Full text of Wartime Food Purchases : Bulletin of the United States Bureau of Labor Statistics, No. 838

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

UNITED STATES DEPARTMENT OF LABOR
L. B. Schwellenbach, Secretary
BUREAU OF LABOR STATISTICS
Isador Lubin, Comm issioner (on leave)
A. F. Hinrichs, A ctin g Comm issioner

W artim e Food Purchases

Bulletin 7s£o. 838

For sale by the Superintendent o f Documents, U. S. Government Printing Office
Washington 25, D. C. - Price 10 cents

Letter of Transmittal

U nited States D epartment op Labor,
B ureau of L abor Statistics,
Washington , D. C., July 13 , 1945.
The S ecretary of L abor:
I have the honor to transmit herewith a report on food purchases by city
families during one week in September-October 1944 as compared with one
week in March-June 1942. The bulletin was prepared by Lenore A. Epstein
under the general supervision of Dorothy S. Brady, Chief of the Cost of Living
Division.
This is the first of several reports to be prepared on wartime purchases, based
on data collected in the Bureau’s survey of prices paid by consumers in cities in
1944. Others will deal with clothing, housing, medical care, transportation, etc.
A summary report will present, by income class, data on expenditures, tax pay
ments, and bond purchases in 1944 for a representative cross section of city
families.
A. F. H inrichs, Acting Commissioner.
Hon. L. B. Schwellenbach,
Secretary of Labor.

Contents

Summary---------------------------------------------- ------- ------------------Changes in outlay for food, 1942 to 1944-------------------------------Amounts of food purchased-------------------------------------------------Food expense in relation to income---------------------------------------Food expense in relation to family size----------------------------------Appendix.—Survey method:
Sampling procedure:
Selection of sample for 1944 study----------------- ---------Selection of sample for 1942 study--------------------------Comparison of the two samples______________ _______
Adjustment of 1942 data for comparability as to family size.
Collection of food data-------------------------------------------------Definitions____________________________________________
(ID

Page
1
4
5
12
14
16

23
24

Bulletin T^o. 838 o f the
U nited States Bureau o f Labor Statistics

[Reprinted from the M onthly L abor R eview, June 1945, with additional data]

Wartime Food Purchases

Summary
In the fall of 1944, just prior to the current meat shortage, the
diet of city families compared very favorably with the diet of families
with similar incomes in the spring of 1942. The lowest income groups,
in particular, bought more meat, milk, sugar, flour and bakery goods
than in 1942 when the drain of war needs on the civilian food supply
had barely been felt. Purchases of fruits and vegetables declined but
the total consumption of these items was supplemented by increased
home production. At the highest income level, the amount of meat,
butter, and cheese obtained was also smaller than in 1942, despite a
greater food expenditure.
The improved diets of the lowest income families meant a very heavy
drain on family resources, with food outlays taking about 70 percent
of their income as contrasted with 50 percent in 1942. Food costs pre
sented an especially serious problem for large families, not only in the
lowest income group, but at income levels up to $3,000. In the upper
portion of the income scale, the average family paid considerably more
for food in the later period, despite smaller purchases.
These are the findings of a survey of prices paid by consumers in
1944, which was conducted in two parts for the primary purpose of
comparing price changes reported by city consumers with price
changes indicated by urban store reports. The first part of the survey,
made in the fall of 1944, provided detailed information on food pur
chases during 1 week, purchases of clothing and household textiles
during the first 8 months of the year, tenure and rental in August 1944,
and sufficient information on family composition, living arrangements,
and income to provide a basis for classification. The second part will
provide data on food purchases during 1 week early in 1945, on pur
chases of clothing and other textiles during the last 4 months of 1944,
and of other goods and services throughout that year. As a by
product of the reports required for analysis of prices, certain data are
available on family expenditures and quantities of selected goods pur
chased. This article deals with the information obtained on food
purchases in the fall of 1944 by housekeeping families and single per
sons at different income levels. It is the only information of this
nature that has been obtained since the spring of 1942.
For this earlier period, the Bureau of Labor Statistics, in cooperation
with the Bureau of Human Nutrition and Home Economics of the
U. S. Department of Agriculture, conducted the survey of spending

(1)

and saving in wartime for the primary purpose of providing national
estimates of expenditures and savings by income class for the year 1941
and#the first 3 months of 1942. In conjunction with the main survey,
detailed information was obtained on food purchases and food con
sumption during 1 week in the spring of 1942. The reports on food
consumption have been analyzed and published for urban as well as
rural areas by the Bureau of Human Nutrition and Home Economics
in U. S. Department of Agriculture Miscellaneous Publication No. 550
(Food Consumption in the United States). The data collected on
food purchases have not been published.
The methods used in the two surveys were practically identical.
The samples are very similar as to coverage; both related to the civilian
noninstitutiona.1 population in cities of 2,500 or more scattered through
out the country. Information was obtained by personal interview in
each case. The figures on purchases of various types of food, from
both studies, were compiled from reports on purchases of a list of ap
proximately 200 food items.1
Since the findings of the survey of spending and saving in wartime
on weekly food purchases in the spring of 1942 have not been pub
lished, they are presented here to permit comparison between the fall
of 1944 and the earlier period, shortly before widespread rationing was
instituted2 and before significant shortages of certain foods had
developed. When making comparisons, it must be recognized that
many of the foods that are under the OPA rationing program were
temporarily off the ration list during part or all of the fall of 1944.
Some of these foods were nevertheless in short supply in several sec
tions of the United States.
Food-purchasing habits are by no means the same in all sections of
the United States. Furthermore, price movements during the war
have differed somewhat by region and by city size. Establishment
and expansion of shipyards and of war plants, while causing a con
siderable migration to the South and West, also improved very
greatly the employment opportunities in those areas. The distribu
tion of the housekeeping families represented in the urban sample
surveys at each income level, shown in table 1, clearly illustrates this
point. Thus, at low income levels, the highly industrialized North
east was much more heavily represented in the fall of 1944 than in
the spring of 1942, and the South and West proportionately less. At
the opposite end of the income scale, the importance of the Northeast
declined sharply during the period, and the representation of the
South nearly doubled, although it remained below the average for
all income classes combined. The West was better represented at
the top of the income scale than in the population at large, and had
greatly improved its position during the war.
There is similar, though less striking, evidence of an improved
income position for cities having a population below 25,000 in 1940.
i See Appendix (pp. 16 to 22) for a statement on the sampling procedure used in the survey of prices paid
by consumers and an evaluation of the differences between the sample for that survey and the sample for
the survey of spending and saving in wartime. The former covered approximately 1,700 families, of which
1,500 were keeping house and provided information on purchases of food for use at home. In the survey of
spending and saving in wartime the corresponding numbers were approximately 1,300 and 1,100.
The reports on food purchases pertain in each survey to the 7-day period immediately preceding the inter
view, or the previous calendar week. The proportion of food reports that covered a week in each of the desig
nated months was as follows: 1944—September, 69 percent, October, 30 percent, and November, 1 percent;
1942—March, 5 percent, April, 42 percent, May, 45 percent, and June, 10 percent.
a Sugar rationing was put into effect in March 1942, but meats, fats and oils, cheese, and processed foods
were not covered until a year later.

3
T able

1.—Percentage Distribution of City Families and Single Persons Surveyed in
1944 and in 1942 by Region, City Size,1 and Annual Money-Income Rate
September-October 1944
Region and city size1

All cities_________________________
Northeast__________________ ______
South.......................................................
North Central ___________________
West_____ _____ _________________
500,000 or more population__________
100,000-500,000 population__________
50,000-100,000 population.....................
25,000-50,000 population......................
Under 25,000 population____________

All
families
and
single
persons
100.0
32.4
24.4
29.1
14.1
33.6
20.7
10.1
9.0
26.6

Housekeeping families and single persons by annual
rate of money income2
All8
100.0
31.6
24.4
29.6
14.4
32.4
21.4
10.2
8.9
27.1

Under $1,000 to $2,000 to $3,000 to $4,000
$1,000 $2,000 $3,000 $4,000 and over
100.0
31.1
25.2
33.3
10.4
30.4
24.4
11.9
5.9
27.4

100.0
28.7
32.5
25.7
13.1
26.0
21.8
14.0
10.4
27.8

100.0
32.8
23.1
29.6
14.5
35.7
18.7
10.1
9.9
25.6

100.0
30.9
20.9
33.3
14.9
33.4
22.5
7.2
7.2
29.7

100.0
31.6
19.6
29.8
19.0
34.8
21.8
6.6
8.9
27.9

100.0
48.6
14.1
24.6
12.7
38.8
21.8
14.1
6.3
19.0

100.0
44.2
10.9
32.7
12.2
34.7
21.1
10.2
9.5
24.5

March-June 1942
All cities___________ _____________ 100.0
36.2
Northeast________________________
South _____ ___________________
20.1
32.6
North Central____________________
West___ _______________ _________ 11.1
500,000 or more population__________ 31.0
100,000-500,000 population.... ................ 20.9
9.1
50,000-100,000 population_____ _____
25,000-50,000 population ....................... 10.6
Under 2.^000 population
_ _ _ 28.4

100.0
35.7
19.4
33.8
11.1
30.6
20.3
9.2
11.2
28.7

100.0
19.4
32.9
32.9
14.8
19.9
18.0
9.3
13.0
39.8

100.0
35.3
21.2
34.0
9.5
31.4
20.6
6.5
9.5
32.0

100.0
37.6
14.7
38.9
8.8
31.4
20.6
9.1
14.7
24.2

» Cities were classified by size as shown by the 1940 census. Each place covered was classified by region
and size, regardless of whether or not it was part of a metropolitan district.
* See p. 25 for definition of income.
*Includes a small number that did not report income.

