Policy Sciences 14 (1982) 117-135 l 17 Elsevier Science Publishing Company, Amsterdam Printed in the Netherlands

The Social Policy of the Firm* MARTIN REIN

Department o f Urban Studies and Planning Massachusetts Institute of Technology. Cambridge, MA 02139. U.S. ,4.

A B S T R A C T

The "welfare state" concept hides an important aspect of modern industrial societies. In capitalist countries welfare is provided through a mixture of public and private initiatives. The author suggests that the concept "welfare economy" more fully captures the economic interpenetration of public and private sectors. The growth of fringe benefits illustrates the extent to which private enterprise performs the welfare function.

Government increasingly intervenes through processes of mandating, stimulating, regulating, and supporting, using private enterprise as the vehicle for delivery of welfare services. Government's traditionally conceived role as welfare service provider is also changed through recognition that it is both an employer and purchaser, significantly impacting society's original income distribution.

Discussions of the welfare funct ions of the state mistake the nature of modern

industr ia l societies. Dist inctions are frequently made between the 'publ ic ' and the

'private, ' between the 'social ' and the 'economic, ' and between an 'original d is t r ibut ion '

of incomes and the post- t ransfer d is t r ibut ion produced by government welfare policy.

None of these dist inctions fit reality.

Conven t iona l analysis treats the economy as being composed of two sub-systems:

the private sector, where the market and the voluntary sector operate, and the public

sector, where the activities of government are carried out. This view fails to take into

account the subs tant ia l role of gove rnmen t as employer, the role of government as

consumer , or the role of government in mandat ing , st imulating, suppor t ing and

regulat ing private enterprise.

Conven t iona l conceptual iza t ions treat wages as an economic re turn for labor, and

sharply dist inguish such economic compensa t ion from transfers in the welfare system.

* An earlier version of this essay was presented as a lecture at the University of Melbourne and published in Ronald F. Henderson, The Welfare Stakes." Strategies for Australian Social Policy, Melbourne: Institute of ,Applied Economics and Social Research, 1981.

0032 2687:'82,' 0000 0000' $02.75 �9 1982 Elsevier Scientific Publishing Company

l l8

However, fringe benefits overlap these categories, directing attention to the welfare

components of wages and the economic functions of welfare.

Conventional economics treats the economy as generating 'an original' distribution

of income, through the operation of the market principles of supply and demand and marginal productivity, and a redistribution of income through government's active

intervention to realize social objectives. This conceptualization fails to recognize that

the private sector also functions to realize social objectives. It also fails to take account of government's role in shaping 'original income'.

This paper proposes to explore these issues in the conceptualization of modern

industrial societies through an inquiry into the phenomenon of fringe benefits.

Although we conventionally distinguish between 'core' income called wages, inter-

preted as a return for productive labor, and 'fringes' which do not directly relate to or

affect production (Reid and Robertson, 1965) in practice the boundary between wages

and fringes proves to be fuzzy. Many fringes are convertible back to cash wages. In

many countries, sick pay, for example, is often both cumulative over time and

convertible into a lump sum payment at the end of a year or upon termination of employment. Furthermore, wages themselves include an element of fringe.

If we press the distinction between wages or salaries and occupational welfare which

is on "the fringe" of the core return of wages (Reid, 1961) we see that wages are set not

only in terms of compensation for productivity, but also in a welfare support context,

where wages are linked to family need. For example, the Australian Arbitration Commission in 1911 established a convention for thinking about wages as the income needed to support a family of four persons. The Commission determined that women

needed only 54% of men's wages. They were considered single persons who did not

have to support a family, thus implying that almost half the wages should be

interpreted as a welfare fringe [1]. We also conventionally distinguish between 'welfare' which is provided by govern-

ment and 'fringes' which are provided by firms as a supplement to compensation from

work. But this distinction also proves to be fuzzy. Government is an employer, providing wages and fringe benefits, as well as a provider of welfare. About one fifth of the labor force in the United States works directly for government. Its indirect

influence on the 'original' income distribution is even greater. To suggest the magnitude of its influence, Thurow (198 i) points out that government in the United

States, directly and indirectly through purchase of goods and services produced by private industry, employs two thirds of the female professional workers. In general, Thurow believes that government is basically a producer of middle and upper middle class jobs. Hence, the state's action sharply influences its 'original' income distribu- tion.

There is no good, direct quantitative estimate of the extent of comingling of the public-private sectors. The degree of penetration of fringes into the core of wages and salaries for productive labor remains elusive. Governments have only recently begun to take interest in the range of employment-related benefits which complement,

l I9

duplicate or obviate the necessity for public transfers or services. For example, federal

employer wage rates are set by surveys which compare public and private com-

pensations. The government has been evaluating the desirability of shifting from pay comparabil i ty to total compensation comparabili ty (i.e., wages plus fringes) because it

believes that total costs will be reduced by such a change. By taking these social

benefits, whether provided by the public or the private sector, as the unit of our

analysis, we redefine the role of 'welfare' in industrial society. This more com-

prehensive definition of social protection calls attention to 'those policy decisions

whose implementation costs are carried by the private sector of the economy, and are not explicitly identified in the national accounts' (Klein, 1978).

It is widely recognized that the countries of Western Europe and North America

have changed over the past centuries from where economic welfare (security and

adequacy of income) was primarily a function of a market economy and money wages,

to where an elaborate range of both public and private institutions serve as intermediaries between the economy and the family to insure particular levels of

economic welfare throughout the life-cycle. The scope of this range can be grasped

from National Income Accounts which treat the whole economy as if it were a single household and explore where this ageless household (aggregated across the life cycle)

acquires its income. The results of this approach are fascinating because they provide a

very different picture of the economic resources of the whole society than does the

examination of cash incomes of households in the active labor market. After taxes, wages and salaries account for no more than one third, and in many countries only one

quarter, of the economic resources of the household sector. The rest is accounted for by public and private transfers and services plus income derived from household

savings [2].

