Content uploaded by Ilja Kristian Kavonius
Author content
All content in this area was uploaded by Ilja Kristian Kavonius
Content may be subject to copyright.
Does our nation have the patience to become prosperous? Finnish
income, consumption and saving in the beginning of the 1950s
Ilja Kristian Kavonius*
University of Helsinki, Helsinki, Finland
The purpose of this article is to analyse the possibilities of Finnish households
saving and their consumption patterns in the beginning of the 1950s. As a tool for
this analysis, the article estimates unbalanced sector accounts and private
consumption for 1950s Finnish households. These accounts allow analysing the
structure of the Finnish households’ income and the structure of household
consumption. As these accounts are estimated for the whole country and in
addition for three different categories of municipalities, they allow an analysis of
welfare differences between different regions. This article is not limited to the
macro analysis. Based on the individual replies of the cost living studies, some
households are analysed in more depth to get a broader view on their way of
living. The analysis of individual household replies is also based on the sector
accounts, i.e. accounts similar to those on the aggregated level are estimated for
individual households.
Keywords: households; household sector; sector accounts; growth studies;
historical national accounting; household consumption expenditure; disposable
income; saving; saving ratio
Introduction
In 1952, Dr Urho Kekkonen, the Prime Minister of Finland and from 1956 onwards
the President of the Republic, published a book called Does Our Nation Have the
Patience to Become Prosperous? Finland was in those days still recovering from the
war against the Soviet Union and paid its last war reparations in 1952. There was a
lack of capital as the capital markets were restricted and it was difficult to get
financing abroad. Therefore, financing for the investments of industry had to be
found from other domestic sectors.
The Prime Minister emphasised that the industrialisation has to be speeded up.
Especially, the effective use of raw materials in northern Finland should be
guaranteed. This required capital. Dr Kekkonen emphasised the role of the
government in this project, as there was a lack of private capital.
1
The purpose of this article is to analyse the possibilities of Finnish households
saving and their consumption patterns in the beginning of the 1950s. The analysis of
households is essential, as it is the most important sector in financing the
corporation sector either directly or indirectly. It receives a large part of primary
incomes. The role of the government sector is in redistributing income rather than
creating new primary income.
*Email: Ilja.Kavonius@Helsinki.fi
ISSN 0358-5522 print/ISSN 1750-2837 online
#2010 Taylor & Francis
DOI: 10.1080/03585520903516346
http://www.informaworld.com
Scandinavian Economic History Review
Vol. 58, No. 1, March 2010, 2542
As a tool for this analysis, the article estimates unbalanced sector accounts and
private consumption accounts for 1950s Finnish households. These accounts allow
for the analysis of the structure of the Finnish households’income, consumption and
saving. As these accounts are estimated for the whole country and, in addition, for
three different categories of municipalities, they allow for the analysis of differences
in welfare between different regions.
This article does not limit itself to the macro analysis only. Based on the
individual replies of the cost of living studies, some households are analysed in more
depth to get a broader view on their way of living.
This is the first attempt to build sector accounts or a part of sector accounts for
1950s Finland. This supports the work of the Finnish Growth Studies.
2
The Finnish
Growth Studies cover only GDP, its main components and industry breakdown by
six main industries. In some countries historical accounts by the institutional sector
are already estimated. To this end Peter Vikstro
¨m
3
estimated historical balanced
sector accounts for Sweden.
4
This article is structured as follows. The next section provides a theoretical
background and description of the applied methodology. The third section describes
the results of this article. The final section draws some conclusions.
Theoretical background and the method used
System of the accounts, saving and the relevance of the system
The system of sector accounts aims at recording all the transactions between the
various economic agents during a certain period of time. Thus, defining the
economic transactions and balancing items is done through the accounting system.
5
The transactions are grouped into various categories. In turn, these categories of
transactions are shown in a sequence of accounts, each of which covers a specific
economic process. This ranges from production, income generation and income
(re)distribution through the use of income, for consumption and saving, and the
investment, as shown in the capital account, for financial transactions such as net
borrowing and lending.
Each non-financial transaction is recorded as an increase in the ‘resources’of a
certain sector and an increase in the ‘uses’of another sector. For each type of
transaction, the total resources of all sectors and the rest of the world equal total
uses. Each account leads to a meaningful balancing item, the value of which equals
total resources minus total uses. Typically, such balancing items, such as gross value
added (at total economy level GDP) or net saving, are important economic
indicators and they are carried over to the next account.
The production account records the output of goods and services as its main
resource, to which taxes, less subsidies on products, are added to obtain total
resources of the production account at market prices. The main use in the production
account is ‘intermediate consumption’such as the consumption of fuel within a
production process. The difference between resources and uses is the balancing item
‘gross value added’. This gross value added is then carried over as a resource to the
subsequent set of accounts, the generation and distribution of income accounts,
which eventually yield ‘disposable income’as a balancing item.
26 I.K. Kavonius
After the distribution of income, disposable income has to be used in the use of
disposable income account. This is the account where final consumption expendi-
tures are deducted from disposable income and the adjustment factor of pensions is
taken into account. The balancing item of this account is saving. Saving can be
derived from both directions: financial and non-financial accounts. In order to
define saving by using the SNA93 standard terminology, it can be presented as
follows:
6
1. Household disposable income (B6) Final consumption expenditure (P3)
Adjustment for the change in equity of household pensions funds (D8)
Net lending/borrowing (B9)Capital formation (P5)Consumption of fixed
capital (K1)Acquisition less disposals of non-produced non-financial assets
(K2) Capital transfers, receivable (D9R)Capital transfers, payable (D9U)
Household saving (B8)
In the real economy, saving can be seen as a difference between disposable income
and final consumption expenditure and the correction factor of pension funds. The
financial approach calculates this in a sense ‘from the other direction’. It can be said
that saving is the difference between sold and bought financial and non-financial
assets.
