INSTITUTE of STATISTICSOxford

WORKING CLASS INCOME & HOUSEHOLD EXPENDITURE

PART I: INCOME & EXPENDITURE ON NON-FOOD ITEMS

(i) INTRODUCTORY

The inquiry into the household expenditure of working-class familiesundertaken by us in the summer of 1946 was the eighth in this series, whichstarted in 1940. Tne method of sampling employed was similar to themethods adopted for our previous inquiries, and the household accountswere obtained for a consecutive fortnight in June or Julyl from families ofurban workers with at least one dependent child at home.2

The original purpose of this series of surveys was the collection andanalysis of data indicative of the main economic effects of war-time conditionson working-class housekeeping, and it was intended to conclude it at the endof the war. However, the continuation of serious shortages and of rationingmade it desirable to carry on beyond this date, and we decided upon twomore inquiries, in 1946, and, we hope, in 1947.

Since our samples are small and full co-operation from the housewivescould be secured only once a year for a fortnight, the resulting data call forcareful interpretation. Only two sets of the figures obtained by us can beconsidered fully representative in themselves, viz., those referring to thepurchases of food and those referring to rent.3 So long as we obtained nostatements of family income the figures of outlay on clothing and miscellane-ous items were important mainly as a check on the figures of total house-keeping money. Their value has been greatly enhanced since, in more recentyears, a substantial proportion of the families were willing to reveal their totalincome. The different groups of expenditure can thus be set out against thislimiting background, which removes much of the inherent indeterminatenessof some of the data. We may not always know the complete expenditure on agroup of items; but we now do know the limits of potential expenditure onit in relation to the expenditure on other items.

A few accounts cover a later period.2 Except in those instances where the family had contributed to earlier samples and the

children have since left school and become earners.'Since our samples refer to the summer months, the figures of outlay on fuel and lighting

show, of course, a seasonal bias, although this is to be found probably mainly in the liguresof consumption of gas and electricity, as most of the families seem to purchase during tuesummer coal stocks for the winter.

Bulletin Vol. 9 May, 1947 No.

THE BULLETIN

(z) THE STRUCTURE OF THE SAMPLEGeneralThe collection of our sample was again made possible through the

generous assistance of voluntary helpers-university teachers, heads ofsettlements, etc.-in the several towns. Unfortunately, many of the house-wives approached felt too tired and overburdened with housekeepingproblems to co-operate, and only i i families supplied a completed householdaccount for a fortnight, as compared with 155 families in 1945, and zzô in1944. Nevertheless, the number of persons covered by the accounts wasslightly larger in 5946 than in 1945, viz., 727 as against 712. No accountswere obtained from Lincoln, St. Helens, and Dundee, which between themwere represented in 1945 by x8 families. On the other hand, Reading, whichhad dropped out of our sample in s 945, came in again with ii accountscovering the household expenditure for 55 persons.

In 5945, the total number of families that had kept an account for us alsoin the preceding year, was iii; of these, 8 had remained unchanged instructure while z6 had changed, either on account of the birth of a child, thedemobilization of a member of the family, the marriage of a son or a daughter,or for some other reason. In 1946, the total number of household accountswhich were thus comparable for two consecutive years, was 83. There were59 unchanged and 24 changed1 families. Sixty-eight accounts came fromfamilies that had not contributed to our sample in 5945; they are styled inthis paper new families.

The Geographical Distribution of the SampleIn Table lis set out the geographical distribution of the families that kept

accounts for us in 1946, subdivided into unchanged, changed, and new families.lt presents a first outline of the sample, drawn up by means of the data of

TABLE I Geographical Disiribution of Families

* Excluding the following 19 families living rent free or buying their houses: Leeds, 1Liverpool. 1; Manchester. 2; Newcastle, 2; Southampton. 7; Swindon, 6.

fAverage for 138 families.'Besides other changes in their structure, 7 children were born to these families since our

survey in 1945.

Un- Chang- New Total No. ofchanged ed Persons

perFamily

Man'Valueper

Person

Foodp. 'Man'p. Wk.s. d.

Rent5 FuelandLight

p. Fam. p. Weeks.d. s.d.(1) Birmingham 4 2 8 14 4.9 0.75 15 44 10 04 8 94(2) Glasgow 4 - 8 12 4.2 0.71 16 6k 8 5 8 31(3) Leeds 9 1 - 10 4.2 0.82 15 04 12 7 8 5(4) Liverpool 11 7 8 26 5.6 0.81 13 74 13 04 9 84(5) London 1 1 9 11 5.7 0.71 15 54 15 71 8104(6) Manchester 7 4 4 15 5.1 0.75 14 101 12 04 7 7(7) Newcastle 1 2 2 5 4.0 0.76 15 44 12 0 4 2(8) Oxford 3 2 2 7 5.9 0.76 12 54 12 0 7 0(9) Reading - 11 11 5.0 0.75 15 54 94 66

(10) Sheffield 5 - 7 12 4.8 0.79 15 11 10114 5 04(11) Southampton lo 5 2 17 4.0 0.79 16 4 19 34 9 94(12) Swindon 4 - 7 11 3.7 0.78 15 84 18 3 8 01

Total. 1946 59 24 68 151 4.8 0.77 15 01 12 51 S 14

Total, 1945 85 26 44 155 4.6 0.77 14 24 13 0f 7 104

WORKING CLASS INCOME AND IIOOSEHOLD EXPENDITURE 135

family structure and of expenditure on food, rent, and fuel and lighting.The largest difference in average expenditure on food prevailed between the7 families from Oxford and the j z families from Glasgow, the former spendingon food zs. sid. per 'man" per week and the latter i6s. 61d. The familiesfrom Glasgow paid the lowest rent-8s. 5d. per family per weekthose fromOxford about the average rent, and the families from Southampton thehighest, viz., 19s. 3rd. per family per week. It is, perhaps, necessary to repeathere a warning given by us on previous occasions: the average figures of rentobtained from these household accounts must not be taken as an indicationof the relative levels of working-class rents in the different towns. The figurefor Glasgow falls probably roughly into perspective, since in general rentsare lower in Scotland than in England. The accommodation of some of thefamilies in Glasgow is, however, very unsatisfactory, and several of the house-wives stated emphatically that they would gladly pay considerably more thantheir present rent for more adequate housing. On the whole, the average rentpaid by the families of our sample is largely determined by the locality fromwhich the sample has been drawn. In some instances, it comes mainly fromnew housing estates where rents are relatively high, while in others it wascollected from families living in old parts of their town where lower rentsprevail.2 No explanation can be offered for the differences in the outlay onfuel and lighting, except as regards the families from Newcastle, most ofwhich purchased coal at reduced prices, since their main earners are employedin the coal mining industry.

The average size of the family should be considered together with the'man' value per person, for the lower this 'man' value the larger the pro-portion of young children. The families from Leeds and Liverpool thuscontained on the average the largest number of persons over 24 years old,and those from Glasgow and London the smallest number.

The two bottom rows in Table I compare the averages for 1945 and I946.rrhey bring out the fact that the average family in 1946 was slightly largerthan in 1945, but that its relative structure was approximately the same, sincefor both years the 'man' values per person are identical. The averagefigures for the two years of outlay on food, rent, and fuel and lighting, mustbe analysed in some detail before an interpretation can be attempted.(c) The Econoinic Outline

In the absence of full information regarding family income, standardizeddata of expenditure on food supply a serviceable gauge for an estimate

1 Thefigures of averageS man' values are based on the following scale: adult male, 1.00;adult female, 0.85 children, from 0.33 to 0.70, with an average of 0.55. The values forchildren are set out in more detail in Part II of this paper. This scale of 'man' valuesis employed by us always unless, in some sections of Part II, explicit reference is made toa different scale.

Our helpers in the different towns arrange for the collection of household accounts fromfamilies with whom they are acquainted through social work. Only through such intimateconnexion with the individual household is it possible to attain the precision of data neces-sary to give validity to a small sample.

1945 and 1946 are used in this paper as a brief reference to our sampling periods in theseyears.

136 THE BULLETIN

of the standard of living of a family. The averages of food expenditureper 'man' of the families from the different towns, given in Table I, may,of course, disguise wide ranges of food expenditure of individual families,but in Table II the families are graded systematically according to their

TAßLE IIExpenditure on Food Compared with Other Household Expenditure

1946 1945 1946Outlay on Food Number of Number of Man Av. out- Total Outlay

per Man' Families Persons Fams. Persons Value lay on House- or Rentper Week per per per Food hold out- p. fam.'

Family Family Person p. 'Man' lay' p.Week

'Including outlay on food,2 Outlay on housing comprises outlay on rent (including rates), mortgage charges, and

rates.Excluding 19 families living rent free or buying their houses, in the following groups of

food expenditure: lis.-12s., 1 family; 12s.-13s., 2 families; 13s.-14s., 6 families14s.-15s., 3 families ; 15s.-16s., 1 family ; 16s,-17s., 1 family ; 17s,-18s., 2 families19s.-20s., 1 family; 20s,-23s., 1 family; 23s. and over, 1 family.

outlay on food 'per man'. With outlay on food we compare outlay on rentand total household expenditure including outlay on food but excludingoutlay on housing. By 'housing' we mean expenditure on mortgages andhouse purchase instalments as well as outlay on rent and rates, and on ratesonly, while 'rent' refers only to the payments made to landlords; 'rent'in this context includes rates.

Only ¶32 of the 151 families lived in rented accommodation. Sinceoutlay on housing, and in particular outlay on rent, is determined by thetype of accommodation available to a family, it remains not only unaffectedby variations in the family income but often also bears no relation to theabsolute size of this income. In Table II we find that the two families withthe lowest outlay on food per' man ' paid the highest average rent per family,viz., x5s. 5d:per week. The next two groups with the highest average rentsare composed of the families spending from 17s. to 19s. per ' man ' per weekon food, while the k west average rent- Ios. 3d-was paid by two families

p.Week excludingHousing'p. 'Man'p.Week

Under 9s. - - 2 5.0 0.78 7 104 12 10 15 549s.-lOs. 4 5.8 - - - - - -

lOs-lis. 9 5.3 9 7.9 0.73 10 4 18 14 13 34lls.-12s. 19 5.7 9 6.7 0.77 11 64 20 64 12 7412s.-13s. 20 5.5 15 4.9 0.79 12 8 23 104 13 0413s.-14s. 18 5.0 25 5.0 0.77 13 51 24 54 11 2414s.-15s. 22 4.5 16 4.7 0.78 14 5 23 114 12 7415s.-16s. 20 4.1 10 4.2 0.72 15 54 28 4 13 1416s.-17s. 7 3.7 16 4.4 0.76 16 64 28 7 10 5417s.-18s. 12 3.8 15 4.3 0.76 17 7 30 5 14 34iSs.-19s. 8 3.4 10 4.2 0.82 18 54 31 94 14 10419s.-20s. 6 3.2 10 4.5 0.78 19 6 36 10 11 1020s.-23s. 7 3.3 8 3.5 0.75 21 44 45 114 12 023s. & over 3 3.7 6 3.3 0.75 24 94 41 1 10 3

Total ... 155 4.5 151 4.8 0.77 15 04 26 104 12 54

la

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 137

with a food expenditure of over 23s. per 'man' per week. On the wholethere is little indication of any connexion between outlay on rent per familyand outlay on food per' man'.

Total household expenditure bears, of course, some relation to outlay onfood, since in general outlay on food constitutes a very large proportion oftotal household expenditure, especially if cost of housing is omitted from it.Household expenditure has been calculated by us also per 'man' per week.It is interesting to find that if we deduct outlay on food from total householdoutlay, the residueexpenditure on non-food items other than housing-varies on the whole much less, except in the three highest expenditure groups,than outlay on food, and shows by no means always the same trend. Thisapparently has little or nothing to do with the structure of the family; for if,for example, we consider the food expenditure groups from jzs. to 135.and from 14s. to 15S., we find that they both relate to practically the sameaverage household expenditure, as defined above. There were only slightdifferences in the number of persons per family and in the ' man ' values perperson. Differences in the expenditure on non-food items of one familyas compared with another of about the same structure and range of incomemay arise from various causes, but perhaps the most general one seems to bea family's preference either for better food or for semi-luxuries of the non-food kind, which latter, as regards household expenditure, are representedmainly by tobacco and visits to the cinema.

