Full employment in a market economy.

Britton, Andrew

Introduction

The occasion for this conference is the 50th anniversary of the white paper on Employment Policy, Cmnd. 6527, presented by the Minister of Reconstruction to Parliament in May 1944. It is a short paper, just 31 pages long, and much of that length is devoted to the special difficulties in the labour market that were expected during the transition from war to peace. The detailed policy proposals set out in the white paper were not implemented in quite the way that was foreseen. Nevertheless this slim document did signal a new approach to economic policy quite different to the approach of governments pre-war. The most important sentence is the first: 'The Government accept as one of their primary aims and responsibilities the maintenance of a high and stable level of employment after the war'. The paper would have been a landmark in the history of economic policy if it had said no more than that.

To understand the significance of the commitment contained in the white paper, we should turn to a very influential book, also published in 1944: 'Full Employment in a Free Society' by William Beveridge. There we can find the reason for the commitment being made and also a fuller account of the methods by which it was hoped that full employment would be maintained. That will be my starting point for an overview of the successes and failures of economic policy, then and now. Why was full employment adopted as an objective of policy? Why was it achieved and maintained so successfully for more than twenty years after the white paper was published? Why has it not been maintained since the 1970s? Should the commitment to full employment be reiterated today? What does it mean in today's circumstances? And, since there is no point in making commitments which cannot be fulfilled, what policy actions should follow?

I have called my paper 'Full Employment in a Market Economy' to emphasise the difference between the circumstances of today and those in which Beveridge wrote fifty years ago. Whether we welcome the fact or not, the organisation of economic life is very different now from that of the post-war years. There is no question of turning back the clock. If there is something in the commitment to full employment which still seems right, even compelling, in 1994, then we need to face some hard and difficult choices if we are to convince a contemporary audience that this is more than a political slogan.

Post-war consensus

The long periods of high unemployment between the wars had been the cause of poverty of a kind we no longer know in this country. The unemployed and their families had experienced severe hardship, even hunger. One of the main concerns of Beveridge and of the wartime coalition government was that a social security system should be established which would prevent this happening again. Such provision, it was recognised, would be difficult to finance unless the level of unemployment was kept low. But that was not the main reason for making the commitment to full employment.

To quote Beveridge: 'Idleness is not the same as want, but a separate evil, which men do not escape by having an income. They must also have the chance of rendering useful service and of feeling that they are doing so'. He also wrote 'A person who has difficulty in buying the labour that he wants suffers inconvenience or reduction of profits. A person who cannot sell his labour is in effect told that he is of no use. The first difficulty causes annoyance or loss. The other is a personal catastrophe'.

He gave three further reasons why full employment should be maintained: to prevent the growth of restrictive practices; to make structural change in the economy more acceptable; and to provide a stimulus to technical progress and the more productive use of labour. No doubt behind this lay also a profound unease as to the political implications of high unemployment. On the title page of his book Beveridge put the quotation 'misery generates hate'. That had indeed been the experience of the inter-war years. The rise of the dictatorships owed much to the perceived failure of liberal democracies to provide prosperity and jobs. It had to be demonstrated that a free society was capable of delivering full employment. The alternative to a free society was to imitate either Nazi Germany or Soviet Russia. The political consensus over economic policy after the war rested on the belief that full employment was essential to the survival of the freedoms that were believed to be fundamental by all the major political parties, and which the war had been fought to preserve.

The importance of wartime experience should never be underestimated as an explanation of post-war economic policies. In war, of course, full employment had been achieved. Useful work was found for everyone as part of a national effort, even those who had been thought unemployable turned out to be good soldiers or munitions workers after all. It could all be put down to good planning and a general willingness to co-operate. Workers and management found themselves, most of the time at least, on the same side. Much of economic life was regulated in the interests of the war effort, and despite some absurdities, regulation seemed to work well. The stimulus for efficiency or initiative came mainly from the need to serve a common purpose. Those who made profits for themselves were regarded as selfish and antisocial. The sense of a common purpose remained after the war was won and helped to support the continued existence of regulations and restrictions on economic life which might otherwise have been found to be intolerable.

Beveridge was well aware of the need for voluntary co-operation if full employment was to be made compatible with economic freedom. He wrote: 'The degree of liberty.. which can be left to agencies independent of the State, without imperilling the policy of full employment, depends on the responsibility and public spirit with which those liberties are exercised. There is no reason to doubt that that responsibility and public spirit will be forthcoming'.

But the main requirement for full employment, according to the white paper or to Beveridge, was neither regulation nor public spirit; these on their own would not be sufficient. What was essential, according to the post-war intellectual consensus, was an adequate level of aggregate demand. The diagnosis put forward by Maynard Keynes of the cause of unemployment between the wars led to an obvious prescription. The second paragraph of the white paper begins thus: 'A country will not suffer from mass unemployment so long as the total demand for its goods and services is maintained at a high level'. The implications of this are set out later in the paper: 'The Government are prepared to accept in future the responsibility for taking action at the earliest possible stage to avert a threatened slump. This involves a new approach and a new responsibility for the State'.