Small-city families were much less numerous in the low-income popu
lation in 1944, but they did not show a- corresponding increase at the
highest income level. Housekeeping families in cities of 25,000 to
100,000 in 1940 made some gains also, though less significant ones,
relative to those in cities of 100,000 or more. The shift of small-city
families out of the lowest income group meant that they were no
longer overrepresented in that group, in comparison with the entire
population of housekeeping families in cities. Among families with
incomes of $4,000 or more, those in large cities remained relatively
more numerous than in the 1944 population at large, though the dis
parity was less marked than in 1942 when large-city families comprised
51 percent of all city families. The middle-sized cities, by contrast,
were not so well represented in the highest income class as in the gen
eral population in 1944.
Two significant facts emerge from the comparison of weekly food
purchases by housekeeping families 3 in cities throughout the United
States in the fall of 1944 with those reported in the spring of 1942.
First, there was a striking increase over the period in the food expense
of families and single persons with incomes below $1,000. Second,
a leveling off occurred among income groups in the quantities of many*
* The term, “family,” is used in this to include single persons. All data presented are based on reports by
housekeeping families. See p. 24 for definitions.

basic foods purchased for home consumption, with increased purchases
by the low-income groups and reduced purchases of many foods by
higher-income families. These developments result in part from the
population shifts just described and the changes in the prices and
supplies of various foods between the two periods.
Changes in Outlay for Food 1942 to 1944
At the lowest income level there was a rise of 45 percent in outlays
for food to be served at home or carried from home, and a rise of
50 percent if the increased amounts for food purchased away from
home are taken into account (table 2). This contrasts with increases
of 14 percent in amounts spent for food at home by families with
incomes of $1,000 to $3,000 and about 16 percent in their total food
expenses. The increases in expenditure exceed by a considerable
margin the increases in quantities purchased. For the middle income
groups, the rise in expenditures closely approximates the 13-percent
increase over the period shown by the Bureau’s index of retail food
costs in large cities. For the higher income groups, expenditures for
food at home increased less than might be expected in view of the
13-percent rise—6 and 8 percent, respectively, at the $3,000-$4,000
and $4,000 and over levels. The difference appears, however, to
have been made up in part by more frequent restaurant meals.
The allocation of the budget for food at home among the major types
of food was remarkably similar throughout the income scale in each
period (table 3). Cereal and bakery products, fats and oils, and (in
1942) sweets each tended to decrease slightly in relative importance
as income increased. In 1942, proportionately more went for meat,
poultry, and fish at successively higher income levels, 25 percent at
the lowest, and 30 percent at the highest; but in the fall of 1944, out
lays for this food group represented 28 to 29 percent of total expense
for food at home at all income levels.

T able 2.—Average Expense for Food and Alcoholic Drinks, and Family Size,1 House
keeping Families in Cities, by Annual Money-Income Rale,2 Week in 1944 and 1942
September-October 1944
Item

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to
to
to and
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over

Total food 8.............................. $8.42 $11.87 $17.04 $18.70
Food at home.................... 7.57 10.44 14.43 15.08
Food away from hom e... .85 1.43 2.61 3.62
Alcoholic drinks....................... .22 .48 .90 1.16
Average family size *............... 2.10 2.52 3.24 3.56

$23.51 $5.61 $10.32 $14.54 $16.41 $21.00
18.49 5.24 9.16 12.68 14.21 17.11
5.02 .37 1.16 1.86 2.20 3.89
1.96 .10 .27 .52 .67 1.13
3.93 2.10 2.52 3.24 3.56 3.93

1 For comparability, purchases reported in the 1942 study have been adjusted to represent the spending
of families of the sizes found in the sample population in September-October 1944. See p. 22 for description
of method.
*Annual rate of income was based on reports for August 194' in the recent study and on reports covering
the first 3 months of 1942 in the earlier one. In both instances, family income represents the sum of all types
of income received by family members during those periods: Wage and salary earnings, entrepreneurial net
income or withdrawals, and nonearned income from all sources, exclusive of inheritances, large gifts, and
lump-sum insurance settlements. For the 1942 study, total wage and salary earnings less occupational
exoense was use d ir computing family’s total money income, but for the 1944 study, wage and salary earnings
after pay-roll deductions was u?ei. See p. 25.
8 Food expense includes cost of all food purchased during the week, whether or not paid for at the time.
See p. 25.
* Family-size figures are based on the number of family members living at home during all or part of the
week covered by the food report. See p. 24.

T able 3.—Distribution of Expense for Food at Home, by Food Groups, Housekeeping
Families in Cities, by Annual Money-Income Rate, 1 Week in 1944 and in 1942 1
September-October 1944
Commodity

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to and
to
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over

Average expenditures for food
at home................................ $7.57 $10.44 $14.43 $15.08 $18.49 $5.24 $9.16 $12.68 $14.21 $17.11
Percentage distribution of expense
Meat, poultry, and fish..........
Dairy products and eggs........
Fats and oils....... ....................
Fruits and vegetables.............
Sugar and other sweets...........
Cereals and bakery products.
Other food and beverages___

28.3
21.2
3.8
20.5
4.2
14.1
7.9

28.5
22.5
3.5
20.6
3.4
13.3
8.2

27.9 27.6
22.6 23.4
2.8 2.9
21.8 21.1
3.7 3.4
13.0 12.9
8.2 8.7

29.2
22.1
2.7
21.1
3.9
12.9
8.1

25.2 26.9
22.1 24.4
4.4 3.8
24.4 22.5
3.0 3.3
13.6 11.9
7.3 7.2

27.4
23.6
3.2
23.9
2.8
11.5
7.6

28.9
24.4
2.5
22.7
2.5
11.5
7.5

30.2
22.5
2.2
24.6
2.4
10.8
7.3

I’Sce tables 5 through 11 for components of each food group. See p. 25 for income definition; p. 22 for note
on adjustment of 1942 reports for comparability with respect to family size.

Differences in price movements of various foods between the spring
of 1942 and September-October 1944, together with changes in quan
tities purchased, resulted in some changes in the allocation of food
expense at corresponding income levels. Thus, although dollar ex
penditures at each income level for every food group (except fruits
and vegetables at the two highest income levels) were larger last
fall than in the spring of 1942, the increases were not in the same
proportion. Slightly less went for dairy products, considerably less
for fruits and vegetables, slightly more for sugar and sweets, cereal
and bakery products, and miscellaneous foods and nonalcoholic
beverages. Fats and oils became less important at the low and more
important at the high levels. Amounts spent for meat, poultry, and
fish represented a larger share of the total at the lower income levels
in 1944 than in 1942, and a smaller share at the higher levels.
Amounts of Food Purchased
The second significant fact revealed by the survey of prices paid by
consumers—the leveling off among income groups in home consump
tion of many of the basic foods—is strikingly illustrated by the
figures in table 4 on weekly per-capita purchases of meats and poultry.
T able 4.—Per-Capita Purchases of Meats and Poultry by Housekeeping Families, by
Annual Money-Income Rate, 1 Week in 1944 and in 1942 1
Average per-capita purchases (in pounds) of—
Meats

Annual money income
1944
Under $1,000____________________________________
$1,000 tn $2,000 ______
_____________________
$2,000 to $3,000.......................................................................
$3,000 to $4,000.......................................................................
$4,000 and over_________________________________

1.99
2.21
2.30
2.04
2.44

Poultry
1942
1.78
2.37
2.52
2.54
2.70

1944
0.48
.55
.55
.55
.60

1942
0.20
.20
.34
.46
.63

The meat figures may well reflect the rationing program which is
designed to distribute short supplies equally among all groups. They
suggest also a tendency for the low income groups to buy close to the
limit of their red-point supply, even though they have customarily
bought less meat. Since poultry has never been rationed, the c ange
in purchasing habits cannot be explained in quite the same manner.
Chicken was unusually plentiful in the early fall of 1944 and it is
possible that the attitudes developed with respect to meat may have
carried over to poultry. Since the price per pound was higher for
poultry than for meats as a group, it cannot be argued that poultry
purchases by low-income families were an economy measure.
Beef made up a smaller proportion of meat purchases in 1944 than
in 1942 except for families with incomes below $1,000. This group
bought somewhat less pork, proportionately, perhaps because South
erners comprised a smaller segment of the low-income urban popu
lation in 1944 than previously (see table 5). At higher income levels
pork products represented about the same proportion of meat pur
chases in the two periods. Veal was consistently purchased in some
what greater quantity by all families, though the differences were not
T able 5.—Purchases of M EAT , POULTRY , AND FISH by Housekeeping Families in
Cities, 1 Week in 1944 and in 1942, by Annual Money-Income Rate 1
September-October 1944
Item

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to and
to
to
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
Average expense in 1 week

Meat, poultry, and fish.......... $2.14 $2.98 $4.03 $4.16 $5.39 $1.32 $2.46 $3.48 $4.10
Meat......................................... 1.52 2.15 2.99 2.98 " T iT iTIcT T o e * 2.89 3.25
Beef.................................... .59 .78 1.10 1.15 1.51 .43 .87 1.23 1.31
Veal.................................... .12 .18 .24 .20 .34 .04 .12 .15 .20
Pork 1*3................................ .50 .67 .89 .82 1.11 .43 .69 .89 1.07
Lamb................................. .09 .16 .23 .27 .34 .07 .12 .26 .33
Variety meat and game *. .22 .36 .53 .54 .63 .13 .26 .36 .34
Poultry..................................... .44 .64 .81 .92 1.12 .13 .25 .38 .56
Fish and seafood..................... .18 .19 .23 .26 .34 .09 .15 .21 .29
Fresh and frozen.............. .15 .13 .15 .13 .24 .06 .09 .13 .19
Canned4....... .................... .03 .06 .08 .13 .10 .03 .06 .08 .10

$5.16
4.08
1.77
.24
1.30
.30
.47
.86
.22
.14
.08

Average quantity purchased in 1 week (in pounds)
Meat, poultry, and fish.........
Meat..........................................
Beef....................................
Veal.................................. Pork3.................................
Lamb..................................
Variety meat and game 3_
Poultry.....................................
Fish and seafood......................
Fresh and frozen...............
Canned4............................

