Can we really cut state spending to Asian levels?

There may be precious little between the main parties nowadays when it comes to macro-economic policy, but one big theme threatening to divide them in the coming election campaign concerns the size of the state. Although not yet clearly delineated in terms of specific policies, the battle lines are being drawn on the share of GDP that should be absorbed by public spending.

Mainstream Conservatives are increasingly grouped around the previously "extreme" right-wing position that the share of the state in national income should be sharply reduced from its present 40 per cent, perhaps to around the 20-30 per cent level currently observed in newly-industrialised economies such as Singapore.

This does not seem to be the position of the Chancellor himself, who has consistently argued that there is no urgent need to slash public spending below the 40 per cent mark. Labour has taken a similar view, though Tony Blair and Gordon Brown have seemed willing to reduce public spending (and taxation), provided this can be done by cutting the costly burden of unemployment. But the idea that vast swathes of public activity could be transferred to private operators has not yet enthused Labour. Could it enthuse the electorate?

It is tempting simply to deny the feasibility of reducing the share of the state to below 40 per cent or so of GDP. Virtually all of our neighbours in Western Europe already exhibit much higher figures, commonly in the 50-55 per cent range. But Mrs Thatcher performed the trick several times in the 1980s and her erstwhile supporters believe it can be done again. Certainly, history suggests there is nothing inevitable or magical either about the 40 per cent threshold or the inexorable uptrend in the size of the state that has characterised the post-war years. We could choose, if we so wished, to reduce the figure substantially.

As the table shows, the mega-state is a surprisingly recent invention - the artefact in roughly equal measures of the two world wars, the Great Depression and the post-war growth of welfare systems. A hundred years ago, it would have been thought absurd in the fledgling industrial democracies of the day to suggest that the state should occupy more than 10 per cent of the economy. As recently as 1960, about 30 per cent was the norm. And even today, the ratio in Japan and the US remains well below 40 per cent. Most countries, though, are in the 42-55 per cent range.

What have countries gained from this huge rise in public spending? A recent IMF working paper by Vito Tanzi and Ludger Schuknecht (see table) argues that high spenders have gained surprisingly little, at least for the rise in spending above the 30 per cent threshold. In order to establish this, the IMF authors chart the progress of several indicators of social advance - unemployment, infant mortality, death rates, schooling and income distribution - for the industrialised countries.

From 1870 to 1960, there is no room to doubt the improvements in all these indicators across the board, gains consistent with the view that increased state intervention was highly worthwhile. (Of course, this evidence does not constitute absolute proof, since the social advances may have been due to other factors, but it is at least indicative.)

Depressingly, from 1960 onwards the picture changes. The extra improvements won in social indicators since then have been limited; countries with large government sectors gained no more than those with small ones. It seems governments tend to do first those things that have large social gains, but that they hit sharply diminishing returns to state activity after they have absorbed about 20-30 per cent of national income.

Since most of the growth in government has come in social security and other transfer payments (including pensions), it is particularly disappointing to discover that income distribution has been little affected. Countries with big governments in 1990 allocated 24 per cent of their income to the poorest 40 per cent of the population, whereas countries with small governments allocated 21 per cent of income to the same group. Such a small difference in the lot of the poorest section of the population seems an inadequate return for an extra 20 per cent of GDP passing through state hands.

This is particularly the case since there seems to be a price to be paid for the extra public spending in terms of economic performance. This comes not so much in reduced GDP growth or higher inflation, but instead in the form of a larger black economy, a higher unemployment rate and a lower rate of innovation in the private sector. All these symptoms would be consistent with a market economy that functions less well, possibly because of the prevalence of the higher marginal tax rates needed to finance extra public spending.

That, anyway, is the conclusion of the IMF authors. They do not consider an alternative hypothesis, which is that economies which perform badly for entirely separate reasons may generate demands for extra social security and public services, so the direction of causation may run from the economy to the share of the state rather than the other way round. But let us forgive them this omission, and agree with them that increases in the state share above about 30 per cent seem to involve limited economic costs without ensuring large social gains. The question is: can the process be reversed in a modern democracy?

Although society as a whole may not gain much from increasing the state's share above 30 per cent, certain interest groups have gained a great deal. Take pensioners, for example. They constitute an increasing part of the problem, but they would not take kindly to having their rights curtailed. Nor would it be remotely fair to change pension obligations after the people concerned have lost the ability to earn their way out of the problem.

But if the current pensioners are not to be hit, where does that leave the present working population? They are already financing the pensions of their grandparents' generation. They certainly would not be happy simultaneously to finance their own pensions through a private pension scheme that was fully funded from individual contributions.

Pensions may be an extreme example of the political problems that are inevitably involved in extricating the state from areas of economic life once it has become fully involved. But the other obvious areas for curtailing state activity - health and education - are no less politically charged. The best that can be done in a democracy is to implement reforms with excruciating slowness, probably taking several decades to make a measurable difference. And wherever the state retreats from direct service provision, there would be thorny new questions of how to regulate the replacement services from the private sector. The population would have to be forced by law to purchase these in many cases, so they could not be left unregulated.

During this lengthy period of slow reform, it would always be open to the electorate to change its mind and kick out the government that had embarked on reform in the first place. To prevent this, many economists have argued that permanent constitutional changes are needed to cut the role of the state, but these would not be possible in our parliamentary system.

So it may be economically desirable to cut public spending towards the 30 per cent mark. It may even be feasible, done slowly. But any party that too quickly embarks on this course in Britain is assured of only one thing - political ruin.

The size of the state

Government Expenditure as % of GDP

About About About 1960 1994

1870 1913 1937

UK 9.4 12.7 30.0 32.2 42.9

US 3.9 1.8 8.6 27.0 33.5

Germany 10.0 14.8 42.4 32.4 49.0

France 12.6 17.0 29.0 34.6 54.9

Italy 11.9 11.1 24.5 30.1 53.9

Japan 8.8 8.3 25.4 17.5 35.8

Singapore - - - - 19.0

Source: "The Growth of Government and the Reform of the State in industrial Countries", Vito Tanzi and Ludger Schuknecht, IMF Working Paper, December 1995