Labour markets and inflation.
There appears to have been no growth in OECD employment in 1991, and the area-wide unemployment rate has risen to 7 per cent. Nominal wages decelerated as unemployment rose, but the deceleration in prices has been less marked so far. The unemployment rate is projected to rise still further to around 7 1/2 per cent in the first half of 1992; the projected recovery in employment may subsequently result in unemployment easing back slowly to around its present level by end-1993. With slack in both labour and product markets expected to persist at or slightly above the levels experienced in 1991, there are good prospects for a further reduction in OECD wage and price inflation over the next two years.
Recent trends in employment and unemployment
OECD employment fell slightly during the first half of 1991 (Table 12), for the first time since the second half of 1982. Job losses were severe in some countries, exceeding 3 per cent (at an annual rate) in the United Kingdom, Canada, Australia, Finland and Sweden. Employment declines have not been cushioned appreciably by reductions in working time: data for industry/manufacturing, which are available for around one-half of OECD countries, indicate little or no decline in average hours worked. Indeed, average factory hours increased during the second and third quarters in the United States. [Tabular Data Omitted]
During the second half of 1991, OECD employment is estimated to have stabilised. Employment continued to rise in Japan and Germany as well as in some small European countries (albeit at a much slower rate than earlier in the year), and was unchanged in North America. Elsewhere, employment prospects appear to have continued to deteriorate. In the United Kingdom and Finland, in particular, there were further sharp falls in employment.
The OECD labour force grew by 1 per cent (at an annual rate) in the first half of 1991 - slightly faster than in the preceding half-year: an acceleration in labour force growth in the United States, Japan, Germany and several smaller countries offset the easing of participation rates elsewhere in response to the weakening in labour demand. OECD labour force growth appears to have slowed in the second half of 1991, affecting many countries.
The OECD unemployment rate, which had begun to rise in the second half of 1990, reached 7 per cent in the first half of 1991 - adding 3 1/4 million persons to the unemployment total over a 12-month period. The largest increases in unemployment in the first half of 1991 were recorded in the United Kingdom, Canada and Finland. By contrast, the rate fell to its lowest level since 1981 in Germany and remained unchanged in Japan. OECD unemployment continued to rise in the second half of 1991, although more slowly than during the first half, to reach 7 1/4 per cent.
OECD unemployment and labour market policies in
Although the OECD unemployment rate fell to a ten-year low of 6.2 per cent in the first half of 1990, this was a full point above the 5.2 per cent registered in 1979, just prior to the sharp recession in the early 1980s. Thus, the protracted cyclical upswing from 1983 to 1990 failed to break the rising long-term trend in OECD unemployment - a trend which has been in evidence since the early 1970s (Chart K). The picture is, however, far from uniform across regions. An upward trend in unemployment from cycle to cycle in most evident in the European Community and Australia/New Zealand; it is much less pronounced in the Nordic countries; and in North America the upward trend of the 1970s appears to have been reversed, bringing the recent cyclical low (in 1989) below its 1979 level. In Japan, the unemployment rate has moved within a relatively narrow range of 1 to 3 per cent since 1970.
During the 1980s, OECD governments experimented with a wide variety of labour market policies, with the explicit aims of reversing the trend rise in unemployment, redistributing the burden of unemployment and promoting a more efficient labour market. Labour market measures currently absorb large amounts of public spending and cover significant fractions of the labour force: the typical OECD country spends over 2 per cent of its GDP on labour market programmes and in 1988 11 million persons (equivalent to 3 per cent of the OECD labour force) participated in such programmes(1). It is too early yet to assess rigorously the impact of such policies on unemployment, but a few tentative conclusions can be drawn from the experience of the 1980s.
One lesson which many OECD countries have drawn is that it is desirable to promote so-called "active" labour market policies, such as training, placement, subsidised employment and rehabilitation programmes for the unemployed, as opposed to "passive" measures of income maintenance (e.g. unemployment benefits and incentives for early retirement). Cross-country analysis suggests that more generous unemployment benefit systems - as measured by high "replacement rates" (the ratio of benefits to previous earnings in employment), few pressures on the unemployed to accept available job offers and relatively open-ended duration of benefits - prolong the duration of unemployment, even though the relationship is not a simple one(2).