5.67 7.43
4.17 5.57
1.65 2.01
.31 .46
1.39 1.81
.22 .43
60 .86
1.01 1.38
.49 .48
.43 .36
.06 .12

9.79 9.81 12.65 4.69
7.44 7.28 9.57 3.73
2.79 2.82 3 65 1.45
.61 .48 .80 .12
2.36 2.14 2.89 1.49
.52 .58 .78 .27
1.16 1.26 1.45 ,40
1.77 1.96 2.36 .42
.58 .57 .72 .54
.41 .35 .54 .33
.17 .22 .18 .21

7.44 10.36 11.91
5.97 8.18 9.04
2.57 3.49 3.74
.36 .44 .55
1.96 2.49 2.93
.37 .76 .88
.71 1.00 .94
.73 1.10 1.65
.74 1.08 1.22
.38 .52 .78
.36 .56 .44

14.05
10.61
4.57
.62
3.30
.89
1.23
2.47
.97
.60
.37

1 See p. 25 for income definition; p. 22 for note on adjustment of 1942 reports for comparability with respect
to family size. Averages are based on all housekeeping families and single persons in the class, whether or
not they purchased the item during the week.
3Includes bacon and salt pork as well as fresh and other smoked or cured pork.
3Includes in both periods liver, bologna, frankfurters, luncheon meats, cold cuts, head cheese, scrapple,
etc., and also all types of game purchased. In 1944 includes all tongue, heart, kidney and other entrails;
in 1942 includes such meats only when the type as beef, lamb, etc* was not known; if the type was known,
such meats were included with other beef, lamb, etc.
4Includes also smoked or cured fish, but purchases of such fish were very small, on the average, in both
periods.

7
great enough to make up for the smaller beef purchases. Variety
and ready-to-eat meats were more important both in amount and in
relation to the total in the later period. Since there was no point
value on variety or ready-to-eat meats last fall, their purchase was
undoubtedly encouraged. A slight difference in classification of
variety meats (other than liver) in the two surveys 4 was not suffi
cient to account for more than a small part of the larger purchases.
Fish has never had an important place in the average city family's
diet. As might be expected, canned fish, which was difficult to obtain
and had a relatively high point value in September-October
was bought much less often then than in the spring of
Pur
chases of fresh fish and shell fish were not increased correspondingly;
in fact, they were smaller last fall at every income level except the
lowest. As a result, purchases of all types of fish averaged less than
one-fourth pound per person.

1942.

1944,

At every income level, purchases of fluid milk were somewhat
greater last fall than in March-June 1942 (table 6), but there was
little evidence of leveling off among income groups. Approximately
2 quarts per person per week were bought by the lowest-income
families as compared with about 3 per person in families that had
$3,000 or more. Butter and cheese purchases were, of course, lower
in 1944 than in 1942. In this case, as with meats, the effect of ration
ing was to cut much more deeply into the consumption of the high
than the low income groups.
Purchases of other fats and oils, most of which had a zero point
value last fall, were about the same for families with incomes under
$1,000 in the two periods, but relatively larger in 1944 at higher income
levels (table 7). On a per-person basis, purchases declined with in
creasing income in both periods. The increase with income in 1944
in purchases (per family) of oleomargarine, which is the most direct
butter substitute, is in striking contrast to the sharp drop, as income
increased, in the period before the butter shortage. The reduction in
butter purchases was not fully compensated for, however; butter,
oleomargarine, lard, and other shortening together averaged 0.2 to
0.4 pound less per family in the later period.
More sugar was purchased in the fall of 1944 than in 1942 at every
income level, with the increase most marked among families that re
ceived less than $1,000 (table 8). In 1942, per-capita purchases
showed no consistent relationship to income, but in 1944 they were
greater the lower the income. The relatively larger purchases of
sugar last fall are undoubtedly explained in part by the severe sugar
shortage in the spring of 1942 when sugar rationing was instituted, and
in part by requirements for home canning of fruits which come on the
market in late summer and fall and were purchased in considerable
quantity in September-October 1944. The amount of other sweets
bought was similar at most income levels in the two periods.
Bakery products were bought in slightly greater quantity in Septem
ber-October 1944 than during the earlier period in 1942 (table 9).

Since the use of commercially baked goods instead of home baked
foods serves to extend a family's supply of shortening and sugar, and
saves the time of an employed housewife, a considerable increase was
to be expected. That they are more expensive, however, is evidenced

4See table 5, footnote 3.
660111°—45----2

—Purchases of D A IR Y PRODUCTS AND EGGS by Housekeeping Families
in Cities, 1 Week in 1944 and in 1942, by Annual Money-Income Rate 1

T able 6.

September-October 1944
Item

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to and
to
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
Average expensejn:Lweek

$1.60 $2.35 $3.27 $3.53 $4.09 $1.16 $2.24 $2.99 $3.46
Dairy products and eggs
Dairy products.............
1.11 1.72 2.50 2.62 3.06 .88 1.75 2.40 2.84
Fluid milk________ . . . . . .59 .98 1; 53 1.52 1.79 .43 .87 1.27 1.37
.03 .05 .09 .15 .14 .01 .08 .12 .23
Cream........................
.04 .12 .18 .21 .28 .02 .10 .18 .20
Ice cream..................
.20 .28 .35 .36 .47 .22 .41 .53 .66
Butter........................
.09 .14 .19 .23 .23 .07 .17 .20 .26
Cheese.......................
.15 .15 .15 .14 .15 .11 .10 .08 .11
Evaporated milk___
.02 .02 .02 .01
.01
.01 .01 09
Condensed milk2. ..
.49 00.63 .77 .91 1.03
.28 .49 .69 .62
Eggs..................................

$3.85
3.12
1.44
.27
.22
.75
.29
.14
.01
.73

Average quantity purchased in 1 week
Dairy products:
Fluid milk................ -qt.. 4.01 6.78 10.27 10.06 11.81 3.56 6.54 9.19 9.80
Cream........... ........... -P t~ .11 .17 .31 .49 .46 .05 .31 .48 .82
Ice cream.................. _qt_. .07 .26 .40 .41 .53 .04 .23 .40 .38
Butter....................... ..lb.. .39 .54 .68 .69 .90 .53 .98 1.24 1.56
Cheese.........................lb.. .24 .33 .44 .49 .52 .26 .57 .64 .81
Evaporated milk_14U oz._ 1.37 1.41 1.46 1.28 1.41 1.32 1.06 .98 1.17
Condensed milk 2__...lb.. .06 .02 .03 .04 .02 .15 .15 .16 .06
Eggs..................................doz_. .93 1.22 1.43 1.67 1.86 .80 1.39 1.62 1.71

10.01
.98
.43
1.72
1.03
1.66
.03
1.89

T able 7.—Purchases of FATS AND OILS, by Housekeeping Families in Cities, 1
Week in 1944 and in 1942, by Annual Money-Income Rate 1
September-October 1944
Item

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to and
to
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
Average expense in 1 week

Fats and oils............................. $0.29 $0.36 $0.41 $0.44 $0.50 $0.23 $0.35 $0.40 $0.36 $0.38
Oleomargarine.................. .06 .08 .08 .10 .11 .04 .03 .03 .02
.01
Lard................................... .07 .06 .06 .05 .05 .10 .06 .06 .06
.05
Other shortening.............. .08 .10 .12 .13 .11 .04 .07 .09 .07
.08
Salad dressing, salad and
cooking oil...................... .08 .12 .15 .16 .23 .05 .19 .22 .21
.24
Average quantity purchased in 1 week (in pounds)
Fats and oils............................. 1.26
Oleomargarine................. .27
Lard.................................... .35
Other shortening............... .34
Salad dressing, salad and
cooking oil2.................... .30

1.49
.30
.33
.45
.41

1.68
.33
.32
.52
.51

1.72
.37
.26
.56
.53

1.96
.44
.28
.50
.74

1.24
.23
.58
.22
.21

1.37
.13
.34
.35
.55

1.61
.15
.37
.40
.69

1.49
.12
.34
.31
.72

1.68
.05
.38
.27
.98

* See p. 25 for income definition; p. 22 for note on adjustment of 1942 reports for comparability with respect
to family size. Averages are based on all housekeeping families and single persons in the class, whether or
not they purchased the item during the week.
2A pint contains approximately 1 pound.

—Purchases of SUGAR AND OTHER SWEETS by Housekeeping Families
in Cities, 1 Week in 1944 and in 1942, by Annual Money-Income R ate 1

T able 8.

