Keynes and the mandarins

Samuel Brittan

If there is one aspect of Keynes which unquestionably dates him, it is his attitude to the democratic process. In his economic writings he was concerned with the extent to which the pursuit of self-interest in the market place did or did not promote the general interest. But it never occurred to him to see the political process as a market place, governed by the self-interest of politicians, officials and voters. He took it for granted that decision would be ultimately made by a small group of the educated bourgeoisie, who were inspired by a disinterested concern for the public good. He assumed that wrong decisions were taken out of intellectual error or, at worst, narrowness of vision; and that if the correct ideas were promulgated with sufficient clarity and vigour, they would eventually win the day.

These characteristics can be illustrated by the present-day arguments between those economists who call themselves Keynesian and those who call themselves monetarists. One fundamental argument is whether demand management of any kind, whether through monetary means or through the budget, is destabilising and actually aggravates the booms and slumps it is meant to cure. The issue is only in part technical. Those monetarists who follow Milton Friedman believe that we would have a much more stable economic environment if government did not adjust the controls so often and contented themselves with a moderate steady annual growth of the money supply and a budget calculated to balance at constant employment levels of business activity.

This approach is not in the spirit of Keynes at all. It is easy to imagine Keynes pointing out the perverse effect of would-be stabilisation measures. But the idea of abandoning discretionary management and going by fixed preordained rules was entirely foreign to his temperament. Both in personal and in public life he had an unquenchable faith in men's ability to work out directly the effect of each of their actions and behave in good faith. In contrast to David Hume, he was an 'act utilitarian' rather than a 'rule utilitarian'.

It is, however, a second argument between Keynes and the monetarists which most directly challenges Keynesian political assumptions. Friedman's most central challenge to the prevailing orthodoxy has little to do with the technicalities of monetary policy. It is on whether it is possible to achieve a target level of full employment by manipulating total spending — whether through the budget, monetary policy, the exchange rate or import controls. Friedman returns to an earlier tradition in insisting that the minimum sustainable level of unemployment—the so-called 'natural rate'— is determined by underlying economic forces such as the speed of industrial change, training and mobility, union and other monopolistic restrictions, and a host of other influences. Any attempt to run the economy persistently below the sustainable unemployment level will need to be supported by an ever increasing rate of monetary expansion; and the end result will be not merely inflation, but an accelerating increase of prices.

The Friedmanites have not simply gone back to the old orthodoxy. They fully accept that a Keynesian boost to demand can reduce unemployment below the sustainable level for a temporary period. 'Temporary' might have been a decade or even more in the 1950s, when people still thought that a pound was a pound and a dollar was a dollar. Today when 'money illusion' has practically gone and people bargain in real terms, a 'temporary' boost will, with good fortune, last for two or three years if that.

It is this temporary success, but ultimate futility, of so-called 'Keynesian' spending boosts that is such a trap for a democracy. Not only do such episodes bring no lasting benefit to employment ; on the contrary unemployment will subsequently have to be abnormally high if governments wish to reverse the ensuing inflation, or perhaps just prevent it accelerating.'

Although it would be absurd to press the question 'On whose side would Keynes have been ?', the idea of a sustainable level of unemployment, which could not be reduced further by 'a boost from the centre' occurs several times in Keynes's own papers. He put this level at 800,000 for the prewar period when the total labour force was much less and social security benefits far more restricted. Postwar economic models geared exclusively to real output and employment, in which wages and prices depend entirely on the mood of the unions and incomes policy, and in which any increase in the money supply is treated as a minor byproduct, seem to me foreign to the spirit of anyone as obsessively interested in money and financial flows, as Keynes was.

The difficulties that the 'natural rate' doctrine pose for Keynes's approach are political and not ones of economic theory. It is not difficult to think of rules which would prevent unnecessary fluctuations of For a fuller explanation see: S. Brittan : Second Thoughts on Full Employment Policy, Centre for Policy Studies, 1975.