Public expenditures on labour market programmes are a substantially higher share of GDP in the EC and Nordic countries than they are in Japan and the United States (Table 13). The share of GDP absorbed by these programmes declined slightly in most countries between 1985 and 1990, as falling unemployment went hand in hand with less public spending on income support. Within the total, public expenditures on income support are typically two to three times greater than expenditures on active labour market measures in the EC, the United States and Japan. By contrast, the Nordic countries attach greater weight to active programmes compared with income support. Over the period 1985 to 1990, the EC countries increased slightly their spending on active measures as a fraction of GDP, whereas the opposite trend was recorded in the United States, Japan and the Nordic countries. [Tabular Data Omitted]
It is difficult to draw conclusions from these data as to the efficacy of labour market policies. One element of an efficient labour market, a high degree of occupational and geographic job mobility in response to market signals, does not seem to have been enhanced during the 1980s(3). Labour market policies appear, however, to have achieved some success in redistributing the burden of high unemployment across different labour force groups. For example, the sharp reduction in the relative (and absolute) rate of youth unemployment in Europe over the past decade can be partly attributed to deliberate efforts by certain European countries to target a wide variety of labour market schemes specifically to young people. During the latter part of the 1980s, France, the United Kingdom and Denmark each spent around 1/4 per cent of GDP on youth measures, Italy and Ireland twice as much, whereas outlays on similar measures in the United States were negligible. In addition, youths benefited disproportionately from other policies such as subsidised employment in both public and private sectors. Moreover, in some countries the negative effects of legal minimum wages on youth employment were reduced either by allowing minimum wage increases to fall behind increases in average wages or by developing a wide range of employment and training opportunities paying below minimum rates to youths. At the same time, these measures may have decreased the likelihood of the prime-age adult unemployed, especially females, obtaining a job by encouraging employers to substitute in favour of the young unemployed. There is some evidence of this mechanism at work in many European countries. Again, the sharp fall in the participation rates of older Europeans has no doubt been influenced by various incentives offered for early retirement. During the latter part of the 1980s, France, Belgium, and Denmark spent on average around 1 to 1.3 per cent of GDP on early retirement schemes. No such publicly-financed incentives exist in the United States(4).
In sum, it seems that labour market policies can influence the structure of unemployment; by judicious targeting of certain groups such as youths they can redistribute the burden of high unemployment away from the target group to other groups. The evidence is less clear as to whether labour market policies can reverse the trend rise in unemployment without leading to accelerating wages, i.e. by lowering the so-called non-accelerating inflation rate of unemployment (NAIRU). Despite spending over 2 per cent of their GDP on labour market policies and switching resources towards "active" measures, the EC experienced a rising trend unemployment rate and a growing incidence of long-term unemployment - by 1989 the long-term unemployed accounted for over 50 per cent of total EC unemployment (Table 13). The Nordic countries, on the other hand, spent slightly less as a share of GDP on labour market measures and unemployment did not follow a rising trend. A major challenge facing policy makers in the 1990s is how to lower long-term unemployment in Europe. The evidence suggests that this will require an integrated approach to active labour market policies and unemployment benefit and other social welfare systems, together with a vigorous commitment to pursue structural reforms more widely in economies; this would contribute to higher productivity, greater labour mobility and a better balance between labour costs and economies' ability to pay.
Recent trends in wages, profits and prices
Wage increases moderated during the first half of 1991 in response to growing labour market slack. At the level of the OECD area, average wages in the business sector decelerated by around 1 percentage point compared with the previous half-year (at annual rates). Lower wage increases were, however, offset by rising non-wage labour costs, so that total compensation per employee accelerated by around 1/4 percentage point, bringing its growth rate to 5 1/2 per cent at an annual rate in the first half of 1991 (Table 14). The rise in non-wage labour costs appears to have moderated in the second half of 1991, and the growth of total compensation per employee slowed to 4 3/4 per cent. [Tabular Data Omitted]
In some countries, weak demand reduced the extent to which higher labour costs could be passed into prices. Consequently, the profit share in business-sector value added declined in the first half of 1991 in Italy, the United Kingdom and Canada, while it remained at the lower level reached in the previous semester in the United States. However, profit shares increased in response to the more buoyant economic climate in Japan and Germany. Profit compression may have eased in the second half of 1991, with profit shares unchanged or increased slightly in most of the major countries except Italy and Germany, where weaker activity put downward pressure on profits.
Prices were affected also by indirect tax increases in a number of countries, notably the United Kingdom, Canada, Germany and Sweden. During the first half of 1991, net indirect taxes (indirect taxes minus subsidies) contributed around 1/2 percentage point to average GDP-inflation (measured at market prices) of 4 1/4 per cent in the major seven economies. But, despite higher indirect tax rates in Germany, there appears to be no further contribution from net indirect taxes to inflation in the second half of 1991.
The net result of these forces was a rise in the average OECD GDP deflator in the first half of 1991 of 4.5 per cent (annual rate), the same rate as in the first half of 1990. With oil prices falling back to levels that prevailed before the Gulf crisis, and non-oil commodity prices continuing to decline, the rate of increase of the total domestic demand deflator slowed, dropping below that of the GNP/GDP deflator. A deceleration in labour costs and the absence of an area-wide effect from indirect taxes may have led to a slowdown in GDP inflation to just under 4 per cent in the second half of 1991; continued terms-of-trade gains may have kept the increase in the total domestic demand deflator below that of the GDP deflator.
In sum, while nominal wages and earnings decelerated steadily through 1991, it has taken some time for this to be translated into slower price increases. However, monthly data for consumer prices are suggestive of some deceleration over the summer and into the latter part of the year (see Chart O in the section on detailed projections and other background information). Overall, the rate of increase in non-food, non-energy prices - a proxy for the "underlying" rate of consumer price inflation - may towards the end of 1991 have returned to, or even fallen below, the rate which prevailed in the second half of 1989, before its increase through 1990. A slowdown in the underlying inflation rate is particularly visible in the United States, Canada, the United Kingdom (even allowing for the effects of mortgage interest), Australia, New Zealand and some Nordic countries. However, in other European countries, upward trends do not appear to have been broken yet.