September-October 1944
Item

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to and
to
to and
to
to
to
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
Average expense in 1 week

Sugar and other sweets_____ $0.32 $0.35 $0.53 $0 51 $0.73 $0.16 $0.30 $0.36 $0.36
Sugar................................. .18 .19 .23 .24 .23 .09 .13 .15 .15
Jam, jelly, molasses, sir
up, candy, etc— ......... .14 .16 .30 .27 .50 .07 .17 .21 .21

$0.42
.lfi
.24

Average quantity purchased in 1 week (in pounds)
Sugar and other sweets_____ 3.16
Sugar................................. 2.67
Jam, jelly, molasses, sir
up, candy, etc— ......... .49

3.49 4.25 4.51 4.74 1.74 2.69
2.82 3.33 3.54 3.34 1.26 1.83
.67 .92 .97 1.40 .48 .86

3.10
2.14
.96

2.99
2.02
.97

3.51
2.53
.98

T able 9.—Purchases of CEREALS AND B A K E R Y PRODUCTS by Housekeeping
Families in Cities, 1 Week in 1944 and in 1942, by Annual Money-Income Rate1
September-October 1944
Item

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to
to and
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
Average expense in 1 week

Cereals and bakery products™
Flour...... ............. .............
Ready-to-eat and un
cooked cereals3..............
Other cereal products *___
Bakery products...............
Bread and rolls_____
Crackers.......... ..........
Cake, cookies, pastry,
etc.............................

$1.07 $1.39 $1.87 $1.94 $2.38
$0.71 $1.09 $1.46 $1.64
.20 .18 .21 .20 .22 .08 .08 .09 .10
.17 .19 .26 .22 .24 .11 .12 .16 .14
.04 .05 .07 .08 .10 .04 .06 .09 .06
.66 .97 1.33 1.44 1.82 .48 .83 1.12 1.34
.41 .53 .71 .79 .90 .30 .48 .60 .80
.07 .07 .08 .10 .09 .05 .06 .09 .08
.18 .37 .54 .55 .83 .13 .29 .43 .46

$1.84
.12
.19
.05
1.48
.81
.10
.57

Average quantity purchased in 1 week (in pounds)
Cereals and bakery products.
Flour..................................
Ready-to-eat and un
cooked cereals 3_............
Other cereal products8...
Bakery goods....................
Bread and rolls_____
Crackers......................
Cake, cookies, pastry,
etc.............................

10.00 11.39 14.37 14.47 16.97 8.00 9.92 12.58 13.11
3.07 2.73 2.98 2.82 3.06 1.75 1.88 2.00 1.53
1.67 1.53 1.92 1.46 1.54 1.55 1.05 1.33 1.08
.30 .38 .53 .58 .78 .41 .45 .77 .55
4.96 6.75 8.94 9.61 11.59 4.29 6.54 8.48 9.95
3.89 5.08 6.69 7.27 8.35 3.37 4.82 6.20 7.53
.36 .38 .41 .52 .43 .31 .38 .54 .43
.71 1.29 1.84 1.82 2.81 .61 1.34 1.74 1.99

14.70
2.11
1.56
.46
10.57
7.64
.48
2.45

1 See p. 26 for income definition; p. 22 for note on adjustment of 1942 reports for comparability with respect
to family size. Averages are based on all housekeeping families and single persons in the class, whether or
not they purchased the item during the week.
*Includes rice, hominy grits, corn meal, cornstarch, and cereals prepared for infants.
*Includes macaroni, spaghetti, noodles, etc. Popcorn was included here in 1944, but with flour in 1942.

by the smaller per-capita purchases by families with incomes below'
$1,000 than by higher income families. Purchases per person of
flour and other cereal products, by contrast, dropped off markedly as
income increased. As between 1942 and 1944, flour purchases rose
by at least 40 percent at each income level and by 75 percent at the
lowest, whereas purchases of other cereal products, which are an imingredient of meat-saving dishes, showed no change at that
!)ortant
evel and a relatively small increase at higher levels.
Although seasonal differences in the supply of fresh fruits and
vegetables limit the significance of any comparison of purchases of
such foods during September-October 1944 and March-June 1942, a
few striking facts are noted. The average family in 1944 bought less
than half as many cans of fruit and vegetables—presumably because
of lack of ration points for more—and much less dried fruit and
vegetables, but purchases of fresh produce also totaled somewhat
less than in 1942 (table 10).
The comparatively small purchases of citrus fruit in the fall of 1944,
which undoubtedly reflect a seasonal low in supplies, were more than
compensated for by the large amounts of other fresh fruit bought.
It seems probable, however, particularly in view of the large sugar
purchases, that a portion of this fruit was canned for winter use and
not eaten at the time of purchase. Certainly every effort has been
made to encourage home canning during the war.
The smaller quantity of potatoes, a relatively cheap and filling food,
not rationed, is difficult to explain, particularly in the case of low-income
families. It may have been due to shortages in certain areas in 1944,
and to a considerably higher price per pound. Purchases of other
fresh vegetables were consistently smaller than in the spring of 1942,
although their average price was not a great deal higher. Victory
gardens may easily have made up the difference, and possibly meant
even more fresh vegetables for home consumption than in the spring
of 1942. One-third of the families occupying dwellings with house
keeping facilities reported consumption during the week of interview
of some home-produced food. It is probable that without this sup
plement, food expenditures would have been somewhat greater than
they were in 1944.
Taken in combination, fresh (and frozen) fruit and vegetable pur
chases provide a further illustration of the leveling that has occurred.
In the spring of 1942, about 7.5 pounds were bought per person by
families with incomes below $1,000 and 11 pounds by those at the top
of the income scale. By contrast, purchases in the fall of 1944 aver
aged 7 and 9 pounds per person in the lowest and highest income
groups, respectively.
Although beverages and miscellaneous foods accounted for about 8
percent of all expenditures foi food at home in 1944, it is not practical
to analyze the purchases in terms of quantity. The increases shown
in table 11 in average expense for beverages at the two lower income
levels and for soups and prepared foods throughout the income scale,
though striking, contribute a negligible share to the increase in total
food expenditures.

11
10.—Purchases of FRUITS AND VEGETABLES by Housekeeping Families
in Cities9 1 Week in 1944 and in 1942, by Annual Money-Income Rate 1

T able

September-October 1944
Item

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to and
to
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000
$4,000 over
Average expense in 1 week

$2.15
Fruits and vegetables............. =$=1.55
===
™:
Fresh and frozen..................... 1.28 1.72
Fruits................................. .67 .87
Citrus.......................... .20 .27
Other fresh................. .47 .60
Other frozen __........... .00 (2)
Vegetables........................ .61 .85
Potatoes...................... .22 .30
Other fresh................. .38 .53
Other frozen ............... .01 .02
Canned................................
.22
.37
.09
Fruits............................
.16
Regular canned.__
.08
.03
Strained and chopped
for infants................ .01 .01
Juices........................... .05 .07
Vegetables......................... .13 .21
Tomato products and
juices8...................... .05 .05
Other canned (reg
ular) *....................... .07 .14
Strained and chopped
.02
for infants...........
.01
Dried......................................
.05
.06
.03
.03
Fruits..............................
Vegetables (excluding
canned)........................
.02
.03
■V

3.14
2.57
1.41
.47
.94
(i2*4)
1.16
.39
.74
.03
.50
.20
.10
.01
.09
.30
.07
.20
.03
.07
.04
.03

3.19
2.60
1.44
.44
1.00
(2)
1.16
.34
.79
.03
.52
.22
.10
.02
.10
.30
.09
.18
.03
.07
.04
.03

$
: ,~ .

3.90 $1.28 $2.06
: ==_., ==
3.13
.87 1.48
1.68
.30
.58
.55
.16
.31
1.10
.27
.03 } *14
.57
1.45
.90
.41
.21
.24
.95 } .36
.66
.09

$
:--- :

.32
.68
.09
.30
.15
.02 1 • •
.02
.13
.23
.38
.09
.08
.27
} . »
.02 I
.09
.09
.04
.06
.03
.05

3.03
2.13
.83
.43
.40
1.30
.32
.98

$
..........

3.23
2.30
.96
.52
.44
1.34
.31
1.03

■-

4.21
3.24
1.39
.69
.70
1.85
.36
1.49

.50
.17
.13
.04
.33
.10

.79
.24
.19
.05
.55
.13

.84
.35
.26
.09
.49
.17

.90
.35

.23

.42

.32

.35

.08
.04
.04

.11
.06
.05

.09
.05
.04

.07
.05
.02

.28
.07
.55
.20

Average quantity purchased in 1 week
Fresh and frozen...............lb—
Fruits..........................lb..
Citrus...... ............lb—
Other fresh...........b r
other frozen____lb—
Vegetables..................lb—
Potatoes................lb—
Other fresh...........to
other frozen.........1b—
Canned:
Fruits:
Regular canned
No. 2H canStrained and chopped
for infants____ oz—
Juices....... No. 2 can..
Vegetables:
Tomato products and
juices8—No. 2 can.Other canned (reg
ular) *_No. 2 canStrained and chopped
for infants..........oz_.
lV fA d
_________ t o
Fruits...........................lb—
Vegetables (exclu d in g
canned)..................... to..

14.73
6.67
2.17
4.50
.00
8.06
3.85
4.19
.02

20.65
9.42
3.04
6.37
.01
11.23
6.14
5.04
.05

30.94
15.77
5.15
10.61
.01
15.17
7.63
7.46
.08

31.83
17.38
4.91
12.46
.01
14.45
7.08
7.26
.11

.11
.46
.29

.27
.54
.39

.31
.71
.53

.35
.47
.55
.25
.13
.12

.38
.81
.82
.42
.16
.26

.42
1.23
1.32
.48
.25
.23

15.76
4.67
2.98
} 1.69
11.09
6.25
} 4.84

22.95
8.82
5.97
2.85
14.13
6.88
7.25

32.55
13.28
8.75
4.53
19.27
8.81
10.46

.30
1.08
.58

.45 1
} .29
1.14 1
.67 .25

.51
.29

.75
.44

1.03
.81

1.16
.48

.61
.98
1.49
.44
.23
.21

.53 .71
1.60
!• 1.18
1.12 J
.52 .78
.29 .26
.23 .52

.79

1.05

1.51

1.58

1.65
.67
.26
.41

3.33
.96
.43
.53

2.19
.75
.36
.39

2.15
.45
.29
.16

35.54
17.17
6.27
10.79
.11
18.37
8.20
9.89
.28

36.31 43.05
16.58 17.74
12.02 11.46
4.56 6.28
19.73 25.31
8.70 11.17
11.03 14.14

i See p. 25 for income definition; p. 22 for note on adjustment of 1942 reports for comparability with respect
to family size. Averages are based on all housekeeping families and single persons in the class, whether or
not they purchased the item during the week.
8Less than 0.5 cent.
8 Includes canned tomatoes, catsup, chili sauce, tomato sauce, puree, tomato juice, and mixed vegetableJuices.
4Includes a very small amount of vegetable juice without a tomato base

—Purchases of MISCELLANEOUS FOODS by Housekeeping Families in
Cities, 1 Week m 1944 and in 1942, by Annual Money-Income Rate1

T able 11.