The new approach is well described by Beveridge as the creation of a 'seller's market' for labour. He defined full employment as being a state in which there were 'more vacant jobs than unemployed men'. Moreover these should be 'jobs at fair wages of such a kind, and so located that the unemployed men can reasonably be expected to take them'. It is important to recognise that this is what full employment meant for Beveridge: not a state of balance or equilibrium in the labour market, but rather a state of labour scarcity, so that anyone who wanted work would not have to look for long to find it. Of course he realised the danger of inflation in such a situation, but he relied on the public spirit and responsibility of employers and especially of trade unions to keep this tendency in check.

The main instruments for securing an adequate level of demand were according to Beveridge to be extra public spending, both current and capital, and the regulation of private investment by a National Investment Board. He criticised the white paper as too timid, in its reluctance to intervene in the private sector and for its continuing concern with the scale of public sector debt. Very little was said either by Beveridge or in the white paper about monetary policy. It was widely believed at the time that the rate of interest had little effect on saving or investment. Hence it was on fiscal policy that the architects of full employment relied for the management of aggregate demand.

Demand management was the main post-war policy innovation that supported the commitment to full employment, but it was not the only one. The white paper also discussed regional or local unemployment and proposed to tackle it in three ways: by encouraging firms to locate in 'development areas'; by removing obstacles to labour mobility; and by providing facilities for training.

Looking back 50 years later we cannot but admire the courage with which the commitment to full employment was made. It was by no means clear at the time that the objective could be achieved. Indeed, all recent history suggested that it could not. In fact the white paper was quite cautious in its wording, speaking of 'a high and stable level of employment' rather than the more emotive words used by Beveridge. And Beveridge himself thought that a reasonable target to aim at would be 3 per cent unemployment, or 550 thousand. 'This margin would consist of a shifting body of short-term unemployed who could be maintained without hardship by unemployment insurance'. In fact unemployment was below that level throughout the 1950s and for most of the 1960s as well.

The years of full employment

In 1964, twenty years after the white paper was issued, the National Institute published a historical study 'The Management of the British Economy, 1945-60' by Christopher Dow. In his conclusions he wrote: 'In terms of its fundamental aim--the desire so to manage the economy as to prevent the heavy unemployment that accompanied the pre-war trade cycle--modern economic policy has clearly been a success. For some years after the war, high employment required no specific intervention: wartime arrears of demand were more than enough to ensure full employment. In the decade of the fifties, however, there probably would have been more unemployment if the government had not intervened to increase demand when unemployment showed signs of increasing: and, perhaps equally important, if the world of business had not acquired some confidence that governments could and would so intervene when necessary'. He then goes on to point out other economic problems, such as slow but persistent inflation, frequent balance-of-payments weakness, and growth rather slower than in many other countries. But he adds: 'Failure in these respects has been relative failure only'. At that stage the main emphasis was, quite rightly, on what went right, and not on the early signs of what would later go sadly wrong.

We cannot be certain even now why it was that demand remained so strong for so many years after the war. It was not because the government, following the teachings of Keynes and Beveridge, borrowed heavily to finance extra spending. They did not need to, because private expenditure was buoyant most of the time. Initially private-sector demand may have been strengthened by the low level of interest rates and by the devaluation of sterling in 1949. But this is not enough to explain the continuing buoyancy of expenditure well into the 1960s, by which time the authorities were more inclined to rein it back than to urge it forward.

We are concerned here with a phenomenon which affected most advanced industrial countries, not just the United Kingdom. Initially it may be attributed to making good the damage done by the war to building and equipment both in the public and private sectors. The Cold War and the Korean War stimulated military spending. Consumers for many years were running down the savings accumulated during the years of rationing, then borrowing to acquire new durable goods as they became available. The expansion of international trade was catching up with opportunities lost, not only in wartime but also as a result of trade restrictions in the 1930s.

The point I would wish to stress however is that demand and supply were not matched exactly together by some extraordinary skill on the part of policy-makers, nor indeed by any flexibility in the workings of the market mechanism. What seems to have characterised the period rather was a condition of persistently high, and at times excess, demand. We did have, as Beveridge intended, a 'seller's market' for labour.

It becomes important therefore to ask what were the restraints which prevented wages and prices rising to remove the excess demand, as would be expected in a market economy. Part of the answer may be with the size of the public sector, as well as the continued regulation of credit, foreign exchange, and investment. But I am also prepared to recognise the importance of what Beveridge called 'responsibility and public spirit' as a restraint on inflationary pressure. There was a very general agreement on the priority to be given to full employment, and memories of the inter-war years were still fresh. In national wage bargaining there was some recognition of the potential conflict between the national aims of full employment and price stability. Perhaps the trade unions did not at first realise the strength of their bargaining position in a 'seller's market' for labour, or they chose not to use that strength. Either way, they were contributing to the persistence of what now appears to us as a golden age. By the mid-1960s it was already becoming clear that the problems with inflation, the balance-of-payments and international competitiveness were getting progressively more serious. Since the early-1970s the commitment to 'a high and stable level of employment', although never formally abandoned, has never actually been fulfilled.

