Labour reserves and the risk of labour market tightening

typically in the case of market services, may trigger a rise in real wages (real labour costs), exerting inflationary pressure.

1.2 Labour reserves and the risk of labour market tightening

The labour reserves for the extensive expansion of the private sector are to be sought within certain unemployed and inactive groups, as well as, in respect of certain sectors, within the labour force of the public sector. Due to the nature of economic growth in Hungary,effectivelabour reserves primarily consist of highly educated and skilled groups and/or people who have only been out of work for a short period of time. However, by the late nine-ties, economic growth seems to have absorbed the unemployed and inactive groups that were easy to reactivate. This necessitates an analysis of the possibility of labour market bottlenecks emerg-ing. Such an analysis can be conducted partly by measuring the level of available labour reservesand partly by studying the in-tensityof labour utilisation.

The utilisation of the capacity provided by the total domestic labour force can be described using several indicators. First and foremost, the economic activity or participation rate is frequently used as an indicator of potential labour supply, showing the por-tion of the 15–74 age group that are part of the labour market, ei-ther as employed or unemployed people. Employment growth and the decline in inactivity have taken place simultaneously with a decline in this age group, causing this indicator to rise to 53.5% during the third quarter, still far below the initial levels seen in 1993 or those in more advanced economies. As noted above this does not imply the existence of voluminous labour re-serves, as not each and every inactive or even unemployed group can be regarded as part ofeffectivelabour supply.

Theemployment rateis different inasmuch as, being a mea-sure of the employed proportion of the 15–74 age group, it indi-cates the sum of the unemployed and the inactive as the compo-nents of the unutilised labour force. By the third quarter of 2000, employment growth had pushed this rate above 50%, close to the highest value (52.5%) measured since the statistics were intro-duced in 1992.

Finally, the degree of labour capacity utilisation can be de-scribed in terms of theunemployment rate, which regards the unemployed only as the reserve force of labour. According to this indicator, derived from the internationally (ILO) compatible LFS of the CSO, the unemployment rate has fallen to 6.2%. The impli-cation is that the utilisation of the potential labour capacity has reached (100 – 6.2) = 93.8%. In a narrower interpretation, only the unemployed group that constituteseffectivelabour supply should be taken into account. However, no detailed data are available on education and skill levels which would be required to separate outeffectivelabour supply. One sign of the existence of regional labour market bottlenecks is that the rate of unem-ployment in the western areas of Hungary, the centre of indus-trial recovery, has fallen to 4–5%.

In addition to the CSO data, the level of labour reserves can also be measured using supplementary statistics published by the National Labour Centre. Claimant countunemployment re-corded in the administrative database has been falling steadily, bringing the third-quarter rate of unemployment⁴down to 9%, compared with 10–11% in 1998. As noted earlier,⁵a flow concept may be a better guide to labour market developments than the stock rates. The number ofcollective redundanciescould serve as an approximate indicator for the number of peopleentering unemployment. Collective redundancies represent a part of ef-fective labour reserves which, by definition, have only been out of work for a short period of time and are therefore relatively easy to reactivate. At the same time, the administrative nature of the data restricts the scope of inferences that can be drawn. Nev-ertheless, the tightness of the labour market is reflected in the fact that, as a result of the steady decline seen since late 1999, the ratio of collective redundancies to the number of employed people has dropped to a “historical” low of one-tenth of the initial level seen in 1993(see Chart IV-2).

As noted above,public sector redundancies may add to avail-able effective labour reserves in certain sectors of the economy.

The labour force shed off due to the contraction of the public sec-tor labour market could be rechannelled into those private secsec-tor services which rely on white-collar labour, requiring at least sec-ondary school education, but no specialised skills. Our data show a continuation in the decline in public sector employment, primarily affecting the manual labour force (in contrast to white-collar employment, which seems to be even expanding in certain sectors). Thus, developments in the public sector are not likely to ease labour market tightness in the private sector.

The shrinkage of labour reserves will primarily entail a more intense utilisation of the employed labour force. Wage rises, which pose the threat of cost-push inflation, are likely to ensue only when the first response has run up against ‘natural’ or regu-latory (labour laws) restrictions. The intensity of labour utilisa-tion is measured by average hours worked by each employed person. As noted above, long data series are only available in re-spect of the manufacturing manual labour force, while data for the private sector as a whole are only available for the post-1998 period. Based on the former source one can say that monthly hours worked by manual workers in manufacturing have risen substantially, by roughly 1 hour on average, since 1993. The cur-rent average intensity of 37.4 hours per week appears to be high in international comparison, with only the Czech Republic, known for exceptionally high rate of manual labour hours worked in manufacturing, showing higher figures than Hungary (see Box IV-1).

Developments over the past few quarters have reflected a generally rising trend in average hours worked in key manufac-turing sectors since the Russian crisis. Hours worked in the tex-tile, wood and paper industries have surged, which signals re-structuring efforts and improvement in efficiency rather than a tight labour market, considering this year’s large-scale layoff of manual workers (the textile and clothing industry recorded espe-cially low numbers of average hours worked compared with other countries)(see Chart IV-3).