In the higher ranges of expenditure discrepancies between relative house-hold expenditure and expenditure on food become less marked. This is asone would have expected it to be; for families with a fairly high income areable to enliven their diet with certain higher priced foods without havingto seriously curtail, or cut out altogether, their consumption of non-foodsemi-luxuries.

Since it appears rather startling to find two families with an expenditureon food of less than 9S. per week and a household expenditure which, ascompared with that of the other families, is even disproportionately lower, aword of explanation seems indicated, in order to dispel the suggestion thatin these cases the low outlay reflects exceptionally low incomes. One of thesefamilies actually contains four earners, and expenditure out of housekeepingmoney on non-food items was low since they all paid for their personalrequirements1 out of their pocket money, the amounts of which were notgiven. The other family only expended 35 per cent of their income on food,and there was a margin between their total stated outlay, including outlayon rent, and their total income, of approximately 25 per cent. Both thesefamilies contributed accounts to our sample for several years, and it seemsthat they are catered for by exceptionally careful housewives.

Diagram A compares the distribution of the families with that of thepersons of our sample, in relation to a given expenditure per' man ' on food;i.e., it compares the standard of living with the average size of the family,

It seemed possible that they also purchased additional food. The housewife couldsupply no information on this point

138

7

'0

PcncC $CO

THE BULLETIN

DIAGRAM A

Cxper,dit.Jro w, rood pe

II(716

:oÓ 7/'9240

M,n' pf Wt&k

7h',5 V. jÇ0,.4 P'_.

Cumulative Frequency Curves of Families and Persons in Relation to theirExpenditure on Food

using expenditure on food per 'man' as a rough gauge. The first quarter'of the number of persons is composed of 28 families, having an average of 6.5persons per family; the second consjsts of 37 families, with 4.9 persons perfamily, the third of 42 families, average 4.6 persons, and the last quarter of

families, with an average of 4.1 persons.2 Thus the central 50 per cent ofthe persons lived in families that, on the average, closely corresponded to theaverage size of the families of the whole sample, viz., 4.8 persons per family.Their outlay on food per 'man' per week ranged from izs. 8d. to 175. 6d.,with an average of 14s. 74d., as compared with an average of 158. old. for allthe ,i families.(d) The Composition of the Families

The composition of the families according to age and sex groups is givenin Table III. The data for the unchanged and changed families are shown for1945 and 1946.

As regards the unchanged families we find, of course, a slight rise in the'man' value per personfrom 0.77 to 0.79as a result of the advancingto a higher age group of a numuer of persons. The average 'man' value ofthe changed families was the same for both years, the relative increase in the

The quarters are, of course, only approximations, since individual families could not besplit up. The first quarter' thus comprises 183, the third 184, and the second and fourthcomprise each 180 persons.

'Diagram A as shown has been cut so as to shape it more conveniently for reproductionin the BULLETIN. There were 8 families, consisting of 26 persons, with a food expenditure per

man' of over 21s. 5d. The highest outlay per man' per week was 25s. 10d.. spent by afamily of 3 persons with a total' man' value of 2.35.

120 I4

Age Unchanged Fame. Changed Fams. New All

number of adults being counterbalanced by a relative increase in the numberof young children. The lowest average' man' value obtained for the personsof the new families. These families contained nearly as many children ofschool age and under, as adults and juveniles, viz., 2.4 as compared with 2.5,in an average family of 4.9 persons.

() HOUSEHOLD EXPENDITURE AND THE COST OF LIVING INDEX

It is interesting to note that the new families had a higher average outlayon food than either the changed or the unchanged families, in spite of theirsmaller number of earners and, in particular, of adult male earners. Therespective figures will be found in Table IV, which gives the total outlay,

TABLE IVHousehold Expenditure of Three Groups of Families

per 'man' per week, on the five main groups of hou old expnditureemployed by the Ministry of Labour for the calculation of their cost of livingindex. Besides the actual outlay per ' man' per week in 1946, of the threegroups of families, there will be found, for the unchanged and changedfamilies, its relation to the corresponding outlay in 1945, in the form ofindices, taking the outlay in 1945 on each group as loo. These indices showthat the total household expenditure per' man' of the families that suppliedaccounts for both years, was about the same in ¡945 as in 1946. The actualfall per family in the cost of housing of the unchanged families was only about

Per' Man' per Week

Unch. Fam. ChangedFam. New Fam.

1946 1946 1946

Index of Expenditure1945-100

Unchanged ChangedFamilies Families

Food 14 111 14 24 15 44 1021 1011

Housing 3 84 2 104 2 111 914 984Clothing 2 0 2 2 2 11 87 112Fuel and Lighting 2 41 2 04 2 14 101 109Miscellaneous 8 9 5 64 6 81 974 914

Total 31 94 26 104 30 01 984 1004

Group (59) (24) Fams.(68)

Fame.(151)

1945 1946 1945 1946 1946 1946Males ... ... 18 & over 1.10 1.19 1.04 1.29 1.03 1.13Females... ... 18 & over 1.05 1.14 1.50 1.38 1.09 1.15Males ... ... 15-17 0.19 0.12 0.21 0.12 0.16 0.14Females... ... li-17 0.22 0.34 0.29 0.67 0.22 0.34Children ... 6-14 1.15 1.08 1.46 1.13 1.54 1.30Children ... under 6 0.73 0.57 0.62 0.87 0.87 0.75

Persons per Family 4.44 4.44 5.12 5.46 4.91 4.81'Man Value per Person 0.77 0.79 0.78 0.78 0.74 0.77

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE '39TABLE III

Average Structure of Families

140 THE BULLETIN

6 per cent, 2, per cent of the reduced index figures resulting from theincreased 'man' value per person of these families, from 0.77 to 0.79. Forthe changed families cost of housing per family had risen from 100 to 105,the decline by only i per cent shown in Table IV being due to thelarger number of persons per family. The figures for this particular item ofexpenditure as given in Table IV are therefore somewhat unrealistic. Butoutlay on housing is probably the only major item of household expenditurethat, at least in the short run, is of necessity unaffected by variations in thesize or age structure of a family. For all the other main items the trends ofexpenditure from one year to another of different groups of families can becompared only after the families have been standardized, which we havedone in terms of 'man' values.

A comparison of the relative outlay on the items enumerated in Table IVwith the weights of the Ministry of Labour index, adjusted for rises in pricesfrom 1914 to 1945 and 1946, is made in Table V. There appears to be some

TABLE VWeekly Family Expenditure

'Corrected for the rise in prices from July Ist, 1914, to June-July, 1945 and 1946.2 Rent only (intl. rates).'In 1945. the family consisted of 4.6 persons with an average' man' value per person of

0.77.

approximation between the relative figures obtained by us for outlay on foodand housing and those of the Ministry of Labour index, if the outlay on'miscellaneous' items is included. Yet 'miscellaneous' items took up inthe Ministry of Labour index less than 6 per cent, in both June-July 1945and 1946, while the proportion of household outlay obtained by us for thisgroup was z6 per cent in 1945 and 24 per cent in 1946. Moreover, our figuresin Table V only refer to total household expenditure, while the Ministry ofLabour index purports to cover total expenditure of working-class income.The margin found by us to exist between total household expenditure andtotal income, i.e., the proportion of income retained by the earners, issubstantial. We shall revert to this point later; it suffices here to say that

1945155 Families

mcl. exc!.Miscellaneous

0/ 0//0 f0

1946151 Families

mcl. excl.Miscellaneous

0/ 0/f0 /0

1945 1946Ministry of Labour Cost of

Living Index Weighting'mcl. exc). nc). exc).Miscellaneous Miscellaneous

0/ 0/ 0/ 0//O /0 /0 /0

(1) Food ... 47 63.4 50 65.7 50.5 53.5 49.9 53.0(2) Housing ... 12 16.2 11 14.1 12.9' 13.7' 13.2' 13.9'(3) Clothing ... 8 10.5 8 10.6 20.3 21.5 20.4 21.6(4) Fuel and Light 7 9.9 7 9.6 10.7 11.3 10.8 11.5

(1)-(4) 74 100.0 76 100.0 94.4 100.0 94.3 100.0(5) Miscellaneous 26 24 5.6 5.7

Total Outlay ...... 100 100 100.0 100.0Weekly Outlay ...... s. d. s. d.

per Family ...... 106 61' 111 5per'Man' ...... 30 21 30 11

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 141

there can be no doubt that average expenditure on food and housing are muchoverweighted in the official index.

At the bottom of Table V will be found the actual weekly householdoutlay, per ' man' and per family, of all the families of our samples in 1945and 1946. The outlay per family was larger by about 4s. iod. in 1946, butowing to the somewhat larger size of the average family in 1946, the outlayper' man' was virtually the sàme in both years.

The analysis of the expenditure on food of the families of our samplewill be deferred to Part II of this paper. In the following sections of Part Iwe shall consider mainly the data of non-food expenditure.

() EXPENDITURE ON NON-FOOD ITEMS

Cost of Housing and RentIn comparing the data of household expenditure with the 'weights' of

the Ministry of Labour index we employed the more comprehensive term'housing' instead of 'rent', since the outlay of the families of our samplerepresents in some instances not rent and rates but mortgage charges, housepurchase instalments and rates, or, when the house is owned, rates only.Three of the zi families in 1946 paid nothing at all for their housing accom-modation, which was obtained in connexion with their work, and twofamilies paid rates, jut the amounts were unknown to the housewives. Thetotal outlay on housing accommodation obtained from our sample, dividedby the total number of families or the total number of' men ' represented bythem, produces our data of average expenditure on 'housing'. Thisamounted per family per week to sis. ud., as against the outlay on rent andrates per family paying these charges, of izs. 5d. If we omit the five familiesthat stated no outlay on housing accommodation, we obtain an average costof housing, per family per week, of izs. 5d., virtually the same as the averageoutlay of the 132 families that rented their houses.

There was a decline in the outlay on housing per family of the unchangedfamilies, from 1945 to 1946, of about 6 per cent, which was mainly due toreductions in mortgage interest and purchase instalments. The reason for theincrease of 5 per cent in the housing charges of the changed families, is lessclear. In a few instances rates seemed to have risen, and some of the familiesapparently paid reduced rents or purchase instalments during the absence onmilitary service of an earner. The outlay on housing per family per weekwas, in 1946, for the unchanged families, 13S. ad., for the changed families,12s. 3d., and for the new families, ios. 9d. Of the five families that stated nooutlay on housing, two belonged to the group of unchanged, two to the groupof changed, and one to the group of new families. With these families omittedfrom our calculations, we obtained the following figures of outlay: un-

changed families, 13s. 7d., changed families, uzs. 9d., and new families, ios.i id.per family per week.

ClothingOutlay on clothing out of housekeeping money amounted, per man per

10*

1 Including firewood, paraffin, candles, matches, firelighters.

Cent since 1945, while that of the changed families had increased by I54percent. During the same period' the Ministry of Labour index of cost of fuelhad advanced by about 0.4 per cent. In the case of the changed families theincrease in the number of infants on the one hand and of adult male earnerson the other may account to some extent for the larger expenditure, sincemore fuel may have been required for washing and possibly also for cooking.The largest increase occurred in the outlay on gas of the changed families,and, next to it, in their outlay on electricity, the former amounting to nearly34 per cent and the latter to nearly za per cent of their outlay in 1945. The

June-July, 1945 to June-July. 1946.