Most analysts who have attempted to classify these mixed public-private institu-

tions accept the distinction developed by Titmuss in his famous essay on the social

division of welfare. Titmuss (1958) distinguishes between the fiscal and social welfare

activities of the state and the occupational welfare of the firm. His important insight

was the recognition that the social policy implicit in the indirect system of tax forgive-

ness which encourages social welfare activities of the firm was very similar to the system of direct distribution of government goods, services and transfers provided in the public sector. He insisted that similar activities should be grouped together, because fiscal

and occupational welfare serve the middle class and thus undermine the redistributive effects of direct public welfare programs. In this framework, 'occupational welfare'

(i.e., fringe benefits) signifies those activities which arise from the initiatives and

interests of the private market, and which threaten redistributive aims. This reclassification of occupational, social and fiscal welfare is helpful, but it fails

to give sufficient attention to the extent to which occupational welfare is directly brought about through the action of the state. For example, some countries have

developed a social contract between the state, the firm and the workers where the state trades off fringes for wages in the hope of reducing the rate of growth of inflation.

120

State initiative is crucial in understanding the origin and development of a wide range of employee benefits, including paid vacation, sick pay, disability, retirement,

maternity leave, etc. Management may turn to fringes as a way of augmenting the real

value of employees' wages, because the fringe is a super dollar, worth more than a

regular dollar, since it is exempt from taxation. So extensive is this active role of the state in shaping the welfare activities of the firm, and so extensive are the employee

benefits which arise f rom the state's role as employer rather than provider, that the

traditional distinction between the public and private welfare sectors is blurred. The

dichotomy between public and private social protection should be reconceptualized. We do not have adequate statistics to identify the full extent of the state's influence

in the private sector. However, the available evidence shows how the inclusion of private sector welfare dramatically alters our insights about the distribution of

economic welfare in industrial society. Enormous sums of money are involved: we are

not dealing with an esoteric and trivial aspect of modern economy. So vast is the state's

role as employer and as indirect and direct initiator of employee benefits that it is

necessary to think in terms of a welfare-economy to highlight the close integration

between economic and social policy, on the one hand, and the activities of the public and private sector, on the other hand.

The Welfare State Perspective

The welfare economy focus proposed here should be contrasted with the more conventional interpretation that social policy is carried out by the welfare state. There

is, of course, no universally accepted definition of the welfare state. A very broad

f ramework is proposed by Cazes (1979) who defines the welfare state as 'the sum total

of civilian governmental furnishings of services, or formulating norms of behavior,

and of providing transfer payments in order to increase the level of well-being of the

community as a whole or to change its distribution' [3]. His emphasis on consumption, investment, subsidies and transfers designed to modify behavior and alter the

distribution of income is much broader than many definitions of welfare as the provision of entitlement to insure minimum standards of health, housing, education

and income as a political right. It still does not include the non-budgetary operations of the state such as regulation and tax expenditures [4] and off-budget outlays for loan

guarantees, but these exclusions are widely understood and accepted. A commission of the European Economic Communi ty is now in the process of developing a measure of tax expenditures for all of its member countries. While there is no comparable

effort to measure the scope of governmental regulation, increasing interest in the size of the public debt has contributed to an increased awareness of the importance of government loans.

At present there is no statistical series broad enough to capture the meaning of 'welfare' and the 'state' which is implied in the concept of the welfare state [5]. However, despite the limits of existing measures, published statistics do provide

TABLE 1

r State Expenditures in 1974-1976"

Country Public as Social transfers Social transfers Welfare state Social protection accounts in % of GNP % of public budget % of GNP % of GNP EEC** and U.S.. % of GNP.

1977

12l

France 38.7 46.0 17.8 27.1 23.9 Italy 39.3 43.8 17.2 28.8 23.1 Belgium 39r5 40.6 16.0 28.7 25.1 Netherlands 50.5 46.5 23.5 37.7 28.8 Norwa 3 44.8 34.9 15.5 28.7 Austria 38.0 33.2 12.8 27.5 Germany 41.5 34.5 14.2 31.1 27.4 Sweden 49.0 33.5 16.4 36.7 Denmark 44.0 30.0 13.2 33.8 25.3 U.K. 44.6 24.3 10.8 26. I 19.7 Switzerland (33.5) 33.5 8.8 22.9 U.S. (35.1) 33.4 11.9 30.0 16.0

* Three 3'ear average 1974-1976 ** EEC: European Economic Community

Source: For Social Protection Accounts in EEC countries: Les Comptes de la Protection Soeiak Methodes et Series 1959 1978, Paris. Institut National de la Statistique et des Etudes Economiques (INSEE), 1979. For the United States: U.S. Office Management and Budget. The Budget o f the Egfited States Government, Fiscal Year 1980. Appendix, (Washington, D.C.: U.S. Government Printing Office), 1980. For all other countries: Organization for Economic Cooperation and Development (OECD), Public Expenditure Trends, Paris, June, 1978.

information on expenditures for social transfers and other welfare state functions [6]. Table I sets out the available evidence about welfare expenditures for a number of

countries of the mid-1970s. When services, goods and transfers are added together, the size of the welfare state in Europe and in the United States accounts for between one quarter and more than a third of the GNP, and between 24 and 46 percent of the public budget. Transfers alone account for between one tenth and a quarter of the GNP.

One of the persistent ambiguities in the analysis of welfare state expenditures is the treatment of transfers, services and goods generated in the public sector from government's role as employer [7]. When the government provides fringes for its employees, they are better interpreted as occupational pensions and should be grouped together with other occupational fringe benefits. It is misleading to treat the fringe benefits of public employees as part of the expenditures of the welfare state. How large is the occupational welfare generated by the government as employer rather than as provider? The figures in the United States are informative. In 1975, about 13% of all public cash transfers were for public employee retirement. These figures exclude veterans' benefits which amounted to some 14.5 billion in 1975, or 8% of public transfers. Disaggregating these figures by the level of government, we find that at the local level, more than a third of total transfers are accounted for by public employee retirement programs.