However, data for financial assets are hardly available for this period. The reason
is that in those days the government was more interested in income and consumption
of households rather than almost non-existing household investment policy in
financial assets. Additionally, the money markets were strictly controlled. Therefore,
this article focuses on this approach, which is also presented in the upper part of the
formula.
As the saving is defined as the difference between disposable income, the pension
adjustment factor and consumption, the saving ratio can be defined as saving divided
by household disposable income and pension adjustment factor. The denominator
can also be rewritten as a sum of household saving and household final
consumption. The formula can be presented as follows:
2. Household saving (B8)/[Household disposable income (B6)Adjustment for
the change in equity of household pensions funds (D8)]
Household saving (B8)/[Household saving (B8)Household final consump-
tion (P3)]
The link between the non-financial accounts and the financial accounts is established
by balancing the item ‘net lending/net borrowing’.‘Net lending/net borrowing’is
derived from the capital account by comparing ‘gross capital formation’plus the net
acquisition of ‘non-produced, non-financial assets’(such as land or licences) with
‘gross saving’plus net ‘capital transfers’(such as an investment grant). If saving
plus net capital transfers received exceeds non-financial investment, a sector has a
surplus of funds and becomes a net lender to other sectors and/or the rest of the
Scandinavian Economic History Review 27
world. In the financial transactions account, this means that this sector acquires
more financial assets than liabilities.
7
In the system of sector accounts, the transactions flow between institutional
sectors. The institutional sectors combine institutional units with broadly similar
characteristics and behaviour: households and non-profit institutions serving
households (NPISHs); non-financial corporations; financial corporations; the
general government; and the rest of the world.
8
As indicated earlier, this article focuses on the households sector. The households
sector comprises all households and includes household firms. These cover sole
proprietorships and most partnerships that do not have an independent legal status.
Therefore, the households sector, in addition to consumption, also generates output
and entrepreneurial income. NPISHs, such as charities and trade unions, are often
grouped with households.
9
Their economic weight tends to be relatively limited in
relation to households.
10
The relevance of applying these definitions to the 1950s Finland can be
questioned. This issue can be seen from two points of view. First, as the SNA93 is
built for today’s use rather than for the 1950s’use, there might be instruments or
transactions which did not exist in the 1950s. Second, the definition or borderline of
transaction does not necessarily correspond to the actual situation in the 1950s.
The non-existence of transactions in the 1950s is more an issue on the financial
side rather than non-financial side. In the 1950s, the financial markets were strictly
controlled and the financial markets were not as developed as today. On the non-
financial side, this is less of an issue as the basic principles of production, generation
and re-distribution of income already existed at that time.
The relevance of definitions is an issue for this study. In the case of income items,
this is not very obvious, but when the borderline between investment and
consumption is drawn, the definition and its practical implication is not necessarily
very clear. The SNA93 defines durable goods as consumption. However, households
often treat durable goods, which are currently recorded under household consump-
tion, as an investment.
11
Presumably, if a household has little money, as the case was
in 1950s Finland, the effect is larger than in the case of wealthier households. This
issue is acknowledged in this study and possible consequences are discussed when the
results are reported.
Data used and the practical application
For the practical application, three sets of accounts are compiled: national economy;
towns; and micro accounts. These three sets of accounts also serve three different
purposes. The core part of analysis is based on the accounts of towns. The purpose
of the micro accounts is to deepen the picture of the consumption, saving behaviour
and the possibilities for the Finnish households to consume or save. As the micro
data are used rather to deepen the analysis than to provide absolute information, the
whole set of the accounts are not published.
12
The results are used in the descriptive
parts of this article. The main purpose of national economy accounts is to verify the
results.
From the methodological point of view, these accounts follow the methodology
as described above. From the compilation point of view, there are methodological
28 I.K. Kavonius
similarities between different sets of accounts as well account specific issues. Overall,
the idea is to bridge the source data to the concepts of national accounts.
Micro approach
This article makes a first attempt to analyse the available consumption survey micro
data. For the purpose of this, article 15 households
13
are analysed in more depth and
two average families are used as an example in the article. On the one hand, as the
families are from Helsinki, it can be argued that the families are not very
representative. On the other hand, Helsinki was the wealthiest area in Finland,
and thus the standard of living in Helsinki can be viewed as the highest among
average families in Finland. The purpose of the micro section is to deepen the
understanding of macro results.
14
These two households are selected randomly. In the process of writing this article,
consumption booklets are also analysed. The replies are recorded as an open booklet
formula, e.g. the replies are written in a booklet by hand rather than being already
recorded according to fixed classifications. On the one hand, this makes the analysis
of the data difficult, as the classification has to be done immediately. On the other
hand, the booklets provide for instance information on the weekly pattern of
consumption, information on the actual products consumed and also overall
information on the standard of living. Additionally, these booklets are used to get
more detailed information on the consumption: what was consumed and who was
consuming. Finally, information on the overall standard of living like occupation,
the standard of accommodation, the size of family, etc. is used in the analysis.
The written booklets include the household consumption and income in four
two-week periods. These also include the whole-year consumption of other goods
excluding food and cleaning stuff. Therefore, the data have to be annualised. This is
done by assuming that these four two-week periods represent the typical consump-
tion and income of the family, and then other food and cleaning stuff are added. The
assumption proves to be almost correct as the four periods represents different times
of the year. Thus, the seasonality of income and consumption is captured. The same
method was applied in the original cost of living study.
15
These individual household accounts can be seen as an attempt to analyse 1950s
Finnish households from the welfare economic point of view.
16
The approach of
welfare economics is to focus on the efficient and fare allocation of resources.
However, the macroeconomic approach does not provide an aspect to distribution.
Currently, there is neither Finnish economic nor economic historical analysis on the
income/consumption distribution before the middle of 1960s. As, for instance,
Tanninen and Tuomala
17
emphasised in their article, the reason for this is that there
is hardly any useable survey data available before the middle of 1960s. By using
taxation data, some approximations are done concerning the earlier periods.