Group ofFamilies

Total Fuel & LightingOutlay Index

1945=100

CoalOutlay

GasOutlay

ElectricityOutlay

Other'Outlay

Unchanged (59)1945 ... 7 111 100.0 3 64 2 24 1 84 541946 ... 8 34 104.6 3 9 2 51 1 94 31

Changed (24)1945 ... 7 7 100.0 3 5 2 14 1 81 441946 ... 8 91 115.5 3 64 2 94 2 04 44

New (68) 1946 7 9 3 2 2 10 1 34 5All (151) 1946 8 14 3 54 2 81 1 74 4

All (155) 1945 ... 7 101 3 74 2 2 1 74 44

142 THE BULLETIN

week, for the unchanged families to 2S., for the changed families to zs. ad.,and for the new families to zs. ud. These figures however only representpart of the actual expenditure on clothing. In most instances they excludeall substantial purchases of clothes for adults, although some of these may becovered by payment into clothing clubs out of housekeeping money. Thesepayments are included under the heading of 'clothing'; they are mainlysavings or instalments for purchases of children's clothes. The housewivesappear to pay mostly only for small items of apparel, children's wear, andrepairs, in particular of shoes, for themselves and the children. Clothes andshoes for adults are as a rule bought by the earners, even the housewife'sclothes being often paid for out of the husband's savings.

It is doubtful whether a sample covering a longer period than a fortnightcould materially improve the data relating to outlay on clothing, unless itwere extended ovea at least one whole year. Failing this, the most satis-factory course is, perhaps, to consider outlay on clothing in conjunction withoutlay on ' miscellaneous' items and, if possible, total income. This at leastwill show whether and to what extent the outlay on clothing disbursed by thehousewife has a bearing on the general standard of living of the family.(c) Fuel and Lighting

The outlay on fuel and lighting is given in some detail in Table VI. Totaloutlay on fuel and lighting of the unchanged families had risen by over 44 per

TASLE VIWeehly Family Expenditure on Fuel asd Lighting

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 143

Ministry of Labour index contains no reference to electricity, but for theprice of gas it showed a rise by z.6 per cent,.from JuneJuly 1945 to JuneJuly1946. The increase in the volume of gas consumed by the changed familieswas thus approximately 30 per cent, while for the unchanged families it wasover 94 per cent, of the families' Consumption in JuneJuly 194.5. The slightrise in the purchases of coal, of about 5 per cent for the unchanged and about4 per cent for the changed families1, may merely indicate a somewhat largeraccumulation of stocks in 1946 than in 1945. Since, however, this wascertainly not experienced in general, the higher figures for both groups offamilies in j 946 suggest also a somewhat higher consumption of coal by thesefamilies, during our sampling period in 1946, as compared with that in 1945.(d) Miscellaneous Items

The group of ' miscellaneous' items covers a wide range. It includesexpenditure on goods and services, the main of which are: all kinds ofhousehold requirements, seeds and plants, garden implements, animal foods,cleaning materials, laundry, voluntary insurance, and clubs, medical attention,education, transport, amusements, newspapers, drink, and tobacco. Onlypart of the family expenditure on these items can be ascertained by us, sincemany of them are paid for direct by the earners. Indeed, as will be seen fromTable VIII, not even outlay on such a general items as newspapers wasstated by all the housewives.

The outlay on ' miscelláneous' items is given in Table IV, per 'man'per week, for the unchanged, changed, and new families. In Tables Villa andVilib, the 151 families of our sample have been split up into those statingtheir total income (75)2 and those stating household outlay only (76). TheseTables also show the outlay per' man' per week. We are not yet concernedwith the data in Tables Villa and Vilib in general, but is more helpful todiscuss expenditure on 'miscellaneotis' items with reference to them thanto Table IV, since from the analyses öf our samples in previous years we havelearned that, on the whole, families that reveal their income will tend toshow a higher outlay on 'miscellaneous' items out of housekeeping moneythan those that only account for expenditure of housekeeping money. Thereason for this seems to be that when the housewife knows how much every-body earns she is more frequently entrusted with purchases of the earnersprivate requirements.

The figures per ' man' of outlay on « miscellaneous' items in TablesVilla and Vilib show, for the changed families that stated their income in1946, an outlay but ad. higher than that of the changed families that onlystated their household expenditure. However, in families with about the samestandard of living total outlay on ' miscellaneous' items will probably beaffected less by the number of persons in the family than by the number of

1 The Ministry of Labour index showed ap increase in the cost of coal of about 0.3 percent, while the increase in outlay of the uncha,zged families amounted to approximately 5.3per cent, and that of the changed families to 4.3 per cent.

a' New ' families in Tables Villa and IXa include also all those unchanged and changedfamilies that divulged their total income in 1940 but not in 1945.

144 TIlE BULLETIN

earners, since the demand for many items in this group comes mainly fromadult earners.' Tables IXa and IXb show the structures of the families,whose expenditure is set out in Tables Villa and Vilib. From them it willbe seen that the 'o changed families that only stated their household expendi-ture had, per family, 0.4 earners over i years more than the 12 changed

families that divulged their income. If we consider the outlay on ' mis-cellaneous' items per family, we obtain the following figures, the first one ineach group for the families that stated their income and the second for thosethat stated household outlay only: unchanged families, 1945, 29s. 7d.-29s. id. ; unchanged families, 1946, 32s. Iod.-24s. 9d.; changed families,1945, z8s. 5d.I9s. 9d.; changed families, 194.6, 25s. 2d.Zos. 6d.; 'new '2families, 1946, 25s. IOd.-25s. 7d. It should be noted that the 'new'families that stated household outlay only had per family about 0.3 earnersover 15 years more than the' new' families that stated their income. Amongstthe unchanged families that only declared their household expenditure in1945, there were a few paying substantial instalments on furniture, whileno such payments were made by any of the unchanged families that statedtheir income in 1945 and 1946. This may partly account for the relativelyhigh outlay per family of the former group, in

In Table VII we have analysed the outlay on some of the more generalmain items of' miscellaneous' expenditure, viz., drink, tobacco, newspapers,

TABLE VIIExpenditure on Miscellaneous flems and of Housekeeping Money

Familiesbuying as% of 75Families

1946 194675 Families Stating Income 76 Families Stating Household

Outlay OnlyWkly. Outlay p. Fam. Families Wkly. Outlay p. Fam.

Actually Average of buying as Acthally Average ofbuying Group of % of 76 buying Group of

75 Fams. Families 76 Fams.

34 Unchanged Families Stating Income1945

Families Wkly. Outlay p. Fam. Familiesbuying as Actually Average of buying as% of 34 buying Group of % of 34Families 34 Fams. Families

1946Wkty. Outlay p. Fam.Actually Average ofbuying Group of

34 Earns.

Under this we may even include treats' for children, since they are non-necessities.Z We shall denote new' families by putting the adjective in inverted commas whenever

we are dealing with the group of families that stated their total income for the first timein 1946, and which includes 7 unchanged and 2 changed families.

1)rink* 43 3111 1 84 33 2 64 010Tobacco 76 8 54 6 51 57 7 8 4 4Newspapers 97 1 74 1 74 95 1 6 1 54Amusements 79 5 8 4 54 70 4 64 3 2Insurances and Clubst 96 4 31 4 14 100 4 71 4 74

Total Av. Outlay 18 4 14 44

Drink* 50 3 04 1 64 47 4114 2 4Tobacco 71 9 3 6 64 76 8 94 6 81Newspapers 97 1 9 1 84 100 1 84 1 84Amusements...... 76 4 9 3 74 79 611 5 6Insurances & Clubst 100 4 54 4 54 94 43 40Total Av. Outlay 1710 20 34

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 145

18 Unchanged Families Stating Household Outlay Only1945 1946

Families Wkly. Outlay p. Fam. Families Wkly. Outlay p. Fam.buying as Actually Average of buying as Actually Average of% of 18 buying Group of % of 18 buying Group ofFamilies 18 Fams. Families 18 Fams.

* Including soft drinks, t Excluding compulsory insurances and clubs' for specificpurposes, viz., clothing, coal, furniture clubs, etc.

amusements, voluntary insurances and clubs (excluding specified clubs forpurchasing clothes, household requirements, and coal), per family per week,of the families that stated their income as well as of those that only statedhousehold expenditure. The first group of averages under each separateheading was obtained by dividing the total weekly outlay on each item by thenumber of families that made the purchases, while the second group ofaverages represents the total weekly outlay divided by the total number offamilies in the group. The average outlay on the five items of all 75 familiesthat stated their income was 4S. higher than that of the 76 families that onlystated their household expenditure. Total outlay on 'miscellaneous' itemsout of housekeeping money amounted for the latter group of families to24s. 9d. and for the former to z8s. iod. per family per week, i.e., it wasfor each group about ros. 6d. more than their average outlay on the itemsspecified in Table VII.

Table VII also shows the comparable figures for 1945 and 1946, for thetwo groups of unchanged families. As regards the unchanged families thatstated their income, it will be seen that about the same numbers of familiesaccounted in both years for expenditure on the several items, and in generalthe pattern of expenditure indicates that the figures convey a fairly reliablepicture of expenditure on these items out of housekeeping money. Thepurchases made by the housewives of those unchanged families that onlystated household expenditure, appear more erratic for the two years. Mostlikely in these cases, where the housekeeping money tends to be a more rigidsum than in the former, the housewife's expenditure on drinks, tobacco, andamusements, for herself and her family, depends largely on the margin that isleft after more essential purchases have been met, and varies accordingly.A sub-sample of only i8 families is, of course, too small to allow us to expectthat variations in the outlay of individual families will cancel out.

() HOUSEHOLD EXPENDITURE AND INCOME(a) Comparison between Household Expenditure and Family Income

Of tile 151 families that supplied household accounts in 1946, 75 statedthe family income.I Only through this is it possible for us to attempt an

The larger the number of earners in a family the more difficult it is on the whole toobtained a statement of family income. The 75 families that divulged their total income in1946 had, per family. 1.4 earners over 15 years old, while the 76 families that only statedhousehold expenditure contained on the average 1.6 such earners.

Drinks 50 3 3f 1 7f 28 4 2f 1 2Tobacco 44 10 11 4 5f 39 8 9f 3 5Newspapers 100 1 71 1 7f 100 110 110Amusements...... 56 6 9f 3 9f 72 5 2f 3 9fInsurances & Clubsf 94 5 41 5 0f loo 47 47Total Av. Outlay .,.. 16 6f 14 91

146 THE BULLETIN

interpretation of the figures of household expenditure on clothing and mis-cellneous items, and to obtain a picture of the interrelation of expenditureon food and expenditure on non-food items.

In Table Villa we have set out the household expenditure of the 34unchanged, 12 changed, and 29 'new' familiesthe latter including 7 un-

changed and z changed families that did not state their income in 1945thatstated their total family income in 1946. For these families, average expendi-ture on food was found to have taken up between 33 and 40 per cent of thetotal net income,1 expenditure on housing, in 1946, between 6 and 9 per cent,and expenditure on fuel and lighting, between 5 and 6 per cent, while theaverage proportion of income left for clothing and miscilaneous items, includ-ing the unspecified margin of income retained by the earners, representednearly 50 per cent of the total income, the actual amounts available in ¡946per 'man' per week ranging, for the three groups, from about i6s. to 23s.,and, per family, from about 56s. for the 'new' families to over 88s. for thechanged families.

The figures of household outlay of the remaining 76 families, which didnot state their income, are given in Table Vilib. The proportion spent outof housekeeping money on food was substantially larger only for the unchangedfamilies in Table VIlIb, as compared with the unchanged families in TableVilla. For all the groups of families in Table Vilib, except the group ofunchanged families in 1945, the average outlay on 'miscellaneous' itemsrepresented a smaller percentage of the housekeeping money than for thecorresponding groups of families in Table Villa.

The data shown in Tables Villa and Vilib should be considered inconjunction with the data of average family structure that will be found inTables IXa and IXb. Since the average number of earners over 15 years oldin the sub-groups of the families that only stated household expenditure, wasin no instance substantially smaller and in some decidedly larger than in thecorresponding sub-groups of the families that revealed their incôme, itseems reasonable to assume that, at least approximately, the relative marginsshown by the latter families between income and household expenditure alsoexisted with respect to the former families. Indeed, the larger the proportionof earners in a family the larger will tend to be the proportion of incomethat is not surrendered into the common pool of housekeeping money, butretained by the earners. It seems therefore that roughly 50 per cent of theaverage income of the families of our sample was expended on food, rent, andfuel and lighting, while the other half of the income was available for familyexpenditure on 'lothing and 'miscellaneous' items, the earners' personalexpenditure, and savings. The small amounts of savings given in TableVilla mainly refer to savings by means of street collections and children'sschool savings.