What went wrong?

I hardly need to rehearse again the history of unemployment in this country since 1970. The trend has been strongly upward, at least until the mid-1980s. The fluctuations from year to year have also become much more severe. For much of the time inflation has been very high and very variable, although it is low and stable now. The experience of other countries has been varied, but also generally unhappy. In most European countries the trend of unemployment has been strongly upward; in the United States the level has generally been high although the increase is not so marked. There is no consensus amongst economists about the explanation of these developments; on the contrary the period has been one of confusion, controversy and disarray.

It is helpful to distinguish between cyclical and structural unemployment. There is plenty of evidence that the pressure of demand rises and falls from year to year and that unemployment moves in sympathy with that cycle. For example, unemployment fell sharply following the boom in the late-1980s, rose again during the recession of the early-1990s, and is now falling again. This cyclical movement does seem to be related to the adequacy of demand in the economy as a whole, in much the way that Keynes described. However, this cyclical movement, troublesome though it may be, is not the main issue. The main issue is the high level around which this fluctuation occurs, and the dramatic increase in the average level of unemployment in the 1970s and 1980s.

It is difficult to argue that this increase was attributable in any simple way to a deficiency of aggregate demand, since it was not matched by other indications of increasing spare capacity and since it was accompanied by bursts of rapid inflation. We must therefore address the problem of 'structural unemployment', or an increase in the NAIRU (the level of unemployment at which inflation is held constant) or an increase in the equilibrium or 'natural' rate of unemployment. But attaching labels to unemployment tells us little or nothing about its origin, or its remedy. I shall summarise the possible explanations of structural unemployment under three headings: hysteresis, sclerosis and skedasticity. If these sound like the names of serious diseases, that is appropriate enough, because they all represent something going seriously wrong with the workings of the economy.

Hysteresis in this context means the long-lasting, perhaps even the permanent, effect of an accident or stress to the system. Thus the shocks caused by the two increases in world oil prices during the 1970s each produced a sharp contraction of activity in most countries, including the United Kingdom, cyclical downturns much larger than those of the-1950s and 1960s. The hysteresis theories turn on the idea that the associated sharp rises in unemployment were difficult even impossible to reverse. For example the investment, the training and even the technical improvements which should have taken place during the years of recession may not have been made good, even when growth was resumed. Another theory of the same kind suggests that the workers made redundant in a recession lose contact with the labour market, are no longer protected by their union, are regarded with suspicion by employers and become demoralised or demotivated. If this kind of theory is correct we do not need a different theory for the trend from the cycle. A succession of severe cycles would be enough to explain the whole story. My own view is that this explanation, although it is important, is not enough to account for all of the upward trend.

Sclerosis, blames the rise in unemployment on a lack of flexibility. It is the main explanation given in the recent report published by the OECD, and in the Mais lecture by the Chancellor of the Exchequer this year. According to this view the difficulty is one of matching demand and supply for labour. Relative wages do not change quickly enough to clear the market, if indeed they move at all. Workers are reluctant to change their location or occupation, or to learn new skills. According to this view there does exist a labour market equilibrium at which full employment would be restored, but the market takes a very long time indeed to reach it, especially in Europe.

Those who favour sclerosis theories would claim that the problem has got much worse since the 1960s. The need for adjustment has become greater, but the ability to adjust has become less. They point to rigidities introduced into labour markets, especially in Europe, in the 1960s and 1970s. These include increased union power, increased regulation of terms and conditions of employment, greater job security (which could ironically increase unemployment by making firms reluctant to take on new staff) and the statutory enforcement of minimum wages. The weakness of the theory is that the economy was highly regulated, and in a sense inflexible, during the war and for some time afterwards; yet then we had full employment. I am in favour of flexibility, of course, but I am not convinced that an increase in inflexibility can explain the rise in unemployment since the 1960s, at least so far as the UK is concerned.

Skedasticity is a term I have invented to describe those theories which emphasise the variation or inequality of earning power in the labour force. Suppose that this inequality has increased for some reason, but that the inequality of actual wages cannot increase, because of minimum wage provisions or the level of unemployment benefits. The result will be higher unemployment. The possible reasons for a widening in the dispersion of earning power are many and various. Some blame international trade, especially imports from low-wage countries; some blame new technology, which benefits the average worker but not the least skilled; some blame the increased participation of women especially in relatively low-paid jobs, others blame an actual deterioration in training and education. The situation could be made worse by increased competition between firms obliging them to eliminate any job which is not essential, and to make quite sure that none of their employees are paid any more than a strict assessment of their contribution to the business would justify.

