Public Finance Course Notes 4. The growth of the public sector 1. Introduction Let us begin by taking for granted a large growth in public expenditure as a proportion of output in all western industrialised economies. We shall then concentrate on attempts to explain the growth of the public sector.
We need to remind ourselves, however, of the distinction between resource-using public expenditure and transfer payments since some of the theories below are concerned only with resource-using expenditure (that is, public expenditure on goods and services and on domestic fixed capital formation), arguing that the reasons for the growth of transfer payments might be quite different. 2. A classification of theories of public sector growth Bailey (1995) (p. 43) divides models of public expenditure growth into macro models and micro models.
In this classification, macro models attempt to account for the long term growth of public expenditure whereas micro models attempt to explain changes in particular components of public expenditure. 2. 1 Macro models of the growth of resource-using public expenditure 2. 1. 1 Wagner’s Law It is customary in discussions of public sector growth to start with Wagner’s law. In 1883, Adolph Wagner, a German social scientist, put forward an idea which became known as Wagner’s law of increased government activity.
The most usual interpretation of this law is that Wagner thought that there would be an inevitable increase in the share of government expenditure in total output, although he did recognise some limits to this increase. Essentially, he was arguing that an expanding government would necessarily accompany social progress and rising incomes. Such a notion, that societies inevitably change according to particular rules, is an example of historical determinism.
In thinking about Wagner’s law, one has to understand Wagner’s view of the relationship between the state and its citizens – that the state can be seen as existing independently of the individuals in society and has a general responsibility for society as a whole. This contrasts strongly with the standard neo-classical view that the state should merely reflect the views of individual citizens. Wagner recognised three functions of the state: (i) providing administration and protection; (ii) ensuring stability; and (iii) providing for the economic and social welfare of society as a whole.
According to Wagner, public expenditure on the first of these (which included law and order and defence) would grow because the increasing division of labour would lead to the breakdown of communal relationships, requiring the state to take over functions previously carried out by families and local communities. In the process, administration would become more centralised and administrative units larger. The increased division of labour would be accompanied by the development of new technological processes which would lead to the growth of monopolies in the private sector.
In Wagner’s view, private sector monopolies would not adequately take into account the social needs of society as a whole and would therefore need to be replaced by public corporations. Further, if private sector companies became too large, the economy would become unstable because problems for individual companies would become problems for society as a whole. Finally, government would need to expand to provide social benefits and services which Wagner saw as not open to economic evaluation. We could include education and health care under this heading.
The principal criticisms of Wagner’s law have concerned his view of history and of the relationship between the state and its citizens. Peacock and Wiseman also queried whether Wagner’s ideas could be applied to all societies at all times and suggested that the time pattern of actual public expenditure growth did not fit well with Wagner’s law. 2. 1. 2 Musgrave and Rostow’s Development Model The economist, Musgrave, and the economic historian, Rostow, (separately) suggested that the growth of public expenditure might be related to the pattern of economic growth and development in societies.
Three stages in the development process could be distinguished: (a) the early development stage where considerable expenditure is required on education and on the infrastructure of the economy (also known as social overhead capital) and where private saving is inadequate to finance this necessary expenditure (in this stage, government expenditure must thus be a high proportion of total output); (b) the phase of rapid growth in which there are large increases in private saving and public investment falls proportionately; and (c) high income societies with increased demand for private goods which need complementary public investment (e. . the motor car and urbanisation). The increasing need in high income societies for skilled labour leads education to become increasingly an investment good for society as a whole. Increased population movements lead to the development of urban slums. Such factors and others lead once again to an increase in public expenditure in relation to total output. These views are interesting in relation to theories of growth and development but are rather too general to provide much of a guide to recent experience in developed industrial countries. . 1. 3 The displacement effect Therefore, Peacock and Wiseman invented the displacement effect. This involves a number of separate ideas:- (a) societies not subject to unusual pressures have fairly stable ideas about the tax burden which they regard as tolerable. These ideas dominate those of desirable government expenditure and hence limit the extent to which government expenditure can grow. (b) however, large scale social disturbances weaken these ideas of tolerable tax burdens.