The higher figures in Table Villa of income per family of the unchanged

I All the Íiures of income used by us exclude direct taxes, contributions to compulsoryinsurances, and trade union contributions.

* Including Unchanged and Changed families that did not state their income in 1945.'Contributions to street and school savings clubs out of housekeeping money.

Net Income: excluding taxes, compulsory insurances, and contributions to TradeUnions.

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 147

TABLE VillaExpenditure of Income of 75 Families

34 Unchanged Families1945 1946

Per' Man' per Week

12 Changed Families 29 New Fams.11945 1946 1946

Per' Man' per Week per 'Man'per WeekFood ......... 14 114 15 44 14 2 13 54 14 84

Housing 4 6 4 11 2 34 2 2 2 114Clothing 2 84 2 34 2 34 2 44 1 104Fuel and Lighting 2 44 2 6 1 10 2 24 2 24Miscellaneous 9 14 9 94 6 114 5 64 7 44

Total Household 33 74 34 1 27 64 25 10 29 04Savings' 0 44 0 11 0 94 0 104 0 5Unspecified ...... 10 9 11 41 8 5 11 7 6 74

Total Income' ... 44 94 46 44 36 84 38 34 36 14

Relative Expenditure of IncomeFood ......... 44.5 33.4 45.2 33.2 51.5 38.6 52.1 35.1 50.6 40.7Housing ...... 13.4 10.1 12.0 8.8 8.3 6.2 8.6 5.8 10.2 8.2Clothing ...... 8.0 6.0 6.7 4.9 8.4 6.3 9.2 6.2 6.3 5.1Fuel and Lighting 7.0 5.3 7.3 5.4 6.6 5.0 8.6 5.8 7.6 6.1Miscellaneous 27.1 20.3 28.8 21.2 25.2 18.9 21.5 14.5 25.3 20.3

Total Household ... 100.0 75.1 100.0 73.5 1000 75.0 100.0 67.4 100.0 80.4Savings' 0.9 2.0 2.1 2.3 1.2Unspecified 24.0 24.5 22.9 30.3 18.4

Total Income' 100.0 100.0 100.0 100.0 100.0

Per Family per WeekTotal Income' 145 54 155 5 150 3 173 34 126 114

Housing 14 74 13 84 9 44 10 11 10 44Fuel and Lighting ... 7 84 8 44 7 54 10 0 7 94

TABLE VIlIb Expenditure of 76 Families Stating Household Outlay Only

18 Unchanged Families 10 Changed Families1945 1946 1945 1946

Per' Man' per Week

48 New Fams.1946

Food 14 44 15 0 13 21 15 34 15 44Housing ... 3 84 3 54 3 84 3 104 2 104Clothing ... 2 04 1 94 1 1 2 04 3 2Fuel and Lighting 2 01 2 14 1 9 1 114 2 04Miscellaneous 8 04 6 84 5 14 5 44 6 84

Total Household 30 24 29 14 24 104 28 7 30 24Relative Expenditure of Income

Food 47.6 51.5 53.0 53.6 50.9Housing 12.3 11.8 15.0 13.6 9.6Clothing 6.7 6.2 4.3 7.2 10.5Fuel and Lighting 6.7 7.4 7.1 6.8 6.8Miscellaneous 26.7 23.1 20.6 18.8 22.2

Total Household ... 100.0 100.0 100.0 100.0 100.0Per Family per Week

Housing 13 44 12 84 14 5 14 94 13 1Fuel and Lighting 7 34 7 114 6 91 7 54 17 04

148 THE BULLETIN

and changed families in 1946, as compared with those for 1945, cannot betaken as an indication of a general increase in individual earnings. The com-position of the changed families had altered too much to allow of a directcomparison of the figures of income for the two periods, in particular inview of the smallness of this sub-group, while the moderate rise of ios. perfamily per week, shown for the unchanged families, may largely reflect thefact that the young earners of these families, as they improved their workingexperience in the course of a year, also improved their earnings. It is truethat the unchanged families had a smaller number of female earners in 1946than in 1945, but the female earners in the age group of i8 years and over in1945 were nearly all housewives, who mostly did part-time worka womanwho worked half a day being counted by us as half an earnerand whosefull-time earnings amounted in general to little more than 5os. per week. Themost spectacular rise in earnings occurs as a rule in the wages of young menat the end of their training.

(b) Relative Outlay of Income and Size of FamilyThe smallness of our sample makes it imperative to verify as carefully

as possible the general validity of our computed data, for families of theselected type. The fairly close approximation of the figures obtained for thedifferent sub-groups is in itself reassuring in this respect. Another test hasbeen applied regarding the main data of relative expenditure, in Tables Xa,Xb, Xc, Xd, and Xe. Since these data yield a much fuller information if theyare related to total income, only the accounts of the 75 families that statedtheir income, have been considered in this context. The families have beenarranged in three groups, the first comprising 45 families consisting of lessthan five persons, the second comprising 23 families with from five to sevenpersons, and the third represented by seven families of over seven persons.The patterns obtained of the relative expenditure of these 75 families onfood, housing, fuel and lighting, and other items, suggest in every case asymmetrical distribution with a distinct central clustering, in most instanceseven for the sub-groups of families, at least with respect to the typical house-hold expenditure on food, housing, and fuel and lighting. For the remainingpart of the income we can scarcely expect a ' normal' pattern for very largefamilies, since, as we have pointed out before, the relative structure of thefamilyi.e., the number of earners in relation to the number of dependentsand this residue are not independent factors.

Tables Xa, Xb and Xc are self-explanatory. Tables Xd and Xe analyseoutlay on 'miscellaneous' items, clothing, and the unspecified marginbetween household expenditure and income. Household expenditure on'miscellaneous' items only is dealt with in Table Xd. The picture obtained,although interesting, is not very revealing. In Table Xe we have widened thefield of expenditureor potential expenditurefirst, by including theunspecified margin together with household outlay on 'miscellaneous'items, and secondly, by adding to this the outlay out of housekeeping money on

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 149TABLE IXa

Structure of Families Staling Family Income

* Including Unchanged and Changed families that did not state their income in 1945.

TABLE IXbStructure of Families Staling Household Outlay Only

TABLE XaOutlay on Food Compared With Income

Families Food Expenditure Representing of IncomeSize Total under % % % %

No. 20% 20-30 30-40 40-50 50-60Under 5 Persons 45 2 12 19 12 -5-7. Persons 23 - 4 5 8 6Over 7 Persons 7 - - 2 2

Average 4.7 Persons 75 2 16 26 23 8

TABLE XbOutlay on Housing Compared with Income

Families Expenditure on Rent Representing of IncomeSize Total under % % % %

No. 4% 4-8 8-12 12-16 16-20Under 5 Persons 45 5 13 17 8 25-7Persons 23 2 10 8 1 2Over 7 Persons 7 1 2 3 - 1

Average 4.7 Persons 75 8 25 28 9 5

34 Unchanged Families1945 1946

Fer Family

12 Changed Families1945 1946

Per Family

29NewFams.11946

Per FamilyEarners, 18 and over:

Male 1.12 1.18 1.08 1.00 0.97Female 0.18 0.12 0.46 0.63 0.05

Earners, 16-18:Male ...... 0.06 0.06 0.25 0.08 0.07Female ... 0.09 0.06 0.17 0.33 0.10

Earners, under 16 0.15 0.12 0.08 0.25 0.07Non-earning House-

wives ...... 0.82 0.96 0.88 0.88 0.95Dependent Children 1.73 1.65 2.33 2.75 2.55

Total 4.15 4.15 5.25 5.92 4.76

Total 4.78 4.78 4.80 4.70 .5.08

18 Unchanged Families 10 Changed Families 48 New Fams.1945 1946 1945 1946 1946

Earners, 18 and over:Per Family Per Family Per Family

Male 1.11 1.11 1.10 1.50 1.06Female 0.19 0.31 0.60 0.30 0.15

Earners, 16-18:Male ......- - 0.20 - 0.08Female ... 0.11 0.06 0.10 0.30 0.21

Earners, under 16 0.11 0.06 0.10 0.30 0.15Non-earning. House-

wives ......0.98 0.96 LOO 1.00 0.98Dependent Children 2.28 2.28 1.70 1.30 2.45

15Ó THE BULLETIN

TABLE Xc

Outlay on Fuel and Lighting Compared with Income

Families Expenditure on Fuel and Lighting Representing of Income

TABLE Xd

Outlay on Miscellaneous Items Compared with Income

TABLE Xe

Outlay on Miscellaneous Items, Clothing, and Money Retained by Earners, compared withIncome

Expenditure on Miscellaneous Items and Money Retained by EarnersFamilies Representing of Income

Expenditure on Clothing, Miscellaneous Items, and MoneyFamilies Retained by Earners Representing of Income

Size TotalNo.

under20%

%20-30

%30-40

%40-50

%50-60

%60-70

%70 andover

Under 5 Persons 45 1 4 8 16 8 7 15-7 Persons ... 23 1 6 6 5 4 1 -Over 7 Persons 7 - 2 4 - 1 - -Average 4.7 Pers. 75 2 12 18 21 13 8 1

Size TotalNo.

under2%

%2-4 %4-6 %6-8 %8-10

%10 and

overUnder 5 Persons ... 45 3 9 13 11 5 45-7 Persons ... 23 1 4 8 5 3 2Over 7 Persons ... 7 - 1 3 - 2 1

Average 4.7 Persons 75 4 14 24 16 10 7

Size Total under %No. 20% 20-30

%30-40

%40-50

%50-60

% %60-70 70 and

overUnder 5 Persons 45 - 2 5 14 14 7 35-7 Persons ... 23 - 3 7 6 5 2 -Over 7 Persons 7 - 2 2 2 - 1 -Average 4.7 Pers. 75 - 7 14 22 19 10 3

FamiliesSize

Expenditure on Miscelicanous RepresentingTotal under % %

No. 10% 10-20 20-30

of Income%

30-405Under 5 Persons 45 3 22 15

5-7 Persons ... 23 3 10 7 3OverlPersons 7 - 5 1 1

Average 4.7 Persons 75 6 37 23 9

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 151

clothing. The unspecified margin includes all savings; thus the data in thesecond part of Table Xe represent that share of the income not taken up bythe items dealt with in Tables Xa, Xb, and Xc. It will be seen that nearly43 per cent of the 75 families spent on food, rent, and fuel and light, lessthan half of their income, and over 17 per cent-13 familiesless than 40per cent of it. Of these 13 families, six consisted of three persons, including achild of school age or under. In one of these families the wife also was anearner.' Four of the families comprised each four persons; one, however, hadno dependent children, one had one such child, and two had two dependentchildren. In one of the families with two dependent children and in thatwith one dependent child the housewives orked part-time. The threelarger families consisted, one of five persons with one dependent child, anotherof six persons with two dependent children, and the largest one of nine personswith three dependent children.

PART H: OUTLAY ON FOOD AND NUTRITION(i) THE TREND OF EXPENDITURE ON FOOD FROM 1945 TO 1946

In order to make possible a comparison of the outlay on and the con-sumption of food of families of different sizes and different age and sexstructures, we have resorted to the expedient of standardizing the families bymeans of expressing the persons in terms of' man' values.2 If, however, weare comparing for two different periods figures of expenditure relating to thesame set of families, and if the composition of these families has remainedunchanged, standardization is not required, and it is more natural to comparechanges in expenditure per person than in expenditure per ' man'.

There were in our sample for June-July, 1946, 59 families that had alsosupplied household accounts in June-July, 1945, and that had remainedunchanged in structure. Another 24 families that had kept accounts for bothperiods had altered in structure, some being larger in 1946 than in 2945 andsome smaller. These families are styled changed families. Besides, 68 newfamilies contributed to our sample in 1946.