Additionally, for instance Hjerppe
18
has analysed household consumption at a
macro level from the 1860s onwards.
There are some household surveys available before the middle of 1960s. As
Tanninen and Tuomala (2007) indicated, these have hardly been used. There are
several reasons for the lack of analysis. As these data are not computerised, they are
difficult to use. In addition, these data are focused on the working class families in
Scandinavian Economic History Review 29
towns or alternatively on the farmers in countryside rather than on the whole
population.
Accounts of towns
Appendix 1 shows the bridge table between the cost of living study
19
and the SNA93
income concepts. The income concepts used in the cost of living study do not
correspond directly with any standardised classification as such. Appendix 2 presents
the bridge table between cost of living study and the SNA93 expenditure concepts.
The accounts are compiled for three types of towns: small; middlelarge; and
Helsinki. The threshold between small towns and middlelarge towns is 20,000
inhabitants.
20
The towns included in the survey were selected carefully so that
different parts of country as well as different industries in the country were
represented. The breakdown used is defined by the source data as this breakdown
was used in the cost of living study as well.
The data set used is the cost of living survey data from 195051. The selected time
span can be criticised for the fact that it does not represent typical 1950s’
consumption behaviour, as the war’s end was still close. However, as argued also
in the methodological part of the original cost of living study, the regulation of
foodstuff was mostly over in 1949, and it can be said that the consumption of
households was near to the normal peacetime consumption.
21
Coffee, rents and
alcoholic beverages were still under regulation. To some extent the regulation of
rents continued until the end of the 1980s. The regulation of alcohol was based more
on the social policy rather than on a shortage.
22
The survey does not cover the whole population. It includes only families in
which at least one person is permanently working and which included at least
husband and wife. The methodological section of survey states that the consumption
habits of the single people are so different that they might disturb the results of the
survey. The survey does not cover the non-working and entrepreneurial population.
However, most of the working population were employees. In those days, most
entrepreneurs were farmers.
The sample includes approximately 650 families. Representative areas, industry,
as well as the size of families were pre-defined by using existing statistics. Thus, the
sample covers all types of families in the population. The families were selected by
the local authorities, which were often assisted by the local trade union representa-
tives. The sample, however, was created so that all essential industries and families
of different size were represented. Only families which had children over the age of
15 years or other adult guests living at home were excluded from the sample.
Even though the sample is small and the methods used in the survey do not fully
correspond to the sampling methods used today, the results are plausible. The results
of the survey are also cross-checked against already existing quantitative data.
Macro approach
The bridging between national accounts
23
and the SNA93 concepts of institutional
sector accounts is done similarly as in the case of the cost living study. However, this
was slightly simpler than in the case of the cost of living study as the income and
30 I.K. Kavonius
consumption concepts of the SNA68 do not differ much from the SNA93
concepts
24
.
The old national accounts do not include institutional sector accounts, but the
components of the household sector accounts are available in the publication. The
concept of the households is also slightly different from the concept of the cost of
living study as it included NPISHs. As the national accounts data were more
aggregated (especially in the case of expenditure items) than the cost of living data,
the comparison of these two data sets had to be done at an aggregated level, as can
be seen in Table 3.
Results
Household saving and income
The beginning of this article questioned whether our nation has patience to save. In
the beginning of 1950s, when this question was raised, it looked like the Finnish
households did not have money to save. As can be seen in Table 1, all the estimations
in this article seemed to give negative saving ratios. The differences between saving
ratios are caused by the differences in the underlying population. The national
accounting saving ratio was slightly negative in 1950 and slightly positive in 1951.
The saving ratio based on the national accounts is slightly higher than the saving
ratio based on the cost of living study as the national accounts estimate include
NPISHs. This category includes organisations like trade unions, foundations,
churches and other organisations, which often have positive savings. Additionally,
entrepreneurs were included in this population, which do have to have a positive
saving if they want to invest in their businesses.
The saving ratios of the cost of living survey estimation are slightly negative. The
overall saving ratio was around 2.6%. In middlelarge towns and Helsinki saving
ratios were negative, but in the small towns saving ratios were slightly positive.
Mostly owner-occupied housing causes this difference. This view is supported by the
fact that 61% of the whole population in the survey lived in rented housing, 20% in
housing owned by the employer and 19% in owner occupied housing. Proportionally
the largest share of owner-occupied housing as well as housing owned by the
employer were in small towns. The smallest share was in Helsinki, where 75% of
families lived in rented housing.
25
The microanalysis supports this view. Family ‘Meika
¨la
¨inen’
26
lived approximately
five kilometres outside the centre of Helsinki. The area was typically suburb where
working class people were living in those days. The apartment was 46 square meters
Table 1. Household saving ratios according to the estimations based on the national
accounts and the cost of living survey estimations.
National Accounting Cost of Living Survey
1950 1951 195051 Group I Group II Group III
Saving ratio (%) 1.03 0.96 2.58 3.17 5.38 0.78
Note: In the case of the consumption survey, Group I refers to Helsinki, Group II refers to middlelarge
towns and Group III refers to small towns.
Source: Author’s estimations.
Scandinavian Economic History Review 31
in size, which was already fairly spacious for those days in Helsinki. The apartment
had one room and a kitchen and there were toilets shared with neighbours. The
apartment was built before the WWII and it had electricity. The husband of
the family, who was 35 years old, worked as a car mechanic in Arabia Oy
27
and the
36-year-old wife was housewife. Additionally the family had two daughters aged five
and seven.
The wife of the family took care of the household economy. She also completed
the cost of living survey booklet. The booklets include several entries which are
defined as pocket money for the husband. This kind of distribution of labour was
typical in working-class families. It is not obvious how the pocket money was spent.