I should emphasise that these theories do not suggest that trade, technology, competition and so on are bad for the economy as a whole. On the contrary, the point is that they do raise average living standards and average pay, but this means that the minimum level of pay which employers will offer or workers accept also goes up--and that leaves a significant minority with no work at all. I suspect that theories which emphasise the problems of inequality and low skills do explain a good deal of the increase in unemployment particularly in the UK.

So we do not lack explanations of the rise in unemployment, nor suggestions for policies to reverse it. As inflation and unemployment rose from the mid-1960s to the late-1970s the initial response of policy was to try to repair the post-war consensus. By this stage it was evident that the problems of rising prices and balance-of-payments deficits were so serious that the government could not restore full employment simply by adding to demand. For the twenty years or so after the war it was possible to maintain a 'seller's market' for labour, as Beveridge had wished. But this could continue only so long as a combination of regulation and co-operation kept the lid on price and wage inflation. The lid was blown off in the 1970s and no attempt to put it back on again has succeeded for long. We have had to reduce the temperature, that is to say the pressure of demand in the economy. A market economy cannot operate with a persistent excess of demand.

A great deal has changed in the labour market since we last experienced full employment. New developments of technology, especially information technology, have transformed working conditions requiring new skills and making old skills obsolete. The participation of women has increased and with that has grown the practice of part-time working. Self-employment has become much more widespread. Job changes have become more frequent, and employment has become less secure. The influence of trade unions in wage bargaining has declined, although they still have an important role to play and still have popular support. Management has become more scientific (at least the jargon it uses sounds more scientific) and its approach to employment has become more businesslike--perhaps because competition between firms has become more intense. To sum up, we have seen the evolution of a market economy in this country, much more like the market described in economic textbooks, and it is in that context that the issue of full employment must now be addressed.

What does full employment mean now?

Since Beveridge wrote about full employment much has changed, but much also remains the same. It is still true that enforced idleness destroys self-respect and that 'misery generates hate'. There is still the same need to belong to society, to serve and to be valued. For most people, indeed for more of the population than in earlier generations, this need to belong can be satisfied only by participation in the economy, in paid work, part-time or full-time, permanent or temporary, for an employer or for your own business. High levels of unemployment have not destroyed liberal democracy as seemed possible in the 1940s, but they have divided society and alienated a substantial minority. We have to ask what kind of society we want to live in. I think that most people would say that they want a society in which everyone is able to participate, in which there is effective access for all to work in exchange for an income. The work is important as well as the income, because it is still true, as Beveridge said, that idleness is a separate evil from want and that no-one should be told by society that they have nothing useful to contribute.

Looking into the far distant future we can perhaps imagine a world in which the need for work has been almost eliminated, a world in which the machines have taken over most of the jobs now done by human hands or brains. In such a world if it ever exists, and I am not sure it will, most of the population would in a sense be idle most of the time. But it would be idleness of a very different kind from that now experienced by the unemployed. It would be voluntary idleness, a life of leisure chosen in preference to work, or indeed a life of voluntary work undertaken for its own sake and with no financial reward. Because it would affect the workforce as a whole it would not be socially divisive. It may be useful to speculate about such possibilities and to consider their implications for society, sometime in the twenty-first century. But this does not help us to solve the urgent problem of involuntary unemployment today and the need for all who wish to participate in economic life to be given an opportunity to do so.

We cannot be certain that this need to participate will necessarily be satisfied in a market economy. In the property market buildings can stand empty for years; in the market for consumer goods some of the stock has to be written off. If we treat labour as simply another commodity to be bought and sold, then it is human lives that will sometimes be written off and declared to be redundant. It is appropriate to use emotive language, because the issue here is one of feelings rather than calculation. We want the labour market to be humane in its treatment of individuals, as well as efficient, and we may fear that it is becoming less so. This is in part a matter of the way in which employers and employees choose to behave towards one another, but it is a matter of public policy as well.

For many professional economists unemployment is the most important policy issue of all. We see our role as contributing to the solution of a social problem, not just making the economy more productive. We would be most reluctant to abandon the objective of full employment, because it points beyond economics to a goal which is not just increasing individual utility but also the cohesion of society as a whole.

Setting full employment as a policy objective, even in the context of a market economy, makes it clear that society as a whole has an interest in the way that individuals are treated. It does involve the concept of a community and something which Beveridge called 'public spirit'. We need, in the very different circumstances of today, to find the institutional setting within which that common purpose can be achieved.

Clearly there is no way in which society can underwrite the continued existence of particular jobs or even occupations. Neither can there be a right to a job of one's own choosing at a wage one considers fair. A commitment to full employment will leave many individual ambitions unsatisfied and hopes disappointed. There will still be closures and redundancies, but perhaps we can find a better way of dealing with the consequences. Perhaps we can prevent people being too dependent on the continuation of a particular job and widen the opportunities they have to recover from misfortune when it hits them.

