Economic effects of working time cuts examined

In March 2002, Sweden's National Institute of Economic Research published the findings of research into the economic effects of a general cut in working time. It predicts generally negative effects, at least in the short term, on production, economic growth, consumption and wages. The Institute's report will feed into the work of a parliamentary commission examining a possible statutory reduction in working hours.

In 2001, the parliamentary commission examining new working time and annual leave rules (SE0101176N) asked the National Institute of Economic Research (Konjunkturinstitutet, KI) to examine the macroeconomic effects of a legal cut in working time - the current statutory normal working week stands at 40 hours. The results of the study were published in March 2002. The KI economists examined four alternative models:

• a cut in statutory normal working time from 40 to 38 hours over 2004-7;

• a cut in statutory normal working time from 40 to 35 hours over 2004-13;

• five extra days' statutory leave per year introduced over 2004-7 through amendments to the Annual Leave Act (semesterlagen); and

• five extra days of leave for a variety of reasons (paid as ordinary working time) introduced over 2004-7 by a new Act on Paid Leave (lag on betald ledigheit).

In basic economic terms, the first two alternatives are the more expensive ones for society in terms of reduced economic growth, whereas the fourth alternative is the cheapest, KI states.

Having conducted similar research in connection with various government-appointed commissions on working time over recent years (SE0004139F, SE0011173F and SE0101176N), KI states in its analysis that, in the short term, a cut in working time would lead to a higher number of employed people. The resulting somewhat lower unemployment would, however, tend to raise wage levels. It is also probable that employees would seek through higher hourly wage increase demands to prevent their annual pay decreasing in line with the working time reduction.

Higher pay in Sweden than in other European countries will lead to reduced investments, subdued growth and less employment, states KI. After some time, employment levels would fall back and, for a transitional period, they would be lower than in a situation where there had been no working time cut. Therefore, a working time reduction may not lead to permanently lower unemployment. The long-term unemployment situation is governed by the conditions for wage bargaining, labour market policy, education policy and tax and transfer (social insurance etc) systems.

Reduced production and growth

A cut in working time will, in the long run, decrease the use of capital in production as well as production itself, to about the same extent as working hours are cut, KI states. A decrease in weekly statutory normal working hours from 40 to 38 hours would possibly decrease 'real' working time (faktisk arbetstid), and definitely cut production, by more than 3.5% in the long run. 'Real' working time means the actual working hours in a given week, where the standard working time laid down by law and collective agreements has been adjusted for absence of various kinds and overtime work

A transition from a 40-hour normal working week to a 35-hour normal working week would, according to KI, mean in the long run a decrease in real working time and production by about 9%. A cut in working time by means of five days of additional leave is calculated by the economists to decrease production by 1.2%-1.5%, in the long run.

The effect, states KI, would be a reduction in annual growth in GDP during the period that a cut in working time were introduced. This reduction would vary according to the method chosen, but between a quarter and a half of the anticipated economic growth would be eaten away by the working time cuts.

Lower wages

The decline in production brought about by working time reductions would affect the whole Swedish population in the form of lower annual wages, lower transfer payments (ie tax and social security) and lower public consumption, KI states. The level of public services provided - eg schools, higher education, sick care, childcare or care for older people - would be lower than it would have been without a cut in working time, with the reduction corresponding roughly to the working time cut implemented for public employees. With retirement pensions currently based on income developments, there would automatically be lower growth in pensions. Lower wages and lower transfer payments would lead to weaker private consumption. On the other hand, employees would have more time off in which to spend - though this would not apply to groups such as pensioners, unemployed people and students.

Profits in industry would be subdued by a working time reduction, at least initially. Business owners, both Swedish and foreign, would have to bear part of the cost of the cut in working time. Profits in Swedish industry cannot, however, deviate from the profit level in the surrounding world, states KI. A cut in working time would thus lead, in the long run, to a lower annual wage for employees, whereas the Swedish and foreign owners would eventually be compensated.

Demographic changes

In the coming decades the number of people in Sweden aged between 16 and 64 will decrease, at the same time as the total population and especially the number of older people increases. The change will be particularly evident in about seven to 10 years time, when the large age cohorts born in the 1940s reach pensionable age. From a demographic point of view, states KI, this development will be characterised by a conflict between a diminishing labour force and an increasing demand for services from pensioners, especially on the sickness and healthcare sector. A cut in working time would risk making this conflict worse, according to KI.

KI has delivered its report to the working time commission, which is preparing proposals to be submitted to parliament. A possible, even probable, proposal for a cut in the working time is expected within a few months, and at any event before the general election in September 2002.

Commentary

In December 2000, the Swedish government announced the establishment of a new working time commission to examine the entire system of legislation on working hours and leaves. This was the latest development in a long-running but inconclusive public and political discussion, lasting over 10 years, which included the establishment of several different commissions to look into the issue of possible working time cuts. The National Institute of Economic Research was involved in the earlier commissions' work, giving a rather pessimistic, or maybe realistic, picture of the possible effects of a working time reduction.

From the economic point of view, it is argued that an overall cut in working time, for example from the current 40-hour normal week to 35 hours, it would not be favourable for Sweden. One could, however, argue that the current very costly high levels of ill-health and sickness absence among workers reflects increasing stress in working life, which might be addressed by decreased working time. This is not discussed in the KI report, but will certainly be considered in political discussion.

In April 2002, the three central trade union confederations - the blue-collar Swedish Trade Union Confederation (Landsorganisationen, LO), the white-collar Swedish Confederation of Professional Employees (Tjänstemännens Centralorganisation, TCO) and the professional Swedish Confederation of Professional Associations (Sveriges Akademikers Centralorganisation, SACO) - presented a joint proposal for a statutory cut in working time. The unions want five extra days off per year for all employees. It should, however, be possible for collective agreements to deviate from this approach, for example trading off the days of leave for higher pay. Sectoral unions, however, do not all appear to be convinced of this solution. The influential Swedish Metalworkers' Union (Svenska Metallindustriarbetareförbundet, Metall) would prefer a sixth week of annual leave, something that is opposed by the central confederations. The legal norm for annual leave in Sweden is currently five weeks. (Annika Berg, Arbetslivsinstitutet)