The money was probably spent on the husband’s personal consumption like coffee at
the canteen of his working place, etc. Porter Benson
28
also observed that in the US
before the WWII husbands had the role of breadwinner and wives run households
accounts. From the standard of living point of view, it looked like the US working-
class households in the 1920s and 1930s had a similar standard of living to that in
Finland in the beginning of 1950s.
The estimated saving ratio for the family was almost 10%. Misreporting can
affect the result but the message in the booklet was clear. Everything that was earned
was also consumed and additional money would have been useful. However, there
were not any signs of borrowing money. This does not of course mean that family
Meika
¨la
¨inen did not do this it only indicates that they have not reported it. The
short-term borrowing of money from other families was also a common phenom-
enon for wage earners. This was apparent in several cost of living booklets. Porter
Benson (2007) observed the same phenomenon also in the pre-WWII US.
The saving ratio of family ‘Virtanen’
29
was also around zero. The family had
several short-term loans from other families as well as from the Helsinki City
Financial Services Office.
30
Father Virtanen was a 30-year-old tram chauffeur
working for the city of Helsinki. His 24-year-old wife was working as well.
Additionally, they had a 3-year-old daughter. The family lived in an apartment of
23 square meters in the district of To
¨o
¨lo
¨, which is located practically in the centre of
Helsinki. The apartment had electricity, central heating, its own separate toilet,
running cold water and a gas stove.
Table 2 shows the income structure of the households. Different income
components are presented in relation to the estimated disposable income. Thus,
different populations are comparable. There would not be an analytical meaning of
showing simply the values at current prices.
As can be seen in the Table 2, the largest share of disposable income is wages and
salaries. In the cost of living study estimations they represent approximately 115% of
disposable income. The proportions were slightly lower in small towns. The reason is
that the salaries are lower in smaller towns and, due to progressive taxation, the
share of taxation is also lower. This also reflects as a slightly higher share of social
benefits. These are actually the main components of disposable income. In those
days, Finnish working families paid or received hardly any property income
households had only some interest payments due to loans. This also indicates that
these families did not have any financial wealth.
Families Meika
¨la
¨inen and Virtanen received most of their income from wages
and salaries. As mentioned earlier, family Meika
¨la
¨inen’s wife was housewife but she
32 I.K. Kavonius
Table 2. Household income according to the estimations based on national accounts and cost of living study estimations.
National accounts Cost of living study
195051 Total Group I Group II Group III
Code Transaction Uses Resources Uses Resources Uses Resources Uses Resources Uses Resources
D11 Wages and saleries 0 69.16 0 115.17 0 115.14 0 115.86 0 113.69
D12 Employers’social contributions 0 6.92 0 0 0 0 0 0 0 0
D41 Interest 0.15 0.05 0.04 0 0.02 0 0.05 0 0.06 0
D421 Dividends 0 2.18 0 0 0 0 0 0 0 0
D44 Property income attributed to insurance
policy holders
0 0 0.01 0 0 0 0 0.01 0 0.01
D45 Rents 0 2.08 0 0 0 0 0 0 0 0
Entrepreunerial Income 0 31.86 0 0 0 0 0 0 0 0
D5 Current taxes on income, wealth etc. 16.64 0 17.69 0 17.51 0 18.46 0 16.39 0
D6111 Employers’social contributions 6.92 0 0 0 0 0 0 0 0 0
D6112 Employees’social contributions 0.85 0 0.29 0 0.27 0 0.23 0 0.48 0
D62 Social benefits other than social
transfers in kind
0 7.12 0 3.6 0 3.53 0 3.52 0 3.93
D71 Net non-life insurance premiums 0 0 0.09 0 0.05 0 0.11 0 0.15 0
D72 Non-life insurance claims 0 0 0 0.08 0 0.04 0 0.09 0 0.13
D75 Miscellaneous current transfers to NPISHs 0 0 0.75 0 0 0.64 0 0.69 0 0.75
D75X Miscellaneous current transfers within
households
0 0 2.19 2.19 2.49 2.49 2.22 2.22 1.43 1.43
D75Y Miscellaneous current transfers from/to
rest of the world
0.03 0.13 0 0 0 0 0 0 0 0
D75Z Miscellaneous current transfers from/to
non-financial corporations
0 0.22 0 0 0 0 0 0 0 0
Notes: Income is indicated as in relation to the estimated disposable income (disposable income100). In the case of consumption survey, Group I refers to Helsinki,
Group II refers to middlelarge towns and Group III refers to small towns. The left hand column indicates the corresponding SNA93 code.
Source: Author’s estimations.
Scandinavian Economic History Review 33
worked occasionally. However, the working was unofficial and she did not pay taxes
on the income. Both families received an allowance for children as a social benefit.
In addition, current transfers between households represent quite a large part of
the income. These are financial help and transfers in kind from one household to
another. At the aggregate level, these incomes do not affect the disposable income as
they net out in the aggregate. At household level, however, this effects the disposable
income.
31
Family Meika
¨la
¨inen also received direct financial support from their
parents. Additionally, both families occasionally got potatoes, vegetables and berries
from the countryside.
In those days households had some current transfers to NPISHs. These were
trade union fees and taxes to either the Evangelical-Lutheran Church or the
Orthodox Church. Families Meika
¨la
¨inen and Virtanen were members of both a
trade union as well as the Evangelical-Lutheran Church.
As mentioned earlier, the national accounts estimation covers the whole
population. As the national accounts also include entrepreneurial income, the share
of wages and salaries is also lower than in the estimation is based on the cost of living
survey. Additionally, as mentioned earlier, NPISHs are included in the accounts
based on the national accounts. As some of the NPISHs are foundations and other
types of institution which own financial assets, the national accounts’estimation also
includes some property income.