Emergency government expenditure is accepted and so too are the higher rates of taxes needed to pay for it. People become used to the higher tax rates and their notions of the tolerable tax burden are displaced upwards. After the disturbance, there is thus increased scope for government expenditure and this does not fall back to its old level. (c) as well as the taxation constraint being eased by the social crisis, there is also an `inspection effect’ of the crisis – people observe social needs during the crisis and accept that there is a case for greater social spending. d) finally, the crisis also leads to an increase in the concentration of power in the hands of central government and this is also not reversed after the crisis. But what are these `social crises’? Wars and major depressions obviously qualify and Peacock and Wiseman used these to explain the increase in public expenditure in the UK in the 1930s and after World War II. But how bad does a depression need to be before it qualifies as a `social crisis’? The theory does not provide any answer. A certain amount of empirical evidence was produced to support the displacement effect hypothesis in a number of countries, but problems emain. The notion of the inspection effect is rather vague and the definition of a `social crisis’ presents problems. In many western industrialised countries, public sector expenditure surged from the mid-1960s to the m-d-1970s. What social crises caused this? 2. 1. 4 Permanent influences Several of the ideas above and some others were grouped by Peacock and Wiseman under the heading of permanent influences on government expenditure from the changing nature of society. We can list these as: (a) the level of development (see Section 2. 1. 2 above); b) population size and the age composition of the population – population growth leads to overcrowding and congestion, increased interdependence and external effects of private consumption and production decisions; changing age composition of populations may lead to increases in the proportion of the old or very young in the population, both of which groups may require greater state support (health expenditure, old people’s homes, education, expenditure on housing). (c) increased output per capita and increasing population lead to growth of conurbations and urbanisation is associated with increasing interdependence and externalities. d) social insurance: again, urbanisation and affluence lead to the breakdown of the extended family, requiring increases in public expenditure, especially on health care. (e) increased mobility of society can be seen as part of the same process. (f) all of these factors and others may mean that the income elasticity of demand for public sector services is higher than that for private sector services, although this begs the question of why public serctors provide the services which they do provide; (g) effects of war, war-related and defence expenditure.
It has been argued that the single best explanation of the huge growth of the public sector which took place in the USA was the dominant influence of direct and indirect expenditures on wars and threats of war. This may have been true up until the 1960s and again during the Reagan administrations, but between 1965 and 1975, US federal civilian expenditure increased twice as fast as defence spending. According to Peacock and Wiseman, the problem lies in the permanent nature of all of these influences.
They argued public expenditure growth had not been steady but had involved spurts of growth followed by long static periods. 2. 1. 5 Baumol’s disease This idea was adapted from a model of unbalanced growth put forward by Baumol in 1967. It assumes two sectors in the economy – a progressive sector and a non-progressive sector. The definition of `progressive’ here has nothing to do with standard economic notions of efficiency but relates to the nature of the product being produced. The progressive sector produces goods where the demand for labour is derived from the demand for the goods themselves.
In other words, labour is only needed to produce the goods and is not itself part of the product. Capital can then be substituted for labour without affecting the quality of the product. Thus, there is considerable scope in progressive industries for increases in labour productivity. On the other hand, the non-progressive sector produces commodities where the labour itself forms part of the product which is demanded. In these cases, labour cannot be replaced by capital, without changing the nature of the rpoduct.
This means that there is little scope for increases in labour productivity in this sector. Next, assume that wages in the two sectors increase at the same rate. It follows that the unit costs of the products of the non-progressive sector must increase relative to those of the progressive sector. We further assume that the public sector provides a high proportion of society’s non-progressive products and use this to account for the relative growth in public expenditure. From here we can take two routes. Firstly, let us assume that the size of the non-progressive sector is determined by consumer demand.
As relative prices for the products of that sector rise, we should expect demand for them to fall. However, IF the price elasticity of these products is low AND IF the income elasticity for them is high, then demand for them will not fall as income increases and total expenditure on the products of the non-progressive sector will rise. The second route involves assuming that the government determines the level of public sector provision of goods and sets out to maintain its share of the final output of goods by the economy as a whole.
With wages increasing at the same rate in the two sectors but with labour productivity increasing faster in the private sector, the government will only be able to achieve its aim if there is a continuing transfer of labour from the private sector to the public sector. This is all very interesting, but two problems arise. Firstly, why should the public sector contain more non-progressive production than the private sector? Secondly, should governments attempt to maintain the public sector’s share of total output if public sector prices are rising faster than private sector prices?
