
mostly vegetables — that can be grown on the sandy soils of Mich-
igan. The other is the wider use of alfalfa in rotations. Also, the rise
in production coincides with the increased use of DDT and other
insecticides.

Actually, Minnesota and Wisconsin, as well as Michigan, have
increased their production of alfalfa seed over the previous decade.
Information as to the influence of the so-called all-crop harvester on
the growth of seed production in this area would be interesting. It is
ossible that alfalfa seed production in these states is taking on the

10 BULLETIN No. 582 [December,

pattern that red clover has in the corn belt (page 8), which is
that no other crops are seriously competing for labor and harvesting
equipment at harvest time and that the costs of these factors of
production are therefore comparatively low.

An opposing tendency is the fact that a seed crop reduces the
amount of hay produced. On dairy farms, hay is the important crop
except when seed prices are high in relation to milk prices. In recent
years (1948-1952), price relations have not favored the production of
alfalfa seed, and production has decreased. In the future, it is likely
that production in this area, while potentially significant, will vary
depending upon relative prices.

Lespedeza. Changes, if any, in production areas of lespedeza seed
are not readily determined. The leading producing states include the
area just north of the cotton belt and south of the corn belt (Fig. 2G).
There is some indication that the crop is increasing in importance
west of this area, in Kansas and Oklahoma.

HISTORICAL ANALYSIS OF PRODUCTION, ACREAGE,
YIELD, AND PRICES

The purpose of this part of the study is to analyze long-run factors
that have influenced the production and prices of the major field seeds.
This has been done by comparing series of seed production, yields, i
acreages harvested for hay and for seed, and price relatives.

The following three aspects of seed production and pricing were :
studied: (1) the relative importance of acreage harvested for seed in
determining total production; (2) rate of increase, or decrease, of
acreage harvested for hay and the respective rate of change in acreage
harvested for seed; and (3) comparison of the prices received for!
these seeds with prices received for all crops by farmers.1

Production, yield, acres harvested for seed, and acres harvested for "
hay are shown on ratio charts so comparisons of rates of growth, or
decline, can be made easily (Fig. 3).

Lespedeza. Of the six seeds studied, lespedeza showed the most
rapid growth in production during the 1919-1951 period (Fig. 3).
Actually, the really rapid growth occurred before 1939 — since that
time, production has increased slowly. The acreage harvested for hay
has followed pretty much the same pattern, with the leveling off
starting after 1942.

1 Data pertaining to these series are given for each seed in Tables 8 through
15 of the supplement (see footnote 1, page 4).

1954]

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

11

SCJNVSnOHi — 03J.S3AMVH S3MOW

W.TJ

6 .M
13 £

•o

cs

* cd

o

2
'oS

TD
C ^-,

rt •-'

§o" o"
to In

SQNVSnOHi— AVH «0d Q3iS3A«VH S3MOV

SQNnOd JO SQNVSnOHi— 033S JO NOI10na08d

SQNVSnOHi — 033S HOd Q31S3AMVH S3MOV

a §

•o I

<U C

v g
w °

S'*

of e»

1 1 1 i i i r

800 O

r^ ift m

S c

- di

c
S3

SONOOd
3MOV 83d OT3IA 033S

SONOOd JO SQNVSnOHi — 033S JO NOUOnOOMd

12

BULLETIN No. 582

[December,

SONVSnOHO. — Q33S UOJ 03J.S3AMVH S3UOV

I I I I I

o w

«r ^^

i- U

S> *

SONOOd JO SONVSnOHl — 033S JO

— 3HOV U3d Q13IA Q33S

SONVSnOHl— AVH aOJ Q3J.S3AyVH S3UOV

SQNVSnOHl— Q33S «0d 031S3AdVH S3UOV

— 380tf M3d QH3IA 033S -

O" O" o" o o"

CO (^ IO <f )O CM

SQNnod do soNvsnoHj.— Q33S do Nouonaoad

1954^

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

13

SONVSnOHl — 033S UOJ 031S3AUVH S3HOV

SONVSnOHl — AVH UOJ 031S3A«VH S3dOV

I I I i I i

s^

II
3 -3

~ T3 C

10 ij

•* O

o 6

-* 73

'o

T3

S13HSD9
— 3MOW d3d Q13IA Q33S

SONOOd JO SQNWSnOHl — Q33S JO NOIiOnQOMd

SQNVSnOHl— AVH dOd 03iS3AdVH S3UOV

SONVSnOHl — Q33S dOJ 03J.S3AdVH S3WOV

9 -S

10

C

5 «J

*3

S oo

8 "8

— 3HOV U3d 013IA 033S

SONHOd JO SQNVSnOHl — 033S JO NOIlOnQOdd

14 BULLETIN No. 582 [December,

Probably the decreasing rate of growth in seed production since
1939 is due to the continuing low price of seed compared with the price
of all crops. This response of production to changing relative prices
was caused by changes in acreages harvested for seed, as yields have
shown a steady upward trend during the entire period.

Imports and exports of lespedeza seed during the period were too
small to affect the trends noticeably and so are not considered here.

Sweet clover. Production of sweet clover seed decreased slightly
in the 1924-1951 period, with three major peaks occurring during that
time (Fig. 3). Acreage harvested for hay decreased noticeably from
1941 to 1947 (the latter was the last year the U. S. Department of
Agriculture reported acreage of sweet clover hay). One important
cause for this decline is the sweet clover weevil.

Partially offsetting the decline in use of sweet clover for hay is its
increased use in recent years for soil improvement. The demand for
sweet clover as a green-manure or catch crop would logically account
for the fact that the acreage harvested for seed (and hence total seed
production) has declined less rapidly than the acreage harvested for
hay.

When a crop is used for soil conservation, it is not so integral a
part of the farm business as when it is used for hay. Accordingly the
demand for sweet clover as a soil-improvement crop can be expected
to vary with farm income. As farm income rises, large quantities of
the seed will be bought for soil conservation, erosion control, and
maintenance of organic matter, for then farmers feel they can "afford"
the added expense. In times of falling farm incomes, this expense is
likely to be one of those eliminated early. Under these circumstances,
the change in the character of demand for this seed has caused the
income elasticity of demand to increase.

It would follow then, that if incomes of farmers fell materially
while the production of seed remained the same, the price of seed
would fall relative to the price of all crops. Conversely, the price
would rise as income went up. In spite of the greater fluctuations in
the prices of sweet clover seed, they still tend to follow the same
general trend as the price index of all crops.

For a detailed price analysis of this seed, data are needed showing
the quantity of sweet clover seed used for soil conservation or the
acreage of sweet clover devoted to this purpose. Another influence
which must be considered when analyzing the demand for this seed is
the cost of nitrogen in commercial fertilizer, as commercial nitrogen
and sweet clover seed compete with each other to a certain extent.

1954} PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 15

Yields per acre did not change significantly in the period studied,
except that from 1938 to 1947 they were generally lower than in later
or earlier years. One cause of this decline appears to be the large
acreage harvested for seed. That is, as acreage increased during this
period, it expanded through the use of land that was not too suitable
for the production of sweet clover seed.

Foreign trade in sweet clover seed is mainly confined to imports,
which, during the period studied, were equal to about 13 percent of the
total production. Since the depression years, imports have been
increasing, thus contributing to the situation already mentioned —
that supplies are larger than are necessary to maintain the acreage
of sweet clover hay.

Alfalfa. The trend in production of alfalfa seed has been upward
at the rate of about 4.0 percent a year; at the same time the acreage
harvested for hay increased at the rate of only 2.2 percent (Fig. 3).
This means that an increasing proportion of the crop is used for green
manure or pasture — assuming that the density of the hay stands and
the length of the rotation have remained the same. There is, however,
some evidence that many alfalfa stands are left down only two or three
years instead of the six or seven years common a decade or two ago.