Household consumption expenditures
Table 4 shows the breakdown of household consumption expenditure. This is
presented as shares of the total consumption. Table 3 shows a comparison between
the national accounts results and the cost of living survey results. The comparison
between the national accounts and cost of living study results show that in the
national accounts the share of food and beverages is approximately eight percentage
points lower than the results based on the cost of living survey. As NPISHs and
Table 3. Household consumption expenditure as percentage of total household consumption
expenditure (100); a comparison between the national account estimation and the cost of
living survey results.
National
accounts
Cost of living
survey
COICOP 195051 195051
1 Food, beverages and tobacco 42.58 50.34
2 Clothing and footwear 21.40 17.90
3 Housing, water, electricity, gas and other fuels 10.02 9.15
4 Furnishing, household equipment and routine
household maintenance
8.17 4.98
510.1 Health and personal care 2.93 4.15
7 Recreation and culture 5.95 1.09
10.210.6
89
Miscellaneous goods and services, education and
restaurants and hotels
8.95 12.39
Note: The left hand column indicates the corresponding COICOP code.
Source: Author’s estimations.
34 I.K. Kavonius
entrepreneurs are included in the national accounts estimation, the share of food and
beverages and health care is typically lower and items like furnishing and fuels, which
are typically used as intermediate consumption, are higher than in the statistics, which
describe the household consumption as such. Nowadays this intermediate consump-
tion can be separated from the private consumption, but in the 1950s and 1960s
statistical sources were not so developed that the separation could have been made.
As can be seen in Table 4, consumption on food and beverages represent
approximately 50% of total consumption. In large towns the share of food and
beverages were slightly less than in smaller towns. It is typical that the wealthier the
households, the smaller their consumption relative share of food and beverages.
Naturally, households prioritise consumption around the necessities (e.g. food). When
income increases they have the possibility to choose whether they consume some non-
necessities or luxuries or whether they want to save it. From this point of view, saving
can be seen as an alternative for the consumption to non-necessities. When we discuss
the patience to save, it is important to look at the structure of consumption and to try
to analyse whether households have the possibility to save. Slightly less than 1% of the
consumption went on alcoholic beverages and between 2% and 3% of consumption
went on cigarettes. However, it is arguable whether these products can be classified as
non-necessities because these products cause physical dependence.
Almost 70% of our example family Meika
¨la
¨inen’s consumption went on food and
beverages. The consumed food was typical for the working class family at that time.
The family bought pork, black bread or French bread, Baltic herring, coffee and
sausages. In the consumption booklet it was separately indicated that the family
bought ‘cheap sausages’.
32
This is an indication of a certain type of sausage but it is
also cheap as its name indicates. In the other consumption booklets this type of
sausage appears very often. Family Meika
¨la
¨inen rarely had money for ‘accessory’
consumption. Sometimes the consumption booklet indicates that the family has
bought Finnish plain coffee bread
33
or salmiak.
The food consumption of family Virtanen was slightly exceptional. A large share
of their expenditure, approximately 45%, went on food and beverages. A large share
of expenditure went on basic groceries as in the case of family Meika
¨la
¨inen.
However, sometimes the family bought luxurious products like beef and imported
fruits like oranges. Additionally, the family bought Finnish plain coffee bread, cakes
or sweets several times a week. In other families, consumption of such products was
rare, and from this point of view this can be considered to be luxurious consumption.
Families Meika
¨la
¨inen and Virtanen rarely consumed alcohol. Before some
national holidays like Christmas, they bought small amounts of alcohol. The
pattern of alcohol consumption can be quite easily followed from the consumption
booklets. In some families it can be observed that at the end of the week a bottle of
alcoholic beverage was bought, and in some families alcohol was just bought around
national holidays.
The husbands of families Meika
¨la
¨inen and Virtanen were most likely smokers.
Family Meika
¨la
¨inen consumed almost two packages of cigarettes a day and that
corresponded almost 6% of the total consumption. The husband of family Virtanen
was not so addicted to smoking, and his consumption corresponded to around 3% of
the total consumption. Especially in the case of family Meika
¨la
¨inen, this was one of
the rare items which were non-necessary consumption. One of the few ways of
getting out of the everyday life was to have a cup of coffee and a cigarette.
Scandinavian Economic History Review 35
Table 4. Household consumption expenditure as percentage of total household consumption expenditure (100) according to the estimations based
on the household cost of living survey.
Cost of Living Study
COICOP 195051 Group I Group II Group III
1 Food, beverages and tobacco 50.34 49.1 50.55 50.74
1.1 Food 46.91 45.31 47.59 47.2
1.2 Beverages (alcohol) 0.89 1.03 0.78 0.78
1.3 Tobacco 2.54 2.76 2.18 2.76
2 Clothing and footwear 17.9 18.82 18.36 18.08
2.1 Clothing 15.04 16.05 15.49 15.15
2.2 Footwear 2.86 2.77 2.86 2.93
3 Housing, water, electricity, gas and other fuels 9.15 7.85 9.36 11.47
3.1 Actual rentals for housing 5.9 5.14 5.77 7.8
3.4 Electricity, gas and other fuels 3.25 2.71 3.59 3.67
4 Furnishing, household equipment and routine household
maintenance
4.98 5.37 4.43 5.12
4.1 Furniture and furnishing, carpets and other floor coverings 3.62 3.96 3.23 3.55
4.4 Glassware, tableware and household utensils 1.36 1.4 1.21 1.57
5 Health 1.93 1.88 1.94 1.94
7 Recreation and culture 1.09 1.28 0.98 0.82
8 Education 2.12 2.01 2.18 2.18
9 Restaurants and hotels 1.95 2.72 1.43 1.19
9.1 Catering services 1.95 2.72 1.43 1.19
10 Miscellaneous goods and services 10.54 10.97 10.78 8.47
10.1 Personal care 2.22 2.26 2.22 2.05
10.210.6 Other goods and services not elsewhere classified 8.32 8.71 8.56 6.43
Notes: In the case of consumption survey, Group I refers to Helsinki, Group II refers to middlelarge towns and Group III refers to small towns. The left hand column
indicates the corresponding COICOP code.