September-October 1944
Item

March-June 1942

Un $1,000 $2,000 $3,000 $4,000 Un $1,000 $2,000 $3,000 $4,000
der to
to
to and der to
to
to and
$1,000 $2,000 $3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
Average expense in 1 week

Miscellaneous foods................ $0.60 $0.86 $1.18 $1.31 $1.50 $0.38 $0.66 $0.96 $1.06
Beverages.................................. .34 .49 .61 .66 .71 .21 .40 .59 .60
Coffee12*............................... .21 .28 .34 .33 .37 .15 .24 .31 .33
Soft drinks......................... .07 .14 .18 .23 .25 .03 .11 .18 .18
Other 3................................ .06 .07 .09 .10 .09 .03 .05 .10 .09
Nuts and peanut butter......... .02 .06 .08 .10 .15 .02 .04 .06 .07
Packaged desserts4............ — .02 .03 .05 .06 .05 .02 .04 .05 .07
All soups, etc............. ............. .13 .16 .25 .28 .35 .03 .05 .07 .08
Soup, canned..................... .08 .09 .17 .16 .17
Soup, dehydrated............. .01 (*)
.01 .01 .01
Prepared and partially
prepared food and food
mixtures......................... .04 .07 .07 .11 .1 7
Relishes, pickles, olives.......... .02 .04 .06 .07 .06 .02 .05 .0 7 .11
Seasonings, flavorings, and
other accessories................... .06 .07 .08 .08 .14 |
Foods prepared for infants,
except vegetables, fruit, and
\ .08 .08 .12 .13
cereals.................................... .01 .01 .05 .05 .03 I
Other food................................ («)
.01 .01
(»)
(*)

$1.25
.84
.38
.39
.07
.10
.05
.06

.0 6

.1 4

Food Expense in Relation to Income
It appears that most Americans in cities were eating much better
in the fall of 1944 than might be expected under wartime conditions.
The similarity of per-capita purchases at low and high income levels
must not be overemphasized, however, because the relatively large
expenditures for meals away from home by the high-income families
provided an important supplement to the food they brought home.
Moreover, the diet of the low-income groups, though improved, was
barely adequate, if that, and it was obtained at a relatively high cost.
Increased food purchases by low-income families may be attributed
in some part to the fact that in 1944 a smaller proportion of this group
lived in small communities and in the South and West than was the
case early in 1942. Since small communities—particularly in the

South and West— offer families more opportunity than do large cities
to produce food for their own use, relatively greater purchases would
be required on the average by this group to maintain their customary
diet. Similarly, since proportionately more of the middle and upper
income families were living in small cities and in the South and West,
they could maintain the same level of living with somewhat smaller
purchases.
A more significant factor may be the types of families represented^!
different income levels in the two periods. The under-$l,000 group
comprises— in varying proportions— newly employed single persons,
young couples, retired persons, and persons requiring public assistance.

The latter two groups presumably spend less for food than the former,
since they have little expectation of improving their economic status.
In 1944, because of high wage levels, the proportion of young persons
with incomes under $1,000 was probably smaller than in 1942, but
those that were in the group may have felt that employment prospects
were so bright that they could spend freely. In addition, there must
have been represented in 1944 a sizable group of servicemen’s wives
living on allotments, but accustomed to incomes that permitted a
much better scale of living. Information collected in the survey of
spending and saving in wartime suggests families do not immediately
adjust their diet to correspond with a decline in income.
The converse operates also, according to the findings of the 1941-42
study, i. e., families whose incomes have increased do not immediately
increase their expenditures for food to an amount that is customary
among families that have been living at the same income level for a
long period of time. There is no doubt that many of the families
with incomes of $3,000 or more in 1944 had never before earned so
much.
Regardless of the reason for the high food expenditures by families
with incomes below $1,000, it seems obvious that they must have run
up large debts, cut seriously into any savings they had, or reduced
their purchases of other living essentials to a very low point.
In the spring of 1942 families and single persons that cooked at
home spent $5.61 per week for food, out of an average weekly income
of $10.90—51 percent; this left them heavily in debt.5 In the fall of
1944, in contrast, families of the same size in the same income group
spent $8.42 per week for food, or about 71 percent of their weekly
income of $11.85.®
At the three middle income levels, the proportion of income spent
for food was about 4 percent greater in the recent period than in the
earlier one, ranging from 39 to 36 to 28 percent as contrasted with 35
to 31 to 25 percent at successively higher income levels. The some
what larger food expenditures in 1944 than in 1942 by families in the
highest income group were covered by their higher average income.
Hence, in both periods, the food bill absorbed about 17 percent of
income. This group probably saved a larger proportion of their in
come in 1944, since new automobiles and many other durable goods,
which normally absorb a significant share of their spending, were
not available.
Although there were fewer city families with incomes below $1,000
and $2,000, respectively, in 1944 than at any previous time in our
history,7 this fact does not reduce the seriousness of the situation for
that group, particularly if present price levels are maintained or
increased.*•
• See Bureau of Labor Statistics Bulletin No. 724, tables 9 and 10, for data on expenditures and saving?
or debts in relation to income, for all city families and single persons, including the nonhousekeeping group.
• This income figure and those used in the discussion that follows are preliminary estimates based on the
findings of the second part of the survey of prices paid by consumers as to income (after pay-roll deductions)
received during the entire year 1944. Income averages are not available by class for 1944, based on report?
for August, since in ** considerable number of cases the information provided for August was sufficient only
as a basis for classification.
• Estimates of the distribution by income of city families and single persons will be available in a few
months from data collected in February 1945 in the study of prices paid Dy consumers in 1944.

Food Expense in Relation to Family Size
Family outlays for food are of course directly related to the number
of persons to be fed, but amounts spent per person at each income
level are smaller as family size increases (table 12). In the case of
food bought for use at home, there is some economy in large-scale
buying. However, the small savings that may be effected are by
no means sufficient to make up the differences in diet suggested by
the lower per-capita expenditures for food at home.
T able 12.—Average Expense for Food at Home and Away by Housekeeping Families in
Cities, by Family Size and Annual Money-Income Rate, Week in 1944 and 1942 1
September-October 1944
Number of persons in
family

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to and
to
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
Average total food expense in 1 week

All families...............................
1 person.............................
2 persons............................
3 persons............................
4 persons............................
6 persons............................
6 or more persons..............

$8.42
5.16
8.26
14.36
(*>
(2)
(2)

$11.87
9. 33
10.17
13.59
14.23
18.53
19.01

$17.04
9.76
14.42
16.53
18.58
20.55
24.51

$18.70
(2)
17.10
17.88
19.43
21.23
25.33

$23.51
(2)
18.21
21.77
23.07
27.05
29.58

$5.61
4.12
5.42
8.36
7.95
9.36
(2)

$10.32 $14. 54
8.09 (2)
9.93 12.87
11.52 14.02
12.30 16.63
10.90* 17.04
13.65 21.20

$16.41 $21.00
(2) 15.98
(2)
12.93
16.34 18.19
17.82 21.05
18.48 23.96
22.41 30.34

Average expense for food at home in 1 week
All families...............................
1 person______________
2 persons............................
3 persons............................
4 persons............................
6 persons............................
6 or more persons..............

$7.57
4.41
7.44
12. 92
(2)
(2)
(2)

$10.44
7.29
8.85
12.32
12.96
16. 93
18.06

$14.43
7.54
11.21
14.03
16.17
18.24
22.77

$15.08
(2)
12.57
14.82
15.99
18.13
22.07

$18.49 $5.24 $9.16 $12.68 $14.21 $17.11
6.68 (2)
(2) 3.66
(2) 12.03
(2)
11.82
5.19 8.88 10.99
10.12
16.95 7.90 10.20 12.50 14.03 14.36
18.42 7.27 11.61 14.40 16.05 16.69
22.67 8.29 10.12 15.23 16.58 20.22
24.52 (2) 12.14 19. 32 20.32 26.71

Average expense for food away from home in 1 week
All families...............................
1 person _____________
2 persons............................
3 persons................ ...........
4 persons............................
5 persons............................
6 or more persons..............

$0.85 $1.43 $2.61
.75 2.04 2.22
.82 1.32 3.21
1.44 1.27 2.50
2.41
(2) 1.27
2.31
(2) 1.60
.95 1.74
(2)

$3.62
(2)
4.53
3.06
3.44
3.10
3.26

$5.02 $0.37 $1.16 $1.86 $2.20
.46 1.41 (2)
(2)
(2)
6:39 .23 1.05 1.88
2.81
4.82 .46 1.32 1.52 2.31
4.65 .68 .69 2.23 1.77
4.38 1.07 .78 1.81 1.90
5.06 (2) 1.51 1.88 2.09

$3.89
(2)
3.95
3.83
4.36
3.74
3.63

1 See p. 25 for income definition; p. 22 for note on adjustment of 1942 reports for comparability with respect
to family size.
*Averages not shown for fewer than 10 cases.