The trend of expenditure on food from one period to another can, ofcourse, only be judged by means of a small sample if we are able to comparethe outlay of identical families for both periods. In order to ascertain thistrend from the summer of 1945 to the summer of 1946, we have considered,in Table XI, the outlay of the unchanged families. The families are rangedaccording to their food expenditure per person per week in June-July,

'This family and two others consisting of husband, wife, and one child, with the husbandas the sole earner, represented the group of families spending less than 30 per cent of theirincome on food, housing, and fuel and lighting.

I The two scales of' man ' values employed by u are set out in Table XIX. The In1iiu1eof Slatistics Scale has been used always unless explicit reference is made to the Ministry ofHeal/h Scale.

152 THE BULLETIN

TABLE XI Changes in Food Expenditure since 1945

Food Expenditure 1946 Index of Food Expenditure per Person: 1945 = 100 TotalPer Person per Week under 18 85-94 95-104 105-114 115-124 125 and Fams.

overrunch. Fam. 1 2 1 2 - - 6Under 9s. ic¼d. Fam. 3 1 - - - - 4

(Unch. Farn. - 2 1 1 2 - 69s.lOs. Chgd. Fam. I - 1 1 1 - 4

I Unch. Fam. - - 1 3 2 - 6lOs.lls. ¶Chgd. Fam. 1 1 1 1 2 - 6

runch. Fam 2 4 2 3 - 1 12lls.-12s. ¶.Chgd. Fam. - - - 2 - - 2

lUnch. Fam. 2 3 4 2 1 - 1212s.-13s. 1Chgd. Fam. - - - - 1 - 1

f Unch. Fam. - - 2 2 1 2 713s.-15s. 1_chgd. Fam. - - - 1 - - 1

lUnch. Fam. - 1 - 2 2 2 715s.-18s. Chgd. Fam. - - - 2 1 3 6

18s. and lUnch. Fain. - - 2 - 1 3over lChgd. Fam. - - - - - - -Total f Unch. Fam. 5 12 11 17 8 6 59

)Chgd. Fam. 5 2 2 7 5 3 24

1946, and the indices show whether this expenditure was the same as inJune-July, or whether it was higher or lower. In each instance theoutlay on food in 1945 has been equated to ioo. It will be seen that ofthe 59 unchanged families, ii spent about as much per person in 1945 as in1946, while 17 spent less and 31 more.

Table XI also gives, in italics, the corresponding indices for the changedfamilies. Of these families two only spent per person per week approxi-mately the same amounts during both sampling periods; seven of thechanged families spent less on food per person in i6 than in and fifteenspent more. It must, however, be realized that for the individual family ofthis group the ' man' value per person may vary considerably from one yearto another. This may be illustrated by two rather extreme examples:the average ' man' value per person of a couple is 0.93. If the wife hastwins, the four persons of the family will have an average 'man' value of0.63. Or, if a family with z children under 3 years and one between 3 and 5years old, is joined by two sons over 14 years oldwho during a previoussampling period may have beer. staying with relationsthe average 'man'value of this family is raised from o.6o to 0.72. Thus, of the 7 families thatspent on food over per cent more in 1945 than in 1946, four had a childborn to them since our survey in 1945, and the decline in their outlay on foodper person does not represent a decline in the standard of living of three ofthese families. The fourth of these had increased from four persons in i toeight in 1946, three children of school age having returned home in 1946.Since there was no change in the housekeeping money, actually less was

260

240

220

200

C --

140

r -

120-C

o.I'

loO -

80-

Average l946l39id

s80

'

../

The outlay on food per person per week of the 59 unchanged and the 24changed families in ¡945 and ¡946 is pictured in detail in Diagram B. TheDiagram has been drawn on a double logarithmic scale ; the unchangedfamilies are represented by dots and the changed families by crosses. Theaverage rise in the outlay on food per person amounted for the 83 families toapproximately 4.3 per cent, and the diagonal divides the Diagram in such away that the dots and crosses lying on it indicate a rise in food expenditure ofthe respective families by this percentage. The average rise in expenditureis obviously well defined by the data of expenditure of the individual families,which, on the whole, cluster fairly closely around the diagonal.

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE ¡53

spent by the family on food in 1946 than in 1945, when the outlay on food forhusband, wife, and two children 12 and 14 years old, amounted to i8s. 9d.per person per week.

DIAcLwi BChanges in the Expenditure on Food per Person per Week of 83 Families

280- -

180 200 220 240lOO 120 140 160

Exper,diture an Pence 1945

(Outlay in pence: ¡945 horizontal, 1946 vertical. Logarithmic scales)

Bread and Cereals1 ......Fats' ............Meat and Meat Products'Fish' ............Dairy Produce'Groceries' .........PotatoesOther Vegetables and Fruit'Other Foods'

WeeklyFood Expenditure:

per Personper Man'

Total Expenditure:per Personper'Man'

Unch. Families Changed Fams. New All(59) (24) (68) (151)

Fams. Fam,.1945 1946 1945 1946 1946 194615.8 15.9 17.4 16.6 15.6 15.9

0.7 0.6 0.6 0.7 0.7 0.619.0 17.8 19.7 17.8 19.1 18.4

5.3 4.9 5.1 5.5 4.3 4.820.2 18.3 20.8 19.1 18.1 18.4

8.0 8.1 7.4 8.9 8.3 8.35.4 6.0 6.1 6.4 5.6 5.9

12.1 13.6 11.9 11.0 14.5 13.513.5 14.8 11.0 14.0 13.8 14.2

Expenditure on Food and Total Expenditure

11 24 11 104 1011 11 04 11 54 11 6414 74 14 11 14 0 14 24 15 44 15 04

24 94 25 3 20 104 20 104 22 44 23 1432 34 31 94 26 94 26 104 30 04 30 14

1 Including flour, cake, biscuits, breakfast cereals, oatmeal and rolled oats, rice, semolina,etc.

'Cooking fat, lard, suet, dripping.'Butcher's meat, bacon, cooked, tinned, and potted meat, offal,, sausages, meat pies,

fowls, and rabbits.4 Wet, cured, bottled, fried, tinned.'Butter, margarine, milk, cheese, egga (in shell and dried).Sugar, tea, cocoa, jam, syrup, honey.Including dried legumes, tinned vegetables, and dried fruit.

'Including all meals bought away from home, condiments, fancy foods, and sweets.

The relative rise in the outlay on food, from 1945 to 1946, of the changedfamilies, was approximately the same per ' man' as per person-a little overi per cent-since the 'man' value per person of these families was the sainein both years. The 'man' value per person of the unchanged families, how-ever, had risen by about z per cent, and the increase in the food expenditureper 'man' was proportionately lower than the increase in the food expendi-ture per person, the former amounting to 2.3 per cent and the latter to 5.9per cent.1

The figures of relative outlay in Table XII show that the outlay on foodwas distributed broadly in the same way during both sampling periods. Aslight decline could be observed in particular in the relative outlay on meat

1Tbe structure of the three groups of families is set out in detail in Part I, Table HI.

154 THE BULLETIN

(z) THE PArTEEN OF FOOD EXPENDITURE

The changes in the outlay on food per person per week are compared inTable XII with the changes in the outlay on food per 'man' per week.Table XII also shows the weekly outlay, per person and per 'man', of thetotal housekeeping money and, further, the relative outlay on the severalfood groups.

TABLE XIIRelative Expenditure on Food Groups

Total 100.0 100.0 100.0 100.0 100.0 100.0

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 155

products and also on dairy produce, the former being mainly due to reducedpurchases of cooked meat and the latter to reduced purchases of dried eggs.Slight increases occurred in the outlay on vegetables and fruit, on mealsbought away from home, and on fancy foods. The average outlay on potatoeswas a little higher in 1946 than in 1945, owing to a somewhat larger consump-flou of new potatoes during our sampling period in 1946, as compared withthat in 1945. New potatoes were generally on sale earlier in 1946 thanin1945, since in 1946, stocks of old potatoes were exhausted earlier.

() CONSUMPTION IN 1945 AND 1946(a) Specifically Rationed Food2 and Milk

The consumption of the main rationed foods is surveyed in Table XIII.Bread, flour, and flour confectionery, were still unrationed during our samp-ling period in 1946, and hence are not included. Purchases of unrationed fats,however, are shown, since they represent a much prized supplement to theallowance of rationed fats.

The weekly amounts of the specifically rationed foods available per ordin-ary ration book were, in June and July, 1946: meat, is. zd. worth; bacon,3 ounces; butter or margarine, 3 ounces'; margarine, 3 ounces; cooking fat,i ounce; tea, z4 ounces1; sugar, 8 ounces; jam or sugar, 4 ounces; cheese,z ounces. In June and July 1945, the allowances were the same for most ofthese foods, but the maximum butter ration amounted to only z ouncesand the supplementary ration of margarine was, therefore, always 4 ounces.The tea ration was only z ounces per ordinary ration book per week. Duringboth periods the meat ration of children in receipt of Government milk-supplied to children up to the age of five yearswas half the adult ration,and these children had no allowance of tea. The weekly amount of fresh milkthat could be obtained by an adult varied in general between 3 and 3 pints,although some people were able to purchase considerably more. A childup to the age of 5 years was entitled to 7 pints per week at zd. a pint,3 andchildren and juveniles under 17 years old were allowed 3 pints per week, atthe ordinary price of between 44d. and 5d. a pint.

During our sampling periods in both 1945 and 1946, a family could buyadditional sugar issued for making fruit preserves; this partly accounts forthe relatively large purchases of sugar shown in Table XIII. Moreover, ourfigures for jam also include the purchases of tinned jam and marmalade,'rationed on' points'. It was not possible for us to distinguish between theseand the specifically rationed preserves. There was in neither year a guaranteedration of eggs, but about one egg per ordinary ration book per week was avail-able to most people during our sampling periods in 1945 and 1946. Our figuresof consumption of eggs in Table XIII refer to eggs purchased and to eggs fromown hens, the latter amounting, per person per week, for the unchanged

1On the average: 2 ounces one week and 4 ounces the next week.the average: 8 ounces for one four-weekly period and 12 ounces for the next period.

3Free, if the family was poor.'Tinned jam was obtainable on points' only in 1946, but tinned marmalade was

points 'rationed in June and July of 1945 and 1946,

156 THE BULLETIN

TABLE XIIIWeekly Consumption of Foods for Ordinary Ration Book and Children's Consumption of Milk

Meat1BaconTot. Rat. FatsBuller onlyUnrat. Fats3TeaSugarJam'CheeseEggs, shell'Foods Rationed on 'Points':Eggs, dried"Cooked Meat11Tinned Meat"Tinned FishTinned Peas and BeansCereals" .........Dried Legunes"Dried FruitSyrup and Treacle

Milk, total16Of this:OrdinaryGovernmentSchoolTinned and Household"

Under 6 years"Over 14 years"

Unchanged Fam,.

(59)1945 1946I7id. l7id.

2.9 02. 2.9 02.7.1 02. 7.0 02.2.0 o:.0.3 02.2.0 05.9.8 os.4.3 02.1.8 os.'1.8 eggs

1.3 eggs0.9 02.

d.id.*d.

1.3 oz.0.3 oz.2.1 oz.1.0 oz.

2.8 o:.0.3 02.2.2 oz.

10.4 02.4.3 oz.2.0 02.

1.5 eggs

0.6 eggs0.4 os.

*d.id.ld.

2.1 02.0.5 oz.1.6 02.0.8 02.

Changed Families NewFamilies

(68)1946

(24)1945 1946l7id. 161d.

2.7 02. 2.9 02.6.8 os. 7.0 os.2.Oo:. 2.60:.0.3oz. 0.4 os.1.8 oz. 2.5 oz.9.4 oz. 11.5 05.3.9 os. 3.6 oz.2.0 os. 2.0 oz.'

2.1 eggs 1.7 eggs

1.1 eggs1.1 Os.

hi.id.

1.3 os.0.6 02.1.9 02.0.6 os.

0.5 eggs0.3 Os.

4d.lid.Id.

1.5 os.0.9 Os.1.2 02.0.7 os.

171d.2.9 os.7.0 os.2.8 os.0.4 02.2.6 os.9.8 os.3.6 os.2.0 os.'

1.0 eggs

0.7 eggs0.5 os.

id.Id.id.

