On the fringe of the labour market
Movements in and out of the ordinary labour market

Jens Clausen, August 2009

• Download AKF Working Paper in pdf-format (Danish)

Summary

The purpose of the working paper is to provide an overview of and analyse the background to movements on the fringe of the labour market:

• How has the number of people on the fringe of the labour market, who are fully or partly supported by transfer incomes, developed?
• Has the number of transfer income recipients outside the ordinary labour market developed according to the demographic changes or have other factors been at play?
• If the number of persons receiving certain transfer incomes increases, does that mean that the number of people receiving other transfer incomes declines?

The study shows that between 1997 and 2006 the number of persons outside the ordinary labour market grew despite a significant decline in unemployment. It is noteworthy that the number of persons leaving the labour market to receive sickness benefits increased by 50 per cent, although no demographic changes warranted such an increase. There are signs that the labour market has become less inclusive, as the risk of slipping out of the labour market has increased without a correspondingly higher chance of returning. This reduced inclusiveness is for example seen relative to sickness, as fewer recipients of sickness benefits have returned to the labour market. The conventional wisdom seems to be that if the number of recipients of certain benefits increases, the number of recipients of other benefits falls. Overall, however, the tendency of the benefits system to work as communicating vessels is relatively weak.

In the following, recipients of disability pension, cash benefit recipients unable to take on a normal job, people in flex-jobs and recipients of sickness benefits are collectively referred to as persons outside the ordinary labour market. The group of persons within or close to the ordinary labour market consists of people employed in ordinary jobs, students, benefit recipients fit to take on a normal job and persons with unemployment insurance (collectively termed as persons in the ordinary labour market). The idea of this division is that if people in the latter group are not already employed, most of them are well positioned to get a job, either after they have completed their education or based on their previous connection to the labour market.

The main analytical method uses a statistical model to determine the extent to which developments in individual transfer incomes can be explained by demographic changes, i.e. people characteristics such as age, gender, ethnicity, education, health and previous connection to the labour market. That which cannot be explained by demographics is ascribable to economic developments, changes in rules and practice regarding benefits etc. Thus it is particularly interesting to identify the points in time when demographics cannot fully explain the trends in a particular benefit scheme as it may be a sign that changes in the rules and awarding of benefits in the relevant period may have been of significance. Very detailed data from each individual transfer income case were sourced from Statistics Denmark’s personal registers for the period 1997 to 2006 for 20-59-year-olds. The following provides an overview of the findings of the working paper.

From 1997 to 2006, the number of persons spending most of the year outside the ordinary labour market increased by 62,000 in spite of a substantial drop in unemployment during this period, see Table. This is primarily ascribable to the introduction of flex-jobs in 1998, as 40,000 held flex-jobs in 2006. Despite the introduction of the flex-job scheme, the number of recipients of other transfer incomes (excluding rehabilitation) increased.

Table 1
Growth in number of persons predominantly spending the year outside the ordinary la-bour market, 19-59-year-olds, 1997 and 2006, 1,000 persons  

The table shows net figures i.e. the sum of those who respectively leave and return to the labour market. The proportion of persons leaving the ordinary labour market has increased over the period, which is not warranted by the demographic changes. The entire increase in exits from the ordinary labour market is due to increased departures to sickness benefits and the introduction of flex-jobs, as departures to other transfers is either constant or has decreased over the years surveyed. Return to the ordinary labour market largely corresponds to the expected level based on demographic changes. Thus, the ordinary labour market seems to have become slightly less inclusive, as individuals face a higher risk of slipping out without having a correspondingly higher chance of returning to the labour market.

Below we expand on the underlying reasons for the contribution of individual transfer incomes to the growth in number of persons outside the labour market.

For flex-jobs both entries and exits over time are explained by demographic changes. Since the introduction of flex-jobs, the numbers coming from the ordinary labour market have been increasing, and the proportion of people returning has been falling. Flex-jobs seem first of all to pull persons on sickness benefits away from disability pension and the ordinary labour market, and second, to pull persons who have been rehabilitated away from cash benefits and the ordinary labour market. Whether this has a combined positive effect is hard to say; flex-jobs represent a more active state than benefits and disability pension, but it is still subsidised employment, which from a socio-economic perspective is less attractive than employment in the ordinary labour market. There are signs that limited-resource groups have been allotted flex-jobs. When a flex-job ended, fewer therefore returned to the ordinary labour market, while many more were granted disability pensions during the period from 1997 to 2006.