Source: Author’s estimations.
36 I.K. Kavonius
The average Finnish working family consumed approximately 18% of their total
consumption on clothing and footwear. Furniture and household equipment made
up for around 5%. In Helsinki, the share was slightly higher than in other parts of
Finland. Our example families spent slightly less on clothes. Family Meika
¨la
¨inen
bought clothes only when they really needed them and spent approximately 8% on
clothes. Family Virtanen spent around 10% on clothes. They mostly bought clothes
for their child. Both families also made clothes themselves. Actually, this was
apparent in many Finnish families, as many families spent money on skein, wool and
other materials. Our example families also spent less on furniture than an average
family. Family Virtanen spent around 4% on furniture.
As mentioned earlier, rents were strictly controlled in those days in Finland.
Therefore, households spent approximately 6% on rents and around 3% on gas and
electricity. The control also affected the relative shares of the rents. As in Helsinki,
the level of received income was higher, the share of the rents was lower: around 5%.
In smaller towns, households received less income, and thus their consumption
expenditures were lower. Therefore, the share of rents was higher.
Family Meika
¨la
¨inen spent approximately 8% of their income on rent and
electricity, which represents the average consumption in Helsinki. To
¨o
¨lo
¨, where
family Virtanen lived, has always been an expensive area, and therefore the share of
rents and electricity was around 10% of the total expenditure, even though the
apartment was small.
The expenditure share of healthcare was around 2%. Family Meika
¨la
¨inen spent
less than this on heathcare. Family members needed to visit a medical doctor a
couple of times during the year. Additionally, they bought aspirin twice. Family
Virtanen’s daughter was often ill, and therefore their medical expenses were slightly
higher than average.
Both families also used miscellaneous goods and services. Overall, households
were spending around 10% on these kinds of miscellaneous goods and services. Both
families needed to use washing machines outside their apartment, as they did not
have their own washing machine. In those days it was typical to go public saunas as
very few apartments had own saunas. That was done also by families Meika
¨la
¨inen
and Virtanen.
Additionally, family Virtanen bought building materials. This indicates that they
had either a building site, which was given for free to war veterans, or an allotment
garden. Theoretically, these should be included in investments, but as these were
quite small they were accounted as a part of consumption.
34
This type of investment
or consumption is an indication of saving, because investing is not logically possible
without saving. There are indications of this kind of consumption, investment in
durable goods or investment in very few survey booklets. This supports the overall
view of this article, that in the beginning of 1950s ‘normal Finnish families’did not
have enough money to save. Also, different treatment of durable goods would not
have changed this result considerably as Finnish families did not invest much in
durable goods.
Averages families were spending approximately 1% on recreation and culture.
Restaurant and hotel bills represented around 2% of total consumption expenditure.
In Helsinki, the proportion was higher than in small towns. Family Meika
¨la
¨inen
spent very little money on these kinds of services. The family went to movies only a
couple of times during the whole year. Family Virtanen spent around 3% on movies,
Scandinavian Economic History Review 37
prom or watching a car race in Helsinki. They also had very few restaurant bills.
Overall, restaurant bills in the families consisted either of a few beers in a local bar or
working place canteen bills.
Conclusions
In this article, institutional sector accounts for households are estimated. The results
of these accounts can be criticised because they do not include all the households.
However, these households can be assumed to represent ‘normal consumption and
saving behaviour’. There were few rich elite in Finland, and therefore they would not
definitely have changed the overall picture of saving. The people who lived alone
without work certainly did not do better than the population captured in this
analysis, and therefore they definitely did not have any possibility of saving.
Therefore, this paper captures the large mass which would theoretically have the
possibility of influencing the aggregate saving. The weakness of this data is that it
excludes the people who lived in the countryside.
The beginning of this article questioned whether Finnish households had the
patience to save. The answer is that they did not have the possibility of saving. The
normal middle class in the beginning of 1950s was still poor and all the money which
was received was also consumed. The financial markets were also controlled, and
therefore it was difficult to get a loan or mortgage. There are some indications of
saving or actually investing, as materials for building houses or allotment gardens
were bought, but to a larger extent any investment could not be seen in this analysis.
In the beginning of the 1950s, the Finnish households were living a very simple
life. They lived in small apartments, ate basic food and spent hardly any money on
entertainment. Every now and then, a little bit of change to the everyday life brought
a cup of coffee, cigarettes, Finnish plane coffee bread or a few beers or other
alcoholic beverages. A small difference between Helsinki, middlelarge towns and
small towns can be seen in the sense that in larger towns people spent slightly more
money on leisure activities. The salaries were slightly higher and it was easier to go to
the movies, for instance.
In order to invest, national economy needs to either save or borrow money from
abroad. As the financial markets were restricted, the latter was not possible to a large
extent. This article studied the households’possibility of saving and confirmed that
Finnish households had limited possibilities of saving in the beginning of the 1950s.
Therefore, President Kekkonen’s policy to involve the central government in the
industrialisation was virtually the only tool to industrialise the country. The idea was
that the government practically had the leading role in these projects, and it financed
the projects finally by collecting taxes.
Acknowledgements
I would like to thank the Finnish Cultural Foundation (Suomen Kultuurirahasto) for
financially supporting my research. Additionally, I would like to thank an anonymous referee
and the participants of the 3rd Sound Economic History Workshop, Gothenburg, Sweden, 34
October 2008, for helpful comments without implicating them in any remaining errors.
38 I.K. Kavonius
Notes
1. See, for instance, Urho Kekkonen, Onko maallamme malttia vaurastua? (Helsinki: Otava,
1952), 11619.