122

Another way to look at the figures is to compare the relative size of social security

with private and occupational pensions available to public employees. In 1975, 79

billion was spent on social security and Railway Retirement programs. About half that amount was spent on occupational sectors (11.2 billion for private pensions and

annuities, 12.6 billion for federal government and military retirement, and 4.7 billion

for state and local government employee retirement).

The Welfare Economy Perspective

As suggested earlier, we propose the concept of a welfare economy as a better

perspective for understanding the nature of industrial societies. If we shift our focus

from measuring the expenditures in the welfare state to an analysis of the varied

institutional forms by which society carries out its welfare function, we gain an

altogether different understanding of the pattern of welfare expenditures in Europe and in the United States. To highlight how the broader framework alters our

understanding of the present structure of social protection, we present a typology of

four different ways that the state exerts an active role in the welfare activities of the

private sector. These include the following: mandating, stimulating, regulating, and

supporting.

Mandating

Mandating is the procedure by which the state passes legislation which requires that

private enterprise carry out a particular social objective specified by the state.

Obviously, the state can mandate with or without providing direct financial reimbursement. Naturally, the firm itself may try to pass on the additional costs of the

mandating to the consumer in the form of higher prices. The public mandating of

private initiative in the field of welfare sets in motion a complicated chain of events in

which the firm tries to offset the costs of the required social programs. This makes it

difficult to trace in detail who in fact bears the cost of mandating. Perhaps an important reason that governments have been relying upon a mandating strategy is that it

effectively 'hides' public expenditures. But without speculating on the motives behind mandating, it seems clear that mandating is very extensively used by governments to

realize welfare objectives. In Europe, for example, it is very common for a governmeht to require that firms provide paid vacations for employees. The government may mandate procedures for hiring or firing workers, with the former being an aspect of the

government ' s affirmative action program and the latter being an attempt to reduce government unemployment benefits by preventing firms from firing or laying off

workers during recessionary periods. Governments may also actively mandate the payment of benefits. Two examples

help illustrate the point. In Sweden, national superannuation benefits are mandated by legislation which requires that the firm pay the total cost of wage related pensions.

123

Moreover, in the Swedish case, not only does the state mandate pensions, it also

controls the investments of the trust funds which are created by the legislation. In this

situation, mandating can be simply understood as a tax on the firm for resources

which are administered and controlled directly by the state. In other words, pensions in Sweden are privately financed and publicly controlled. A different pattern

prevails in the United States where the government mandates that only a portion of a

pension program be financed by employers, but the principle of mandating is similar

in both situations. A second example of mandating involves sick pay. Most countries

have a parallel system of sick benefits provided by the state and sick pay provided by

the firm. Sick pay can be won through collective bargaining, but it may also be

mandated by government. In West Germany in the early 1970s we find a rather dramatic and fascinating shift from a public to a private benefit. In 1973 new

legislation required that the first six weeks of sickness benefits be paid directly by the

firm; for longer periods of time the public sickness benefits system would assume

responsibility for the payment of benefits. This new legislation sharply reduced, at least for a while, the expenditures of the public sickness fund.

There is not a sharp line between benefits which arise through collective bargaining

between workers and managers and those mandated by the state. These processes

interplay. Unions may actively lobby governments to pass legislation which requires

firms to provide certain benefits, as well as bargain directly with individual firms or industries.

Regulating

Regulation involves the establishment of procedures for overseeing the activities of the

firm to ensure that they have a benign and beneficial effect on the workers' well-being.

The field of occupational safety and health is an illustration of the regulatory

approach. Firms are required to provide a safe working environment for their employees. Governments may also be concerned about the health of employees and

require that firms provide periodic medical check-ups [8]. The government may send

its own inspectors to determine compliance with these regulations or may empower unions to play the overseeing role.

Stimulating

Stimulating refers to the devices which the government uses to provide incentives for

firms to pursue public social policies and objectives. The most prevalent form for creating incentives is taxation policies designed to encourage private business to carry out activities which government regards as worthwhile and desirable. Of course, the government can use taxat ion to provide disincentives as well. When government relies upon stimulation, it pays part of the cost of the fringe in the form of foregone taxes. The field of manpower training is one example of stimulation through tax

124

incentives. In the United States and in many European countries, governments have tried, through a variety of tax incentives, to encourage enterprises to develop

apprenticeship programs, manpower training programs and employment programs

for special disadvantaged groups.

Supporting

Government can actively support a firm that is not economically viable to make sure

that the employees do not lose their jobs. The state can use various devices to promote

job security such as loans which permit the firm to continue to function or large scale purchase contracts to assure a market for the firm's product. The state's strategy of

support is particularly visible in the case of large firms, when it cannot politically

afford to let a large employer become bankrupt and displace thousands of workers f rom jobs. The U.S. government 's financial rescue of the Chrysler Corporation arose

from a recognition that Chrysler's value to the economy had to be understood not in

terms of the product it produced, because there was a glut of cars available on the

market, but in the income, jobs and security which Chrysler provided for workers.

When the state assumes a supporting role for private enterprise, the productive system

becomes converted into a distributional system, that is to say, we maintain the system

of producing cars so that it can function as a system for the distribution of income.

The management of systems of production to realize distributional aims has

recently become a political issue in Sweden. Through a Parliamentary act in 1969,

Sweden established the Statsf6retag, a holding company for more than thirty state run firms. These companies can be divided into two types: those that operate under normal

commercial conditions and in fact make a profit, and those that government classifies

as 'special programs' which require injections of state capital to keep them operating.