The increased production of seed has been due primarily to in-
creased acreage, and this increased acreage has in turn been due to a
favorable price for alfalfa seed relative to the prices of all crops. The
additional acreage harvested for seed brought into use land which was
not likely to be very productive for seed. An example of this has
already been cited — the use of sandy land in the north central states
(page 9). As a result of using less productive land, yields per acre
tended to decline throughout the 30's and early 40's. There has, how-
ever, been a noticeable increase in yields since 1944, when they were
at their lowest level.

Imports have averaged about 11 percent of total production dur-
ing the period while exports have averaged only about 1 percent.
Imports fell from high levels in the 1920's to a negligible amount in
the early 1930's, but since 1935 they have risen steadily. The relatively
high price of alfalfa seed is partly responsible for this increase.

Alsike clover. There are no data to show the utilization of this
seed for hay. The significant fact is that production of this seed has
been fairly constant over the 32 years studied. The acreage harvested
for seed has had a downward trend, reflecting the movement of pro-
duction toward the Northwest, where higher yields can be obtained

16 BULLETIN No. 582 [December,

(page 8). Yields per acre have shown a fairly steady growth since
the mid-1920's (Fig. 3).

At the same time the price of alsike clover seed declined relative
to the price of all crops. This factor has also contributed to the
decline in acreage, especially in regions such as the corn belt, where
yields are relatively low.

Foreign trade in alsike clover seed has not been so important in
recent years as it was in the 1920's. From 1919 to 1951 imports
averaged about 15 percent of production, fluctuating between zero and
65 percent.

Timothy and red clover. Because the harvested acreages of timothy
and red clover hay are reported as one crop, the seeds of these crops
are discussed together. The trend in production of timothy seed has
been decidedly downward since 1919, while the trend in production of
red clover seed has been upward (Fig. 3). The acreage of clover and
timothy harvested for hay has shown a downward trend. Since pro-
duction of timothy seed has declined at a faster rate than acreage of
harvested hay, it is reasonable to conclude that the percentage of red
clover in the hay mixture is increasing. In fact, there has been a shift
toward stands of pure red clover hay.

For timothy seed there has been no significant trend in yield. Red
clover seed yields, however, declined from 1940 to 1946. The increase
in acreage harvested for seed during the war years tended to result in
lower yields as areas were drawn into production which did not have
the greatest advantage for growing this crop. Since 1946 yields have
increased and are now back to normal.

Prices of red clover seed followed the fluctuations of the prices of
all crops but tended to average higher in 1926-1928 and in 1934-1936.
Timothy seed prices have also followed the index for all crops rather
closely, with these exceptions: In 1934 and 1949, when production was
unusually low, price relatives increased to 424 and 554 respectively.
And since World War II, prices have declined relative to prices of all
crops. This decline reflects the decreasing demand for this seed, which
is due to the declining horse population and the introduction of superior
grasses that are replacing timothy in the hay mixture. The character
of these factors indicates that production of timothy seed will con-
tinue downward for some years to come.

The importance of foreign trade for red clover seed has varied
tremendously over the period studied. In 1923, imports were more
than 50 percent of the year's production, while in 1932, for all prac-
tical purposes there were no imports. Over the period studied, imports

1954^ PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 17

averaged about 10 percent of total production; exports averaged less
than 2 percent.

Little timothy seed has been imported except in 1934, a year of
very low production, when imports were almost 40 percent of total pro-
duction. Even in 1949, also a year of low production, imports were
less than 4 percent. Timothy, however, is the chief seed crop exported,
and from 1922 to 1950 exports averaged over 15 percent of total pro-
duction.

FACTORS INFLUENCING THE ANNUAL
PRODUCTION OF FORAGE SEEDS

Variability in production of major crops is due mainly to fluctua-
tions in yield.1 But for the six seed crops studied fluctuations in
acreage were more important than yield in causing major changes in
production (Table 2). Yield, however, is more important than might
at first seem apparent from Table 2, for fluctuations in acreage are
caused partly by high or low prospective yields. When yields are
expected to be low, the acreage harvested decreases because the only
areas harvested are those with yields high enough to pay costs. But

Table 2. — Average Changes From the Preceding Year in Production,
Acreage Harvested, and Yield of Seed Crops, United States, 1931-1950"

Seed croo Average change Average change Average change

in production in yield in acreage

Alfalfa

21.77

percent
12 46

22 25

Red clover. . .

42 53

12 90

53 18

Sweet clover . .

22 45

12 02

25 51

Alsike clover

22.71

12.99

25 87

Lespedeza

39.77

11.55

27 17

Timothy

. . . . 117.81

22 37

61 30

* Calculated by averaging, for the 20-year period, the percentage by which production, yield,
or acreage changed each year from the preceding year.

if the seed crop promises to yield high, more acreage is likely to be
harvested. As a result, we are unable to determine the total effect of
changes in yield on total production.

If we could predict changes in acreage harvested for seed we could
estimate with some degree of certainty the total production of seed
for a coming year. (This does not, of course, hold true of other crops
as yields fluctuate and are not readily predictable.) It is therefore

1 Waite, Warren C., and Trelogan, Harry C. Agricultural market prices, 2d
ed. John Wiley and Sons, Inc., New York, 1951. p. 65.

18 BULLETIN No. 582

important to determine the factors that influence farmers in deciding
what acreage to harvest.

As already mentioned, prospective yield is important in determin-
ing changes in acreage. It would seem likely that prospective price
is also a deciding factor, and that the previous season's price might be
a measure of what farmers would expect the next year's price to be.

Studies were accordingly made with red clover to indicate the
extent to which the variation in acreage harvested may be explained
by the previous season's price. The acreage of red clover harvested
for seed in Illinois was expressed as a percentage of the trend value.
The purchasing power of red clover seed was calculated by dividing the
price relative of red clover seed in Illinois by the Illinois price index
of feed grains and hay (indexes adjusted to 1935-1939 = 100).

If the acreage of red clover seed harvested is dependent upon the
purchasing power of last year's crop, there should be some correlation
between the two. But inspection of a scatter using the purchasing
power as the independent variable shows little correlation between this
variable and the acreage harvested for seed the next year. If the
residuals from this regression were plotted with red clover seed yield
as a second independent variable, one would expect that a higher
percentage of the variability of acreage harvested would be explained.
This, however, was not the case.

One reason for these negative results is that previous season's price
is not the same thing as prospective price; nor is yield received the
same as anticipated yield.

Studies made at Purdue University indicate that the acreages of
other crops are affected by the previous season's price, although the
changes may be very small. Some evidence was found that truck and
special crops respond more to this factor than other crops, but even
so, only a small percent of total year-to-year variation in acreage
could be explained by previous season's prices or purchasing power of
the commodity.1

Other things no doubt affect acreage harvested to a greater extent
than the purchasing power of seed the previous year. But, like pro-
spective yield and prospective price, some of these factors are not
statistically measurable. Significant statistical results may be obtain-
able with other factors, but an intensive study would have to be
undertaken. Because of the characteristics of field seed crops as
opposed to other major field crops, we believe it may be possible
to predict production more accurately than is possible for other crops.

1 Kohls, R. L., and Paarlberg, Don. The short-time response of agricultural
production to price and other factors. Ind. Agr. Exp. Sta. Bui. 555. p. 7.

1954^ PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 19

SEASONAL PRICE VARIATIONS

Seasonal price indexes were calculated for five of the seed crops
studied — not enough data being available to calculate an index for
the price of alsike clover seed. Prices received by Illinois farmers were
used for calculating indexes for red clover, timothy, sweet clover,
and lespedeza; United States average prices were used for common
alfalfa.