Since average expenditures for food away from home by families
in the same income group show no consistent relation to family size,
it is clear that restaurant meals do not serve to round out the diet of
large families as compared with small ones. Rather, it appears that
persons in large families carry their lunches to work and to school
much more often than single persons or members of small families.
Information is not available on the amounts of various foods pur
chased by families of different size. It is logical to assume, however,
that families with incomes of $4,000 or more, and most of those receiv
ing $3,000 to $4,000, were able to allocate their income in a manner
that allowed sufficient money for food to purchase a good diet eveft
when the family consisted of 5 or more members.

Among large families with smaller incomes, adequate diets were
probably the exception rather than the rule, even in 1942, and cer
tainly in 1944. Furthermore, the proportion of income required to
cover their food bills was far in excess of that required by small
families. The general problem posed above on the basis of class aver
ages is thus in fact much more critical for large families. Since this
usually means families with young children, it is particularly serious.
Families of one or two and, at income levels above $2,000, somewhat
larger families do not, of course, face the same problem.
The percentage increase in food expenditures between the spring of
1942 and the fall of 1944 tended to be greater among large than
among small families at each of the lower income levels. This indi
cates merely that the usual food purchases of small families provided
more leeway for purchase of cheaper—or less—food in the face of
price increases, while this was not true for the larger families whose
diets were already restricted.

Appendix.—Survey Method
SAMPLING PROCEDURE

Selection of Sample for 1944 Study

The Survey of Prices Paid by Consumers in 1944 was based on a
sample of approximately 1,700 families and single persons in 28
metropolitan districts and 20 cities with populations under 50,000
outside of metropolitan districts. These places were selected to
represent, with respect to region, State, and city size, all cities in the
United States with a population of 2,500 or more. Each metropolitan
district with a population in excess of 1,275,400 was automatically
included in the sample. Each of the smaller metropolitan districts
and the cities with populations under 50,000 represented a stratum of
cities of approximately the same size in the same region.
For classification of cities by size, the November 1943 population of
each city in the United States was estimated by applying to the 1940
census population figure the percentage change in the population of
the county in which the city is located, as indicated by the registration
for OPA Ration Book No. 4. The number of dwelling units 1in each
stratum was estimated for 1943 by dividing the estimated population
of all the cities in the stratum by the average size of private house
hold,2 as shown by the 1940 census, for the city selected to represent
the stratum. The quota of 1,400 dwelling units for the survey was
distributed among the 48 cities (28 metropolitan districts and 20 small
cities) in proportion to the number of dwelling units in the stratum
represented by the sample city, for the purpose of determining the
sampling ratio for each city. This was obtained by dividing the dwell
ing-unit quota for the city by the estimated number of dwelling units
in that city in 1943.
The quota for each metropolitan district was distributed between
the central city or cities and a sample of other cities and towns in the
district. To select the sample places within a metropolitan district,
all urban places in the district were divided into four strata on the
basis of their population. The dwelling-unit quota for each metro
politan district was distributed among the four strata in proportion
to then* population. Within the stratum of the largest communities,
all cities with sufficient population to warrant a dwelling-unit quota of
3 or more were automatically included in the sample. Within the other
strata, the dwelling-unit quota of the stratum was allocated to a city
i A dwelling unit was defined in the 1940 census as the living quarters occupied by, or intended for occu-

ancy by, one household. A dwelling unit may be a detached house; a tenement, flat, or apartment in a
Eirger
building; or a room in a structure primarily devoted to business or other nonresidential purposes. It

may be a superintendent’s living quarters in a public building; a watchman’s living quarters in a factory; or
a tourist cabin, trailer, railroad car, boat, tent, etc., if occupied by persons having no other place of residence.
The following special types of dwelling places were not considered dwelling units: Hotels for transient
guests, and similar places maintained by the Y. M. C. A. or kindred organizations; missions; cheap onenight lodging houses; dormitories for students; nurses’ homes; educational or religious institutions; military
institutions; penal institutions; soldiers’ homes; homes for orphans and for the aged, blind, deaf, infirm,
or incurable; Civilian Conservation Corps camps; Coast Guard stations; and lumber camps, and railroad or
other construction camps, in which the workers live in a common barracks.
2 The term private household was used in the 1940 census to include the related family members and the
lodgers, servants, or hired hands, if any, who regularly lived in the home.
( 16)

selected at random. Eighty-two places within metropolitan districts
were chosen, making a total of 102 separate urban places in the final
sample. (See table A.)
A sample of blocks was selected in each of the 102 urban places.
For cities which had over 50,000 population, census “block statistics”
were used,3 but for smaller places, for which block statistics are not
available, block maps were used. The block sample for the former
group of cities comprised blocks that contained dwelling units in 1940
and blocks that contained no dwelling units in 1940, i. e., undeveloped
blocks, business blocks, blocks containing only hotels, YMCA's,
dormitories, etc.
For each city having a population of 50,000 or more, the number of
sample blocks with dwelling units was equal to the dwelling-unit
quota for the city. These blocks were selected at random in such a
way that each represented a group of blocks having approximately
the same total number of dwelling units in 1940. They were grouped
according to density of dwelling units and the within-block sampling
ratio for each group was determined in such a way that each block
would yield one dwelling unit, on the average, provided the number of
dwelling units was the same as in 1940. For each large city, the sample
of blocks without dwelling units in 1940 was 50 to 100 percent as large
as the sample of blocks with dwelling units. The exact number of such
blocks to be included in the sample for each city was determined in
such a way that the within-block sampling ratio would not be exces
sively low or excessively high, the limits being fixed by the city sam
pling ratio, i. e., the product of the within-block sampling ratio and the
block sampling ratio.
For cities under 50,000, sample blocks equal in number to the
dwelling-unit quota for each were drawn at random from city maps on
which were numbered all blocks except those known positively to
contain no dwelling units. The within-block sampling ratio was
accordingly the same for all blocks and equaled the city sampling ratio
divided by the ratio of the number of sample blocks to the total num
ber of blocks.
For each sample block a listing was prepared of all residences,
(i. e., of all dwelling units occupied by private households and of all
rooms in lodging houses, hotels, nurses’ homes, etc.) in the block.4
Thus, while the sampling unit used for selection of blocks was a
dwelling unit, a residence was used as the unit for selection of assign
ments for interview. For scheduling and analysis, an economic
family (see definition, p. 24) formed the basic sampling unit.
In cities with a population of 50,000 or more, the listings for blocks
having the same sampling ratio were grouped and every nth residence
was selected, n being determined by the within-block sampling ratio
for the group of blocks. In smaller places, the block lists were com
bined and every nth residence selected, n being determined by the*
* Sixteenth Census of the United States, 1940: Housing, Supplement to First Series—Block Statistics.
Block statistics for 1940 were prepared by the Census only for cities which had a population of 50,000 or more
in 1930. Of the cities, with a population of 50,000 or more in 1940 covered in the Survey of Prices Paid by
Consumers, block statistics are not available for Corpus Christi, Tex.; of those with a population below 50,000
in 1940, block statistics are available for Jackson, Mich., and Kenosha, Wis. All subsequent references to
sampling procedures in the two city-size groups relate to cities of the designated size in 1940, with these
exceptions.
< Residences as defined for purposes of the survey included dwelling units as defined by the census (except
that the separate rooms in lodging houses rather than the lodging houses themselves were treated as resi
dences) and all rooms in hotels, nurses’ homes, etc.

18
T able

A.—Cities Included in Sample for Survey of Prices Paid by Consumers, by
Region

Cities within metropolitan districts
Central city or cities

Other cities and unincorporated
places

Cities outside metropolitan
districts

New England
Boston, Mass.
Providence, R. I.

Bridgewater, Mass.
Cambridge, Mass.
Chelsea, Mass.
Dedham, Mass.
Malden, Mass.
Manville, R. I.
Newport, R. I.
Pawtucket, R. I.

New London, Conn.
North Adams, Mass.

Middle Atlantic
Tonawanda; N. Y.
Buffalo, N. Y.i
New York, N. Y., and Jersey City and Fort Lee Borough, N. J.
Hoboken, N. J.
Newark, N. J.3
Long Branch, N. J.
Williston Park Village, N. Y.
Bridgeport Borough, Pa.
Philadelphia, Pa.
Norristown Borough, Pa.
Bridgeville Borough, Pa.
Pittsburgh, Pa.
Canonsburg Borough, Pa.
Dickson City Borough, Pa.
Scranton and Wilkes-Barre, Pa.
Moosic Borough, Pa.
Solvay Village, N. Y.
Syracuse, N. Y.

Berwick, Pa.
Clean, N. Y.

South Atlantic
Baltimore, Md.
Jacksonville, Fla.
Roanoke, Va.
Washington, D. 0.

Sparrows Point, Md.
Vinton, Va.
Arlington County, Va.
Hyattsville, Md.

Goldsboro, N. 0.
Kingstree, S. 0 .
Rome, Ga.

East South Central and West South Central
Corpus Christi, Tex.
Houston, Tex.
Little Rock, Ark.
Memphis, Term.

West University Place, Tex.
North Little Rock, Ark.

Gulfport, Miss.
Kingfisher, Okla.
McAllen, Tex.

1 Niagara Falls, N. Y., was not included in the sample for the Buffalo-Niagara metropolitan districts,
although it is one of the two central cities.
2 In New York City, Bronx, Brooklyn, Manhattan, and Queens Boroughs were sampled, and not Rich
mond Borough. Elizabeth and Paterson, N. J., were not included in the sample for the New York—North
eastern New Jersey metropolitan district, although they are classified as central cities in the New Jersey
division.