The OECD unemployment rate could ease back slightly over the projection period, to 7 per cent by the end of 1993, compared with a projected 7.4 per cent in 1992. This reflects projected labour force growth of around 1 to 1 1/2 per cent per annum, while employment growth picks up to around 1 3/4 per cent by 1993. Nevertheless, unemployment is likely to remain high in a number of countries.
Projected unemployment rates over the next two years in many OECD countries would be generally above OECD Secretariat estimates of NAIRUs. Therefore, the short-term prospects point to continued downward pressure on wages. Decelerating wages are projected over the next two years in virtually all OECD countries, with the largest declines in the rate of wage increase occurring in the United Kingdom, Canada, Switzerland and some southern European and Nordic countries.
With relatively stable projected increases in non-wage compensation, the gradual resumption of productivity growth implies further moderation of unit labour cost growth. Between the first half of 1991 and the end of the projection period, unit labour costs in the business sector are projected to decelerate by 1 1/2 percentage points, to grow at an annual rate around 2 1/2 per cent in the second half of 1993. With the aggregate profit share rising slightly as demand growth resumes and growth of labour costs continues to moderate, inflation, as measured by the GDP deflator, may fall by slightly less. By the second half of 1993, the average GDP deflator in OECD countries may be rising at a rate 1 percentage point below its cyclical peak rate in 1990.
Output gaps and inflation
The increase in OECD unemployment since mid-1990 has accompanied a fall of actual output relative to potential output(5), defined as the level of output that is consistent over the medium term with stable inflation. While actual OECD output exceeded its estimated potential level in the period 1988-90, this situation was reversed in 1991. For the major seven OECD economies in aggregate, the size of the negative output gap in 1991 is estimated at around 1 1/2 per cent of potential output (Chart L). There is uncertainty about the precise magnitude of this gap, however, since it depends on an estimate of the level of potential output, which can only be inferred indirectly from the paths of various macroeconomic indicators. Over the period 1992-93, the output gap for the major seven economies is projected to widen slightly to around 2 to 2 1/2 per cent.
There appears to be a relatively close link over time between the size of the output gap and the change in average inflation for the aggregate of the major OECD economies. As a rule of thumb, a negative output gap of 1 per cent has been associated with a reduction in inflation, as measured by the GDP deflator, of slightly more than 1/2 percentage point each year in the major countries taken together. Large divergences from this pattern have not occurred since the first oil shock.
The link between output gaps and changes in inflation is not as clear for all individual countries as it is for the aggregate of the major seven countries. Thus, the projected reversal of the positive output gap for Japan in 1992-93 is not expected to slow inflation noticeably - just as the build-up of the output gap over the period 1986-90 did not lead to a significant acceleration of prices. For Canada and Italy, disinflation may not be as great as suggested by negative output gaps.
On balance, it would seem fair to characterise the projection of the slowdown in inflation as being relatively cautious. Applying the rule-of-thumb noted above, an aggregate negative output gap of 2 to 2 1/2 per cent would be expected to result in disinflation of 1 1/4 to 1 1/2 percentage points. The projection for the major seven countries of a fall in inflation from 3 3/4 per cent in 1991 to around 3 per cent in 1993 may, therefore, be on the low side. However, reductions in inflation from current rates, which in many countries are not far from absolute lows reached since 1970, may meet with more resistance than past falls from higher rates.
(1.) For a detailed analysis of the evolution of labour market
policies in the 1980s, see OECD (1990), Labour Market
Policies for the 1990s, Paris. (2.) See OECD (1991), Employment Outlook, July, Chapter 7,
for a detailed description of unemployment benefit systems
in OECD countries. It also highlights a cross-country correlation
between the structure of benefits and the structure of
unemployment. (3.) On the other hand, there is some evidence that labour market
policies can shift the relationship between unemployment
and vacancies (the so-called Beveridge curve) in a favourable
direction. See Jackman, R., C. Pissarides and S. Savouri
(1990), "Unemployment policies", Economic Policy,
(October), pp. 441-483. (4.) However, there has been a tendency to de-emphasise early
retirement schemes in view of their heavy budgetary costs.
This is clearest in France where such outlays' share halved
between 1985 and 1989 from 1.2 to 0.65 per cent of GDP. (5.) For details on the concept and measurement of potential
output, see Torres, R. and J.P. Martin (1990), "Measuring
potential output in the seven major OECD countries",
OECD Economic Studies, No. 14 (Spring), pp. 127-149.
PHOTO : Chart K. Unemployment developments
PHOTO : Chart L. Price and output gap developments
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Domestic and International Developments; OECD countries|
|Publication:||OECD Economic Outlook|
|Date:||Dec 1, 1991|
|Previous Article:||Financial and fiscal policy developments.|
|Next Article:||Trade and foreign direct investment.|