The prices of all five crops reached their peaks sometime during
the spring and their low points in late summer or fall (Table 3 and Fig.
4). The variations in price were particularly pronounced for sweet
clover, timothy, and lespedeza. The bulk of the seed for these crops is
sown in the spring and harvested in late summer or early fall. Thus
demand is strongest in the spring, after which it declines until fall,
when there is a slight upturn, depending on the extent of fall seeding.
The supply of the seed is largely determined at harvest time although
minor changes occur during the year, depending on exports and im-
ports. The net result of these forces is that the strong demand in spring
raises the price in this season, after which the approach of a new crop,
together with the decrease in demand, causes prices to fall. The
seasonal low for lespedeza comes later than for the other seed crops as
the bulk of the seed is produced farther south and harvested somewhat
later.

As the indexes for alfalfa seed were based on United States prices,
they would be expected to fluctuate less than the indexes for other
seeds. Planting and harvest dates vary over the country, so that their
effects tend to cancel each other. In the South, alfalfa is often fall-
sown, while farther north it is more likely to be spring-sown in oats.
Similarly harvest dates would not be the same in Montana, for
example, as in Arizona.

Inspection of monthly seed prices in Illinois each year indicates
that variation is greater during the months of harvest and immedi-
ately thereafter than during the spring. This is due to the uncertainty

Table 3. —

Monthly Price

Indexes of Five Forage

Seeds*

Crop

Jan.

Feb.

March April

May

June

July

Aug.

Sept.

Oct.

Nov.

Dec.

Sweet clover ....
Timothy

. 101
. . 104

107
110
101
107
99

115 121
116 116
110 117
108 111
104 107

115
118
114
108
108

111
109
110
103
105

101
84
106
99
105

83
79
98
86
100

81
86
94
87
99

84
90
89
92
91

87
92
81
95
92

94
96
87
99
94

Lespedeza

94

104

Alfalfa

. . 97

a Based on Illinois price received except for alfalfa, which is based on U. S. price received.
Indexes for sweet clover, timothy, and red clover calculated by ratio to 12-month moving average
on basis of July, 1934-June, 1950, omitting July, 1941-June, 1947; those for alfalfa calculated in
same way but based on July, 1931-June, 1951, omitting July, 1941-June, 1947; those for lespedeza
calculated by link relative method for January, 1937 -December, 1950, omitting 1942-1946.

20

BULLETIN No. 582

INDEX
120

110
100
90
80
120
110
100
90
80
120
110
100
90
80
110
100
90
110
100
90
80

LESPEDEZA

J I I

TIMOTHY

i iii

RED CLOVER

JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT NOV DEC

Seasonal price indexes in Illinois of lespedeza, timothy, sweet clover, and
red clover seeds, and United States seasonal index of alfalfa seed. (For
periods that indexes are based on, see footnote to Table 3.) (Fig- 4)

1954] PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 21

of both supply and demand at the beginning of a new season and
the corresponding price risk to anybody buying seed in the fall and
holding it until spring.

The large fluctuations between the spring and the fall are due
partly to the fact that there are no futures markets or other effective
devices for hedging price risk. By assuming the price risk and other
storage costs from the date of harvest until seed is sold in the spring,
dealers discount the price to the farmers when they buy seed in the
fall. Since a large proportion of seed is sold to dealers soon after
harvest, it is apparent that they are the ones who benefit from the
usual seasonal price rise.

The extent of seasonal price fluctuations in Illinois indicates that
farmers with seed to sell would do well to store it until spring. Al-
though exact storage costs are not known for the seeds included in this
study, it is recognized that the costs are fairly low as legume seeds
can be stored for two years and grass seeds for three years with a
minimum amount of care, provided they are in a dry shed that is
free from insects and rodents. At least, farmers who can meet these
storage conditions can generally afford to store their seed crops until
spring, if not longer.

Insofar as retail prices for seed follow the same general pat-
tern as the prices received by farmers, those farmers who intend to buy
seed for spring sowing would do well to buy it early and store it until
planting time — provided, of course, clean seed is available. Examina-
tion of retail prices for the four crop seasons of 1950 through 1953
indicates that in three of the four years, farmers would have saved
about $1.90 per 100 pounds of timothy seed by buying it in the pre-
ceding September rather than in March. Buying red clover seed, they
would have been ahead two of the four years, for an average annual
savings of approximately $1.75 per 100 pounds. Data for alfalfa seed
do not all point in the same direction. For improved varieties, the
average annual savings may have been significant, while for common
alfalfa seed there may have been an actual loss.

Farmers cannot always buy early, since much of the seed is not
processed and ready to sell until shortly before planting time. Fur-
thermore, even when they can get seed early, they may not be con-
vinced of the advantage of doing so. The annual cost of forage seeds
is a relatively small cash outlay on many farms and hence farmers
are not likely to scrutinize this cash expenditure as carefully as they
do their feed or labor bills. Farmers who grow these crops for seed are
more likely to be interested in seasonal patterns as the price received
may be a fairly important element in their incomes.

22 BULLETIN No. 582 [December,

PRICE ANALYSIS OF SIX FORAGE SEED CROPS

Demand factors affecting seed prices

A number of factors affect the demand for forage seeds and hence
the price. These include livestock numbers, crop rotations, gross farm
income, and, to a limited extent, personal income. From these factors
it was necessary to choose one that could be used in the price analysis
as a measure of the demand for forage seeds. Such a factor would have
to be both measurable mathematically and reasonably accurate. Anal-
ysis of the different factors led to the selection of gross farm income
as the best available measure of demand. Following are brief analyses
of the major factors affecting seed demand.

Livestock numbers. It is commonly supposed that the demand for
field seeds is closely related to the demand for hay and pasture to
feed livestock. As shown by correlation studies by Henry Schultz,1 the
demand for hay in turn depends on the price of hay, the price of other
feeds, and the number of animal units. Changes in cattle numbers are
followed, with appropriate lags, by changes in the production of hay.
Similarly, Gordon Butler has shown that the amount of acreage
harvested for hay depends largely on the number of hay-consuming
livestock units.2

These studies indicate that changes in the number of forage-con-
suming animal units might be a major factor in the demand for field
seeds. Before giving this factor too much weight, however, it must be
remembered that Schultz and Butler were both dealing with harvested
acreage of hay, not with the entire acreage planted for forage. Harvest-
ing a hay crop to feed a known quantity of livestock during the
coming winter is a different matter from planting seed to be used for
hay several years in the future. The amount of forage seed planted in
any one year may be more closely related to the type of rotation
followed than to livestock numbers.

1 Schultz, Henry. The theory and measurement of demand. Ch. 9: The de-
mand for hay. University of Chicago Press, 1938. (Karl Fox in an analysis of
factors affecting the price of hay included in the demand factor cash receipts
from sales of dairy products and beef cattle weighted in proportion to hay con-
sumption. To the extent that cash receipts are related to total numbers of cattle
in these two classes, this demand index also includes numbers of forage-consuming
livestock. See Fox's article, "Factors affecting farm income, farm prices, and food
consumption," in Agricultural Economics Research, U. S. Dept. Agr. Bur. Agr.
Econ., July 1951.)

2 Butler, Gordon Gray. Changes in feed crop acreages. Field Seed Institute of
North America, Madison, Wis., June 1943 (processed).

1954} PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 23

Length and stability of rotation. The six seeds studied are con-
sumed largely in the northeastern and central portions of the country,
where some kind of rotation is generally followed. As acreage planted
for forage is related to the rotation, a major factor in determining the
quantity of seed demanded is the length and stability of the rotation.

A three-year rotation such as a 1-1-1 (intertilled crops-small grain-
legumes) will require forage seed for 33 percent of the land each year,
while a 2-1-2 rotation will require seed for 20 percent of the land.
Red clover and timothy seed are in greater demand for short rotations
than long ones; while alfalfa and smooth bromegrass are used more in
long rotations.