The major political reasons for maintaining these firms is the state's policy of

maintaining a certain level of employment. Both socialist and the current (since 1976) non-socialist government subsidize industries which are hit by competition from

developing countries, and by the recession of the mid-1970s [9].

Welfare in the Welfare Economy

We have so far criticized two of the critical assumptions of democratic capitalism: first, that we can meaningfully separate the public and the private sectors and treat them as two independent entities; and second, that within the private sector of the economy, the systems of production and distribution are best understood as separate, but interdependent. These two themes are related because our understanding of the politics and management of production is changed when we discover that the production system is driving the distributional system or that the production system is being used directly as the distributional system for maintaining jobs, security and economic well-being. The boundary separating the structure of wages viewed as

125

compensation for productive labor cannot be sharply distinguished from fringe

benefits directed at the welfare of the worker and his family.

The 'social protection accounts' of the European Economic Community (EEC)

provide data which integrate the actual and imputed social contributions of employers and individuals [10] with contributions of government. These statistical series

integrate private social benefits and state outlays. Data from social protection accounts (Table 1) show that welfare states in Western Europe look much more similar

than when one only takes into account those benefits that flow directly through the

public sector.

Another way of approaching the level of expenditure in the welfare economy is to

examine the structure of fringe benefits as they relate to total labor costs. We also want

to distinguish who initiates fringe benefits and why they are initiated. We can identify

three patterns: first, they can be mandated by the state. Here the public sector requires

that the enterprise sector provide fringes as required by law. Secondly, fringes can

arise as a result of collective bargaining. Here the union bargains for fringe benefits by negotiating in collective bargaining for their inclusion as part of compensation. Third,

the firm initiates the fringe on the assumption that it can improve its competitive

position in the industry in which it operates, perhaps through buying loyalty or

through an investment in the firm's human capital. In addition, the trust fund set aside

f rom a pension or retirement scheme often provides a source of capital which the firm can use for purposes of capital investment.

An examination of all fringe benefits expressed as a proportion of total labor costs

in the mid-1970s show the following patterns: high expenditure of 40% or more of

labor costs are found in West Germany, Belgium, France, Austria, Italy, and the

Netherlands; medium expenditure of about 25% of labor costs are found in the United

States and Sweden; low expenditures of 20% or less are found in Great Britain, Ireland and Denmark (see Table II).

A review of trends in expenditures for fringe benefits serves to highlight both the

growing importance of fringes and the variability across countries. Consider trends in West Germany, Britain and the United States. In West Germany from 1966 to 1977,

legally required employee benefits as a proport ion of wages increased by almost two

thirds, while other employee benefits increased by a little more than 40%, for an overall increase of 51%. In Britain, there has been an increased reliance on fringe

benefits, for both managers and manual workers. Statistics prepared for the Royal

Commission on Income and Wealth (1979) make this dramatically clear. In 1960, wages accounted for 85-90% of total labor costs for manual workers. But by 1977 wages accounted for only 67% of total labor costs. Fringe benefits for manual workers grew dramatically during this period from less than 10% to exactly one third of labor

costs. For non-manual workers, fringe benefits grew to 40% of total labor costs. While the British trends are not universal, they alert us to the importance of the development

of fringe benefits in industrial society. In the United States, the figures are equally dramatic. In 1966 wages for time

126

TABLE 2

Average Other Labor Costs as Percent of Total Labor Costs for Industrial Workers in Selected Countries

Country (year) Total other Time (vacations Shelter (other Consumption labor costs (%) and other paid legally required (other contractual

free time*) (%) personnel costs) and voluntary (%) costs) (%)

Italy (1972) 46.6 9.9 25.7 10.8 Austria (1975) 44.2 7.4 16.6 20.2 France (1972) 39.8 8.4 19.3 12.0 Netherlands (1972) 39.6 11.8 14.1 13.7 Germany (1977) 39.6 12.9 16.2 10.5 Belgium (1972) 39.3 12.3 19.5 7.3 Sweden (1975) 28.5 7.9 17.4 3. ! United States (1975)** 26.5 10.1 6.5 9.9 Great Britain (1973) 19.2 9.1 5.6 4.5 Ireland (1975) 18. I 7.4 8.3 2.5 Denmark (1975) 16.9 10.4 3.7 2.9

* Includes legally required minimum vacations in countries which have them (most European countries). Most of these costs are legally mandated. ** Large firms only. Total amount for all industrial workers probably lower, legally required proportion higher, and other two categories lower. Source: Sakowsky, Heinz, "Personalzusatzkosten als Wettbewerbsfaktor," Reht der Arbeit, Mai/Juni 1979.

worked were 83% of to ta l compensa t i on for the pr ivate non- fa rm por t ion of the

economy. Fr inge benefits accoun ted for 17% of to ta l compensa t ion . By 1979, pay for

the t ime worked decl ined to 75% and fringe benefits increased to 25% of the to ta l

compensa t ion . We can fur ther ident i fy which types of fr inge benefits grew dur ing this

per iod. Vaca t ion and hol idays increased by 32%; sick pay, bonuses and severance pay

by 4%; legally required social security, u n e m p l o y m e n t and workers ' c o m p e n s a t i o n

grew by 60%; payments for emp loye r insurance con t r ibu t ions for health, d isabi l i ty ,

accidents and life by 126%; and pensions, re t i rement and savings plans by 83%

(Smeeding, 1981).

Severa l facts emerge in our review of the welfare economy. Firs t , fringe benefits as a

percent of l a b o r costs are impress ively high and rising. Secondly , when these fringes

are a d d e d to the t ransfer paymen t we conven t iona l ly ident i fy as welfare state

expendi ture , the range of out lays a m o n g countr ies is much narrower . Third , the roles

of gove rnmen t as emp loye r and the role of pr ivate fringes are cri t ical in under s t and ing

welfare in the welfare economy.

