Media Briefings

Unemployment Insurance for the Self-Employed: Danish evidence of the impact on business failure

  • Published Date: September 2013

Self-employed men in Denmark who are covered by unemployment insurance are more likely to become unemployed than those who are not covered. But any increased risk of the failure of their businesses that is created by the safety net of insurance is small compared with the overall risk of business failure.

These are the central findings of research by Mette Ejrnæs and Stefan Hochguertel. Their study, published in the September 2013 issue of the Economic Journal, suggests that only a limited proportion of business failure can be attributed to a lack of effort.

The researchers note that during Europe’s recent recession, there has been increasing interest among policy-makers in providing insurance to the growing group of self-employed people. Spain, for example, has very recently introduced a voluntary unemployment insurance scheme for self-employed workers.

This study investigates whether unemployment insurance for self-employed is a good idea. One of the main arguments against an unemployment scheme for self-employed is the fear that self-employed who are covered by an insurance will exploit the system and declare themselves unemployed whenever convenient. This is an example of the phenomenon of ‘moral hazard’.

The results of this study suggest that self-employed men who are insured are more likely to become unemployed, but that this increase in risk is small compared with the overall risk of failure.

The main challenge when analysing the effect of unemployment insurance on subsequent unemployment is the fact that individuals who have a higher risk are also more likely to take up insurance – for example, if they are having health problems.

Self-employed people who are insured also have a higher risk of unemployment than the uninsured. But this does not indicate the causal impact of insurance: it may simply be due to the fact that those who are insured are different from those who are not insured – a ‘selection’ effect.

To disentangle the causal impact, the researchers use three particular features of the Danish unemployment scheme:

· First, it is voluntary to take up unemployment insurance.

· Second, unemployment insurance is also available for the self-employed.

· And third, eligibility for an early retirement programme is closely linked to unemployment insurance.

The first two features make it possible to compare the risk of unemployment for self-employed people who are insured and those who are not insured. Eligibility for early retirement gave many people an additional reason for being enrolled in the unemployment insurance. By exploiting the fact that a substantial fraction of the self-employed enrol in unemployment insurance to be eligible for early retirement rather than to be covered by insurance, it is possible to disentangle the causal effect of insurance.

Using detailed longitudinal data from official registers of the Danish self-employed for the period 1980-98, the researchers estimate the causal effect. They find that:

· The raw data suggest that self-employed men who are insured are more likely subsequently to become unemployed than the uninsured. The difference in transition rates is about 1.8 percentage points, which is very sizeable compared with a raw transition rate into unemployment of about 2% in the sample.

· But this observed difference is due to both selection and moral hazard. Correcting for age, time and cohort effects, the marginal effect of insurance actually increases to 2.2 percentage points.

· Controlling for individual characteristics, the marginal effect of insurance decreases again to 1.8 percentage points. Taking account of the non-randomness in the insurance choice, the marginal effect drops to 0.6 percentage points.

· This suggests that of the original difference only about 30% is due to moral hazard, while the remaining 70% is due to heterogeneity or sorting. But the overall magnitude is not very big, which suggests that only a limited proportion of business failure can be attributed to the lack of effort.


Notes for editors: ‘Is Business Failure Due to Lack of Effort? Evidence from a Large Administrative Sample’ by Mette Ejrnæs and Stefan Hochguertel is published in the September 2013 issue of the Economic Journal.

Mette Ejrnæs is at the University of Copenhagen. Stefan Hochguertel is at VU University Amsterdam.

For further information: contact Mette Ejrnæs on +45 3532 3062 (email:; Stefan Hochguertel on +31 20 598 6033 (email:; or Romesh Vaitilingam on +44-7768-661095 (email:; Twitter: @econromesh).