Once a rotation is established, it is not easily changed — usually
sudden alterations come only as the result of adverse weather condi-
tions or major economic changes, although there may be gradual
changes resulting from technological developments in the use of
legumes, especially for soil improvement. The demand for forage seed
should therefore be relatively constant, reflecting the stability of the
rotation.

Disposable personal income. As most farm products are ultimately
consumed for food or clothing, the measure of demand used for them
has often been disposable personal income. This, however, was not
considered a good measure of the demand for legume and grass seed.

For one thing, seeds are not sold as a final product to consumers;
nor are they sold as raw material to an industry to be converted
directly into consumable products. Legume and grass seeds are
sold to farmers primarily to grow hay or pasture to feed stock
which in turn may be sold for consumption or used as breeding stock
to produce more stock. As the demand for hay and forage seeds is a
derived demand, time lags can be expected between changes in demand
for animal products and in the demand for hay and forage seed. These
lags will shift over time depending on the type of change: When
cattle numbers are being built up, a different lag will probably be
involved than when cattle numbers are being reduced. It would not be
easy to adjust data to these changing conditions.

In addition, it must be remembered that some legume and grass
seeds are used for soil-improvement purposes and that disposable
personal income would not reflect this demand.

Gross farm income was considered to be a better measure of the
demand for forage seed than disposable personal income.

One reason is that it reflects the profitableness of the present farm-
ing system, including the rotation. As already mentioned, the rotation,

24 BULLETIN No. 582 [December,

once it is established, is usually changed only after major shifts in
income. As long as the gross farm income shows only relatively small
year-to-year changes, the demand for forage seed should remain about
the same, provided there are no pronounced technological changes that
would affect the amount of legumes in the rotation.

This measure of demand also includes some indication of the
demand for seed for soil-improvement purposes. Farmers are most
likely to use seeds for soil improvement when their incomes are high.
As incomes decline they are likely to increase production of other
crops and cut their expenses by buying less seed for soil improvement.

As income from field seeds is included in gross farm income, some
intercorrelation between farmers' income and prices of seeds occurs.
The income from seed, however, makes up only a small fraction of
total income.

Variations in demand

So far we have considered general measures of the demand for
forage seeds throughout the country. However, the demand picture
in a particular area or for a particular kind of seed may differ from
the general demand of the country as a whole.

Variations among areas. In the dairy area of the Northeast, the
number of dairy cows fluctuates much less than do cattle numbers
in the Plains area and the fattening areas of the corn belt. In the West,
changes in price and fluctuations in weather and feed supplies cause
large movements of cattle from one area to another within a short
period of time. And even without such changes, the time required to
fatten cattle for market is much shorter than the productive period
for dairy cows. The result is that demand for forage in the West under-
goes periodic changes while the Northeast has more of a constantly
increasing demand. The increase in population, if nothing else, would
tend to increase the demand for milk.

Sudden changes in weather and available feed cannot, of course,
be forecast, but they do condition responses of farmers in later periods.
A drouth in the Plains states will cause feeder calves to be shipped
out to other areas with more favorable climates. As a result, demand
for field seeds increases not only in the drouth areas but also in the
area with more favorable climate, as pastures may be overgrazed and
need to be reseeded the following year.

Variations among different seeds. The demand for a particular
seed may not be adequately represented by an index of demand suit-
able for other forage seed crops. This may be true of sweet clover, for

1954} PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 25

example, which is utilized in large quantities for soil improvement
as well as for forage.

Another problem that must be considered is the substitution
possibilities between the various legume seeds. Seedsmen recognize that
the various seeds are competitive and consequently price them in
relation to one another. Hence prices are adjusted in accordance with
the quantities not only of the particular seed but also of competing
seeds. In some areas, the five major legume seeds all compete with
one another. But climate, soil conditions, and disease problems limit
the adaptability of some legumes to different regions. Only one or two
of the major legume seeds may be adapted to a particular area and
they may compete with a legume that is minor in the country as a
whole but that is of great importance in the area. Such a situation
prevails to a certain extent in parts of the Northeast, where birdsfoot
trefoil may compete with alfalfa and red clover. As a result, the prices
of the two major legume seeds will operate differently than when
they are in direct competition.

Supply factors affecting seed prices

Domestic production accounts for most of the supply1 of seeds. The
importance of foreign trade varies from year to year. Timothy is the
only seed studied which is exported2 in very large amounts, but all
seeds except lespedeza have been imported in significant quantities in
particular years (Table 4) .

Since seeds can be easily stored from one year to the next, stocks
are important. However, data on seed carry-in before 1939 are limited,
and the figures obtained since then do not cover a long enough time to
allow a valid analysis. Table 4 shows the average percentage that
carry-ins were of production and also the range in percentage from
1939 to 1951. 3 The importance of carry-ins is indicated by the fact
that in 1948 the carry-in of timothy seed was greater than the pro-
duction.

Since results from correlations to determine the factors affecting
prices are sometimes more significant when consumption series are
used, it would be desirable to derive these data. However, the fact that
stocks of seeds before 1939 are not known poses a difficulty.

1 "Supply" as here used refers to quantity : It is not used in the sense of a
supply schedule.

2 While exports often represent a measure of demand, in this study they have
been considered as a deduction from total quantity available to domestic
consumers.

3 Complete data for each seed for each year are given in Table 16 of the
supplement (see footnote 1, page 4).

26 BULLETIN No. 582 [December,

Table 4. — Percentages That Exports, Imports, and Carry-in*
Were of Seed Production for Six Forage Seed Crops

(Period covered for carry-in was 1939-1951 ; for imports, 1919-1951 unless
otherwise specified; for exports, 1922-1951)

Percentage imports Percentage exports Percentage carry-in
Seed crop were of production were of production was of production

Range Average Range Average Range Average

Alfalfa.

0 1 — 94

11
1
10
15
13b

0.5—5
2—55
0.3—11

1
15

2

3—21
1—124
1—31
2—32
5—38
1—20

11
38
10
13
15
7

Timothy

0—40

Red clover

0—52

Alsike clover. . . .

0—65

Sweet clover. .

0 — 79b

Lespedeza .

a Percentages for carry-in were calculated from Table 16 of the supplement (see footnote
1, page 4).

b From 1923 to 1951.

One possible method of deriving consumption series would be to
take the average annual harvested acreage of a crop — alfalfa hay,
for example, and divide it by the average number of years in the
rotation. This would give the number of acres planted for the year.
Multiplying this by the average rate of planting per acre would give
the total alfalfa seed planted for hay for the year. It is obvious, how-
ever, that averages for the number of years in the rotation and the
average planting rate are practically meaningless when they cover wide
areas, even so much as one state. Also, seed planted but not harvested
for hay is not included. It is therefore believed that any such con-
sumption series would introduce less reliable data than those obtained
from a supply concept which does not include changes in stocks.

A second alternative for deriving consumption data would be to
relate consumption to a supply or production series derived from the
data since 1939. Domestic consumption would equal (production +
carry-in + imports) — (carry-over + exports). A graph of consump-
tion data of the five major legume seeds derived in this way is shown
in Fig. 5. Inspection of this graph shows that consumption tends to
follow the fluctuations in production as well as in total available sup-
ply, although the relationship between these series is not constant from
one year to the next. In addition, it should be noted that five of the
13 years were war and price-control years and hence not subject to
normal economic forces. This leaves only eight non-war years for
which carry-in data are available. Since the data on carry-ins were so
limited, it was decided to include just production and imports in the
supply concept rather than to estimate consumption on the basis of
only eight years of data.

19541

640

PRICING AND UTILIZATION OP LEGUME AND GRASS SEEDS

27

CONSUMPTION

Relationship of the United States consumption of five major legume seeds
to the total available supply, 1939-1950. (Fig. 5)

28

BULLETIN No. 582

[December,

Price analysis of five legume seeds

To account for the substitution possibility the five major legume
seeds were combined into a composite. The composite price was com-
puted by totaling the value of production for each of the five seeds
and dividing by the total production of the five seeds in pounds. Hence
the resulting composite price received by farmers per pound is weighted
by the production of each seed.