Wha t precisely is the "welfare" c o m p o n e n t in the welfare economy? The g rowth of

non-waged componen t s of economic resources t h rough publ ic and pr ivate enterprises

in the fo rm of d i rec t ly -prov ided cash benefits and ind i rec t ly-provided transfer , goods

and services has two quite different side effects. Fr inges protect indiv iduals f rom the

127

uncertainties of the self-regulated market, producing a sheltered society for most of its

members. But the increase in shelter brings with it a foregoing of freedom of action as

the structure of consumption is restricted. By their very nature, earmarked fringe

benefits are resources which are set aside for the consumption of specific goods and

services like medical care and education. They are not unrestricted cash as are wages.

Thus, increased sheltering can also be combined with a broadened control over the individual. The more generous the fringe, the greater the risk of a reduction in the

scope of personal au tonomy and occupational mobility.

Democratic industrial societies have certainly created a welfare economy, made visible when we use a broader definition of welfare which includes the range of private

institutional forms that provide welfare in addition to those of the welfare state.

However, while the search for shelter may be inspired by communitar ian and

protective ideals such as job security and social protection against risk and

uncertainty, it also appears to have the perverse effect of multiplying inequalities in the

distribution of economic resources. The combination of economic sheltering and

equality varies across countries. In Sweden, they are tightly integrated and we can think of an egalitarian welfare society. In other countries, they are more loosely put

together, and the character of the society can be authoritarian, as was the Bismarkian

welfare state, or paternalistic, as the Japanese reliance on the social policy of the firm illustrates.

The continuing expansion of fringe benefits makes these large questions about the character of the industrial society increasingly important. We can identify three

factors which will contribute to the accelerated growth of these fringe benefits.

1. Inflation. As incomes rise to match the rising cost of living, more and more middle

and upper management personnel find that their income reaches levels of confiscatory

taxation. As a result, executives begin bargaining for, "and in many cases receiving a

variety of prerequisites as inflation eats away at their salaries" [11]. Since most fringe benefits are tax-exempt, their value is worth more than a dollar to the employee. The

fringe benefit becomes a"super-dol lar" because it is tax-exempt. The problem is more

complicated in the U.S. than elsewhere because of the Internal Revenue Service rules

which provide that if benefits are to be tax-deductible by the firm, they must apply to a

"broad cross-section" of employees. This implies that fringes must not discriminate in favor of highly-compensated employees. But it seems plausible that firms will try to

discover legally acceptable ways to re-introduce privilege and undermine the intent of

broad coverage. Attaching fringes to working conditions may be one such device. We might also speculate that in an inflationary period, government concern about

the rise in wage levels fuels inflationary pressure for increased fringe benefits. An implicit standard emerges which converts declining value of real wages into increased fringe benefits. Historically, fringe benefits developed during World War II when there was a freeze on wages. The fringe was thus a way of bypassing this system of wage controls.

2. Nationalbudgets. Many economists believe that the growth of public expenditure

128

inhibits investment and feeds inflation. High property taxes to support local welfare

state activities further fuel the anxiety about the growth of welfare state expenditures.

Yet, the resulting attack on the growth of the welfare state does not lead to a cut-back

in the demand for protection and shelter. What seems most likely is that restriction of national budgets, e.g. Reaganomics, will change the institutional arrangements

through which social protection is secured. One pattern is suggested by the German experience, with its expanded state mandate that private enterprise carry out the

state's social policy objectives. Of course, private employers will in turn try to find

ways of "passing through" to the consumer the increased labor costs associated with

mandating. Perhaps the government will try to block the pass-through by exerting

pressure on firms which increase the costs of their products. But whatever the final net distribution of costs and benefits, anxiety about the growth of the public sector will

contribute to the further expansion of enterprise-dominated social policy. 3. Changing work force composition. The gender composition of the labor force has

changed dramatically since World War II, with women now accounting for about

43% of the total labor force in the United States. The entry of women into the work force expands the demand for fringe benefits so that women can better accomodate

family life and work. This naturally gives rise to demands for maternity leave and day

care services. But an even more important fringe benefit is likely to be the redefinition

of time. The integration of work and family can best be served by redefining the time

boundaries of the job, thus giving rise to part-time work, flexi-time, time away from

work to care for a sick child, time to give birth to a baby, etc. Of course, under current practices when women enter the labor force on a part-time basis, they are ineligible for

other fringe benefits. Paradoxically, then, the use of time as a fringe benefit can

displace other fringe benefits. But as this trade-off becomes recognized as discri-

minatory, there will be mounting pressure to shelter and protect the part-time worker

from exclusion from entitlement to fringe benefits.

Mapping the Terrain

To understand the kind of"welfare" that exists within the welfare economy, we need to

examine fringe benefits from several different perspectives: what are they sub- stantively; their rationales and distributive effects; who controls them and how this

control changes over time. These questions are linked, each providing a different insight into the structure of fringes and the contribution they make to the welfare society. While these questions form a research agenda for further inquiry, still the

broad shape of some of the answers is already visible.

The Substance of Fringes

The task of classifying the types of fringes is made difficult because of a lack of definitional uniformity and the absence of single summary statistical series that

129

describes them. Together the different sources of information provide the basis of a

preliminary typology of fringes.