If this is to be achieved then in practice something may well have to be given up in return. The recipe for full employment proposed by Keynes was in effect a 'free lunch'. By making good the deficiency in demand everyone could be better off. If, however, unemployment is now a structural problem then it is unlikely that a painless cure can be found for it. One interpretation of the rise in unemployment is that society has given up the aim of full employment because it found the cost in terms of other objectives was too great. Perhaps no such conscious choice was ever actually made. If, however, we are now to make the deliberate choice that full employment is to be given priority then we need to know what else is being given up. To make a choice between aims we need to know the means that will be adopted. I turn therefore to the question of how a commitment to full employment in a market economy might actually be fulfilled.

There is no shortage of recent studies of unemployment, certainly no shortage of policy recommendations. In the last two years extensive work has been done on the subject both by the European Commission and by the OECD. The latter organisation has produced not one solution to the problem but sixty. It may be inevitable that international organisations, which have to take account of the very different situations in all their member states, will produce recommendations with no clear single focus. Moreover there is a natural tendency, not confined to international organisations, to include within the policy menu a large number of reforms, which could be thought worthwhile in their own right, but which are at best of marginal value in relation to unemployment. Those who favour causes as diverse as privatisation, European integration, or nursery education will want to argue that they have something to offer as part of the package. Those who are looking around for attractive ingredients to put in the package will find these offers difficult to resist. But if we are serious about achieving full employment then we must concentrate on the search for policies which will really make a big difference.

Macroeconomic policy

It follows from what I have already written that the problem of unemployment cannot be solved by macroeconomic policy on its own, simply by more public spending, by tax cuts or lower interest rates. We have quite recent evidence of the effects of stimulating aggregate demand. In the late-1980s, partly as a result of credit liberalisation, partly as a result of tax cuts, both consumer spending and fixed investment accelerated. The result for a few years was a rapid growth of output. In some parts of the country at least there was briefly a 'seller's market' for many kinds of labour, as national unemployment fell sharply and skill shortages developed. But the consequences were rapid growth of imports, renewed inflation and an increase in interest rates. The expansion had to be stopped and there followed one of the most severe recessions since the war.

We are now in the recovery phase of the cycle, with unemployment falling again and capacity utilisation rising. Macroeconomic policy became expansionary after sterling was forced out of the Exchange Rate Mechanism, and the impetus given by the depreciation and reduction in interest rates is still carrying the economy forward despite the tax increases coming into effect this year. There is some purely cyclical fall in unemployment still to come, but on the basis of past experience I do not think that the level can be held much below 2 to 2 1/2 millions. Several years of really rapid growth of output, unless it was accompanied by improved competitiveness and growth of capacity, would lead only to fresh problems of inflation and for the balance of payments. There are some optimists who believe that the level of structural unemployment, the NAIRU or the equilibrium rate (whatever term one uses) is already much lower than that, perhaps no more than 1 million, thanks to the free market policies introduced since 1979. There is not much evidence from the behaviour of the aggregate economy to support that view. I would like to believe it is true, but I shall proceed on the assumption that it is not.

If structural unemployment is indeed of the order of 2 to 2 1/2 millions then there is not much more that can be achieved by demand expansion. The management of demand does, nevertheless, still have an important part to play in a strategy to achieve full employment. As has already been indicated, one theory of structural unemployment sees it as a legacy of past recessions. If it were possible to keep the economy growing steadily, with no recessions, and no unsustainable booms either, then it is more likely that the damage done by the instability of the last twenty years can gradually be put right.

A great deal has been learnt about the management of demand since the white paper was issued fifty years ago. Far more statistical information is available and econometric models have been developed to improve forecasting methods. Nevertheless, the record of demand management has been disappointing. We know, from many years of experience, the inevitable limitations of economic forecasting, and hence the difficulty of taking timely action to offset the economic cycle. The white paper said in 1944 that 'the Government are prepared to accept in the future the responsibility for taking action at the earliest possible stage to avert a threatened slump'. I think the government is still prepared to play that role if it can, and that a long-drawn out slump could probably be corrected, but clearly it is not always possible to avert a sharp recession.

Given the limits of demand management it would be very beneficial if more stability could be built into the economic system itself. A relatively free market economy may be more prone to cyclical variation than the more regulated economy of the post-war period. The British economy seems to have had a more bumpy ride than most other economies in Europe. Possible reasons for this include the structure of financial markets and the importance of home-ownership. These considerations are relevant to the problem of unemployment, and need to be considered in that context, but obviously more direct action is needed as well, if a commitment to full employment is to be fulfilled.

Industrial policy and international competitiveness

The foreword to the 1944 white paper ends with the following sentence: 'But the success of the policy outlined in this paper will ultimately depend on the understanding and support of the community as a whole--and especially on the efforts of employers and workers in industry; for without a rising standard of industrial efficiency we cannot achieve a high level of employment combined with a rising standard of living'. The point is an important one and still valid today.