1.3

1.2

1.0

o
8.9

•34

•'38

36

3.9

4.0 4.1 4.2 43

LOG OF FARM GROSS INCOME

Relationship of the logarithm of composite price of five legume seeds to
logarithm of farm gross income in the United States, 1929-1950. (Fig. 6)

Figs. 6 and 7 show the relationship of the logarithm of composite
price to the logarithms of farm gross income and of composite supplies.1
Logarithms were used as it appeared that the changes in the variables
were proportional. The logarithm of the composite price is treated as
the dependent variable and the logarithms of farm gross income and
supplies as the independent variables. The war and price-control years

1 Data for the correlation are given in Table 17 of the supplement (see foot-
note 1, page 4).

1954} PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 29

+.2

tu .0

in

-.2

•'36

'30 »'34

5J2 5.3 5.4 5.5 5.6 5.7 5.8

LOG OF COMPOSITE SUPPLY

Relationship of the logarithm of composite price residuals of five legume
seeds to logarithm of composite supplies in the United States, 1929-1950.

(Fig. 7)

from 1942 to 1946 were left out of the analysis, although they were
plotted and placed in brackets.

The regression line through the scatter of the logarithms of com-
posite price on the logarithms of farm gross income indicates that the
price of the composite increases 0.86 percent with a 1-percent increase
in farm income (Fig. 6), while the regression through the scatter of
logarithms of price on logarithms of production plus imports indicates
that the price of the composite decreases 0.46 percent with a 1 -percent
increase in supplies (Fig. 7) .

The percentage relationship between price and supplies indicates
an elastic demand for this composite of seeds. It must be remembered,
however, that supply was necessarily defined as production plus im-
ports, since figures on carry-in were not available before 1939 (page
26). Since carry-in is often a large fraction of the total quantity of
some seeds, the supply concept used does not give a completely
accurate picture. If it were possible to include carry-ins for all years
and if more legume seeds were included in the composite, thus account-
ing for substitution possibilities more completely, the demand curve
would tend to become less elastic and possibly inelastic.

30 BULLETIN No. 582 [December,

The correlation resulting from the use of the two independent vari-
ables— gross farm income and supplies (production plus imports) —
accounted for 86 percent of the year-to-year variation.1

During the analysis many other independent variables were tried
in an attempt to account for more of the variation. Among them were
various price indexes which did not improve the scatter enough to
warrant the added disadvantage of explaining the price of seeds by
the price of another product. Similarly, data concerning roughage-con-
suming animal units were used and discarded, as any improvement in
the correlation came as a result of reducing the scatter in only two
years, and this advantage was offset by the added variable incor-
porated into the analysis.'

The uniformity of the scatter with numerous demand factors sug-
gests that the error lies in the supply concept. In an attempt to correct
this, carry-over was added to the years since 1939 with results shown
as stars on Fig. 7. It is evident that even for these later years, carry-
over does not improve the final results very much. Consumption series
for 1939 to 1950 were also used but the same approximate scatter
remained. This supported the conclusion that it was not worth while to
try extending the consumption series back to 1929 (page 26) .

Some factors influencing prices. In examining Fig. 7, it is found
that the residuals for the years 1934 and 1936 are both high. This
probably was the result of drouths in these years, which caused a
stronger demand for seed than would be expected from income alone.
Reasons for deviations of other residuals are not easily explained, but
it is thought that certain other factors which influence the prices
received by farmers for these five legume seeds should be discussed.

One of these factors is the substitution possibilities of other seeds.
Although, as already mentioned, a composite price was used in this
analysis to account for substitution among the five seeds studied,
other seeds of course enter into the picture. Because of insufficient data,
however, seeds such as Ladino clover could not be included.

In the literature of the seed trade, weather at seeding time and
hay and pasture conditions are often spoken of as important variables
in the demand for seed. If fields are in good condition, farmers may
delay their rotation or plant at a lower rate.

Another point to remember when analyzing price is that not all
seed passes into commercial channels and hence is not directly subject
to price-making forces. Some data on this can be obtained from gov-
ernment seed disposition reports, but they do not go back far enough
for valid analysis.

1 Coefficient of multiple determination = .86.

19641

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

31

Another factor that affects prices received by farmers is the esti-
mated percentage of "clean-out." When this is high, dealers will dis-
count the seed and when it is low, the seed will bring higher prices.
From available data it appears, however, that the amount of clean-out
is fairly stable and thus does not affect prices as much as one might
expect.

Prices of individual seeds. Despite the limitations of this analysis,
certain important relationships can be inferred from the prices of each
seed as related to the composite. Fig. 8 shows the price per pound
of the five legume seeds (value of production -=- production in pounds)
and the composite price. It is evident that the price of alsike clover
has fluctuated around the price of red clover seed. Both remained above
the composite price throughout the period, as did the price of alfalfa
seed. Prices of lespedeza and sweet clover seed, however, were gener-
ally below the composite. Changes in the prices of each of the five seeds
have generally been in the same direction. A major exception occurred

50

45

40

9 35

30

25

20

'5

u.

o 10

£ 5

RED CLOVER,

ALFALFAs. A I

/

/ \

/,- vu

ll^J \!

/ V - v

LESPEDEZA

CLOVER
I i I i I I i i I

-\

i i

1929 1931 1933 1935 1937 1939 1941 1943 1945 1947 1949 1951

Relationship of the price per pound of five legume seeds to the composite
price in the United States, 1929-1950. (Fig. 8)

32

BULLETIN No. 582

[December,

+ 25

*20

8 +15
£ +10

- 5

a.

-20

RED CLOVEF

ALFALFA-

-^

-.

COMPOSITE PRICE

SWEET CLOVER

LESPEDEZA

\

V

1 1 1 1 1 1

1 1 1 1 1 1 1 1 1 1 1

1 1

1929 1931 1933 1935 1937 1939 1941 1943 1945 1947 1949 1951

Price deviations of five legume seeds from the composite price per pound
in the United States, 1929-1950. (Fig. 9)

in 1947, when red clover seed price was at its peak while the price of
alfalfa seed was at its lowest level since 1941.

Fig. 9 shows approximately the same data but in a different form.
In this figure, deviations from the composite price are shown for each
seed. They indicate more clearly the trend in the price of a particular
seed in relation to the composite price. It is apparent that although
prices of alfalfa, red clover, and alsike clover seed rose in relation to
the composite price, prices of sweet clover and lespedeza seed declined.

Forecasting individual seed prices. A possible method of predicting
the price of an individual seed would be to calculate the percentage
which the supply of that seed is of the composite supply and relate this
to the percentage that the price of the seed is of the composite price.
The validity of this method would depend on the accuracy with which
composite prices and supplies are predictable or known.

This procedure was used and shown for each of the five legume
seeds in Fig. 10. * The scales are in logarithms of the percentages. In

1 Data are given in Tables 18 and 19 of the supplement (see footnote 1,
page 4).

1954} PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 33

Table 5. — Data Showing the Relationship Between Composite Supplies
and Composite Prices of Five Legume Seeds

Seed crop

Estimation
equation8

Standard error
of estimate
in logarithms

Coefficient
of determi-
nation

Correlation
coefficient

Alfalfa

Y = 3.13-.70X

057

.56

.75

Red clover ....
Alsike clover . . .
Lespedeza . . .

Y=2.98-.59X
Y = 2.34-.27X
Y = 2.16-.32X

.044
.057

.088

.74
.40
.68

.86
.63

.82

Sweet clover . . .

