De Nederlandse collectieve uitgaven in historisch perspectief[Dutch public expenditure in historical perspective]

This paper discusses the development of public expenditure in the Netherlands since 1850. Why did public expenditure increase from 14% GDP in 1850, nearly 20% in 1921, 30% GDP in 1950 and over 60% GDP in 1983? Dutch public expenditure has fallen to less than 50% GDP in 2003. Why did this reversal occur? In the period 1921-1950, Dutch public expenditure increased with 10% GDP. About one third of this increase is due to expenditure on social security. Also the expenditure for most other functions increased rapidly: defence, transfers to corporations, international cooperation, health care, public administration and safety. This contrasts with the development of expenditure on education: these remained stable for thirty years as a percentage of GDP; expenditure on infrastructure even declined in this period. In the period 1951-1983, Dutch public expenditure increased with 32% GDP. Nearly half of this increase was caused by the expenditure on social security, like benefits for old age, disablement en unemployment. The rest of the increase was due to extra expenditure on public administration, health care, education, interest and transfers to corporations. During the last two decades, Dutch public expenditure decreased with 12% GDP to 49% GDP in 2003. This is the same as the level during the middle of the seventies. Expenditure on social security, like benefits for old age, illness, unemployment and social assistance, contributed a major part to this decline. However, also the expenditure on various other functions, like transfers to corporations, interest, defence and education, were substantially reduced. Their impact on the decline of the share of total public expenditure even surpassed that of the function social security. At the same time, the public expenditure for health care, public administration and safety increased. For the period since 1950, the role of a wide range of determinants have been investigated. Increasing participation in education, extra expenditure on health care due to increasing welfare and lagging productivity growth of public services put a permanent upward pressure on public expenditure. Since 1950, participation in education increased with 0.7% per year and the real public expenditure on health care per capita increased with 7% per year. During the fifties and sixties, the development of the welfare state, the extension of the services provided by the government and the increase in the interest rate have boosted the increase in public expenditure. As a percentage of GDP, public expenditure increased from nearly 30% GDP in 1950 to nearly 45% GDP in 1970. During the seventies, the use of social security arrangements grew rapidly, while labour market participation fell. Together with factors like the introduction of big subsidies on private investments (WIR), this lifted public expenditure to over 60% GDP in 1983. As a consequence, public expenditure as a percentage of GDP doubled in the period 1951-1983. The trends with respect to participation in education, health care and lagging productivity growth of public services pushed public expenditure since 1983 upwards. This makes the decline of public expenditure with over 10% GDP even more remarkable. General wage moderation, cuts in the benefit level of social assistance, reduction in the number of unemployment and social assistance benefits and the abolishment of the subsidy on private investments have contributed substantially to this decline. Nevertheless, the role of four other developments was even bigger: increased labour market participation of women, the savings on education due to ageing, the drastic reduction in the interest rate and the diminished military threat. Labour market participation of women increased from over 40% in 1983 to 63% in 2003. This increased GDP and - via the denominator - decreased public expenditure as a percentage of GDP; this effect counts for a reduction of about 5% GDP. Education pertains in particular to the population between 5 and 24 years. Since 1983, the relative size of this age group declined with 1,6% per year. This generated a budgetary saving of 2% GDP. The decline in the interest rate on government debt from 9% in 1983 to 5% in 2003 was responsible for a similar saving. The end of the Cold War diminished the military threat. As a consequence, the share of defence in national employment could be halved from 2% to 1%. Also this specific factor accounted for a budgetary benefit of about 2% GDP. Finally, also the shift from public to private tasks, like the sale of equity stock, the redemption of the big loans to housing corporations and the abolition of collectively financed paid sick leave, has lowered public expenditure. The shift from public to private tasks and the substantial budget cuts reflect a drastic change in the view on the tools and tasks of the government: the role of steering and protecting by the government have become much less important, while the role of private responsibility, financial incentives and the market mechanism have become much more important. Some lessons from the past Figures on public expenditure as a percentage of GDP are often used to indicate that the government is spending too much or not enough. However, public expenditure as a percentage of GDP is often affected very strongly by external factors, shifts between public and private tasks and shifts between public expenditure and fiscal arrangements. Ignoring the role of these factors can lead to wrong analyses and unintended policy, e.g. the increase in expenditure on education in order to compensate for the decrease in the number of pupils students due to ageing. According to Baumol’s disease, public expenditure as a percentage of GDP will increase due to the combined effect of a lagging productivity growth of publicly financed services, a change in wage rate in the public sector that follows productivity growth in the private sector and a relatively price-inelastic demand for public services. However, during the past fifty years in the Netherlands, wage rates in the public sector did not increase in line with productivity growth in the private sector but in line with the much lower general productivity growth. Innovative methodology An interesting feature of the paper is the method of analysis. The international COFOG-functions were mostly ignored, as they are not well suited for policy analysis. Instead alternative functions were defined based on a mix of national accounts classifications, transactions, industries and sectors receiving transfers. Furthermore, the expenditure by function were systematically decomposed by looking at major underlying schemes (e.g. for social security benefits and subsidies) and by a decomposition of the changes in value into changes in volumes (e.g. employment in the public sector or number of social benefits for scheme x) and prices (e.g. average wage rate in the public sector and average social benefit for scheme x). Such decompositions are essential for a meaningful analysis, e.g. to unravel the impact of demographic changes and to show the role of new social security arrangements and shifts between public expenditure and tax expenditure.