There are those who believe that the whole economy needs to be strengthened and reformed before full employment can be achieved. They stress the need for greater industrial efficiency and competitiveness if Britain is to participate in the open markets of Europe and the world as a whole. This broad approach is particularly attractive if it could make possible an increase in industrial employment at the same time as real wages were rising. Moreover the potential constraint on growth arising from the deficit on the balance of payments would be lifted if Britain's trade performance were improved.

The white paper published by the European Commission last December puts great emphasis on improved industrial competitiveness as the best approach to job creation. Implicitly it is assuming that Europe can gain employment at the expense of other industrial countries for example America and Japan. (The OECD report prepared for the governments of all industrial countries jointly on the other hand could not make recommendations which would help some member states at the expense of others.) In considering policy options for the UK we must remember that considerable scope remains for raising our standards of performance and productivity to match those of our Continental neighbours. Indeed a considerable effort may be needed to ensure that we do not fall further behind.

Having accepted all this, it would nevertheless be wrong to concentrate too much attention in this paper on industrial policy. In the first place there is no agreement as to the actual policy measures which are most likely to achieve the desired result of improving Britain's competitiveness and efficiency. Whilst some economists argue for a 'developmental' approach, which would involve some kind of national planning, others regard that with disdain and see the proper role of the state as confined to regulation--and no more of that than is unavoidable. I do not need to take sides in that debate on this occasion.

The second reason for placing the main emphasis elsewhere is real doubt as to the scale of effects on unemployment which could be achieved by industrial policy of any kind. One could imagine a successful industrial policy which achieved its main goal of improving international competitiveness, whilst leaving the problem of structural unemployment largely unsolved. The theories of what I have called 'skedasticity' attribute unemployment to the distribution within the workforce of skills and other characteristics relevant to employment. If this approach is right we need to concentrate on the lower end of that distribution, not on the midpoint. It is to policies with that focus that I turn next.

Labour market policies

We now have a great deal of experience of special employment measures in this country, and plenty of examples to draw on from the experience of other countries as well if we wish. The list of schemes gets longer and longer, although many of the new ones turn out to be variations on themes which have been tried out before. There have been selective employment subsidies and schemes to promote employment in the public sector. More recently the emphasis has been more on help with searching for work and tightening up the conditions for the receipt of benefit. At the most, these schemes may have reduced unemployment by a few hundred thousand each year since the mid-1970s.

Proposals to introduce similar schemes on a much larger scale run into problems of administrative feasibility. If unemployment was low, say half-a-million or a million in total, then it would be possible to consider each case individually, and to show a genuine care on behalf of society. Training and counselling could be provided on a generous scale. Every effort could be made to find or to create jobs which are within the capability of each unemployed person. It might even be right, in that context, to insist that job offers are not refused. One could readily imagine an employment 'fallback' provision, even an employment guarantee of some kind to cope with a relatively small number of people who for one reason or another have difficulty in finding or keeping employment. But I do not think that mass unemployment, running to two millions or more, can be tackled in this way. The task becomes unmanageable and the expense becomes prohibitive.

It might be possible by something akin to conscription to create work of a kind for even two million unemployed, but there would be no question of giving individual attention to the needs or the potential abilities of such vast numbers. It would mean creating a large regimented sector clearly differentiated from the rest of the economy. This is not at all what Beveridge had in mind when he wrote about full employment. The underlying purpose of renewing that commitment today could not be fulfilled by special employment measures on a gigantic scale. The social marginalisation and alienation caused by unemployment would remain, and the need to feel that everyone belongs to the community would not be satisfied. On a small scale they undoubtedly have an important part to play, and it matters a great deal how that role is performed. But they cannot be the main means by which a promise to restore full employment would be kept.

Taxes and subsidies

In a market economy economic policy works best by influencing relative costs and prices, by creating incentives for individuals or firms to behave in a socially-desired way. Thus the burning of fuel which may threaten the environment is not prohibited or rationed, but it is taxed. Energy saving is not made compulsory, but it is encouraged by subsidies. A similar approach could work in the labour market.

The aim is not to encourage employment as such, but to increase the employment opportunities for those who, for any reason, are likely to earn wages well below the average. This relates to the dispersion or skedasticity theory of unemployment. We want to offset the widening in the range of earning power which seems to be one reason for the upward trend in unemployment. The Chancellor took a small step in that direction last November when he reduced by 1 per cent the National Insurance contributions of lower paid employees and widened the 20p income tax band. Could the same approach, on a very much larger scale, be the centre-piece of a strategy to restore full employment?

Taxes and subsidies apply to the population as a whole, not just to particular individuals identified as requiring individual attention. The state operates at arms length. No-one has to be identified as requiring special help; there is no need to interfere in their lives or question their motivation. Unlike special employment measures, taxes and benefits can operate on a very large scale without the danger of stigma or marginalisation.