References are given in Second Thoughts: unemployment above its natural rate, without spending ourselves into 'runaway inflation'. To understand the political threat posed by short-term temptations of moneY creation and excessive deficit spending, it is helpful to turn to the analysis of democracY of one of Keynes's contemporaries, the Austrian-American economist, Joseph Schumpeter.3 One is not surprised to know that the two thinkers had very little understanding of each other.

Schumpeter's analysis starts with the truism that democracy in a large country cannot be—as the literal-minded see it—the rule of all the people by all the people. There are, at least for the foreseeable future, limits to the ability of technological progress in communications which make continual consultation of all the people all the time impossible. Even if it were possible, most people might well prefer a less incon' venient and burdensome system of organis' ation. In any case, there would be the difficulty of weighting the importance of different views on issues where people's interests, knowledge and feelings were involved to very different degrees.

A more realistic definition of democracY, which would include the essential characteristics of many of the systems of government in the Western hemisphere, was provided bY Schumpeter. He conceived of democratic representatives as akin to other econoMic agents: they deal in votes as steelmen deal in steel or oilmen in oil. The democratic character oi their behaviour results from the competition between different politicians and parties for votes. To gain or to regain power, they must offer policies or, more characteristically, promise results that will attract votes away from other potential governments. To this extent, the views of at least part of the electorate will influence the way in which the country is governed.

The electors are assumed to act according to their own self-interest. This does not imply any view of their motives; as in' dividuals, or within small groups, people can display great generosity. In larger groups, however, any individual who does not look after his own interests is likely t° suffer in comparison with those who do. It seems realistic, therefore, to assume that voters will, at least at the margin, vote according to their perceived interests.

The commercial market place is charaterised by the individual pursuit of self' interest. This is unlikely to cause irresistible demands for more than the economy call provide. Individuals in their own lives are subject to budget constraints; they cannot spend more than they can earn or borrow.

The political market place is characterised by the pursuit of self-interest by members of large groups, where these personal budget constraints are absent. Electors can rather more easily demand an increased slice of the cake without any agreement on the part of those who are supposed to have the thinner

3 Capitalism, Socialism and Democracy (4th posthumous edition, 1952).

slices. The costs of the handouts, whether 'met through taxation or inflation, will not necessarily accrue to the groups who benefit from them. In each individual case, whether a sudsidy is paid to council house dwellers, cheese eaters or car makers, there is a strong incentive for the interest group to press its demands as forcefully as possible without any real limit on the sum total of interest group demands.

Even with fixed rules of public finance, the process will induce a bias toward public expenditure. Nevertheless, because taxPayers or borrowers have votes, an increase in public 'expenditure which requires an increase in taxation or high interest rates is likely to be somewhat. less attractive to a government than' an increase which can be financed effectively by printing money. With traditional rules such as the balanced budget principle, or gold standard limits on money creation, the bias towards government overspending is likely to have some limits placed on it by opposition from taxpayers.

If on the other hand the government is permitted to boost aggregate demand by extra expenditure Without extra taxes or higher interest rates, its trade-off is different. One alternative is to lose votes by failing to offer an interest group as much as other parties will offer. The other is to join the competitive bidding and offer as much.

If the effects of printing money on infla.tion were immediate, the government might have some qualms about substituting, say, temporary job-preservation for stable Prices. But as the inflationary effects are delayed the government is presented with a choice between a certain benefit in increased electoral support in the short run and the uncertain cost of some very unpleasant choices between a slump and runaway inflation in some years' time. It is not surprising that governments, conscious of Keynes's view of long-run mortality, accept the inflationary alternative.

In short, therefore, there are two reasons Why governments are liable to overstimulate demand, or attempt to promote unsustainably high employment: the benefits are short-run while the costs are long-run, and the benefits are specific and easily attributable to the government while the costs are general and less easily attributed to any single cause.