2. Please refer to Ru
¨tta Hjerppe (1988), Suomen talous 18601985, Kasvu ja rakennemuutos,
Kasvu-tutkimuksia XIII, Suomen Pankin julkaisuja, Helsinki 1988. And also to Ru
¨tta
Herjppe (1996), Finland’s Historical National Accounts 18601994. Calculation Methods
and Statistical Tables. Jyva
¨skyla
¨n yliopiston Suomen historian julkaisuja Z4, Jyva
¨skyla
¨
1996. There are also separate studies in which the national accounts framework has been
applied. For instance, an inputoutput framework has been applied to the historical data.
The last application of inputoutput tables is Jari Kauppila’s application for 1928 in The
Structure and Short-Term Development of Finnish Industries in the 1920s and 1930s An
Input-output Approach (Helsinki: Statistics Finland Research Reports 246, Statistics
Finland, 2007). Additionally, Eino H. Laurila, Consumption in Finnish Economy (Helsinki:
ETLA, 1985), made some historical private consumption estimations.
3. Peter Vikstro
¨m, The Big Picture A Historical National Accounts Approach to Growth,
Structural Change and Income Distribution in Sweden 18701990. Umea
˚Studies in
Economic History nr 26/2002 (Umea˚: Umea˚University, 2002).
4. The coverage of the accounts in this article do not correspond with the growth studies and
thus are not consistent with the ones in growth studies, i.e. these accounts are in this sense
independent and the accounts have not been reconciled with the growth studies.
Conceptually, the accounts are consistent as both follow either the SNA93 or the
SNA68 except in the cases in which exceptions have been explicitly mentioned; System of
National Accounts 1993 (SNA93) (Commission of the Brussels/Luxembourg, New York,
Paris, Washington DC: European Communities, International Monetary Fund, United
Nations, World Bank, 1993).
5. In the theoretical literature, the used income concept is called ‘Hicksian Income’; John R.
Hicks, Value and Capital: An Inquiry into Some Fundamental Principles of Economic
Theory (Oxford: Clarendon Press, 1939). The SNA93 recognises the linkage to the income
concept of Hicks, although it is arguable whether the SNA93 disposable income is the
Hicksian income, as Hicks included, for instance, holding gains in his income concept.
From that point of view the income concept is nearer to the one of James E. Meade and
Richard Stone, ‘The Construction of Tables of National Income, Expenditure, Savings
and Investment’,The Economic Journal 51, no. 202/203 (1941): 21633.
The overall national accounting system is based on the Keynesian theory; John Maynard
Keynes, The General Theory on Employment, Interest and Money (London: Macmillan,
1936). From that point of view, this paper estimates the private consumption component
and compensation of employees of the Keynesian macroeconomic equilibrium. According
to the Keynesian theory, the GDP can be defined as follows: GDP(B1G) OP(P1)
IC(P2)TAX(D21)SUBP(D31)CE(D1)OS(B2N/B3N)TAPRI(D2) SUB(D3)
CFC(K1)C(P31)I(P5)G(P32)EXP(P6)IMP(P7) , where OP stands for output,
IC for intermediate consumption, TAX for taxes on products, SUBP for subsidies on
products, CE for compensation of employees, OS for net operating surplus (mixed
income), TAXPRI for taxes on production and imports, SUB for the respective subsidies,
CFC for consumption of fixed capital, C for private consumption, I for investment, G for
government consumption, EXP for export and IMP for import. The codes in brackets are
referring to the standard codes of the SNA93 and the ESA95, (see European System of
Accounts 1995 (ESA95), European Commission Regulation number 2223/1996 (Eurostat,
1996)). The last part of the presented formula is the Keynesian macroeconomic
equilibrium, which is known as goods and services accounts in the national accounts.
6. The standard SNA abbreviations have been presented in brackets. The U and R are,
however, not standard abbreviations. U refers to ‘uses’and R refers to ‘resources’.
7. The full sequence of accounts for households is presented in the SNA93, table A.V.6.
8. See more in detail: SNA93, 1.1314 and 4.14.165.
9. Due to the source data, this is also the case in the total economy accounts, which are
presented later in this article.
10. See more in detail: SNA93, 4.11, 4.132160, 19.1729.
Scandinavian Economic History Review 39
11. More on this issue: SNA93, 9.389.40; Jukka Jalava and Ilja Kristian Kavonius, The Effect
of Durable Goods and ICT on Euro Area Productivity Growth? (European Central Bank,
Working Paper Series No. 940, September 2008); and Jukka Jalava and Ilja Kristian
Kavonius, ‘Measuring the Stock of Consumer Durables and its Implications for Euro
Area Saving Rations’,Review of Income and Wealth 55, no. 1 (2009): 4356. Additionally,
C.R. Hulten, ‘The ‘‘Architecture’’ of Capital Accounting: Basic Design Principles’,in
A New Architecture for the US National Accounts, NBER Studies in Income and Wealth,
vol. 66, ed. D.W. Jorgenson, J.S. Landefeld and W.D. Nordhaus (Chicago and London:
University of Chicago Press, 2006),; and OECD, OECD Manual: Measuring Capital, draft
version 30 May 2008’(Paris: OECD, 2008) support the view that durable goods should be
included in the investments.
12. This would also raise a juristic problem as indicated in Statistics Finland, Minutes of the
Committee on the Ethics of Statistics (Tilastoeettinen lautakunta) 3/2007.
13. Elinkustannustutkimus helmikuu 1950tammikuu 1951 (Cost of Living Study from
February 1950 to January 1951), macro dataset, file number I, the Archive of Statistics
Finland. (The data have been moved in 2008 to the National Archive.)
14. However, as required in the minutes of the Committee on the Ethics of Statistics
(Tilastoeettinen lautakunta) 3/2007, the description has been changed in that sense that the
families cannot be recognised.
15. Sosiaalisia erikoistutkimuksia 21, elinkustannustutkimus helmikuu 1950tammikuu 1951
(Cost of Living Study from February 1950 to January 1951) (Helsinki: Suomen virallinen
tilasto, 1954).