1.4 os.0.1 Os.1.2 oz.1.3 os.

Milk Consumption per Person4.7 pt. 4.6 pt. 4.7 pt. 4.6 pt. 4.6 pt.

2.9 S. 2.9 t. 3.1 p1. 2.8 t. 2.5 p1.1.1 t. 1.0 t. 1.0 Pt. 1.1 t. 1.3Ø.0.6 Pt. 0.6 Pt. 0.5 Pt. 0.6 Pt. 0.7t.0.1 Pt. 0.1 t. 0.1 t. 0.1 t. 0.1 Pt.

Actual Consumption of School and Government MilkPer Child per Week

6.9 pt. 7.0 pt. 7.0 pt. 6.7 Pt. 7.0 pt,2.3 pt. 2.2 pt. 1.7 pt. 2.6 Pt. 2.1 pt.

1 Including off als, but excluding cooked and tinned meat products, and sausages.'Butter, margarine, lard, and cooking fat.'Suet and dripping.'Including honey, and jam and marmalade rationed on ' points'Excluding 9 special rations.'Excluding 5 special rations.'Excluding 1 special ration.'Excluding 3 special rations.The figures refer to consumption per person. No correction has been made for special

allowances for infants under 2 years old, and expectant mothers. The following amounts ofeggs from own hens are included: Unchanged families, 1945-0,5; 1946 0.5; changedfamilies, 1945-1.1; 1946-0.9; new families, 1946-0.l.

"One packet of dried eggs equivalent to 12 eggs in shell.11 Including small quantities of unrationed cooked meat."This group overlaps the preceding one. It also includes tinned mixtures of meat and

vegetables."Excluding rolled oats, but including small quantities of unrationed macaroni and

semolina."Beans and peas."The figures in this section show the total quantities per week divided by the total

number of persons, for each group."Rationed on 'points'. The combined figures for tinned and Household milk are given

in terms of fresh milk." Excluding children between 5 and 6 years old who attend school."Including the children that were excluded from the preceding group, and also those

children over 14 years old who obtained milk at school.

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 157

families to about half an egg, and for the changed families to about one egg,in both years.

The consumption of milk is given in Table XIII per person per week.The data have been computed by adding together the quantities of all themilk obtained per week and dividing this figure by the total number ofpersons. Below these data of consumption for all milk taken together, aswcll as for the several groups of milk, per person per week, we have set outthe data of special milk allowances per child entitled to this milk, viz., thetotal weekly consumption of Government1 milk allotted to the children ofour sample, divided by the total number of children up to the age of 5 years,and the total consumption of school milk, divided by the total number ofchildren at school. It will be seen that virtually all the families entitled toGovernment milk for their children took up their allowance,2 and that thechildren at school consumed there on the average between z and z pintsweekly, in

In general, the figures in Table XIII indicate that the families of oursample purchased all their rations, although a few of them replaced part ofthe butter ration by margarine. The figure of weekly average consumptionof tea in 1946, of the unchanged families, was only 2.2 ounces instead of thepossible 8-weekly average of z. ounces; but the household accounts of themajority of these families were compiled du fing a period when the allowancewas only 2 ounces per ordinary ration book per week.(b) Foods Rationed on ' Points'

The number of' points' available for the purchase of certain foods was5 per person per week in June and July, 2945, and 6 in June and July, 1946.However, our figures in Tables XIII and XIV referring to these foods indi-cate no general rise in consumption from the earlier to the later period.Indeed, the consumption of some of the foods rationed on' points' declined,since of several of these foods the 'points' value was higher in June andJuly, i6, than it had been during the corresponding period in 2945,while dried eggs, which then belonged to the specifically rationed foods,were sold in the summer of 1946 on' points', at 8' points' a packet, equiva-lent to iz eggs in shell.

Table XIV also shows the consumption of cocoa and of wet, cured, anddried fish, which foods are unrationed. All the other foods enumerated inTable XIV were rationed on 'points' in 2946. For all the foods we havecalculated the consumption per family buying the particular food and alsothe average consumption of all families, i.e., the total quantities consumed oramounts spent per week, divided by all the families of the group, irrespectiveof whether or not they actually purchased this food. The proportions offamilies buying the several foods as well as the average quantities bought,

1 By Government' milk we understand the milk sold at 2d. per pintor supplied freefor children up to the age of 5 years and expectant mothers, and by school' milk themilk issued at school at 4d. for a third of a pint. or free.

a One of the changed families stated that they never applied for this milk.Consumption of milk is discussed more fully in the Appendix.

11*

TA

BL

E X

IV

Dis

trib

utio

n Pa

ttern

of

Purc

hase

s of

Cer

tain

Foo

ds

1 N

o qu

antit

ies

stat

ed.

'One

pac

ket e

qual

s 12

egg

s in

she

ll,a

One

tine

equ

als

4 pi

nts

of s

kim

med

fre

sh m

ilk.

'Equ

ival

ent.

No

of P

urch

asin

g Fa

nis.

as

% o

fG

roup

Uni

t

Purc

hasi

ng F

ams,

Con

sum

ptio

npe

r Fa

mily

per

Wee

kA

ll Fa

ms.

Con

sum

ptio

n pe

rpe

r Fa

mily

per

Wee

k

Unc

hang

edC

hang

edN

ew59

Fam

ilies

24

Fam

ilies

68Fa

ms.

1945

1946

1945

1946

1946

Unc

hang

edC

hang

edN

ew59

Fam

ilies

24

Fam

ilies

68Fa

ms.

1945

1946

1945

1946

1946

Unc

hang

edC

hang

edN

ew59

Fam

ilies

24

Fam

ilies

68Fa

ms.

1945

1946

1945

1946

1946

Fish

8892

6779

84oz

.31

.032

.045

.146

.431

.027

.329

.330

.136

.826

.0Fi

sh, t

inne

d46

5142

5446

d.9f

8f13

f14

8f4f

4f5f

7f4

Mea

t, co

oked

5132

7133

28oz

.7.

65.

98.

04.

19.

43.

91.

95.

71.

42.

6M

eat,

tinne

d'19

20-

2528

d.15

f12

f-

12f

173

2f'-

3f4f

Egg

s, d

ried

251

3254

3337

egg'

11.4

8.8

10.2

8.3

8.9

5.8

2.9

5.5

2.8

3.3

Milk

, tin

ned

2022

2542

25tin

1.1

1.0

0.8

0.9

1.0

0.2

0.2

0.2

0.4

0.2

Milk

, Hou

seho

ld'

108

88

7pt

.'2.

32.

42.

02.

03.

20.

20.

20.

20.

20.

2D

ried

Leg

umes

1015

1725

6oz

.14

.713

.317

.017

.312

.01.

52.

02.

84.

30.

7R

olle

d O

ats

2415

1717

21oz

.32

.023

.118

.014

.022

.37.

63.

53.

02.

34.

6B

iscu

its75

6871

7575

oz.

11.7

11.4

16.9

14,8

13.0

8.8

7.7

12.0

11.1

9.7

Coc

oa27

2929

2931

oz.

5.1

5.7

5.7

4.0

5.0

1.4

1.6

1.7

1.2

1.5

Dri

ed F

ruit

5849

5467

53oz

.16

.114

.617

.99.

511

.49.

37.

29.

76.

36.

1Sy

rup

and

Tre

acle

2229

2129

41oz

.19

.112

.214

.413

.715

.14.

23.

53.

04.

06.

2

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 159

indicate a definite pattern of preferences. This pattern is, of course, not arigid one, and it can be modified by changes in the relative' points ' values ofthe individual foods. This may affect even the purchases of unrationed foods.Thus, the increase in the proportions of families that bought unrationed fishin 1946, as compared with the corresponding figures for 1945, suggests thatfish acted partly as a substitute for cooked meat, which was dear in terms of'points'. Dried eggs were bought in 1945 by about half the families thatsupplied accounts to u for both years, but only by about a third of them in1946. In '945, one packet was available per ordinary ration book aboutevery four weeks, while in 1946, dried eggs had to be bought on 'points,'so that a housewife had to decide whether she preferred dried eggs to otherfoods similarly rationed.

The Main Unrationed FoodsOur survey in 1946 was undertaken before the introduction of the

rationing of bread, flour, and flour confectionery. A few families in factsupplied accounts after the rationing of bread and flour had started, but theirconsumption of these foods was probably unaffected by it, since during theinitial stage this rationing was not applied very strictly. The weekly con-sumption, per person and per 'man', of these foods and of potatoes, fish,and sausages, and further, the weekly outlay per person and per 'man'on fresh fruit and vegetables (excluding potatoes), are considered in TableXV. We have there also included the figures of average consumption of rolledoats,1 which were rationed on 'points', since in some families porridgerepresents an important partial substitute for bread at breakfast.

The average total consumption of the five main carbohydrate foods-bread, flour, flour confectionery, potatoes, and rolled oatshad increasedfor the unchanged families, while for the changed families it had slightly fallen.The average amount of calories obtained from these foods was, however,approximately the same in 1946 for the unchanged, the changed, and the newfamilies, and we have no reason to assume that the purchases of flour in1946 of the unchanged families included considerable quantities bought forstock.

There was apparently more soft fruit available during our sampling periodin 1946, as compared with that in 1945, and the increase in outlay on freshvegetables and fruit of the unchanged families seemed to be due to this.

Meals Away from HomeThe average numbers of luncheons and dinners eaten by adults and

children away from home are considered in Table XVI. The numbers ofmeals for adults have been calculated per earner, a housewife working part-time being counted as half an earner, and the numbers of meals for children,

'None of the families of our sample bought oatmeal.'The calorific intake per man 'per week from foods in Table XV was as follows (Ministry

of Health Scale)Foods in Table XV. Unchanged Families Changed Families New Families

Nos 1945 1946 1945 194-6 1946(1) to (6) ......9,960 10,680 10,960 10,570 10,620(7) & () ... 370 300 370 390 30

TABLE XVILuncheons and Dinners A way from Home

Group of No. of Meals per Week Av. Price per Meal Boughtof Per per Child at School Adult Child

Families Adult at bought free d dWork

Unchanged

TABLE XVIIExpenditure on Rationed and Unrationed Foods

per 'Man' per Week59 Unch. Fama. 24 Changed Fama. 68 New Human

Fama. Needs'1945 1946 1945 1946 1946 Nov.1946

1 Including meat ofials, and tinned jam and marmalade rationed on' points'.'Including macaroni and semolina.'Rationed on 'points'.'Including Government milk and children's milk at school.I Including wet fish, kippers, and fried fish, sausages, cocoa, unrationed fats, fancy foods

and condiments, and meals bought away from home,'Cf. BULLETIN, Vol. 8, No. 12.

(1) Specifically Rationed Foods1 ... 4 4 4 31 4 14 4 24 4 5 r(2) 'Points' Rat. Foods (cxci. milk)' 1 14 0 11 1 04 0 101 1 04 . 4 g(3) Tinned and Househ. milk' ... 0 04 0 04 0 04 0 04 0 04 i.(4) All Fresh Milk' 1 81 1 8 1 94 1 81 1 8 1 64(5) Bread and Flour 1 2 1 54 1 4 1 44 1 54 1 54(6) Cake and Buns 0 94 0 74 0 94 0 81 0 7 -(7) Potatoes ............ 0 94 0 101 0 104 0 11 0 101 0 54(8) Fresh Vegetables and Fruit 1 64 1 94 1 5 1 4 2 04 0 104(9) Sundries' ......... 3 04 3 31 2 7 3 04 3 24 1 i

Total ... 14 74 14 114 14 0 14 21 15 44 10 21

59 Unchanged Familiesper per

Person 'Man'1945 1946 1945 1946OZ. 02. OZ. 02.

24 Changed Familiesper per

Person 'Man'1945 1946 1945 194602. 02. 02. 02.

68 New Famaper per

Person' Man'1946 194602. 02.

