An expression which came into general use in economics after the Depression of the 1930s, full employment applies to industrially developed economies in which the majority of the economically active are the employees of firms or public authorities as wage and salary earners.

There has always been some unemployment in the course of development of capitalist economies and views have differed as to its causes and as to the extent to which it was a matter of public concern. In the first part of the twentieth century three principal strands of thought about unemployment can be distinguished. Firstly, the followers of Marx believed that cycles were an integral part of capitalist development and would lead to ever deepening crisis: the attempt to evade this by colonial expansion would only lead to conflict between imperialist powers. A second group of analysts paid particular attention to the measurement and dating of business cycles, distinguishing cycles of different periodicity, but they did not, as a rule, offer systematic theories. The third strand consisted of those economists who argued that in capitalist economies, if the forces of the market were left to work themselves out, there would always be a tendency towards an equilibrium, in modern parlance towards full employment.

Table 1 shows average rates of unemployment in six developed countries for various periods of the twentieth century. National estimates of unemployment are obtained either by sample survey or as the by-product of administration, such as a system of unemployment insurance. There are many problems in counting both the numbers unemployed and the labour force, whose ratio is to constitute the ‘rate’ of unemployment. There have been attempts to standardize rates obtained in different countries by different methods and over different periods. The figures in Table 1, taken from Maddison (1982) and OECD Main Economic Indicators are thought to be reasonably comparable. Only in two cases was it feasible to give estimates before World War I. We have four countries for the interwar years and all six after 1950. It will be seen that in the Depression years 1930–1934 the average rates of unemployment were far higher than in any earlier period in the twentieth century and that even in the later 1930s the rates remained abnormally high except in Germany.

Full Employment, Table 1 Unemployed as a percentage of the total labour force
Full size table

The time was ripe for a theory which could account for the persistence of large-scale unemployment and it was provided by John Maynard Keynes in The General Theory of Employment, Interest and Money (1936), which the author himself said was all about ‘my doctrine of full employment’. The self-equilibrating tendencies expounded by those whom (In the overlapping years 1975–1979 there are small discrepancies between Maddison and OECD for Germany and UK. The latest OECD figures were adjusted to be consistent with Maddison.) (In the overlapping years 1975–1979 there are small discrepancies between Maddison and OECD for Germany and UK. The latest OECD figures were adjusted to be consistent with Maddison.)

Keynes called ‘classical’ economists did not necessarily function in the manner prescribed for them and capitalist economies could get stuck with persistent unemployment. According to orthodox theory, unemployment should entail falling wages which would eliminate any ‘involuntary’ unemployment. Similarly, interest rates would fall, bringing about a recovery of investment. Keynes argued that money wages might be ‘sticky’, and even if they were not, falls in money wages would not entail corresponding falls in real wages, since prices would also fall. As to rates of interest, there was no guarantee that such falls as could occur would give a strong enough impetus to recovery. The analysis points clearly to the idea, which others developed more explicitly, that fiscal policy, that is, the adjustment of the budget balance between revenue and expenditure, could prove a more powerful lever to bring about full employment.

Within less than 10 years, the British wartime coalition government, in a famous White Paper, had accepted ‘as one of their primary aims and responsibilities’ the maintenance of ‘a high and stable level of employment’, and other governments, in Australia, Canada and Sweden, for instance, made similar affirmations. Article 55 of the United Nations Charter called on members to promote ‘higher standards of living, full employment, and conditions of economic and social progress and development’. This remarkable change in public policy cannot be attributed simply to the ‘Keynesian Revolution’ in economic thought. More powerful was the observation that twice in a generation full employment had only been realized in war. How far the new principles were responsible for the performance of economies in the postwar period is a disputed question. The facts are that for the 25 years after 1945 the growth rates of productivity in European countries were much higher, and the average levels of unemployment much lower than they had ever been. Fluctuations in output and employment were smaller than in the past. A group of OECD experts reporting in 1968 said that the results of using fiscal policy to maintain economic balance had been encouraging, though there was room for further improvement. In the United States, the government’s attitude towards the new ideas was initially somewhat cooler. By its own past standards, productivity growth was not exceptional, and unemployment, though much lower than in the Depression, was much the same as in the 1920s and before 1914. The Keynesian battle was not truly joined in the USA until the 1960s. In the majority of countries, the era of exceptional growth and full employment came to an end in the early 1970s, since when longer spells of high unemployment have been experienced.