If the intention is to encourage employment creation then the natural place to start reform is with national insurance contributions, as the Chancellor has already indicated. I shall not attempt to go into any detail, but two general points can be made. The first is that employees' contributions must in the long run influence employment prospects just as much as do employers' contributions. The need is to improve incentives to seek employment as well as incentives to provide it. If the employers' contribution is cut then there will be a tendency for wages to rise because the demand for labour will increase, and if the employees' contribution is cut there will be a tendency for wages to fall because the supply of labour will increase. The net effect on labour costs and take-home pay should be much the same in the end--and both employers and employees would benefit. The second point to make is that the graduation of contributions should relate to pay per hour. The intention is to improve the employment prospects for those most at risk of unemployment, that is those whose earning power is relatively low, whether they work full-time or part-time.

Whilst national insurance contributions are the natural place to start, the reforms could affect other forms of taxation as well. Income tax thresholds could be raised. But if the wish is to focus specifically on earned income then there would need to be a new form of tax allowance designed with that in mind. This too would need to be related to pay per hour--an innovation so far as the tax system is concerned.

Instead of cutting national insurance contributions a very similar result could be achieved by subsidies to employers, also related to pay per hour. For the lowest paid it might be necessary for the subsidy to represent a large proportion of the wage before sufficient growth of employment could be induced.

One of the merits of a general tax cut or employment subsidy for the low paid is that we do not have to know in advance where the new jobs will be created. That can be left to the market to decide, so long as the state does not stand in the way. My own guess is that many of them would be in services, and more generally in sectors which do not compete with imports. This inevitably points to activities, for example health and education, which are largely in the public sector. If this approach is to be effective, then employment by central and local government must be allowed to increase even if total public spending is left unchanged. It would be absurd to adopt a target of full employment and then say that the public sector is not allowed to contribute to its achievement.

If the problem of employment is addressed in this way it is bound to be expensive in terms of revenue; there can be no disguising that. Large sums of money must be involved if a large effect is to be achieved. If the reform succeeded in its aim of reducing unemployment permanently to a tolerable level then there would be substantial savings on benefits now paid to those out of work. That would be an important offset to the gross cost of the reform to the exchequer. Nevertheless I suspect that the net cost would still be large. The political process must indicate whether society is prepared to pay that cost, and if so in what form the revenue should be raised. Options worth considering would include higher NI contributions from the better paid, higher rates of VAT or taxes on energy. None of these would be popular, but many people would say it was a price worth paying for full employment. Over a number of years significant changes in the impact of taxes and spending do take place, so we should not dismiss this type of policy option simply because the numbers involved are big.

The benefit system and minimum wages

In the UK, unlike many other industrial countries, the ratio of unemployment benefits to wages has been falling since the 1960s. If the rates of benefit were to be cut that might well reduce the numbers of claimants but at the expense of increased poverty for those who remain out of work. Most people would consider this a price not worth paying.

The replacement ratio could also be reduced however by paying benefits to workers on low pay, which would of course tend to reduce poverty rather than to increase it. Family Credit already does this and the scheme has many admirers. Some would like to develop this approach much further and make it a major element in a strategy to achieve full employment. Certainly it merits careful examination and it is possible that it is the best option available.

As compared however with the alternative of reforming NI contributions and employment subsidies considered above it has two disadvantages. The first is simply that it approaches the main problem we are concerned with indirectly rather than directly, by focusing on individual or household income when it should focus on job creation. The second disadvantage, to my mind at least, is that most of the schemes considered require some form of means test for the individual or the household so as to keep down their cost. A relationship of dependency is unavoidable between the recipients and the state as the donor. This may be right and proper as a temporary relationship for individuals or households in times of particular adversity, but it is not so appropriate as a permanent relationship between the state and perhaps millions of less skilled workers. I do not think that Beveridge would have approved at all.

The attraction of paying benefits to those in work reflects the increase in the dispersion of wages over the past decade, adding to the prevalence of low pay, often of pay so low that workers gain little compared to their benefit entitlements. The same increase in dispersion has not been observed in much of Continental Europe, partly because of minimum wage regulations.

Where minimum wages are high relative to rates of benefit they must reduce the possibility of creating relatively low paid jobs for relatively low skilled workers. The countries concerned therefore face a difficult choice between better pay and more jobs, with a conflict of interest between the employed and the unemployed to be resolved. The issue has, of course, been much discussed in this country in recent months and I do not think I have much to add to the debate. The main issue so far as unemployment is concerned is not whether there should be a minimum wage or not, but the level at which the minimum is set if we do have one. If it was set high enough to require the wages now paid to a large number of workers to be raised, then some, though not all, of the jobs concerned would be lost.

The situation could be helped by the reforms to national insurance contributions or the introduction of employment subsidies already discussed. By cutting total unit wage costs for those offering relatively low skilled jobs the incidence of very low pay should be reduced. The need for minimum wage regulation would then be less, and so also would be the loss of jobs if a minimum wage was in fact introduced.