Shelter Against Insecurity: The Social Security Administration since 1950 prepares

annual statistics on the coverage, contributions and benefits under employee benefit

plans that arise from the employment relationship and are not paid for directly by

government. This American definition has been restricted to shelter against two forms of insecurity: loss of wages and the high costs of medical care. The series highlights the

income maintenance provided by employers during periods when regular earnings are

cut off because of death, accident, sickness, retirement, or unemployment. In addition to the maintenance of income, the series also covers medical expenses associated with

illness or injury, presumably under the assumption that no comprehensive public

program for medical care exists. Consequently, workmen's compensation and other forms of statutory provision for employers ' liability are not included. To a much lesser

extent, employers provide actual health care rather than financing medical bills. This

type of health care fringe, also not included in the data, apparently has not grown

extensively, though one can find it in specific industries, such as higher education. Yet,

even with its exclusions, the series shows that during the past quarter century, the expansion to benefit plans has been impressive. "Almost every type of employee

benefit plan registered coverage gains in the part quarter century that exceeded the

growth in the paid labor force." (Skolnick, 1976). Time Away from Productive Work: The Chamber of Commerce in the United

States commissions the Conference Board to carry out a periodic survey designed to

assist executives in thinking about the "leading edge" practices which have been

developed by industry in each benefit area. The survey was first begun in 1937 to track down personnel practices of large American firms. [12] The Chamber of Commerce in

its definition of fringe benefits includes most forms of time away from productive

work as a fringe benefit. The survey is limited to paid time away from the job and does

not include time off allowances during paid hours of work for rest periods, coffee

breaks, etc. However, data on vacations, including bonus salaries for vacations and

public holidays are included. The report notes that since 1971 the Federal government attempted to standardize

the system of celebrating holidays on a Monday in order to provide for a long

weekend. Four new holidays for Memorial Day, Veterans' Day, Washington and Columbus ' birthdays were instituted. Paid vacations continue to be tied to tenure on

the job. The two week vacation is conventionally accepted. However, the third and fourth week of vacation are typically available to employees who have been with the

company at least ten or more years. The cost of these holidays and vacations and other paid leave accounted in 1976 for

over 7.4% of the wage and salaries of the private non-farm economy (Smeeding, 1981, p.5). Pay for time worked as a proport ion of wage and salaries has fallen by about 2% between 1966 and 1979, largely because of the increased fraction of wages and salaries attributed to paid vacations and holidays, but also in part because of changing

130

definitions in eligibility criteria. Some countries, like Sweden, permit workers to take

sick leave when their children are sick. Also, as suggested earlier, with the changing

gender composition of the labor force, time as a fringe benefit is likely to become

increasingly important in the future [13]. Women have begun to win the security of a

job while on maternity leave. Still unresolved is the question of whether they should get paid when away from their job bearing children. At issue is the question of who

should bear the burden of responsibility: the parents, because they wanted the child;

the firm, because it responds to the family needs of its employees; or the state, because it must resolve the enduring question of the social reproduction of labor.

Consumption Subsidies: We can distinguish two types of consumption subsidies.

One type is universally available to all workers. These benefits include free or

subsidized meals, sports and recreation facilities, goods at discount prices, and other

benefits in kind. There is no statistical series on the value of these consumption

benefits in the United States. The study by the Royal Commission on the Distribution of Income and Wealth in Britain (1979) refers to them as welfare benefits and they"are

generally available to all employees in a firm. Their structure is often the result of

historical accident; their recent growth has been the result of negotiations between

employers and unions" (paragraph 9 : 20). That these benefits have a long history is illustrated by the example of the "truck" system which was introduced in the latter half

of the eighteenth century in Britain. It was designed to reduce real wages by forcing employees to take part of their reimbursement in kind or requiring that they spend

their income in the company store. The truck system has become transformed to its

modern equivalent product discounting for employers/employees.

The motives for introducing subsidized canteens and playing fields and social clubs

appears to be quite different than that of product subsidies. They seem more designed

to build morale, improve nutrition, make workers more physically able to carry out

their jobs, and prevent workers from leaving the building at lunch hour. In more

modern terminology, they were designed as an investment in human capital, i.e., a way of keeping the human machinery smoothly functioning. Such consumption subsidies

are important because they illustrate the fact that fringe benefits accommodate to the changing norms of society over time. What is particularly interesting is that these

benefits are not altogether dropped when they become historical anachronisms, but

become transformed into a modern equivalent.

A second type of consumption subsidy is much broader: it includes subsidized housing, assistance with housing purchase, educational benefits for workers and their children, low interest loans, bonuses and the use of the company car where its use is not directly related to the job. This differs from the first type in that benefits are restricted to a privileged few. The Royal Commission observes that "they are provided to attract and retain staff and have expanded considerably over the years of income policy to combat the narrowing of the margin of executive rewards caused by the tax structure and pay restraints" (paragraph 9: 21). The report points out that the proport ion of labor costs accounted for these payments has risen sharply from 10% in 1960 to 20% in

131

1977. Of course, most firms do not try to cost out the value of these benefit schemes

which are designed to attract and retain staff. Hence, very little information about

them is available. However, it does seem clear that the very intention of these benefits is to increase wage inequality.

The above typology of fringe benefits is crude because it excludes a number of

important types of fringe benefits such as discretionary cash grants and preventative

forms of social protection. Aside from shelter, time and consumption, the firm also

makes available cash bonuses, both on an ad hoe and regular basis. These lump cash

payments have their origin in in kind benefits such as the turkey for Christmas, which

eventually became transformed into the Christmas bonus. Essentially these benefits

were gifts. Of course, with time, these discretionary gifts came to be expected as part of

routine privileges which are attached to a decent job. Thus the line between informal

and formal bonus has eroded over time. Cash fringes can take many forms: from

holiday gifts and thirteen month bonuses to stock investment plans where fringes take

the form of a capital asset. Also, the firm is mandated by the state to provide

protection for the worker against the risk of illness and accidents. Should the cost of

these preventive programs be regarded as a fringe benefit? More generally, should the

cost of making available occupational health and safety measures be treated as a fringe

benefit? This is obviously a fuzzy category because we conventionally regard fringes as

a direct benefit to the individual and not merely conditions surrounding employment.

Thus, prevention of morbidity and mortality seems to be an elusive benefit. Very, few

systems of classification treat prevention as a fringe benefit.