Full employment does not mean zero unemployment. There can be dislocations where large numbers of workers are displaced from their present employment, and time is needed before new workplaces can be created. This can happen at the end of a war, or following some major technological change. Apart from such special cases, regular allowance must be made for frictional and seasonal unemployment. Policy would not aim, therefore, at zero but at the elimination of unemployment attributable to demand deficiency. Governments targeting full employment would like to know the level of measured unemployment to which this corresponds. Three attempts to answer this question deserve mention. (1) The definition given by Beveridge (1944) was that the number of unemployed (U) should equal the number of unfilled vacancies (V). When U is very high, we would expect to find V low, and vice versa. If, over a number of fluctuations, U and V trace out a fairly stable downward sloping curve, we could pick the point on it where U = V as indicating full employment. (2) Phillips (1958) claimed that for Britain there was a good statistical relationship between the level of unemployment and the rate of change of money wages. By choosing the level of unemployment delivering zero wage inflation, or when labour productivity was rising, the slightly higher level delivering zero price inflation, we could pinpoint full employment. (3) Friedman (1968) objected that in the long run there was no trade-off between unemployment and inflation: instead he argued that there was a ‘natural’ rate of unemployment, such that if the actual level was pushed below this, there would be not only inflation, but accelerating inflation. If this theory could be substantiated, one could choose the ‘non-accelerating inflation rate of unemployment’ (NAIRU) as the target. It is evident that the usefulness of each of the above approaches turns on the closeness and stability of the statistical relationship actually observed. Experience in different countries has varied, and the British evidence should be regarded as illustrative. For the period from the early 1950s to the later 1960s econometric analysis produced reasonably stable relationships for all three approaches, yielding estimates of the full employment level of unemployment of the order of 2–3%. But in the 1970s any stability of the Phillips curve crumbled, and estimates of NAIRU shot up from below two to over ten per cent, but without any clear indication of the institutional or structural changes which must have occurred to bring about so large a shift in so short a time. The UV relationship did not escape entirely unscathed either, but a plausible story can be told in terms of an outward shift of the UV curve. Brown (1985) reckoned that the United States, the United Kingdom and France suffered increases in the imperfections of the labour market in the period from the early 1960s to 1981 which might account in full employment (U = V) conditions for extra unemployment of 2% or less. It would seem that the substantial rises in unemployment, especially in Europe, in the 1970s and 1980s can only be accounted in a smaller part by a rise in ‘full employment’ unemployment and that a greater part denotes a shortfall below it.

If the growth of output of developed economies after 1945 was exceptional, so also was the rate of price increase: in Britain, for example, such a sustained and substantial rise (3–4 % a year on average) had not been seen in peacetime for more than two centuries. Some countries had faster rises, but, in most cases, there was no clear sign of acceleration. A marked change of gear in price inflation occurred between the 1960s and the 1970s, precipitated by two large cost impulses. Around 1969 there was in many countries a distinct surge in wage increases which Phelps Brown (1983) has called ‘the Hinge’ and in 1973 there was the first of the great OPEC oil price rises. Confronted with these spontaneous boosts in costs, the authorities had to choose between allowing their consequences to be worked out within the bounds of the existing monetary and fiscal stance and adjusting that stance to accommodate them, which would mean that final prices would also jump. They began increasingly to opt for the former course. In doing so they received intellectual support from the first wave of the ‘monetarist’ counter-revolution against the now orthodox Keynesian demand management. Firstly, it was said that to push unemployment below the ‘natural rate’ would cause accelerating inflation. In any case, too little was known about the structure of the economy, in particular its time lags, for fine tuning to be a sensible policy. Better to adopt simple rules, such as fixed targets for the growth of the supply of money, which would keep inflation under control, and output and employment would adjust to the level indicated by the ‘natural rate’ of unemployment. Later developments in the new classical economics went further and denied altogether the possibility that governments, by loan financed expenditure, for instance, could effect lasting changes in employment. Instead, it was suggested, the only way to bring down unemployment was to reduce the monopoly power of trade unions, and to take other steps to free labour markets, such as abolishing minimum wage legislation and reducing unemployment benefit. Though not supported by any substantial body of evidence, these new ideas undoubtedly helped to persuade central banks to adopt fixed monetary targets, or rules, and after the second OPEC price rise in 1979, most governments followed restrictive monetary policies with more severe budgets. Calculations of ‘constant employment’ budget balances show a tightening equivalent to several percentage points of GNP in some cases, especially in Europe where unemployment rose considerably after 1980. On the other hand the United States broke ranks in 1983, allowing both actual and ‘constant employment’ deficits to rise, and it was the one major economy to experience falling unemployment.

If there is little evidence of a unique ‘natural rate’ of unemployment, it is nevertheless clear that to bring down a cost-induced inflation by demand restriction may involve high unemployment for a great many years. A wide range of ‘income policies’ has been attempted, and others canvassed, to secure that firms and workers would settle for lower prices and wages than they would seek if they were acting alone, provided others would do the same. It is unlikely that full employment of the kind experienced in Europe in the 1950s and 1960s could return without the aid of such policies. Throughout the great postwar expansion world trade grew at an unprecedented rate. Fixed exchange rates, with permission to change parities if needed, worked well enough for most countries to maintain their external balance. However, the Bretton Woods system crumbled and was succeeded by generally floating exchange rates, while at the same time controls over capital movements were being dismantled. Exchange rates came to be determined as much by capital movements as by trade, and they can diverge widely and for long periods from any level suggested by purchasing power parity. Thus full employment is also seen to depend increasingly on the joint action of all, or of a large number, of countries.

Employment policy has been linked with the welfare state in contradictory ways. On the one hand, higher unemployment is tolerated on the grounds that welfare provision mitigates the economic hardship involved: on the other hand, higher welfare costs are perceived as a growing burden on economies with high unemployment.

See Also