19
T able

A.—Cities Included in Sample for Survey of Prices Paid by Consumers, by
Region—Continued
Cities within metropolitan districts
Central city or cities

Other cities and unincorporated
places

Cities outside metropolitan
districts

East North Central
Chicago, 111.
Cincinnati, Ohio
Cleveland, Ohio
Detroit, Mich.
Racine and Kenosha, Wis.
Toledo, Ohio

Berwyn, 111.
Brookfield, 111.
Gary, Ind.
Hobart, Ind.
Covington, Ky.
Lawrenceburg, Ind.
Berea, Ohio
Garfield Heights, Ohio
Dearborn, Mich.
Grosse Pointe Woods, Mich.
River Rouge ,Mich.

Bluffton, Ind.
Jackson, Mich.
Jacksonville, 111.
Wooster, Ohio

Maumee Village, Ohio
West North Central

Cedar Rapids, Iowa
Kansas City, Mo., and Kansas City, Independence, Mo.
Kans.
Edwardsville, 111.
St. Louis, Mo.
Granite City, 111.
University City, Mo.

Detroit Lakes, Minn.
Grand Island, Nebr.

Mountain
Cheyenne, Wyo.
Deming, N. M.
Pacific
Los Angeles, Calif.
San Francisco and Oakland, Calif.

Beverly Hills, Calif.
Fullerton, Calif.
Glendora, Calif.
Long Beach, Calif.
Menlo Park, Calif.
Redwood City, Calif.

Corvallis, Oreg.
Riverside, Calif.

Seattle, Wash.

within-block sampling ratio for the city. Whenever a residence con
tained more than one economic family, each such family was included
in the sample.
Substitutions were required whenever the occupants of an assigned
residence refused to provide the information requested or when it
proved impossible to contact the occupants after repeated visits.
Substitute residences were drawn at random from the listing of resi
dences between the one requiring a substitute and the next assigned
residence; The only control on substitutions was that a housekeeping
unit should replace a housekeeping unit and a room should replace
a room. The final substitution rate was approximately 12 percent.

Substitutions were not required for vacancies, for residences closed
for 2 or more weeks by virtue of the occupants’ absence from the city,
or for residences (usually rooms) occupied by persons who were
members of an economic family residing elsewhere. When visiting
families were found at an assigned address, a schedule was taken
provided the family did not expect to return to its regular residence
for 2 or more weeks. Persons living temporarily apart from their
family were included as part of that family whenever such a family
wras drawn in the sample.
Use of a sampling ratio based on a dwelling-unit quota of 1,400
yielded a sample of more than 1,700 families of one or more persons.
The excess over the dwelling-unit quota resulted principally from
use of ratio sampling, applied uniformly to all residences—rooms as
well as private dwelling units—and from scheduling the separate
families living in the same dwelling unit. A total of 1,728 residences
wTere selected for the sample. Residences that were vacant, closed for
2 or more weeks, or occupied by families returning home within 2
weeks or by persons who were members of families living elsewhere
totaled 154. This left a total of 1,574 residences which should have
yielded at least one schedule. The number of schedules obtained was
1,719, or an average of 1.09 families to each “ occupied” residence.
Selection of Sample for 1942 Study 5
The urban sample for the Survey of Family Spending and Saving
in Wartime covered approximately
families and single persons
in
cities with a population of
or more in
The com
munities were selected to give proper representation with respect to
size, proximity to a metropolis (for cities under
region and
State, rental level of the city, and racial composition.

1,300
2,500

1940.
50,000),
The number of residences to be visited in each city (totaling 1,200)
wTas fixed on the basis of a quota determined by the number of occupied
dwelling units in the stratum represented, as shown by the census
of 1940.

A sample of blocks was selected for each city, the number of blocks
being the same as the number of residences to be visited. For cities
of
or more, average block rent in
was used as a basis for
stratification in the selection of blocks; only blocks with occupied
dwelling units in
were included. For cities under
for
which census block statistics were not available, all blocks on a
detailed block map were numbered consecutively and every nth
block selected.
A listing was then prepared of all families containing one or more
persons living in each sample block. For the cities of
or more,
one family was selected at random from the listing for each block.
For the small cities, the block lists were combined into one consecutive
listing and every nth family selected, n being equal to the number of
sample blocks in the city. In some instances, an assigned unit proved
to consist of two or more independent economic units. In every
such case, each of the economic families was scheduled separately.
The
original assignments thus yielded almost
economic
families of one or more persons.6

50,000

1940

50,000

1,200

1,300

• For complete discussion see Family Spending and Saving in Wartime, Bureau of Labor Statistics Bul
letin No. 822, Part I. (In press.)
6 In accordance with the quota sampling plan, substitutions were required for vacancies as well as for
families that refused the information requested or that could not be contacted. The final substitution
rate was 7 percent.

Comparison of the Two Samples

The primary differences between the two samples were the use of
ratio sampling in the selection of sample residences in 1944 as against
the use of fixed quotas in 1942 and the coverage in 1944 (in cities of
50,000 or more) of a sample of blocks which in 1940 had no occupied
dwelling units as defined by the census and the exclusion of such
blocks in 1942.
The use of 1940 population figures to determine the distribution of
a sample in the spring of 1942, together with the use of fixed quotas,
resulted in an underrepresentation of war production centers with
marked increases in population and, therefore, a probable under
statement of wage and salary income. While this feature of the
sampling method does not appear to have seriously affected the data
with respect to income distribution,7 it may have resulted in under
estimates of some of the expenditures at the various income levels.
The comparison of the two surveys may, therefore, exaggerate some
what the changes that have taken place. There is no doubt that the
1944 data were greatly improved by including blocks in which there
had likely been home construction in the period since the census, as
well as hotel blocks, and by use of ratio sampling which takes full
account of population changes.
Since considerable migration is known to have taken place during
the 4^-year period from April 1940 to the time of selection of the 1944
sample, the distribution of families at a given income level by region
and city size is somewhat different in the two studies. Table 1 shows
the distribution by region and 1940 city-size class of families and
single persons sampled at given income levels in the two studies.
While it would be desirable for certain purposes to isolate the effect of
changes in regional and city-size distribution, it is nevertheless true
that, given the population changes that have taken place, each survey
provides a reliable picture of urban spending at the respective dates.
One serious limitation to valid comparison arises from the under
representation of single consumers in the 1942 study.8 Exclusion in
the 1942 study of blocks containing no dwelling units in 1940, con
tributed in large measure to the apparent underrepresentation of
lodgers and hotel residents—typically single consumers—as well as
occupants of recently constructed war housing or trailer camps,
which might be expected to have housed a heavy proportion of single
persons.7 A second factor was the selection of blocks with reference
to the number of dwelling units, whereas the general sampling unit
used in the survey was an economic family. Thus, for example, a
block with a few private dwellings and a number of dormitories and
lodging houses (i. e., a large proportion of single persons) had a
relatively small chance of inclusion in the sample since a dormitory
was not counted as a dwelling unit and a lodging house was con
sidered only one dwelling unit. Similarly, a block in which most
private families had a few lodgers had a relatively small chance of
inclusion. This was true in both surveys, but the ratio sampling
plan followed in 1944 gave the lodgers and hotel residents the same
chance of inclusion in the final sample as all other families and single
persons. In 1942, because of the quota sampling plan, lodgers had a
smaller chance of inclusion than did other families and single persons.
7 See Family Spending and Saving, Bureau of Labor Statistics Bulletin No. 822, Part II. (In press.)
8This is less serious when purchases of housekeeping families and single persons are compared, as in the
present report, than when outlays by all families and single persons are compared, since lodgers comprised
a large proportion of the single persons lost to the 1942 sample.

22
ADJUSTMENT OF 1942 DATA FOR COMPARABILITY AS TO FAMILY SIZE

The average family size was considerably smaller in 1944 than in
1942, particularly at income levels under $2,000 and over $4,000, as
a result of the underrepresentation of single consumers in the 1942
sample and the heavy drain of the armed services between March
1942 and August 1944.
The effect of the war on total expenditures and on the amounts of
various foods purchased by families at different income levels can be
measured accurately only by comparing the reports of families of
the isame size. Both surveys were too small to allow presentation
of the detailed data by family size and income.9
Accordingly, 1942 averages presented for families of all sizes at
each income level have been standardized as to family size, i. e.,
adjusted to represent the purchases characteristic of families of the
sizes found in the 1944 study. The adjusted average family expense
for all food at home and away from home, respectively, was derived
by weighting the averages for each of 7 family-size groups (families
of 1 through 7 or more) by the number of such families in the same
income group covered in the 1944 survey. Standardized averages
for expense for food at home at each income level were then converted
to a per-capita basis and the component items (quantity as well as
expense) estimated from the detailed tabulations for 1942 by linear
interpolation, at the point in the income scale where actual percapita expenditures coincided with the standardized estimate.10
The distribution of housekeeping families by family size in the
two studies is given in table B. The average family size in 1942 and
in 1944 is shown in table C, together with the average weekly expen
diture for food at home in the spring of 1942, as reported and ad
justed.
B .—Percentage Distribution of Housekeeping Families in Cities by Size and
Annual Money-Income Rate, 1 September-October 1944 and March-June 1942

T able

Annual rate of money income
Family size

September-October 1944

March-June 1942

$2,000 $3,000 $4,000 Under $1,000 $2,000 $3,000 $4,000
Under $1,000
to
to
to and
to
to
to and
$1,000 $2,000
$3,000 $4,000 over $1,000 $2,000 $3,000 $4,000 over
All families_______________ 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Families containing—
1 or 2 persons__________ 77.0 59.7 32.0 28.1 18.4 70.0 45.7 33.8 25.7
3 or 4 persons..................... 17.8 33.1 54.2 52.6 50.6 22.3 39.1 49.3 54.3
5 or more persons.............. 5.2 7.2 13.8 19.3 31.0 7.7 15.2 16.9 20.0

100.0
13.5
57.4
29.1

1 See p. 25 for definition of income.
9 Comparison on a per-capita basis is not satisfactory because the distribution of families among the size
groups varied considerably and food purchasing habits of large and small families with similar incomes
differ markedly.
10 Previous studies indicate that within certain limits, the distribution of total food expense amofig
items is similar for families with similar per-capita outlays for food.