Using the neo-classical model of the relationship between state and society, the answer to this will relate to the price-elasticities and income-elasticities of public and private sector goods. This, in turn, requires us to examine the argument that public sector goods are price-inelastic but income-elastic. 2. 1. 6 The relative price effect It is certainly possible to show that over quite long periods in many countries public sector prices have risen faster than private sector prices. For example, between 1955 and 1975 in the UK, the price index of consumers’ xpenditure rose from 59 to 179 (1970 = 100) whereas the price index of the current expenditure of public authorities rose from 44 to 212 (1970 = 100). Beck used such facts as these to argue that in real terms the government’s share of GDP fell between 1950 and 1970 in a majority of the thirteen developed countries he studied. The UK was one such country. According to Beck’s calculations, the government’s share of output (as distinct from expenditure) fell from 30 per cent to 25 per cent over the period. Beck’s study provided no indication of why public sector unit costs were increasing faster that those in the private sector.
Baumol’s disease may have been a major part of the answer but other possibilities exist. 2. 1. 7 The changing environment We talked above of the notion of a government attempting to maintain its level of output in real terms. This begs the question of how one measures the output of many government activities. One argument is that the political pressure on governments is for them to maintain output in terms of performance – keeping down the rate of growth of crime; reducing hospital waiting lists; reducing class sizes; reducing rates of infant mortality etc..
It may then be that public sector unit costs rise not because of inefficiency, nor indeed because of Baumol’s disease, but because worsening social conditions make it more difficult to maintain levels of performance – for example, the spread of new diseases, or increases in numbers of one-parent families, migrant children, and the consumption of drugs. It is also possible to think of objectives of public sector output more broadly and argue that the objective of, for example, `improved education’ may be achieved equally by different mixes of activities, some of which may derive from the private sector, others from the public sector. . 1. 8 Political models of public expenditure growth (the Leviathan model) One argument we have heard a great deal of in recent years is the deceptively simple one that the fact that much of public sector output is not sold on the market necessarily means a lack of incentive for public servants to minimize costs and hence to x-inefficiency. Even where public sector output is sold, it is argued that there is often lack of competition and for this reason a lack of incentive to work hard and to produce the sort of products required by consumers at the lowest possible cost.
Some political theories of public sector growth make use of this idea e. g. Niskanen uses a simple microeconomic argument to support this view. It goes as follows: A private firm selling products and attempting to maximize profits will increase its output only to the point where Marginal Revenue = Marginal Cost, but a public sector bureau providing services free of charge will go on providing the product until the Marginal Benefit is equal to zero. Hence it will over-provide the product and its unit costs will be higher than those of the private firm. There are a number of very important assumptions being made here.
The point for the moment is just that Beck’s studies leave open the question of the causes of differences in unit costs between the two sectors. Such models are extended to incorporate the idea of producer capture – the public sector grows in the interests of the state itself and of the people who work for it. We shall deal with these ideas in more detail in a later set of notes. 2. 2 Micro models of the growth of resource-using public expenditure Micro models include models of the behaviour of voters, politicians and bureaucrats. These are all dealt with in a later set of notes. . 3 Explanations of the growth of transfer payments 2. 3. 1 Attitudes to the redistribution of income Next, let us spend a little time thinking about the growth of transfer payments as societies have become more affluent. In theory, two conflicting views of the relationship between transfer payments and increasing affluence are possible:- (a) with increasing affluence, there should be less need for transfers; (b) with increasing affluence, the economy can afford to help the less well off more and therefore transfer payments should increase relative to the total.
One’s view of (a) depends on the view taken of the nature of poverty and hence of the purposes of income redistribution. One could, for example, argue that the principal aim of redistribution is to provide a minimum level of income for people in society. This suggests an absolute notion of poverty. Alternatively, it is possible to argue that poverty is essentially relative and that the aim of income redistibution should be to reduce inequalities in the distribution of income, irrespective of the absolute levels of the incomes of the poor.
If one takes the latter view, the question of what is likely to happen to the amount of transfer payments as an economy becomes richer depends on what happens to the degree of inequality as the average level of income in the economy rises. The argument for (b) depends on the notion that the marginal benefit from increases in income falls as income rises (one could develop the idea of an income-elastic social conscience). There are many other ideas concerning income redistribution. For example, one could argue that as society grows, it becomes more interdependent and risks from unemployment, old age, poor health etc. ecome greater. Society could then be given the role of compensating for these greater risks and their impacts on individual citizens. A variation on this would be to suggest that with increased interdependence and increasing affluence, low income involves greater relative deprivation. Then one could add an insurance principle: people realise that at some future time they might be badly off and are thus prepared to pay higher taxes to support the poor as an insurance against the risk of their becoming so.