Y = 2.16-.39X

.071

.27

.52

• Y := logarithm of the percent of composite price and X = logarithm of the percent of
composite supplies. Supply is equal to production plus imports.

studying these figures, it should be recognized that the supply for each
seed does not include carry-in. This definitely limits the analysis as
carry-ins fluctuate more for individual seeds than for the composite.
The equation showing the average line of relationship for each chart is
given in Table 5 as are the standard error of estimate, the coefficient of
determination, and the correlation coefficient.

The regression line for alfalfa seed explains 56 percent of the varia-
tion.1 However, it is to be noted that all the years from 1929 to 1936
are in the lower right portion of the chart, indicating that when the
alfalfa seed supply was a large fraction of the composite supply, the
price was low. The years from 1937 to 1950, except for 1947, are all
higher and to the left, indicating that alfalfa seed was a much smaller
proportion of the composite supply and hence was higher in price.
Despite the low correlation, the changes in supply of alfalfa seed
relative to the composite suggests the importance of considering sub-
stitution possibilities between legume seeds in a price analysis.

The scatter for red clover shows a definite relationship between the
percentage supplies are of the composite and the percentage price is of
composite price. The regression line explains 74 percent of the vari-
ation.

The analysis of relationship for alsike clover seed indicates a de-
crease in demand. The points for 1929 to 1937 are farther up and
farther right on the chart than are those for later years. The scatter
suggests two possible lines of relationship: a rather steep one rather
far to the right for 1929-1937 and another for the later years. Although
a line showing the average relationship of the scatter does not repre-
sent the demand curve, the decreased demand can be shown by plotting
residuals chronologically. In 1949 and 1950 supplies of alsike clover

1 Coefficient of multiple determination — .56.

34

BULLETIN No. 582

[December,

88

30l«d Q33S 31ISOdWOD JO SI 30IHd 033S VZ3Q3dS31 !N33d3d do 901

33ldd Q33S BlISOdtNOD JO SI

!N3Dd3d JO 901

30ldd Q33S 3itSOdWOO JO SI
30IHd Q33S U3A010 133MS !N30d3d JO 901

19541

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

35

28

•S c*

30l«d Q33S 3J.ISCWWOO JO SI SOIHd 033S W3/\010 Q3« iN30U3d JO 901

4) ~

f J3

o ^3

i-c *^
<U <-M

a o

30lbd Q33S 31ISOdWOO JO SI 30ldd Q33S U3A013 3XIS1V !N30b3d JO 901

36 BULLETIN No. 582 [December,

seed were a relatively small proportion of the composite; however,
prices did not rise as high as they did in the earlier years when supplies
were larger, such as 1934, 1935, and 1937. On considering this from a
different standpoint, we find that in 1940, 1947, and 1948, relative
supplies were about equal to those of 1934 and 1935, yet prices did not
rise as much. Such a result can come about only by a decrease in
demand. This result is logical as alsike clover has in the past been
generally used as a "catch-all" in pasture and hay mixtures. In recent
years, with increased knowledge and use of soil amendments, particu-
larly lime, soils have been made more adaptable to the superior
legumes. Also, Ladino clover has been generally better than alsike
clover in mixtures and has replaced alsike to some extent. In fore-
casting the price of this seed, one should give special attention to the
regression through recent years.

Since 1931 there appears to be some relationship between relative
supplies and relative price of lespedeza seed. As this seed has shown
a very large upward trend in production, the decline in relative price
appears reasonable.

No significant relationship is apparent for sweet clover seed. The
regression through the scatter explains only 27 percent of the variation.

Price analysis of timothy seed

Reliable regressions could not be established for timothy seed
because carry-ins represent a large proportion of the supply, aver-
aging nearly 40 percent of the production for each year since 1939.
The data for carry-ins are summarized in Table 4, page 26. l Data on
carry-ins are not available before 1939 and five of the years since then
were war and price-control years, so that there were not enough data
for a valid analysis. Another complicating factor is that detailed data
on other grass seeds are not available for a long enough time to permit
analysis of the influence of competing seeds.

For all these reasons, the price analysis for this seed is confined to
the description of historical price movements already given (page 16).

GEOGRAPHICAL PRICE AREAS

When we speak of the average United States price of a certain
seed — alfalfa, for example — we tend to think of a particular kind >
or variety of seed which may or may not be the "average." Yet within
the classification "alfalfa" there are a number of varieties which differ

1 Detailed, data are given in Table 16 of the supplement (see footnote
page 4).

19541 PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 37

in their adaptability to different regions. To the extent that a specific
variety or type of seed is adapted to one region and not to another, we
would expect to find different price-making forces at work in each
region.

The purpose of this section is to show what is included in the
average price of a seed and to indicate the degree of homogeneity in
a particular classification of seed. If geographical price differentials
are, on the average, larger than can be explained by transportation
costs, the indication is that different price-making forces are in effect,
or that different seeds actually occur in the same classification but are
averaged together to give a United States average price. Conse-
quently such an average might not be meaningful.1

Alfalfa. There are three rather definite areas of adaptation for
alfalfa seed, depending on the hardiness of the varieties (Fig. 11).
Seeds for hay and pasture cannot be readily interchanged between a
southern and a northern area; hence seed adaptable for each area is
subject to the conditions of supply and demand within the area.

Prices received by farmers in the different states were averaged for
1947-1951 and plotted on the map in Fig. 11. Lines were drawn to
separate states with average prices above $25.50 a bushel, between
$20.00 and $25.50, and below $20.00. It can be seen that the price areas
practically coincide with the areas of adaptation.

While this method gives only approximate location of geographical
price areas, it does indicate the variation in average price and also its
cause. The average price in the northern area is about $3.00 above that
in the central area, and nearly $10.00 above the southern price. This
large differential cannot be explained by transportation costs alone
and is no doubt related to the areas of adaptation.

Lespedeza. Much the same situation exists for lespedeza as for
alfalfa. Average price relates to different varieties of lespedeza, each
adapted to a certain region and hence subject to the influences of
supply and demand within that region. "Common lespedeza, of which
Kobe is the most used variety, is especially adapted to the southern
third of the lespedeza area, while Korean is especially suited to the
more northern rjart."2

Fig. 11 shows the geographical price areas for this seed. Prices
averaged $7.37 per hundred pounds in the central area, $10.98 in the

1 Average prices by states and areas for the six seeds are given in Tables 20-
25 of the supplement (see footnote 1, page 4).

2U. S. Department of Agriculture Yearbook for 1948. pp. 709-711.

38

BULLETIN No. 582

[December.

1954^

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

$ 16.00

°°~~^y/tf _ ^*\

$ 18.00,

39

Average price, by geographical location, that farmers received for six seeds,
1947-1951. Also shown are the areas of adaptation for hardy, non-hardy, and
medium hardy alfalfa varieties. (Fig. 11)

40 BULLETIN No. 582 [December,

south-central area, and $16.41 in the southern area (without Louisi-
ana). The differences are significant enough to indicate that average
prices may not be meaningful for a price analysis unless we know the
proportion of the seed of the different varieties and something as to the
factors influencing the supply and demand.

Other seeds do not show the price differentials by areas that alfalfa
and lespedeza do. Instead, price differences probably reflect to a large
extent surplus and deficit areas with transportation costs accounting
for a large fraction of the differential.

In general, this appears to be the situation for red clover seed (Fig.
11), with the middle Atlantic states being a deficit area. However,
the Northwest, a surplus-producing area, shows a higher average price
than the consuming area of the central states. This can be explained
by the fact that the Northwest produces a larger proportion of im-
proved varieties than does the rest of the country. While certain
varieties of red clover have definite areas of adaptability, this analysis
is not detailed enough to show any differential resulting from this
cause alone.

