Unemployment insurance protects workers from major loss of income during economic recessions and helps to smooth out prosperity over time. It is also an important automatic stabilizer that dampens economic fluctuations. The unemployed often have high consumption rates and spend any additional income during a recession. Income security also supports the creditworthiness of households. Solidly financed unemployment insurance stabilizes consumer demand and strengthens an economy’s resilience.

In a recession, growth slows, demand collapses, companies lay off workers and financial markets suffer. Economic stimulus packages are intended to mitigate these negative effects. In 2008, during the great economic crisis, governments around the world relied on additional public investments, tax cuts, or easier access to credit for companies and households, often at a cost of more than 2% of gross domestic product.

Nevertheless, the crisis hit some regions harder than others. Economic stimulus packages of different sizes are of little relevance to this. Much more important is the fundamental resilience of an economy. This depends heavily on the effectiveness of automatic stabilizers. Unlike stimulus packages, they do not have to be activated in the event of a crisis. You can automatically and without delay significantly reduce economic fluctuations. A particularly important automatic stabilizer is unemployment insurance.

As unemployment rises, more households lose income. It also takes longer to look for a job, so people consume less. Even those who do not lose their jobs suffer from greater insecurity and save more. In all cases, unemployment insurance helps to support demand. On the one hand, it increases the disposable income of the unemployed. In addition, the risk of those who are afraid of losing their jobs is reduced. However, there are also counterproductive effects. For example, more generous unemployment benefits can also lead to higher unemployment in the long term if it reduces the intensity of the job search and the willingness of the unemployed to accept a new job offer.

Such countervailing effects make it difficult to consider how unemployment insurance should be designed in a favorable manner. Despite some influential studies, there is still a lack of practical guidance for policymakers. Economists Marco Di Maggio and Amir Kermani from the Universities of Harvard and Berkeley want to empirically clarify the question of which mechanisms of unemployment insurance can improve the stability of an economy. Above all, they examine the influence on fluctuations in employment and consumption. Unemployment insurance also affects the ability of households to meet their financial obligations. Therefore, they also examine the effect of unemployment insurance on the availability of credit that households can use to sustain their consumption.

The economists exploit a peculiarity of unemployment insurance in the United States. States can largely decide for themselves how generous their unemployment benefits are. These differences allow Marco Di Maggio and Amir Kermani to compare the impact of economic downturns at the local level. They analyze employment and economic trends in counties that differ primarily in their unemployment insurance benefits but are otherwise very similar.

Maggio and Kermani use data on employment and unemployment insurance from 1990 to 2013. They measure unemployment insurance generosity using the replacement rate. This records what proportion of the previous salary the unemployed receive after losing their job. Replacement rates vary greatly between states, ranging from $190 to $400 per week. The average replacement rate was 36.4% with a standard deviation of 3.9 percentage points, or 11% of the average.

How to measure differences in local labor demand? The economists use the concept of the “Bartik shock” for this. You first look at the development of the labor market at the national level by sector. Then they weigh the sector-specific labor demand with the sector’s share of employment in the respective district in 1998. This gives them a measure of the differences in local labor demand. For example, if national service and manufacturing employment declined by 2 and 4%, respectively, and a county has half the workforce in each of those two industries, that equates to a minus 3% Bartik shock in that county. The advantage of this approach is that the local labor demand determined in this way is not falsified by district-specific changes in the labor supply. In the data, the shocks to local labor demand vary between 6.9 and +3.3%.

Generous unemployment insurance can mitigate the job slump in an economic downturn. If the replacement rate increases from 36.4 to 40.3%, the slump in job growth decreased by 9%. This particularly supports employment in the service industries.

Unemployment insurance helps to support employment and consumption during a crisis. For example, a drop in local labor demand (as measured by the Bartik shock) reduces employment growth by 9% less if a county’s unemployment insurance replacement rate is one standard deviation higher, specifically 11%. This supportive employment effect is particularly strong in services such as retail and gastronomy, where generous unemployment benefits even reduce the local drop in employment by 16 to 20%. If higher replacement rates mitigate the decline in disposable income, even the unemployed will not have to reduce their consumption as much. This props up demand in a downturn, reducing the job slump. Unemployment benefits benefit precisely those households with a high consumption rate that spend most of their additional income and do not save.

If the unemployment insurance replacement rate is 11% higher, then in a recession the local decline in consumption for durable goods is moderated by 12% and for non-durable goods by 6%.

The study goes on to show that more generous unemployment benefits after a local shock also mitigate the drop in consumption by an average of 7%. The decline in consumption for durable goods is up to 12% less severe, while demand for consumer goods is 6% less severe. Everyday consumer spending such as food, for example, only fluctuates slightly over the course of the business cycle. The purchase of durable goods such as a new car, on the other hand, can easily be postponed to a later date. Since consumer demand for such goods fluctuates much more, more generous unemployment benefits can provide greater stability, especially for durable goods.

What role does the financial sector play? The economists find that households often take on more debt during an economic downturn due to higher credit card debt. In this way, they can smooth out their consumption over time and maintain their standard of living in the short term.

It turns out, however, that unemployment benefits do not have a significant impact on borrowing behavior. This suggests that while households would like to borrow more, it is difficult to obtain during a recession. As a result, the decline in consumption and demand is more pronounced, which tends to reinforce the downturn. Even if additional debt is not possible, generous unemployment insurance can at least cushion the decline in lending. This is because households suffer less loss of income despite being unemployed and are therefore better able to meet their financial obligations. There are fewer loans. The risk of the lender decreases. They are therefore more willing to keep lending.

A financial accelerator of recession occurs when households are unable to repay loans because of unemployment. More generous support allows them to still meet their obligations and thus prevents high loan defaults and a sharp drop in lending.

The empirical evidence makes it clear that higher unemployment insurance replacement rates have a stabilizing effect in a recession. Two mechanisms are central here. First, higher replacement rates support disposable income and thus consumption and demand. Second, the affected households are more likely to meet their financial obligations. This lowers credit risk and prevents a credit crunch. In both ways, stronger demand weakens the cyclical decline in employment.