16. Susan Porter Benson, Household Accounts Working-Class Family Economies in the
Interwar United States (Ithaca, NY: Cornell University Press, 2007) has done a similar
kind of analysis on households’standard of living in the 1920s and 1930s US which is
based on household interviews.
17. Hannu Tanninen and Matti Tuomala, ‘Tuloerot ja makrotaloustiede’,inKansantalous-
tiede, talouspolitiikka ja hyvinvointivaltio juhlakirja Jukka Pekkarisen kunniaksi, ed. Visa
Heinonen, Markus Ja
¨ntti ja Juhana Vartiainen Helsinki: Palkansaajien tutkimuslaitos,
reports number 11, 2007):15969.
18. Riitta Hjerppe, Kasvun Vuosisata (Helsinki: VAPK kustannus, 1990).
19. Cost of Living Study from February 1950 to January 1951, Sosiaalisia erikoistutkimuksia
21, elinkustannustutkimus helmikuu 1950 tammikuu 1951 (Helsinki: Suomen virallinen
tilasto, 1954).
20. The middlelarge towns in the sample are Turku, Tampere, Lahti, Pori, Oulu, Kuopio and
Kotka. Kokkola, Kajaani, Lappeenranta, Karkkila, Suolahti, Valkeakoski and Rovaniemi
are representing small towns and market towns.
21. The economic growth and the increase of the welfare were rapid at those days. For this
reason, a new cost of living study was already conducted in 195556.
22. Sosiaalisia erikoistutkimuksia 21, elinkustannustutkimus helmikuu 1950tammikuu 1951
(Cost of Living Study from February 1950 to January 1951) (Helsinki: Suomen virallinen
tilasto, 1954), 9.
23. Suomen kansantalouden tilinpito 19481964 (National Accounting in Finland 19481964).
Tilastollisia tiedonantoja no. 43 (Helsinki: Tilastollinen pa
¨a
¨toimisto, 1968).
24. The bridge table has not been added in this article as it does not differ so much from the
one presented in the appendix. The table can be obtained on request from the author.
25. Sosiaalisia erikoistutkimuksia 21,elinkustannustutkimus helmikuu 1950tammikuu 1951,
15.
26. The name of the family is invented and does not appear in the data.
27. Arabia is a Finnish ceramics company which is currently owned by the Iittala Group.
Since 2007, Iittala Group is 97% owned by Fiskars Corporation, while its operative
management holds 3% of the shares. In the 1950s, Arabia used to be an independent
company. Arabia has specialised in kitchenware and tableware, china and sanitary
porcelain.
28. Porter Benson, Household Accounts.
29. The name of this family is invented and does not appear in the data.
40 I.K. Kavonius
30. In Finnish: Huoltokonttori. Helsinki City has a financial services office for its employees. It
provides instance credit with low interest rates. As in the 1950s financial markets were
controlled, the financial service office played an important role in financing its employees.
31. See more on differences between macro and micro approaches for instance: Ilja Kristian
Kavonius and Veli-Matti To
¨rma
¨lehto, ‘Household Income Aggregates in Micro and
Macro Statistics’,Statistical Journal of the United Nations Economic Commission for
Europe 20, No. 1 (2003): 925; Expert Group on Household Income Statistics, The
Canberra Group: Final Report and Recommendations (Ottawa, 2001).
32. In Finnish: halpamakkara.
33. In Finnish: pulla
34. For instance, the ESA95 has a threshold of ECU 500 at 1995 prices. This is an example of
the borderline case of durable goods and investment as referred earlier in this article.
Table A1. The bridge between the SNA93 income items and the consumption survey income
items.
Code SNA93 transaction Consumption survey item
D11 Wages and salaries Permanent labour income of
husband
Extra labour income of husband
Labour income
Labour income of children
Other income
D41 Interest (uses) Other expenses of which: interest
D44 Property income attributed to insurance policy
holders
Estimated to be 10% of net
non-insurance premiums
D5 Current taxes on income, wealth, etc. Taxes levied when income has
received
Central government taxes levied after
income has received
Local government taxes levied after
income has received
Other taxes (not related to
consumption)
D6112 Employees’social contributions Pension, sick and other funds fees
D62 Social benefits other than social transfers in
kind
Child benefit and other social
benefits
D71 Net non-life insurance premiums Other insurance premiums
D72 Non-life insurance claims Estimated to be 90% of net
non-insurance premiums
D75 Miscellaneous current transfers to non-profit
institutions serving households
Taxes of the church
Membership fees including trade
union fees
D75X Miscellaneous current transfers within
households
Other expenses of which: presents
and financial assistance
Appendix 1
Scandinavian Economic History Review 41
Table A2. The bridge between the SNA93 expenditure items and the consumption survey
expenditure items.
Code COICOP Consumption survey item
1 Food and beverages and tobacco 1.11.21.3
1.1 Food Food/nutriment
1.2 Beverages (alcohol) Alcohol beverages
1.3 Tobacco Tobacco
2 Clothing and footwear 2.12.2
2.1 Clothing Clothing
2.2 Footwear Footwear
3 Housing, water, electricity, gas and other fuels 3.13.4
3.1 Actual rentals for housing Rents
3.4 Electricity, gas and other fuels Heating and electricity
4 Furnishing, household equipment and routine
household maintenance
4.14.4
4.1 Furniture and furnishing, carpets and other floor
coverings
Furniture and households
textiles
4.4 Glassware, tableware and household utensils Household supplies
5 Health Nursing
7 Recreation and culture Entertainment
8 Education Learning and reading
9 Restaurants and hotels 9.1
9.1 Catering services Meals outside home
10 Miscellaneous goods and services 10.110.210.6
10.1 Personal care Cleaning and tidiness
10.2
10.6
Other goods and services not elsewhere classified Housemaid, cleaning
person, etc.
Other expenses
Appendix 2
42 I.K. Kavonius