Sweet clover seed shows very little geographical differential in
prices except for a central core of three states (Illinois, Iowa, and
Missouri) , where the average price was $8.24 a bushel compared with
$7.27 for the other 11 producing states (Fig. 11). These three states
were probably deficit-producing states during the period studied.

Timothy seed also shows very little price differential by location.
Iowa shows a fairly low price as it is the heart of the timothy seed-
producing region and is a surplus state.

Geographical price patterns seem apparent for alsike clover seed,
with Minnesota, Michigan, and Ohio having the lowest average price
— $17.58 a bushel — and Indiana, Illinois, and Iowa having the
highest average price—- $19.88 a bushel (Fig. 11). The latter three
states are in the heart of the consuming area and are likely to be
deficit states with the shift in seed production to the Northwest. The
northwestern states and California, on the other hand, are surplus-
producing states and so show an average price below that for the
central corn-belt states.

How geographical areas affect price analyses

It is apparent from the above material that two important con-
siderations in the price analysis of a particular seed are (1) the extent
to which different types of seed are adapted only to particular loca-
tions and (2) the number of possible substitutions for each type of <
seed.

1954] PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 41

Prices fluctuate widely. The effect on price of variations in adap-
tability and substitution possibilities of different varieties can best be
illustrated by considering alfalfa. As we have seen, hardy, medium
hardy, and nonhardy alfalfa seeds are each adapted to a rather defi-
nite region. As the variegated varieties (hardy types) are generally su-
perior to common varieties north of the fortieth parallel, it is reason-
able to suppose that there is a large potential demand for these
superior varieties. As a result, changes in the quantities of improved
adapted varieties of seeds will cause prices to fluctuate considerably
more than if farmers could substitute other varieties more readily.

In years when production of seed of the variegated varieties is low,
farmers tend to bid up the price, even though the supplies of other
types may be large and the total supply normal. The result may well
be to raise the average price for all alfalfa seed so that it would be
higher than expected on the basis of total available supplies. If in a
particular year supplies of adapted variegated varieties are relatively
large, the result would probably be lower prices for the variegated
varieties and hence a much lower average price for all alfalfa seed.

As yet, comparable prices for the different types of alfalfa are not
available for a long enough period to illustrate these points statisti-
cally. It can readily be seen, however, that for a good understanding
of the behavior of alfalfa seed prices, the price of each type, as well as
the proportion of each in the total supply, should be known.

Although red clover does not have as much diversity in types and
varieties as alfalfa has, somewhat the same problem exists in analyzing
this crop. As farmers become better acquainted with the varieties
adapted to their locality they begin to think in terms of a specific
variety and hence each variety responds to particular supply and
demand forces. For example, a short supply of Kenland red clover
seed in a year when production of red clover seed is large may cause
a higher average price than the total supply of red clover seed would
indicate, as certain farmers are convinced of the superiority of Ken-
land and are willing to pay a good price for it.

For other forage seed crops the same situation may prevail in
varying degrees, depending on varieties and strains available.

Elasticity of demand is affected. The substitution possibilities or
lack of substitution will affect the elasticity of demand1 for a particular

1 Degree of elasticity will vary for both supply and demand depending upon
the period involved. Demand is likely to be least elastic when data for a year
are considered, and is more elastic either for an instantaneous or market period
or for a period longer than a year. Here we are considering the elasticity shown
in a year, or the period which will ordinarily show the least elasticity.

42 BULLETIN No. 582 [December,

seed. It is reasonable to conclude that the demand for all seeds is
inelastic as there are no substitution possibilities for forage seeds
except to let the present stand remain for another year. As a result,
prices would be expected to fluctuate proportionately more than the
change in quantity supplied. But when we consider a specific seed the
demand may be elastic if important substitution possibilities are avail-
able. For seeds with few substitution possibilities, the demand will be
relatively inelastic.

A complicating element is that elasticities of demand for the same
crop will vary from farm to farm and from area to area as the result
of differences in soil, climate, and rigidity of the rotation.

While these considerations are not readily measured and hence not
easily analyzed by statistical methods, they should be understood by
those concerned with producing, buying, or selling seed. The analysis
in qualitative terms of these tendencies can be extended to any seed if
knowledge of economic behavior is combined with accurate informa-
tion about the seed.

Future studies desirable

It is suggested that for future investigations, supplies and prices of
different varieties or groups of varieties should be considered, since
prices of each respond to different demand conditions depending upon
the area of adaptation. Such an analysis was not undertaken in this
study because prices of the different types of seeds were not available
for a long enough period to draw valid conclusions.

MARKETING MARGINS

In Table 6 is shown, for the six seeds studied, the average share of
the retail dollar going to the farmer, the retailer, and the wholesaler
during the 1926-1950 period.1' 2 Analyses of year-to-year trends, size
and fluctuations of the margins, and the relationship of retail prices to
farm prices are made in the following paragraphs.

Trends since 1926

Fig. 12 shows graphically for the six seeds the percentages of the
retail dollar going to the farmer, the wholesaler, and the retailer. It is
evident from these charts that since the early 1920's the farmer's share
of the retail dollar has declined for all the seeds except lespedeza.2

1 Data for each year are given in Table 26 of the supplement (see footnote 1,
page 4).

2 Data studied for lespedeza seed go back only to 1933.

1934^ PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 43

Table 6. — Average Share of the Retail Dollar Going to Each Level in
the Marketing System for Six Forage Seeds, 1926-1950

(Omitting the price-control years, 1942-1946)

Seed crop Farmer's share Wholesale share Retail share

Lespedeza* . .

70 4

percent
11 5

18 1

Red clover

68.6

23.1

8.3

Alfalfa

65.1

19 1

15.8

Alsike clover

63 2

25 7

11.1

Sweet clover. ...

53 7

29 9

16 4

Timothy ....

51 8

27 4

20 8

a For the period 1933-1950, omitting 1942-1946.

Margins for sweet clover and timothy seed are characterized by
rather sharp declines in the farmer's and retailer's shares of the dollar
and an increasing share for the wholesaler. Comparing price relatives
for these two seeds with the price received by farmers for all crops
shows that the price received by farmers has tended to decline in
recent years. As marketing costs have remained high or have increased,
the total margin between farmer and retail purchaser would be ex-
pected to increase.

Margins for red clover seed show very little trend of significance:
The downward trend in the farmer's share of the dollar is only a result
of high percentage returns in the 1920's. The wholesaler's share has
tended to increase over the period studied.

Alsike clover seed has shown a slightly but constantly decreasing
percentage return to the farmer. This may have resulted from the shift
of the alsike clover seed-producing region to the northwest with ac-
companying higher costs of transportation to the consuming area. There
is no apparent trend in the relative shares going to wholesaler and
retailer.

The farmer's relative share of the alfalfa seed dollar decreased
fairly rapidly until the early 1930's, but since that time has been
slowly increasing. The wholesaler's relative share increased over the
period studied while the retailer's share increased until the midthirties
and subsequently declined.

The retail dollar for lespedeza seed is characterized by a rising
share to farmers, a declining share to retailers, and a widely fluctuating
share to wholesalers.

From this short discussion several important relationships become
apparent. First, for seeds that are increasing fairly rapidly in produc-
tion, there has tended to be an increasing share of the dollar returned

44

BULLETIN No. 582

[December,

p o

N <o

avnoo

* n CM —

do j.N3o«3d

8 S

jo iN3oa3d

5 -

</>

8 S

S g

3

< -

dvnoa

JO j.N3oa3d

1954^

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

45

P °

f~ u>

bvnoa

jo iN30«3d

46 BULLETIN No. 582 [December,

to the farmer. This has been true of both lespedeza and alfalfa since
the early 1930's. There is further evidence that the increasing demand
for these seeds in the early periods resulted in large markups by the
retail trade with a larger relative share of the dollar going to the
retailer. Seeds with decreasing price relatives (in comparison with
prices received by farmers for all crops), such as timothy and sweet
clover, are characterized by a declining share of the retail dollar going
to the farmer and an increasing proportion going into the marketing
margin.

