Essays on Poverty, Redistribution and Labour Market Policies in Nordic Countries

This thesis consists of an introductory chapter and three empirical essays on public economics and labour economics. The introductory chapter examines the paradox of the Nordic model. It may seem paradoxical how the Nordic countries have combined equality and high taxation with competitive economies. Conventionally, economists have thought that there is a trade-off between equality and efficiency, but the success of the Nordic countries has not been a coincidence. Equality, financial security and dynamic economies can be complementary. The essays examine the evolution of poverty and the effects of social protection systems on income redistribution and employment in Finland and other Nordic countries.

In-work poverty is problematic for the functioning of the Nordic model. The first essay examines the evolution of in-work poverty in Finland between 1990 and 2010. This essay examines in-work poverty with multiple indicators, socio-economic factors behind poverty and income mobility of the poor. The results show that in-work poverty trended upwards but remained at a moderate level. Although poverty risk remained low, poverty risks among risk groups increased substantially. The poverty risk among young persons aged 18–24 increased the most, from four to sixteen percent. Subsistence is more likely to require two earners than before. In this development, single parents were in the weakest position. Social transfers substantially reduced in-work poverty in Finland.

The second essay evaluates the employment effects of a Finnish earnings disregard experiment for social assistance clients using difference-in-differences and coarsened exact matching (CEM). The earnings disregard was introduced as an experiment in 2002, but it became a permanent part of the legislation in 2015. It allowed for earnings disregards up to 100€, and the monthly amount was increased to 150€ in 2005. The earnings disregard is evaluated in an experiment period 2002-2007 when it was applied at the household level. Because the control receiving labour market subsidy did not receive social assistance for one year, the treatment group is to a large extent formed of households that did not receive social assistance for predominantly exogenous reasons or did not apply for it. As robustness checks, the results are estimated without spouses who belong to the control group and using homeowners as a control group. Because the earnings disregard increased eligibility for social assistance, the treatment variable is defined solely on a prereatment period in 2001. On average, the results show no employment effects, but there is some evidence of positive employment effects on women.

The third essay studies the longer-term evolution of key policy outcomes associated with the Nordic model. Based on simulated counterfactuals using EUROMOD, the essay examines the effects of tax-benefit changes on poverty, inequality and employment in Denmark, Finland and Sweden. The Static EUROMOD model is extended to take into account the behavioural effects on employment. The longest examined period covers 2006-2017. The simulated risk-of-poverty rates are supplemented with multidimensional Alkire-Foster (AF) poverty indices. Sweden is expectational among the Nordic countries as both income inequality and relative poverty have increased significantly. According to Eurostat, the poverty risk has reached the EU level. Swedish labour market policies have aimed at efficiency at the cost of equality. The centre-right government’s work line policy 2007-2004 aimed at increasing labour supply with policies such as earned income tax credits, lower benefits and active labour market policies. These policies increased incentives to work, but they widened the gap between labour market insiders and outsiders. The results show that the work-line policy increased poverty risk by one percentage point and Gini by 0.4 point. The poverty risk of the unemployed increased substantially, by 1.6 percentage points. Recent participation elasticity literature implies modest employment effects in Sweden. Still, disposable incomes also increased in the lowest deciles, and Sweden fared well in multidimensional poverty. The results show that inequality and relative poverty increased mainly for reasons other than tax-benefit changes. 1e poverty risk increased by 5.1 percentage points and Gini by 3.5 points during the work-line policy. It is possible that immigration has increased the poverty risk in particular, but the effects of immigration are not isolated in the essay. Compared to Sweden, the Danish flexicurity model has better fulfilled the goals of the Nordic model. The Danish model shows that high benefits can be combined with high taxes and employment. The participation tax rates were 13.6 percentage points and for part-time workers 24.2 points higher in Denmark than in Sweden. Inequality has risen also in Denmark, and policy changes increased Gini by 0.8 point, but they decreased the poverty risk. Finland had the most redistributive policy changes, but it fared least well in multidimensional poverty.