C.— Average Food Expense of Housekeeping Families in Cities, 1 Week in 1942,
as Reported and as Adjusted to Represent the Spending of Families of the Sizes in the
Sample Population in 1944, by Annual Money-Income Rate 1

T able

Annual rate of money income

Item

Under $1,000 to $2,000 to $3,000 to $4,000
$1,000 $2,000 $3,000 $4,000 and over

Average family size—
As reported in 1942..................................................
As reported in 1944...................................................
Food at home in 1 week:
Average expense per family—
As reported.........................................................
As adjusted ............... .....................................
Average expense per person—
As reported.........................................................
As adjusted--.....................................................

2.28
2.10

3.03
2.52

3.30
3.24

3.59
3.56

4.20
3.93

$5.39
5.24
2.36
2.50

$9.71
9.16
3.20
3.63

$12.84
12.68
3.89
3.91

$14.43
14.21
4.02
3.99

$17.83
17.11
4.25
4.35

i See p. 25 for definition of income.
COLLECTION OF FOOD DATA
Information on 1 week’s purchases of food for home use was ob
tained in
and in
by personal interview, using a detailed
list of approximately 200 food items. There were certain differences,
however, in the design of the two food schedules and in the emphasis
placed on the food reports.
The differences in schedule design were relatively minor. A
greater effort was made in 1944 than in 1942 to obtain an exact
description of the foods bought with respect to quality, unit of pur
chase, brand, etc., because of the interest in comparison of prices
reported by consumers with prices reported to the Bureau by retail
stores. The arrangement of items varied somewhat. Furthermore,
prepared and partially prepared foods, and foods put up especially
for infants, were itemized in more detail on the 1944 than on the
1942 schedule, on the ground that purchases of such foods had become
more common in recent years.

1942

1944

In the Survey of Prices Paid by Consumers, primary emphasis
was placed on the collection of information on food purchases, whereas
in 1942 the food schedule was appended to a detailed schedule cover
ing the income and expenditures of the family during 1941 and the
first quarter of 1942. In 1944, therefore, the food report was obtained
early in the interview, immediately after a few preliminary questions
on family composition and living arrangements. In 1942, by con
trast, the week’s food report was usually taken last in the interview
or series of interviews required to complete the schedules. It might
be expected, therefore, that the 1942 food reports would be somewhat
less complete than the 1944 reports. If this were the case, differences
would be expected principally in the case of small purchases in the
miscellaneous group which are made irregularly and might easily be
forgotten. Preliminary examination of the reports on miscellaneous
food purchases, however, suggests that there is no clear bias of this
nature. Reports on dairy product and meat purchases, which would
be expected to be the most complete and accurate in a hasty recital,
generally showed greater outlays in 1944 than in 1942, except in the
case of rationed food purchases by families whose incomes probably
allowed larger meat purchases than the general supply or their red
points permitted.

24
DEFINITIONS
Unless otherwise indicated, the definitions cited below are common
to both the Survey of Prices Paid by Consumers and the Survey of
Family Spending and Saving in Wartime.
Economic family

.—(1) A group of persons, usually related, who
lived together during a designated period, contributing to the family
income or receiving a large part of their support from family funds.

All relatives of the head of the family who ordinarily lived with the
family but were temporarily away from home (at work in a civilian
occupation, at school, in a hospital, or on a visit) were included as
family members, provided they either contributed to the family in
come or received a large part of their support from family funds.
Related persons in military service, living on military reservations,
were not included as family members.

(2) An individual who lived independently, apart from relatives,
as a 1-person economic family. In the 1942 survey, adult sons or
daughters who made regular payments to their parents for room and
board were sometimes treated as 1-person families, but more often as
members of their parents’ family. Inmates of institutions, as well as
residents of military camps and posts, were excluded from the surveys.
Family size.—Family-size figures presented in conjunction with

data on food represent the number of family members living at home
during all or part of the week covered by the food report. Measure
ment of family (or household) size in relation to the number of meals
consumed in the home during the week (i. e., considering 21 meals, to
represent 1 person) provides a more refined tool for analysis of food
purchased for home consumption.

A count of all meals eaten in the home during the survey week was
available for the earlier but not for the later study. Hence the rougher
measure, i. e., the count of family members, was used in both in
stances to insure comparability. It is apparent from the similarity
at each income level in 1942 in average family size as measured by
the two methods, that meals eaten away from home by family mem
bers are largely balanced by meals served in the home to boarders,
guests, and servants:
Average family size in March-June 194&,
as measured by—

Annual rate of money income:
Under $1,000____ _____
$1,000 to $2,000_________
$2,000 to $3,000...................
$3,000 to $4,000____ _____
$4,000 and over__________

Number offamily
members living
at home

2. 28
3. 03
3. 30
3.59
4. 20

Number of mealequivalent
persons
2. 20

2. 98
3. 27
3. 51
4. 05

Housekeeping.—A family occupying a house, flat, or apartment
with regular cooking facilities was defined as housekeeping.11 One occu
pying an apartment without cooking facilities, or a room or suite of
rooms, was defined as nonhousekeeping, even though some of the
latter had hot plates in their rooms or kitchen privileges, which en
abled them to do some cooking. In 1944 about 11 percent of the
roomers had light-housekeeping privileges and prepared some food at

11A few residences were occupied by two or more families that cooked and ate together. In each case,
the week’s purchases of food for home use were recorded on the schedule of one of the families. For
analysis of food purchases, the others were treated as boarders.

home. The family (usually 1-person) that had cooking facilities but
did not cook at home is excluded from the group on which the data
on fodd at home are based.
Of tHe •f;700 families covered in the Survey of Prices Paid by
Consumers in 1944, approximately 1,500 were keeping house and
provided information on food purchases for home consumption.
In the Survey of Spending and Saving in Wartime in 1942 the corre
sponding numbers were approximately 1,300 and 1,100.
Money income.—The annual rate of income was based on reports for
August 1944 in the recent study and on reports covering the first 3
months of 1942 in the earlier one. In both surveys, family income
represents the sum of all types of income received by family members
during those periods: wage and salary earnings, entrepreneurial net
income or withdrawals, and nonearned income from all sources,
exclusive of inheritances, large gifts, and lump-sum insurance settle
ments. For the 1942 study, total wage and salary earnings less
occupational expense was used in computing the family's total money
income, but for the 1944 study, wage and salary earnings after pay
roll deductions was used. Because of the increase in income-tax
rates since early 1942, an income classification for 1944 based on takehome pay is believed to provide a better basis for comparison, at
corresponding income levels, of food purchases in the two periods.
Income averages are not available by income class for 1944 based
on reports for August, since in a considerable number of cases the
information provided for August was sufficient only as a basis for
classification.12 Since the proportion of families with high incomes
was very much greater in 1944 than for 1942, it might be expected
that the average income of the lowest and highest classes would be con
siderably greater in the more recent period. Preliminary figures
obtained in the second part of the survey of prices paid by consumers
on money income received during the entire year 1944 indicate that
the income of housekeeping families and single persons receiving less
than $1,000 (after pay-roll deductions) averaged $616 in 1944 as
compared with $567 in 1942, based on the first 3 months; the income of
those receiving $4,000 or more averaged $7,235 as compared with
$6,513 in 1942. In 1944, as in 1942, the average for each of the middle
income groups was near the midpoint of the class.
Food expense.—Includes the cost of all food purchased during the
week for use at home during that week or at a later time, and all food
purchased away from home during the week, whether or not the items
purchased were paid for at the time.
Food at home includes all food and nonalcoholic drinks purchased
to be served at home or to be carried from home (as box lunches),
whether the food was served to boarders, guests, servants, or family
members. Expense figures exclude amounts paid as sales taxes in
places having a sales tax on food. It is practically impossible to
allocate a sales tax to specific items since a sales tax is customarily
collected on the total expense for all items bought at one time.13
12 Approximately 3 percent of the sample families either failed to report income or provided information
that was inadequate as a basis for classification. Most of them were keeping house. The size of their
families and their food expenditures averaged about the same as for housekeeping families that reported
income.
**In the 1942 study, the sales tax in cities where applicable was computed as a single item on the weekly
food schedules. The amount of these taxes, averaging 7 cents or less for all families at an income level, is
not included in the total food expense figures for 1942 because similar figures are not available for the 1944
survey.

Food away from home includes board, meals, and between-meal
snacks purchased in restaurants and cafeterias, and at counters and
fountains, and ice cream, nonalcoholic drinks, etc., bought to eat
with meals carried from home.
Quantity of food purchased for home consumption.—Every effort was
made in both studies to obtain an accurate measure of the quantity
purchased and the unit of purchase for every item. For summariza
tion, purchases were converted to a common unit. When the diver
sity of purchase units for items included in a particular food group
was very great, however, conversion to a common unit was impractical
and figures are presented only on amounts spent.
Basis of item averages.—Averages are based on all housekeeping
families and single persons in the class, whether or not they purchased
the item during the week.

t . 8 . GOVERNMENT PRINTING OFFICE: 1 9 4 8

Full text of Wartime Food Purchases : Bulletin of the United States Bureau of Labor Statistics, No. 838

FRASER