When considering arguments regarding transfers, one has also to consider the possibility that growth in resource-using public expenditure may be a second-best but politically more feasible means of redistributing income. This assumes that a number of the major public expenditure programmes do actually involve a redistribution of income from richer to poorer. This idea has, however, been attacked, by Stigler amongst others. Stigler’s Directors’ Law suggests that public expenditure benefits the middle class most and is paid for by the sick and the poor. 2. 3. 2 Changing demographic and economic structures
A number of ideas which we mentioned above in relation to resource-using expenditures apply also to transfer payments. These include: (a) demographic structure: increased numbers of older people or of children increase (other things being equal) the size of transfer payments. The total size of pension payments depends also, of course, on the average age of retirement from work. A major recent political problem in Italy has concerned the determination of the Left to maintain existing rules which allow public sector workers to retire on a state pension from the age of 50.
The issue of retirement age is an important one throughout Europe. Equally important, of course, is the extent to which pensions and child benefit payments maintain their real value over time. One possible response to growing numbers of pensioners in an economy is to allow the real value of the pension to fall. It should be noted that there are alternative views of the meaning of the `real value’ of the State pension. One view is that to maintain its real value the pension should be increased in line with the general rate of inflation in the economy.
The alternative view is that pensions should increase in line with the average wage rate in the economy. A rule based on this view would see pensions increase more rapidly and increase the total payments to pensioners more rapidly. Supporters of this view argue, however, that these payments would not constitute an extra burden on wage-earners since there would not need to be any increase in tax rates to pay for the higher pensions. The extent to which pension payments do, in fact, constitute a burden on the current generation of wage-earners depends on the extent to which pension payments are funded.
That is, a government can act like an insurance company, collecting a social security tax from workers and investing this money in a fund, the interest from which would be used to pay for future pension payments. Each generation would be paying for its own future pensions. However, governments have in general treated taxes raised allegedly for social security purposes as part of current revenue and have used the money for current expenditure. Pension payments for one generation are then made from the taxes paid by current wage-earners.
There is then obviously a problem during periods in which the proportion of pensioners in the population rises sharply. One way out of this dilemma is to encourage or require people to take out private pension plans. This brings us back to problems of income distribution since unemployed and low-paid workers are unlikely to be able to afford payments to private pension plans. The question of the balance between private and public pensions is a major one which will be of great political importance in the coming years. b) Household structure: the increasing numbers and proportions of marriages ending in divorce have led to an increasing number of single parent families. Such families tend to be more heavily dependent on state benefits such as income support and housing benefit. This is another controversial area. One outcome in the UK has been the setting up of the Child Support Agency to try to make all divorced parents contribute to the support of their children.
Another outcome has been a significant increase in social and political hostility towards single parents, although the 1997 Conservative Party conference saw some signalling of a reduction in this hostility. (c) Economic recession: higher unemployment leads to higher transfer payments (and to lower tax revenue), although this also depends on the extent to which unemployment benefits retain their real value over time and the extent to which rules regarding entitlement to such benefits remain unchanged. This is generally seen as a cyclical phenomenon.
However, all European countries have seen significant long-term increases in unemployment in recent decades. Reading Beck M (1976), `The expanding public sector: some contrary evidence’, National Tax Journal, March, 15-21 Bird R (1971), `Wagner’s law of expanding state activity’, Public Finance, vol 26 no 1 Brown C V and P M Jackson (1990, 4th edn), Public Sector Economics ch 5 Corry D (ed) (1997), Public Expenditure. Effective Management and Control, London: Dryden Press chs 2 and 3 Gemmell N (ed) (1993), The Growth of the Public Sector: Theories and International Evidence, Aldershot: Edward Elgar
Heald D (1983), Public Expenditure Lybeck J and M Henrekson (eds) (1986), Explaining the Growth of Government, Amsterdam: North Holland Peacock A and J Wiseman, The Growth of Public Expenditure in the UK, Intoduction to 2nd edn. Peacock A and J Wiseman, `Approaches to the analysis of government expenditure growth’, Public Finance Quarterly, vol 7 no 1 Tussing A D and J A Henning (1974), `Long-run growth of non-defence government expenditure in the US’, Public Finance Quarterly, vol 2 no 2, 202-2