Education and training

I have left until last what is perhaps the ideal solution to the problem of unemployment. If only we could make the labour force more productive, then of course the economy would be more competitive, real wages and living standards would rise, and the prospects for employment would improve as well. Everyone is in favour of better education and training, for these reasons and many more.

In the context of structural unemployment however we should look not just at the average level of education and training but at its variation across the workforce. It is striking that in this country the variation of educational achievement is greater than in most other European countries. In the context of the distribution of 'earning power' and its relation to unemployment this is an important and disturbing fact. The widening seems to begin at an early age, with very different rates of progress of children at primary school. It may then be self-reinforcing as the children who have not benefited fully from one stage in their education get left further and further behind. It is also true that in this country we have been relatively good at providing higher education opportunities for the academic elite and relatively bad at organising craft training or vocational qualifications for the population at large. Public awareness of these issues has increased over the past ten years and new policies have been introduced. New initiatives have been announced within the last few months. This is not the occasion for a critique, but the intention behind these reforms is clearly to address long-standing problems that have contributed to the high level of unemployment in Britain.

In reviewing the various forms of policy action which might help reduce unemployment we must keep in mind the very different time horizons over which they might be effective. The reform of education and training for the population as a whole will mainly affect the employment opportunities of the next generation. That does not make them any the less important or urgent, but it means that on their own they are not a sufficient response to unemployment today. If the government made a commitment to full employmemt it would be reasonable to ask to what time period the commitment referred. If the answer was well into the twenty-first century, then the popular endorsement would not be so enthusiastic.

Conclusions

It is remarkable how support of the objective of full employment has survived despite the failure for over twenty years to achieve it. Clearly it accords with a popular perception of the responsibility of the State and the well-being of society. Self-interest may also be involved now that so few people can feel really secure in their possession of jobs. The concerns which led to the commitment made in the 1944 white paper are no less relevant today. A job is still, for most people, the basis for participation in the community, and to be deprived of a job is to be rejected or pushed to the margins.

The full employment of the post-war period reflected a condition of persistent excess demand in the labour market in which inflation was suppressed by regulation and by restraint in the use of market power. It also depended on a consensus, not just between political parties about the conduct of policy, but also between employers and employees about the priority of jobs as against pay.

Those tacit agreements broke down in the 1960s and attempts to replace them with the formal structure of prices and incomes policies have all been unsuccessful. Structural unemployment has developed for a combination of reasons, which may include the lasting effects of three severe recessions, a lack of flexibility in relative rates of pay, and a widening in the dispersion of potential earnings within the labour force. Since 1979 (but not only because of the change in government) the British economy has been transformed from a largely corporatist to a free-market system. This is the context within which a new commitment to full employment would now be made.

It would be a cruel deception to speak of full employment as an objective if there were no way of achieving it. My own view is that full employment is possible, although only at a considerable cost. Some form of taxation would have to be increased or some form of public expenditure cut. The sums involved could be large. The question remains therefore whether political support can be found in the country for the measures which would be needed. There is no lack of support for the objective, but the real challenge is to build support for a sustained effort over many years and for some sacrifice in the interests of social solidarity.

Steady, non-cyclical, growth, if it can be achieved, would be a great help, but on its own it will not result in full employment. A successful industrial policy, if that can be achieved, would also be of value both in supporting real wages and escaping from a balance-of-payments constraint. The employment problem could be greatly eased by improvements in education and training, and for the long term this is the most attractive solution. But the group most likely to be helped by such reforms would consist of new entrants to the labour force, so as a means of achieving full employment, this may take us well into the next century.

Special employment measures, targetted on the unemployed themselves, to improve their job prospects on an individual basis, will always be of value if they are well-designed. But they cannot be expected to cope with mass unemployment as we have known it since the 1970s. Extending them to a much larger scale would run the risk of creating another form of marginalisation, not very different in its social implications from unemployment itself.

In a market economy the main instrument for increasing job opportunities for those at risk of unemployment should be incentives and relative costs. These can be influenced by changes in the system of national insurance contributions, changes in income taxation or public spending on transfers and subsidies. The reforms needed to bring about full employment by this method would be on a large scale and involve substantial costs to the general taxpayer. It is best to aim as directly as possible at the objective of creating more jobs, and for that reason reform of national insurance contributions and subsidies for low-skill employment seem more appropriate than benefits paid to low income individuals or households. They also avoid the intrusiveness, and the sense of dependency, which is always associated with a means test. They would have to be introduced gradually, and the scale of change needed could not be calculated accurately in advance. Suppose however that the reform was continued to the point where the cost of employing the lowest-paid workers fell mainly on the community as a whole rather than on the employer. If one went to that extreme then I have little doubt that full employment would be the result.

There may be other means of reaching the same objective, not covered in this paper. I hope I will have provoked some of my professional colleagues to continue the debate by responding to what I have said. I hope that most of them would agree that full employment is a very worthwhile aim, and that, given a high enough priority, it is an aim that we can achieve.