Rationales, Entitlement and Distribution

This review of the substance of fringes (what they are, who initiates them, and why

they are initiated) makes it clear that there is no single rationale which can encompass

the diversity of benefits whose character and purpose change over time. That no single

principle unifies them is perhaps the distinguishing characteristic of fringes they are a

legitimate arena in which the redefinition of social relations is worked out. They

therefore represent a flexible arena in the system of economic rewards where social

change occurs without a compelling drive for internal coherence. Social confusion represents social change.

Fringe benefits are both an extension or deferral of wages and an instrument for

meeting the social need of the worker and his/her family. For each specific type of

fringe, we face the fuzzy boundary separating compensation from social obligation.

But whatever rationales are fashionable at any point of time, the expansion of fringes

is characterized by a transformation from an informal, discretionary gratuity

provided by employers as gifts, to formal, contractual "vested contract rights" we

identify as legal entitlements which the courts protect against diminution or

revocation. Benefits can be arrayed on a continuum from a gratuity to a vested

contract right where benefits are treated as property and where the courts act as

132

guarantor of entitlement. This t ransformation f rom gratuity to right brings with it a

new set of problems. "To an extent, these reactions against bad old gratuity interpretations have already gone too far, locking in against statutory alteration

benefit-inflating gimmicks such a s . . . New York City's calculation of lifetime

pensions on the basis of a final year wage doubled by overtime" [14].

Whatever intellectual rationales are advanced to justify fringes and the principle of

entitlement to them, the growth of fringes as an economic resource has substantial

effect on the distribution of income. Sometimes fringe benefits can narrow inequali- ties, as was the case with sick benefits legislation in West Germany. The effect of

Germany's move from a public to a private system of payment and administration was

to narrow inequalities. White collar workers had always been entitled to 100%

reimbursement during periods of sickness. This new scheme extended those benefits to

blue collar workers. As a result, after the new law had been introduced, the total

earnings of blue collar workers increased by two to three percent.

While the German case of sick pay is interesting, it is also atypical. Many fringe

benefits do not reach workers with low earnings or those in part-time employment. Hence, they tend to widen inequalities in society. Most of the empirical evidence seems

to suggest that the scope of the coverage and the adequacy of benefits varies sharply

with workers' earnings. Townsend (1979) argued that in Britain employer welfare benefits "are distributed more unequally than either gross or net earnings, and a

substantial proport ion of low-paid employees had no benefits at all or benefits of very

small annual value." S meeding (1979, p. 26) reached the same conclusion in his United States study. "In the private non-farm economy, 37.8% of all workers and 54.6% of all

full-year, full-time workers receive both employer provided health insurance and

pension benefits." The value of the fringe benefits for this subject group of workers accounted for 44% of their wages and salaries. Smeeding then ranked all workers into

ten wage and salary levels and computed the value of fringe benefits as a proportion of each earning level. He found that legally required benefits exert an equalizing

influence on compensation, because all workers receive them and because employer

contributions are at a constant percentage of earnings up to a ceiling. However, by contrast, some of the other fringe benefits showed a regressive effect by earning level and by employment status, i.e., full-time, full-year vs. paid part-time workers. For

example, pension contributions increased sharply with earnings, because they are computed as a constant percentage of earnings for all workers. Health insurance contributions by employees are allocated equally in terms of dollar amounts and thus

are proportionately more valuable to the low income earner. However, the percentage of workers with low earnings who are covered is low. This "cancels the pro-poor distributional impact of health insurance contributions." These contributions are only pro-poor "if we look at those workers fortunate enough to be counted." Smeeding (1981, p. 42) concluded that the distribution of total employer compensation is less equal than the distribution of wages and salaries alone.

133

Conclusion

The literature takes for granted that there is a particular, identifiable kind of society

called "the welfare state" which is distinguishable from other societies. From this perspective, we are led to ask such questions as: Which societies are evolving into a welfare state and whether the rate of development of welfare state activities exceeds that

of economic growth. Of course, embedded in this analysis are normative concerns

about identifying the limits and possibilities for further development of the welfare

state as a social instrument for social protection against the uncertainties of market

economy.

I believe that the focus on the welfare state is not a helpful way to understand modern industrial societies. Several well known facts are not easily interpreted in this

framework. (1) Every mature welfare state contains social provisions which are

distributed and financed through the private sector. Thus we are always dealing with,

at the very least, a two tiered system of government and private initiative. (2) Many developing countries that have little welfare state protection nevertheless have an

extensive system of government-mandated social protection which is attached to

employment. Even when the state's direct role in providing a fall-back system of security is very modest, it can nevertheless be an active agent in promoting such

welfare through the private sector. The state in all modern societies manages the economy to some degree, with welfare

goals among the objectives it pursues. These considerations lead one to conclude that

it is not the evolution of the welfare state which needs to be studied, but the political

economy of industrialized and industrializing states. This requires a detailed systematic study of the interaction between the public and private sectors. I start with

the premise that industrialized societies must be viewed as a unified or joint system in

which the state and the market are viewed as different aspects of political economy. The state is actively involved in shaping the system of social protection that exists in

the private sector; and the private sector is deeply involved in determining the role of

government. More specifically, government is actively involved in determining the

structure of social protection through specification of what risks are to be covered,

which groups are to be protected against these risks, and how much protection is to be enjoyed. Government influences private decisions through a variety of means

including mandating, stimulating, regulating, and direct support. The firms, in turn, through the activities of management and unions, influence government initiatives.