Schumpeter himself was less of a Cassandra than a present-day adherent of his analysis would tend to be. He put forward three main preconditions for the insulation Of liberal representative democracy against the internally generated economic forces that would tend to destroy it. It is intefesting that these conditions were also part of Keynes's implicit view of the political world. But because of the latter's underlying optimism about the political process, he did not bother to spell them out himself or consider the circumstances in which they might be endangered.

Schumpeter's three conditions were (a) the limitation of the area of effectively poli

tical decision making, (b) the existence of a well trained bureaucracy and (c) the exercise of political self-restraint. If these were fulfilled, there would be a greater chance of achieving the Keynesian ideal of policy in which rational ideas and rational men held sway. It is under these conditions that one could realistically assert, as Keynes did, in the most widely quoted single passage of his General Theory (p. 387) that 'The power of vested interests is vastly exaggerated compared with the gradual encroachment of ideas.'

Government by rational men and rational ideas is possible, if at all, only if there are limits to the incursions of voteseeking politicians. In the UK there have existed a number of special agencies whose non-political nature has been constantly stressed both by themselves and by the government of the day. The pre-I914 Bank of England was a leading example. Keynes may well have envisaged an entire government operation in this fashion. However, obscurantist those running such an institution might be, mortality and the permeation of ideas could be relied upon to disseminate enlightened thinking after, at worst, a lag of a generation. Economic debate in such a society would be confined largely to the technical problem of how to manage demand optimally to face existing and expected states of nature, with no danger that demand management would become the tool of competing political teams.

Thus Keynes almost certainly believed that the second Schumpeterian condition, the need for a powerful well-entrenched bureaucracy, was also fulfilled and could keep government in the hands of 'experts'.

Besides the evidence that governments could resist democratic pressure if the experts thought it necessary, the theory of government by experts was reinforced in' Keynes's era by the class composition and attitudes of politicians and civil servants. Politicians of all main parties were, to some extent, enlightened amateurs with sufficient means and independence to resist democratic pressures, while many bureaucrats for similar reasons were able to resist the pressure of politicians. Neither group was under financial pressure to continue at their posts carrying out tasks in which they no longer believed. There was a division of labour between the politicians, who were professionals at dealing in votes and the bureaucrats, who were professionals at policy analysis. Since economic policymaking clearly fell into the domain of the second group, demand management could be insulated from political pressures.

The existence of a large sector of government that could be insulated to some extent from democratic pressures was both made possible and further reinforced by Schumpeter's third condition for the success of democracy; the existence of tolerance and democratic self-control. In the UK in the earlier part of the twentieth century electorates were able to exercise this self-restraint partly because they were slow to realise their power, and partly because of a series of ad hoc events, such as the First World War, which produced an external threat and a patriotic myth to override sectional conflicts or the Great Depression, which weakened the pressure which could be exerted by unions on industry and government alike.

But without the self-restraint of the electorate, the other two preconditions for the successful combination of democracy with demand management, the limitation of political decision and key role of the mandarin class have inevitably crumbled. The area of effective political decisionmaking has been vastly enlarged from its size in the inter-war period, while the independence of the bureaucracy from political pressure has declined. Moreover, insofar as the bureaucracy has been able to promote its own interests, it has itself come to operate in the same direction as other interest groups rather than as an elite which did not have to lobby for jobs and privilege. It thus tends to stimulate rather than to discourage the excessive expectations of the electorate. The public sector has itself become an important lobby for increased public expenditure.

Keynes died in 1946 and we do not have the benefit of his observations on the new social and political environment (which could hardly have failed to distress him aesthetically whatever his final judgment). Unfortunately, a system of economic thinking was developed in his name which rationalised the most self-destructive tendencies of democracy, instead of acting as a bulwark against them. To)escape from our predicament we need, not another revolution in economic theory but a revolution in constitutional and political ideas which will save us from the snare of unlimited democracy, before we find ourselves with no democracy—and very little freedom—left. The widespread call for 'a new Keynes' reflects the worst aspect of the Great Man's legacy: the belief that we can make deep-seated problems go away by a few tactical gimmicks which can be applied costlessly by a few clever men in an office.