People becoming benefit recipients unfit to take on a normal job constitute almost one third of the total number of annual exits from the ordinary labour market. The transition rate to cash benefits decreased slightly during the period, although to a lesser extent than could be expected based on demographic trends. On the other hand, the proportion of persons unfit to take on a normal job returning to the labour market has been stable, corresponding to the general demographic trend. Fewer were rehabilitated following a term of cash benefit, whereas the transition rate to disability pension increased. In total cash benefit contributed 17 per cent of the growth in the number of people outside the labour market, see Table.

The transfer to disability pension accounts for only 5-10 per cent of the annual exits from the labour market. The exit to disability pension was halved from 1997 to 2000, and from then on it was fairly constant, apart from a temporary sharp increase in 2002. This increase can probably be attributed to many people moving to disability pension immediately prior to the disability pension reform in 2003. The reform meant that only people unfit to take on a job on normal terms or a flex-job were eligible for disability pension. It may be that a particularly high number of people were awarded disability pension immediately before the act came into force, just as the tightening implied by the reform may have contributed to the continuing drop after 2003. The halving of exits from the ordinary labour market to disability pension in the period 1997-2000 can only to a limited extent be explained by demographic changes. But it coincided with the more stringent work ability test requirements prior to the processing of an application for disability pension (1 July 1998) and with the reduction of the reimbursement from the government to the local authorities (1 January 1999). The development in the number of people receiving disability pension has only made a modest net contribution to the increase in the number of people outside the labour market.

During the period under review, the ordinary labour market and sickness benefits made a large but declining contribution to the transition to disability pension. On the other hand, the proportion of transitions from cash benefits and flex-jobs increased. Especially the proportion of people from flex-jobs increased significantly – probably because flex-jobs in some cases served as a postponement of disability pension entitlement.

Sickness benefits account for more than half of all exits from the ordinary labour market. The proportion of people in or close to the labour market annually moving on to receiving sickness benefits increased by about 50 per cent during the ten-year period. It is worth noting that if the development had been determined solely by demographic changes, no increase would have occurred at all. At the same time, the proportion of recipients of sickness benefits returning to the ordinary labour market decreased significantly, which can be largely explained by demographics. This suggests that the labour market has become less inclusive relative to sick people. And the trend in sickness benefits accounts for a fifth in net terms of the increase in the number of people outside the labour market.

In 1997 the transition to rehabilitation accounted for 10 per cent of the annual number of exits from the ordinary labour market, whereas in 2006 this transition only accounted for 5 per cent of the annual number of exits. During the period the likelihood of returning from rehabilitation to the ordinary labour market has decreased relative to what could be expected based on demographic trends. Again this suggests that during the period it became harder for people with relatively weak qualifications to return to the ordinary labour market.

People entering rehabilitation for the first time make the transition almost exclusively from the ordinary labour market, cash benefits and sickness benefits. Exits from rehabilitation are primarily into the ordinary labour market and – to a smaller extent – into cash benefits and (at the end of the period) into flex-jobs. Over the period the transition from sickness benefits increased to constitute almost half of the number of people entering rehabilitation for the first time, without any demographic explanation.

Several examples have been found of correlations between internal movements between transfer income schemes and to and from the ordinary labour market, e.g.:

• Following the disability pension reform of 2003, the number of exits from the ordinary labour market to disability pension were reduced, and the likelihood of benefit recipients unfit to take on a normal job returning to the ordinary labour market increased.
• The introduction of flex-jobs shifted the flows to the effect that fewer people transfer from sickness benefits to both the ordinary labour market and disability pension. Instead the transfer to flex-jobs increased.
• Among those leaving flex-jobs, significantly more transferred to disability pension, while fewer returned to the ordinary labour market or transferred to cash benefits. Moreover, for some people flex-jobs seem to serve as a substitute for cash benefits following a rehabilitation process.

However, there are also several examples of a lack of correlation between the transitions. For instance it does not seem to be the case that if the number of people unfit to take on a normal job transferring to a specific benefit scheme following a term of cash benefits falls, the number of people transferring to other selected benefit schemes increases. As regards social assistance, the sole trend seems to be rehabilitation serving as a substitute for remaining on benefits.

The fact that this survey reveals both correlations and non-correlations in the benefits system is consistent with another AKF survey, cf. Andersen (2006). In that survey the theory of whether the income transfer system follows the principle of communicating vessels was investigated, i.e. do municipalities that succeed in reducing the number of benefit recipients in one scheme end up with more recipients in other benefit schemes? This theory was investigated by examining whether municipalities with a particularly high number of people moving on from a term of social assistance or sickness benefits also have a particularly high number of people moving on to other benefit schemes. The conclusion was that the theory was supported, albeit to a limited extent. Municipalities with few benefit recipients in one benefit scheme do not automatically have many benefit recipients in other benefit schemes.

Link to

• Presentation

Publications