The detailed ways in which the public and the private sector interact in the provision

of social protection is critical to an understanding of industrial society. We need to think of the welfare economy in a more precise way, namely as a structure of rules and institutions within which the state and the market interact as instruments for protecting the members of society against the uncertainties of industrial society. The distributional effects of the welfare economy must always be taken as problematic rather than as given. The extent of egalitarianism produced by a political economy

134

c o n c e r n e d w i t h s o c i a l p r o t e c t i o n m u s t i t s e l f c o n t i n u e to b e t h e s u b j e c t f o r e m p i r i c a l

i n q u i r y . H o w e v e r , s y s t e m a t i c u n d e r s t a n d i n g o f t h e d i s t r i b u t i v e e f f e c t o f p r o v i s i o n in

t h e w e l f a r e s o c i e t y r e q u i r e s t h e i n f o r m a t i o n o n t h e s c o p e a n d c h a r a c t e r o f s o c i a l

p r o t e c t i o n in b o t h t h e p u b l i c a n d p r i v a t e s e c t o r s . I n t h e U n i t e d S t a t e s , f r i n g e b e n e f i t s

a r e i m p r e s s i v e l y l a rge , a c c o u n t i n g f o r a b o u t o n e t h i r d t o o v e r f o r t y p e r c e n t o f t o t a l

l a b o r c o s t s f o r t h e t h i r d o f p r i v a t e s e c t o r w o r k e r s w h o r e c e i v e b o t h p e n s i o n s a n d

h e a l t h i n s u r a n c e ; h e n c e t h e y p r o v i d e us w i t h i m p o r t a n t i n s i g h t s as w e a t t e m p t to

u n d e r s t a n d t h e m o d e r n i n d u s t r i a l s o c i e t y .

N o t e s

l Even if wages have part of their origin in meeting need, this does not imply that they do not reflect productivity. Typically decisions about the capital-labor ratios treats the wage as socially defined; capital investment treats wages as the constraint.

2 These figures were calculated from statistics presented in recent National Income Accounts by subtracting all taxes (income, social security contributions, indirect, etc.) from all wages and salaries plus entrepreneurial income.

3 We can, of course, also question the focus on civilian expenditures. The military performs an important social welfare role; for example, it provides an apprenticeship system for working class youth. Moreover, enrollment in the military increases when unemployment rises. When young people cannot get a job, they will stay in school or join the army. Also military spending in the civilian sector has a significant effect on employment rates. This is especially the case in certain key sectors of the economy.

4 Tax expenditures are "a reduction of expenditures for those who benefit from them and a loss of revenue f ~rthe b~dy which grants them.~ A c ~ u n t s ~ f S~ia~ Pr~tecti~n in the Eur~pean C~mmunit y~ ~ 97~ 1975, p. 23.

5 Schultze argues that in the United States, "in the short space of twenty years, the very nature of federal activities has changed radically, the newer programs are different; and the older ones have taken on more ambitious goals" which traditional explanations do not measure.

6 Civil public consumption equates current expenditures for goods and services, excluding defense expenditures; and social transfers include current transfers to households and social security benefits plus social assistance, but excluding subsidies, interest on the public debt, and other transfer expenditures.

7 Public employment as a percent of the labor force, excluding nationalized industries, varies by country. The figures are 25% for Sweden; 20% for Great Britain and the United States, and about 13% for Germany and France.

8 Such examinations can, of course, be used by the firm as part of its own personnel policies in getting some workers to change jobs within the firm or to impose early retirement on other workers. Thus, the regulatory process can have perverse and unintended effects on the well-being of the workers.

9 Financial Times, July 14, 1980, p. 8. l0 These imputed contributions "represent the counterpart of social benefits granted directly. . , by

employers to their employees. . , irrespective of whether these obligations are paid in pursuance of a legal or other statutory obligation, a collective agreement . . , as employer/employee agreement within an undertaking, the contract of employment itself, or even, in certain cases, on a voluntary basis.

l I " 'Perks' Rise with Inflation," New York Times, August 27, 1980, p. D5. 12. Profile o f Employee Benefits, The Conference Board, 1974. 13 Flexi-time extends the privilege of organizing one's work time downward to the rest of the occupational

structure. The flexible use of time permits the worker to respond to the demands made by the family for errands and crises as well as civic demands for jury duty and other public obligations. We need to distinguish privilege from fringes, but there is no sharp dividing line. Part time work may itself be regarded in some situations as a fringe.

14 Correspondence with Lance Liebman, Professor of Law at Harvard University, June, 1980.

135

References

Cazes, B. (1979). "The Crisis of the Welfare State in the Western Economies," a paper prepared for a conference in Mexico City supported by the Fondacion de Sciences Humaines, August, 1979.

Klein, R. (1978). "Public Expenditures in an Inflationary World," a paper prepared for the Brookings project on the politics and sociology of global inflation, November, 1978, pp. 5 and 6.

Reid, G. L. (1961). "The concept of fringe benefits," Scottish Journal of Political Economy 9: 208-218. Reid, G. L. and Robertson, D. J. (1965). FrhTge Benefits, Labor Costs and Social Security. London: Allen &

Unwin, Ltd., p. 21. Royal Commission on the Distribution of Income and Wealth (1979). Report Number 8, Fifth Report on

the Standing Reference, London: HMSO. Schuhze, C. L. (1977). The Public Use o f the Private Interest, Washington, D.C.: The Brookings Institute,

p. 9. Skolnick, M. (1976). "Twenty-five years of employee benefit plans," SocialSeeurity Bulletin, September, p.

4. Smeeding, -F. (1981). "The size distribution of wage and non-wage compensation: employer cost vs.

employee value," a paper prepared for the NBER Conference on Research in Income and Wealth, December, 1981.

Thurow, C. (1981). "Equity, Efficiency, Social Justice and Redistribution," a paper prepared for the OECD Conference on Social Policy in the 1980s, Paris.

Titmuss, R. M. (1958). "The social division of welfare," in Essays in the Welfare State, London: Allen & Unwin, Ltd., Chapter 3.

Tox~nsend, P. (1979). Poverty in the United Kingdom, London: Pelican Books, p. 223.