(1) Bread ...... 64.2 70.2 83.6 88.3 75.4 72.3 96.7 93.2 71.5 96.0(2) Flour ...... 7.3 10.9 9.5 13.7 7.0 5.8 9.0 7.5 5.2 7.0(3) Cake, Buns,

Biscuits ... 10.9 9.1 14.2 11.4 10.0 9.5 12.8 12.3 7.6 10.2(4) Potatoes, old ... 13.1 15.2 17.0 19.1 27.9 14.0 35.8 18.0 19.6 26.4(5) Potatoes, new ... 35.3 46.5 45.9 58.6 34.6 44.8 44.4 57.8 42.4 57.0(6) Roiled Oats ... 1.7 0.8 2.2 1.0 0.6 0.4 0.8 0.6 0.9 1.3

(7) Fish ...... 6.2 6.6 8.0 8.3 5.9 6.7 7.5 8.7 5.3 7.1(8) Sausages ... 3.5 3.9 4.5 4.9 3.9 3.6 5.0 4.6 4.0 5.4

Outlay on d d d d d d d d d dFresh Veg. and Fruit(cxci. potatoes)

141 17 181 211 13 121 17 16 174 241

1945 1.45 1.22 0.33 14 44194.6 1.84 2.20 0.38 141 5

Changed1945 1.72 0.64 0.38 121 51946 1.53 1.25 0.31 144 5

New1946 1.19 1.91 0.19 154 31

i6o THE BULLETIN

TABLE XVWeekly Consumption of the Main Carbohydrate Foods and of Fish, Sausages, and Fresh

Fruit and Vegetables

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE ibi

per child attending school. The figures of average consumption of schoolmeals indicate a decidedly upward trend from 1945 to 1946. For the childrenof the unchanged families, the average numbers of school meals consumedincreased by 63 per cent, nearly half of all the children at school of thesefamilies having their mid-day meal there on five days a week. Of the childrenof the changed families, 27 per cent more had a main meal at school in 1946than in 1945, 31 per cent of all the children at school of this group partakingregularly of school meals in 1946. For the children of the new families thisproportion amounted to 4! per cent.

Nearly every third of the earners of the unchanged families had one mainmeal out on 6 days a week in 1946, nearly every fourth camer of the changedfamilies, and nearly every fifth of the new families. The average prices paidfor meals for adults in 1946 varied from Is. 2d. to IS. 34d., and the averageprices paid for children's meals from 3d. to 5d.

(.) THE OUTLAY ON RATIONED AND UNRATIONED FooDsThe outlay on the several groups of rationed and unrationed foods per

'man' per week is given in Table XVII, and is there compared with theoutlay on the corresponding groups of a 'human needs' diet, compiled byus for a family consisting of husband, wife, and three children of school age,for November 1946.1 The average expenditure of the new families on all therationed foods amounted to approximately jod. per 'man' per week morethan that computed by us for the ' human needs' diet With respect to theunchanged families, however, the difference was only about 6d., and withrespect to the changed families, 5d. These differences are mainly accountedfor, first, by the higher prices paid for bacon by the ii families of oursample, as compared with the price allowed in the 'human needs? diet,2secondly, by the purchases of butter by these families instead of margarine,substituted for butter, in the 'human needs' diet, and, thirdly, by theinclusion of some unrationed meat offals in the figures of outlay on rationedfoods, in the case of the families of our sample. The outlay on milk in the'human needs 'diet excludes expenditure on school milk, which in November1946, was free for all children. Had this milk been paid for at the rate of i4d.per pint, the outlay on milk per 'man' in the 'human needs' diet wouldhave risen to is. 84d.approximately as much as was spent on the averageon this item by the families that supplied the household accounts. Nôpurchases of cakes and buns were allowed for in our 'human needs' diet.

The main differences between the outlay on foods of the families thatsupplied the household accounts and the outlay calculated for a human needsdiet appear for the groups of fresh vegetables and fruit, and of 'sundries',i.e., the groups that include the unrationed semi-luxury -foods, as distinctfrom such a basic unrationed food as potatoes, -the price of which was con-trolled in November 1946, at about a penny a pound. Bread, flour, cake, andbuns were rationed in November 1946, but not in June and July of that year.

1 \Vith a total man value of 3.75. Cf. ISIJLLErIN, Vol. S No. 12.2 2s. 1d. per lb. as against is. 9d. per lb.

x6z TIlE BULLETIN

() NUTRITIONAL INTAKEThe average nutritional intake of the families of our sample is given in

Table XVIII. It has been compiled by means of two different scales. The

TABLE XVIII. Nutritional Intake per Man per Day

1 Pre-war B.M.A. standard. 'Not given.

original scale, termed in this paper ' Institute of Statistics Scale', measuresprimarily the calorific requirements of individuals. We have also appliedit for assessing the relative share in the total household expenditure neededfor the several members of the family, on food as well as on non-food items,and it has been used for this purpose throughout this paper.

In the summer of 1943, the Ministry of Health supplied us with im-proved data of nutritional requirements of the different age and sex groups,and we have employed these ever since in our calculations of nutritionalintake, side by side with our own scale, under the title of' Ministry of HealthScale'. Our own scale has retained its usefulness mainly as a means ofstandardizing the families so as to be able to compare data of householdexpenditure of different families, since the Ministry of Health scale could noteasily be adapted for this purpose. We have also continued to calculate forour samples every year the average calorific intake and the average intake ofprotein on the basis of the Institute of Statistics scale, so as to make possiblean easy comparison of these data in our earlier papers with those in thelater ones. The two scales of 'man' values are shown in Table XIX.

TABLE XIXScales of' Men' Values

Institute Scale(Gross Values)

Calories Protein (gram mes)total animal

Ministry of Health Scale(Net Values)

Calories Protein (grammes)total animal

A verage Requirements : 3,300 100 5O 3,000 70Intake

f1945Unch. Fams.L 1946

3,3503,600

108116

5452

3,1003,350

8594

4140

f1945Chgd. Fams. 119463,3003,550

109114

5251

3,0003,250

8389

3839

New Fams. ... 1946 3,650 117 54 3,400 94 41

Institute of Statistics Scale Ministry of Health ScaleAge Groups Calories Proteins

Adults and Juveniles : 21+, male 1.00 1.00Maie, 14+ 1.00 21+, female 0.83 0.86Female, 14+ 0.85 17-21 1.00 1.29

15-17 1.00 1.43Children : 12-15 0.97 1.36

10--14 0.707-12 0.70 0.93

6-9 0.603-7 0.50 0.71

3-5 0.50 l-3 0.37 0.60Under 3 0.33

6-12 months 0.27 0.330-6 months 0.20 0.26

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 163

Not only are the basic standards different in the Institute of Statisticsscale and the Ministry of Health scale, but so also are the tables of nutritionalvalues of foods used in conjunction with each of them. Much research in thecomposition of foods has been carried out in recent years, and the tables offood values supplied to us by the Ministry of Health in 1943 have alreadybeen superseded, in 1945, by a new set of tables published by the MedicalResearch Council. Since, however, it is intended to conclude this presentseries of surveys in 1947, it seemed better to avoid introducing another set offigures, in particular as the average figures of nutritional intake derived fromthe latest data would not substantially differ from those obtained by meansof the earlier Ministry of Health data. On the other hand, the differencebetween the data of nutritional values used with the Institute of StatisticsScale, and those used with the Ministry of Health Scale, is considerable, forthe former refer to gross values, i.e., nutrients as bought, and the latter to netvalues, i.e., nutrients as consumed. It will be seen from the data in TableXVIII that the average requirements foi' calories and protein are givenconsiderably higher for the Institute of Statistics Scale than for the Ministryof Health Scale.

The average calorific intake of the unchanged as well as of the changedfamilies was higher in 1946 than it had been in 1945, and the figures of intakeof the new families were well in keeping with the figures relating to the twoother groups of families, for 1946. Since under the present conditions ofrationing of fats and sugar a noticeable increase in the calorific intake can ingeneral be attained only through an increased consumption of starchy foods,whose calorific content is partly represented by second class protein, a rise incalorific intake implies a rise in the intake of protein. We thus find for theunchanged as well as for the changed families higher figures of calorific intakein 1946 than in 1945 accompanied by higher figures of protein intake. Noperceptible rise is indicated in the corresponding figures of animal protein.Moreover, the figures of calorific intake and of the intake of second classprotein only suggest a return to the nutritional level of our sample familiesin 1944, after a slight decline in nutritional intake in 1945.

CONCLUSIONThree main conclusions seem to emerge from this survey. First, no

deterioration of income appears to have been experienced by the families ofour sample, although one might have expected a decline of overtime fromJune-July 1945, to June-July, 1946, and, in consequence, reduced earnings.Secondly, there was no apparent change in total household expenditure sinceour last sampling period, but a slight increase was indicated in the outlay onfood and also on fuel and lighting, the latter mainly due to an increased con-sumption of gas and, possibly, electricity. Purchases of coal, too, were some-what higher in 1946 than in 1945. Thirdly, the total expenditure on foodhousing, and fuel and light, seems to have absorbed on the average only abow

o per cent of the income, the remaining half of which was thus available foi

164 THE BULLETIN

expenditure on clothing and miscellaneous household requirements, theowners' personal expenditure, and savings.

A comparison between the purchases of foods of the families of our samplein June-July 1945, and June-July 1946, shows that there were no importantchanges in the consumption of the main foods. Differences were apparentmainly as regards the purchases of foods rationed on' points ', and a distinctdecline could be observed in particular in the purchases of dried eggs and ofcooked meat. The outlay on fresh vegetables, fresh fruit, and on fancy foods,was higher during our sampling period in 1946 than during that in 1945,and the consumption of school meals appeared to have risen sharply. Thecaloriflc intake and, in conjunction with it, the intake of second class protein,was above that in 1945 and corresponded approximately to the average intakeobtained for our sample in 1944, but there was no change in the averageintake of first-class protein from 1945 to 1946.

T. ScHuLz.

APPENDIX: CONSUMPTION OF MILKThe figures of average consumption of milk in Table XIII are about the

same for both our sampling periods in i45 and 1946. In 1946, 4.6 pints ofmilkincluding Government milk and School milkwere obtained perperson per week by the unchanged, the changed, and the new families. Of thistotal, however, the Government milksupplied to children up to the age of

years and to expectant mothersand the milk obtained by the childrenat school, represented, per person per week, i.6 pints in the case of theunchanged families, 1.7 pints in the case of the changed families, and 2 pintsin that of the new families. The new families thus consumed per person perweek about half a pint of ordinary fresh milk less than the unchanged familiesin 1946. Nevertheless, since at least 21 pints of ordinary milk was obtainedper person including the children who also were in receipt of Governmentmilk, as well as those in receipt of school milk, the figures of average con-sumption of milk were satisfactory for all the three groups of families.

Yet the quantities of milk purchased by individual families varied widely.Milk is a quasi-rationed food, but it is also a highly perishable one, and cowsare not machines with a standardized output capacity. Thus, although fami-lies that obtain their milk from a large combine with facilities for utilizingany temporary urplus of milk, may never receive more than their minimumrationin June and July of and 1946, 3 pints per adult per week, and34 per child of school ageothers that are supplied by small dairy farmers,can sometimes buy considerably more than this. It is therefore of interestto analyse the consumption of iJIk in some detail.

Table ja, ib, and Ic, show the consumption of milk of the ii families ofour sample in 1946, in relation to their expenditure per ' man' per week onfood. Table ia is concerned with the 74 families that had nq young children

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 165

TABLE la74 Families with No Children under 6 Years Old

Domestic Consumption of Fresh Milk per Person per Week (Figures in columns: Heavytype, total no. of families; below: average nos, per family of persons over 14; children6-14')

'Including children at school over 14 years old.

in receipt of Government milk, while the 77 families having children in thisage group are dealt with in Tables ib and IC. In all these three Tables, thefigures in heavy print indicate the numbers of families in the respectiverange of food expenditure and milk consumption, while below these figuresare set out the relevant data of average structure of the families. In Table ¡athe families have thus been split up into persons over ¡4 years old and childrenof school age.' The figures of average structure of the families in Table ¡aindicate only for 4 of the 74 families a consumption of milk less than 2kpints per week per adult and 3k pints per child at school. Two of thesefour families spent on food per 'man' per week between ios. and i Is., onefamily spent between i6s. and ¡78., and one spent over ¡98.