6 7 8 9 10 11 12 13 14 15 16

Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 ILO unemployment rate

Claimant count unemployment rate Collective redundancies (right-hand scale) Per cent

1997 1998 1999 2000

Per cent Chart IV-2 Labour reserves:

Unemployment rates and the rate of collective redundancies

* Based on seasonally adjusted series.

4This rate is the Bank’s own calculation, see page 46, 2000 June Report.

5Page 42, 2000 September Report.

It has only been possible to record hours worked inmarket services since1998, in the form of simple annual indices. This sec-tor tends to show a different picture than manufacturing. While in 2000 Q3, the hours each person worked in manufacturing in-creased relative to a year earlier, market services recorded flat-to-falling numbers. Developments over the past two years reflect no trend-like increase in labour force utilisation (indeed, the monthly and quarterly fluctuations are due to movements in the number of working days). On the other hand, it is a fact that the hours worked in services, especially in trade and repair, were significantly higher than in manufacturing, by 2–3 hours monthly, on average. This implies that in the area of market sec-tor services, entrepreneurs have only limited freedom in intensi-fying the utilisation of the employed labour force in an attempt to adapt to the relatively tight labour market, due to ‘natural’ and regulatory restrictions on maximum working hours. Therefore, an acceleration in wage inflation, carrying with it the threat of in-flationary pressure, is a more imminent danger here than in man-ufacturing.

36 37 38 39 40

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 36 37 38 39 40 Food and drinks Textiles, wood and paper

Manufacturing average

Chemicals Metals

Machines

1997 1998 1999 2000

Hours/week Hours/week

Chart IV-3 Average weekly hours worked

* Based on seasonally adjusted data series.

Box IV-1 Hours worked in Hungarian manufacturing in an international comparison

Hours worked is often regarded as a key measure of labour market conditions, giving a better understanding of the relationship between demand and supply and changes therein than the number of employed or unemployed peo-ple. This is because entrepreneurs will tend to respond to information on cyclical expectations on the ‘intensive margin’, i.e. by changing working hours, rather than on the ‘extensive margin’, i.e. by changing the number of em-ployees. This is especially true of the highly regulated environment of European economies, where the costs of dis-missal may be substantial (also incorporated into hiring costs by forward-looking entrepreneurs). Thus, a change in hours worked is one of the key indicators of the cyclical position of many countries. Thelevelof hours worked is an interesting measure of capacity utilisation in its own right. However, it is difficult to determine total capacity utilisa-tion, as neither the natural upper limit of 168 hours nor the ‘statutory’ 40 hours provides much of a clue.

In terms of economic theory, total capacity utilisation could be defined as the level of hours above which there will be a rise in real (hourly) wages or above which wage inflation will begin to accelerate. Theoretically, this level is affected by several factors. An increase in thecapital intensityof production or statutoryregulationmay reduce the hours worked by boosting real wages or imposing costs on overtime or shift-work patterns, respectively; (regula-tion relating to dismissal costs will typically affect the intensity of the cyclical movements of the hours worked). Fur-thermore, social and tax regulations providing for explicit maximum working hours or other type of working condi-tions, and the market positions of companies will also influence the ‘equilibrium’ level of the number of hours.

In the absence of a model required for the application of our definition, an international comparison of the hours worked may be a source of information. Hungarian manufacturing hours worked appear to be ‘in mid-field’ in a re-gionalcomparison: less advanced countries, such as Russia, Slovakia and Croatia, are characterised by lower hours worked. By contrast, countries with better economic performance, although somewhat less advanced in terms of restructuring, such as the Czech Republic or Slovenia, have higher rates. In aEuropeancomparison, Hungarian fig-ures (as well as Czech and Slovenian) appear to be high, while most advanced economies have lower (e.g. Austria and Norway) or similar figures (e.g. Belgium, Portugal, Finland). In a sectoral breakdown, despite the strongly rising trend, the number of hours worked in themachineryindustry fall short of the figures for European countries, to-gether with figures for the chemical industry and non-metal sectors(see Chart IV-4).At the same time, aglobal com-parison reveals that a number of emerging countries which have achieved fast industrialisation (e.g. Mexico and Ko-rea) work considerably higher hours than Hungary.

If the level of total capacity utilisation relating to the Hungarian labour force is identified with the highest number of hours worked in the sector in question, it can be seen that the level of hours worked in Hungary is relatively high as there are few European countries with higher levels. It should be considered, however, that Western European countries incur much higher real wage costs than Hungary, due to high capital intensity and welfare regulation; thus the Hungarian ‘equilibrium’ level of hours worked must be higher. The situation is just the other way round com-pared with newly industrialized Mexico and Korea. Thus, in Hungarian manufacturing the hours worked represent-ing total capacity utilisation(see Chart IV-5)should be seen as lying somewhere in the middle between the figures for Western European countries and emerging economies outside Europe (probably slightly closer to the former).