Relative variability of total margins

Table 7 shows the variability in the total margin for each seed. It
might be expected that seeds having the greatest variability in produc-
tion would also show the greatest variability in the total margin
(wholesale share plus retail share). The variation in production for
each seed crop, already shown in Table 2, is included in Table 7 so
that comparisons can be readily made. The standard deviation is
divided by the average margin and multiplied by 100 to give Pearson's
coefficient of variation. This reduces the standard deviations to terms
which are comparable. Direct comparisons of the various standard
deviations are not adequate, as a standard deviation of 1 percent on an
average margin of 50 percent represents less of a fluctuation than does
the same standard deviation on an average margin of 25 percent.

The data in Table 7 tend to confirm the hypothesis that variations
in production are related to variations in marketing margins. Alfalfa,
alsike clover, and sweet clover seed show about the same variation in
total production and also have coefficients of variation of about the
same magnitude. Timothy seed shows the highest variation in produc-
tion and also the largest coefficient of variation of the margin. Varia-

Table 7. — Variation in Total Margins as Related to Variation
in Total Production, 1926-1950

Seed crop

Average percent-
age change in
production from
preceding year

Average
total margin
in
percentage

Standard
deviation
of the
margin

Coefficient
of variation
of the
margin

Timothy

117.8

48.2

25.99

53.9

Lespedeza"

39 8

29 6

8.66

29.3

Red clover

42 5

31 4

5 09

16.2

Alfalfa

21.8

34.9

4.99

14.3

Alsike clover . . . .

22.7

36.8

5.18

14.1

Sweet clover. . . .

22.5

46.3

6.31

13.6

« For the period 1933-1950.

1954] PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS 47

tions in production of red clover and lespedeza seed fall in between
these two extremes as do their respective coefficients of variation of
the margins. It must be remembered, however, that data for lespedeza
seed are not strictly comparable with the other data.

Size of total margin

In addition to the relationship between variability in production
and variability in margins, one might expect a further relationship
between fluctuations in production and the size of the margin. That is,
seed production which is unstable would be expected to show the high-
est margin as the industry must cover the higher risks involved.
Timothy seed, however, is the only one that meets this hypothesis.

For the other seeds additional factors are evidently involved —
probably variations in the conditions of demand and location of the
production and consumption areas. For red clover, for example, the
area of consumption is identical with the area of production, so that
the total margin would probably be somewhat less than would be indi-
cated by variations in production alone. Because of these various
interrelated forces, no general conclusion is made in regard to the size
of the margins.

Relationship of retail prices to prices received by farmers

In Fig. 13 the retail price of each seed is plotted against the price
received by farmers, and the line of average relationship is fitted by
the method of least squares. This relationship covers the period 1926-
1950, omitting 1942-1946. The variation from this line is shown by the
coefficients in Table 8 for each of the seeds. For all seeds except
lespedeza, the line of average relationship accounts for the retail price
with high correlation if the average price received by farmers is given.

Table 8. — Data Indicating the Relationship Between Price Received

and Retail Price for Six Forage Seed Crops in the

United States, 1926-1950

Crop

Regression
equation

Standard
error of
estimate

Coefficient
of determi-
nation

Index of
correla-
tion

Lespedeza* . .

Y = .633 + 1.332X

1 36

87

93

Timothy .

Y = . 829 + 1. 767X

1 82

94

97

Alsike clover ....

. . Y = . 759 + 1. 533X

2.43

.96

.98

Alfalfa.

Y = 2.082 + 1.423X

1 75

98

99

Sweet clover

Y = .565 + 1.759X

77

98

99

Red clover

Y = 2.033 + 1.352X

1 62

99

99

For the period 1933-1950.

48 BULLETIN No. 582 [December,

These charts may be useful in indicating the retail price as soon as the
crop moves into the marketing channel from the farmer, as the average
price received by farmers will be determined before the retail price.
However, the retail price will be affected to some extent by the eco-
nomic conditions existing at planting time. The usefulness of the charts
in predicting retail prices before seeds are marketed depends upon how
well the farm price can be forecast.

Other forces affecting margin

In conclusion, several other forces should be mentioned which have
a bearing on the margins for all the seeds. Over the period studied the
seed industry has increased its services. Expanding or adding such
services as purity and germination tests, packaging in smaller contain-
ers, and handling improved varieties and certified seeds has increased
cost of handling. These added costs have tended to increase the whole-
saler's share of the retail dollar (Fig. 12). However, an opposing
tendency is the fact that as prices have risen since the depression years,
marketing costs generally have been "inflexible" and hence have not
risen as fast as the price of the product.

SUMMARY

This study was limited to the six forage seed crops having the
highest average money value from 1940 to 1949: red clover, alfalfa,
lespedeza, alsike clover, sweet clover, and timothy.

Red clover and alsike clover seed are produced mainly in the Mid-
west and Northwest; timothy and sweet clover in the Midwest; alfalfa
throughout the western half of the country; and lespedeza in an area
south of the corn belt, north of the cotton belt, and extending from
the Atlantic to Oklahoma and Kansas.

Lespedeza seed production increased rapidly from 1920 to 1939 but
has tended to level off since then. Production has decreased for sweet
clover and timothy seed, increased for alfalfa and red clover, and re-
mained rather constant for alsike clover. Although imports were rela-
tively unimportant on the whole, they were of considerable importance
for individual seeds in particular years. Only timothy seed was ex-
ported in any quantity. Carry-ins averaged 10 to 15 percent of total
production, amounting to as much as 30 percent in some years.

Changes in acreage account for most of the changes in seed-crop
production. Prospective yield and prospective price probably influence
acreage harvested, though these are not statistically measurable.

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

49

* £

2 1

suvinoa— soNnod 001 «3d 0335 do aoidd

N
UJ
Q
Ld
CL
CO
UJ

8 .S

<T 'S

K 8

§ '£

s T3

2Z <u
U
M

O I

> 2!

UJ

SdVTIOa—SQNnOd 001 H3d Q33S JO 30IMd HIV13«

50

BULLETIN No. 582

[December,

a:

UJ

§

LJ
UJ

syvnoa-- saKinod 001 U3d 0335 jo soiad

1

savmoa--saNnod 001 83d 0335 jo soiad

1954}

PRICING AND UTILIZATION OF LEGUME AND GRASS SEEDS

51

o o\
- vi
K 81

13
V

o o

tE 'C

°- a
o

SWITCH!-- SONnOd 001 «3d Q33S JO 30l«d 1IV13H

8

3

O

O
UJ

cr

ft
T3

w C

o .2

syvnnoa--saNnod 001

0335

52 BULLETIN No. 582

Farmers received the lowest prices in fall, and the highest in spring,
the difference being great enough that it would pay Illinois farmers
to store their seed until spring.

Among the factors determining seed prices are: gross farm income
(used as a measure of demand), price of competing seeds, and total
supply, including supplies of competing seeds. Seeds vary in substitu-
tion potentialities from area to area, depending on soil types, type of
farming, and disease problems. These conditions alter the price-making
forces in different areas. Hardy and nonhardy alfalfa varieties, for
example, which are adapted to the northern and the southern part of
the United States respectively, show such great price differentials that
they should probably be treated as two different commodities in a
price analysis.

The portion of the retail dollar going to the farmer has tended
to decline, with only lespedeza showing a marked reversal of this trend.
Seeds showing a wide year-to-year variation in production also have a
wide variation in their margin.

Probably many persons connected with the seed trade have reached
these conclusions from their own observations. This report should
serve not only to confirm these observations but also to make them
available to others who want more insight into the behavior of seed
prices.

3M— 12-54— 55989

UNIVERSITY OF ILLINOIS-URBANA