Economic incentives, employment and health
Not peer reviewed
MetadataShow full item record
The goal of this thesis is to shed new light on the mechanisms behind the high rates of disabilityrelated non-employment in Norway, and to find explanations for the apparent rise of labor market exclusion problems within some demographic groups. The thesis focuses on the role of economic incentives, for individuals as well as firms, and on possible trends in the competitive environment that may (or may not) have changed the health- and productivity requirements in the labor market.
In the first paper we study whether and how recipients of temporary disability insurance (TDI) respond to economic incentives. In order to identify causal effects, we make use of a reform of the TDI system in Norway which was implemented in January 2002. The reform involved a new principle for calculation of the benefits, which changed from being based on the entire income history to being based on income in the last year (or the last three years) prior to disablement. In addition, the minimum level of benefit was raised while the maximum level of child allowance was reduced. This result in changing benefit level, where the benefit level increased for some individuals and reduced for others. We find that the benefit level has a causal impact of the outcome and duration of TDI. According to the point estimates, a 10% cut in the benefit level would induce a 3.3% increase in the transition rate to employment, 2.5% increase in the transition rate to permanent disability, and a 3.9% increase in the transition rate to unemployment. The results are in line with previous findings indicating that there is a significant labor supply potential among temporary disabled people, which can be realized by financial incentives. However, it does not necessarily follow that cutting benefit level is the desired policy from a welfare perspective. Many of the recipients suffer from severe physical and mental illnesses and generous benefits protect them, as well as their dependents, from poverty. Additionally, it provides claimants having the capacity to return to work, with more time to find a suitable and viable job match.
The second paper studies the effect of firm incentives on sickness absence behavior. In most of the industrialized countries (including Norway) the employers are responsible for the costs during an initial period of sickness absence spell, after which the public insurance system covers the costs. Hence, the employers have incentives to prevent short-term absences. But when absence spells stretch beyond the co-payment period, employers may not put much effort in facilitating a quick return to work, since return to work potentially involves new absence spells where the employers are again financially responsible. We examine the impacts of employers’ incentives by exploiting a reform in the Norwegian sick leave insurance scheme. The reform was implemented in 2002 where employers’ pay liability was removed for pregnancy-related illnesses. The intention with this reform was to make it more attractive for employers to hire young women. Our findings show that firm incentives actually affect sickness absence behavior by raising short-term absenteeism significantly, while the duration of long-term spells declined. According to the pointestimates the reform increases the probability of starting a period of sickness absence by 10 %, but the probability of ending a period of sickness declines by 12 % for spells exceeding the earlier copayment period. We also find some evidence indicating that the reform actually affect the jobopportunities for young women positively. By following individuals after graduation at school, the reform raised the employment propensity one year after graduation by around 1.5 percentage points for young women in general and by 3.0 percentage points for those who were pregnant at the time of graduation. This implies that there is a trade-off between incentives for sick-leave prevention and incentives for employing workers with high expected absenteeism.
The third paper investigates how exogenous changes in employment opportunities influence take up of disability insurance. Exogenous variations in employment opportunities are measured by variation in firms’ economic performance – including profitability, downsizing and firm closure – and fluctuations in local industry-specific labor market conditions. The data we use is Norwegian employer-employee registers together with firms’ audited accounts and information from the bankruptcy courts. With data about bankruptcy we are able to distinguish mass layoffs from organizational restructuring, demergers, and takeovers. The estimation results show that job opportunities have significant impact on take up of disability benefits, particularly for men. Job loss, in terms of bankruptcy, more than doubles the risk of entering permanent disability retirement for men while raising entry by approximately 50% for women. Furthermore, it doubles the risk of nonparticipation for both men and women. We also detect that other indicators, as profitability, downsizing and local labor market tightness affect the probability of claiming disability benefits, as well as the probability of being outside the labor force. Putting altogether, the paper shows that there is a considerable element of substitution between unemployment and disability insurance schemes.
In the fourth paper, I explore how employment propensities and earnings of vulnerable groups have developed relative to the population at large. Vulnerable groups are defined as individuals having either poor health, low cognitive ability or coming from low socioeconomic classes. My main indicator of poor health is low birth weight, which is observed for both men and women. In addition, I use information about height, Body Mass Index (BMI) and cognitive ability measured at age 18-19 for men entering the military service. Socioeconomic class is defined according to parents’ earnings rank during their age 50-54. For men, lower birth weight and underweight at age 18-19 has become a stronger predictor of low earnings and non-employment, while there is quite constant effect of height and obesity. For women, where birth weight is the only health-measure I have, I do not find any evidence of changing impact. The influence of cognitive skills on labor market performance has become less important over time, which is due to decreasing returns to high ability. The most striking finding, however, is that poor social background has become a steadily more important determinant of non-employment and low earnings.