The picture supplied by Table ¡b is less satisfactory. The figures in smallprint in this Table give, per family, the numbers of persons over ¡4 yearsold, the numbers of children of school age, and the numbers of children underschool age. In calculating the domestic consumption per person these latterchildren have been omitted, since they all had an allowance of Government

1 Actually, all juveniles up to the age of 16 years old are entitled to 34 pints of milk perweek. But it is sometimes impossible to persuade a young person to consume much milk,especially if he has left school and has perhaps six main meals per week at his works; 24to 3 pints per week may be all that is required for his tea. We have therefore only includedthose children over 14 years old who were still at school.

Food per 'Man'per Week

Lnder lOs.

lOs. +

ils. +

12s. +

+

S +15s. +

16s. +

Ss. +19s.and

over

under2 pints-

4.0;1

4.0

---1

.2.0; 2.0

-1

6.0; 3.0

2 pints +

-2.0; 7.0

24.0; 2.0

32.7; 1.3

33.3; 1.3

--2

3.0; 2.02

2.0; 1.02

4.0; 0.523.0; -

3 pints +

--1

2.0; 5.05

3.6; 2.25

3.0; 1.87

2.3; 1.3

2.02

1.5'4

2.5; 1.32

2.5; 1.012.0; -2

2.5; 2.0

4 pints +

23.0; 1.0

2.0; 1.0

-3

3.3; 1.75

2.6; 1.01

4.0; 1.0

-1

3.0; 1.01

2.0; 1.03

3.0; 0.7

-

5 pints +

----

14.0; -2

3,0; 1.01

2.0; 2.03

2.0; 1.0

6 pints andover

---

-1

2.0; 1.0

34,0; 3.0

173.1; 1.5

292.6; 1.7

172.8; 1.1

72.6; 1.0

12.0; 1.0

Do mestic Consumption of Fresh Milk excluding Government Milk and Children receiv-ing it. per Person per Week (Figures in columns: Heavy type, total no. of families; below:average nos, per fani ily of persons over 14; children 6_141; children under 6

1IncIudng children at school over 14 years old.'2s. 6d. was spent per week on Ostermilk by one family in this group.

milk of one pint a day. The consumption of milk of all the i. families thatpurchased less than between z and 3 pints per person per week was substan-tially below their minimum allowance, as was also the consumption of io outof the 20 families that bought between z and3 pints per week, since on accountof their numbers of children of school age, they could have obtained per weekwell over 3 pints per person over 5 years old.

One might have expected the families with the smallest milk consumptionto be those with the smallest average expenditure on food. This, however,is not the case. In both Tables ia and ib, we find families with a very smallexpenditure on food and a satisfactory consumption of fresh milk, and inboth Tables we also find families whose outlay on food was fairly high, andwhose consumption of milk was low. One family, spending on food per'man' per week over 17s., bought in fact no fresh milk except the Govern-ment milk allowed for their 3 infants, their only supplementary milk con-sumption being one tin of evaporated milk per week, for three adults and onechild of school age. None of the families with a very low consumption ofmilk purchased âny patent milk food for their infants.'

1 National dried milk for infants supplied in lieu of fresh Government milk has beencounted by us as Government milk,

Food perMan' perWeek ...Under

lOs.

S. +

. +

8, +

't"

. +

+S. +

19s.andover

no ordin-ary fresh

milk

---1

2.0;-; 2.03.0;

-1

2.0; 1.0; 2.0

-1

3.0; 1.0; 3.0--

under2 pints

33.0; 3.7;

22.0; 1.5;

12.0;2

3.0; 2.0;2

2.5; 2.0;

--1

3.0; 2.0;

--

1.0

1.0

1.5

2.02.7;

1.54.0;

2.02.0;

2 pints +

-3

3.0; 1.74.0;23.0; 2.04.0;

-5

2.0; 2.6; 1.64

2.5; 1.3; 1.01

2.0; 1.0; 1.01

2.0; 1.0; 1.02.8;2

2.0; 2.02.0;1

7.0; -; 1.01

4.0;-; 1.0

3 pints +

12.0; 3.0; 1.0

11.0; 1.01

2.0; 1.03

2.0; 0.7; 2.03

2.3; 3.3; 1.31

2.0; 1.0; 1.02

1.5; 0.5; 1.54'0.8; 1.31-; 1.02.0;-3'

2.0; 0.7; 1.02.0;

4 pints +

---1

2.0; 2.0; 1.0

-3

2.0: 1.7; 1.7

-2

1.0; 1.5

-4

0.8; 1.52.0;

5 pints +

-

--1

2.0; 1.O;1.01

2.0; -; 2.02.0;

-4

0.8; 1.3

6 pintsand over

---

12.0; -; 2.0

-11.0; 2.02

2.0; -; 2.03

1.7; 0.3; 1.3

2.3;3

0.7; 2.3 2.7;112.4; 1.5

202.8; 2.0; 1.5 2.4;

201.3; 1.3 2.0;

101.2; 1.5 2.0;

60.7; 1.3 1.9;

70.3; 1.7

i66 THE BULLETIN

TABLE lb77 Families with Children under 6 Years Old

WORKING CLASS INCOME AND HOUSEHOLD EXPENDITURE 167

In Table ic we have taken the same families considered in Table ib,but we have now added together the ordinary fresh milk bought by a

TABLE ic77 Families with Children under ô years old

Domestic Consumption of Fresh Milk including Government Milk, per Person perWeek (Figures in columns: Heavy type, total no. of families; below: average nos, perfamily of persons over 14; children 6-14'; Children under 6.)

'Including children at school over 14 years old.'2s. 6d. was spent per week on Osterniilk by one family in this group.

family per week, and the Government milk allowed per week for their youngchildren, and have divided this quantity by the total number of persons in thefamily. The picture obtained by Table ic corresponds much more closely tothat of Table ia than that of Table ib, as is brought out particularly strikinglyin the summary of Tables Ia, ic, and ib (in this sequence), given in Table z.The inevitable conclusion of this is that by a number of families the milkobtained under the Government scheme for infants, is treated as ordinarymilk and used partly for other members of the family.

It would, however, be quite wrong to assume that in most cases this isdone in order to economize in the expenditure on food. The main reasonseems to be that older infants are sometimes given large quantities of foods thatare more expensive but less suitable than milk, and which prevent them fromdrinking as much milk as they should have. It is frequently the most con-scientious mothers, who try to make their infant eat all the rations to whichhe is entitled-an impossible undertaking-that 1nd their child refusingpart of his milk allowance.

Food per'Man' perWeekUnder

lOs.

los. +

under2 pints

-2

3540154020

2 pints +

21.02.0;

3 pints +

-3

3.0; 2.3

4 pints +

12.0; 3.0; 1.0

5 pints +

--

6 pints +

--

7 pints andand over

-ils. + - 1

2.0;-; 1.03.3;3

3.0; 2.04.0;1

2.0; 1.0 - -1 1 1 2I + - 3.0;2.0; 1.02.0;-; 2.02.0; 1.0; 1.02.0; 0.5;2.5 -3 3 4 1

1 S. + - 2.7; 2.3; 1.0 2.3; 3.0; 1.3 2.0; 2.7; 1.8 2.0; 2.0; 1.0 -3 4 114s. + - 2.3; 1.7; 1.7 2.5; 1.3; 1.0 2.0; 1.0; 1.0 -

1 1 3 1 155. + - - 2.0; 1.0; 1.0 2.0; 4.0; 1.0 1.7; 0.7; 2.0 2.0; -; 2.0 2.0; -; 1.052 2+ - - - 2.6; 1.2; 1.42.0; 1.0:1.0 - -

1 1 2 2 1 117s. + - 3.0; 2.0; 2.0 3.0; 1.0; 3,0 2.0; 0.5; 1.5 2.0; 1,0; 1.5 2.0; -; 2.0 2.0; 1.0; 2.01 2S + - 7.0; -; 1.0 - - 2.0; -; 2.0

19s.and 1 3' 6 3 2over - - 4.0;-; 1.02.0;O.7; 1.02.0;0.7;1.32.0;O.7; 1.31.5;O.5; 1.5

2 11 18 19 16 5 63.5; 4.0; 1.5 2.8; 1.8; 1.3 2.8; 1.9; 1.6 2.3; 1.6; 1.3 1.9; 0.8; 1.6 2.0; 0.4; 1.6 1.8; 0.3; 1.7

Total Group No ord. under 2 pts. 3 pts. 4 pts. 5 pts. 6 pts.fresh 2pts. + + andmilk over

0/ 0/ 0/ 0/ 0/ 0/ 0/ 0//0 /0 /0 /0 /0 /0 /0 /0Nos.

Families with no Children under 6 years old (Table 1Families .........74 100 - 4Persons over 14 ......210 100 - 6Children 6141 ......109 100 - 8

Families with Children under 6 year old (Table 1G).Fresh Milk)

Families .........77 100 - 3Persons over 14 ......184 100 - 4Children 6141 ......110 100 - 7Children under 6 ......114 100 - 3

Families with Children under 6 years old (Table lb).Children receiving it: all Children under 6 years old

Families .........77 100 4 14Persons over 14 ......184 100 4 16Children 6141 ......110 100 2 24Children under 6 ......114 100 6 15

'Including children at school over 14 years old.

TABLE 3aConsumption of School Milk and Domestic Consumption of Fresh, Tinned, and Household

Milk (in terms of liquid Milk)

a).23 39 23 10 125 37 23 8 124 44 17 6 1

(Government Milk and Ordinary

14 23 25 21 1417 28 23 17 1118 31 28 12 412 25 22 22 16

(Excluding Government Milk and

'26 26 13 8 930 26 11 6 735 23 11 3 225 23 13 7 11

Domestic MilkConsumption

Per Person per WeekUnder 2 pt.2pints+3 pints +4 pints +5 pints +6 pints and over

Dom. Milk Cons.Per Non-PriorityPerson p. Week

Under 2 pt.2 pints +3 pints +4 pints +5 pints +6 pints and over

Number ofFamilies Children

School1 3

14 2922 4312 20

5 72 2

TABLE 3bConsumption of School Milk and Domestic Consumption of All Non-Priority Milk (in terms

of liquid milk)

Number of Nos, of Children receiving Schi MilkFamilies Children Children none 14 Pt. per 34 Pt. per

under at School week weekSchool Age'

10 10 26 2 22' 6116 23 34 9 18' 12'14 21 31 4 19' 12'9 13 13 4 7 23 3 4 - - 42 4 2 - - 2

54 83 110 19 66' 38''AU the children in this group were priority consumers receiving 7 pints of Government

milk per week.'Two of these children were also in receipt of Government milk.'Three of these children were also in receipt of Government milk.'Six of these children were also in receipt of Government milk.Seven of these children were also ¡n receipt of Government milk.

i68 THE BULLETIN

TABLE 2Domestic Consumption of Fresh MILk er Person per Week

56 104 21 52 31

Nos. of Children Receiving School Milknone 14 pt. per 34 pt. per

week week37 14 82 28 138 5 7

4 21 1

WORKING CLASS IN5OME AND HOUSEHOLD EXPENDITURE 169

The consumption of school milk in relation to the domestic consumptionof milk by the families of our sample is considered in Tables a and 3b.There were i io families having 214 children over 5 years old that couldobtain milk at school. For five other children of school age this was notcertain, and we have disregarded these in our computations for Tables aand 3b. The figures in both these Tables indicate a tendency for a largeconsumption of milk at home to be accompanied by a large consumption ofmilk at school. That is as one would have expected it to be: childrenaccustomed to milk and milk foods at home like, as a rule, the taste of milk,and drink it willingly at school.

We have indicated in Table 3b those instances in which a child receiving7 pints of Government milk per week was alo consuming milk at a nurseryschool. There were six children attending a nursery school for half the dayand seven children attending it for the whole day, these latter childrenobtaining there three and two-thirds pints of milk per week, in addition totheir luncheon, and bread and butter, etc., for tea. In these cases the mothersmay find that in order not to waste part of the milk obtained under theGovernment scheme they have to use it for other members of the family.

12