The basic income experiment in Finland yields surprising results

The results of the Finnish basic income experiment suggest that basic income does not increase employment, and that current employment services are appreciated more than previously thought. Heikki Hiilamo, Professor of Social Policy at the University of Helsinki, comments on the results.

The long-awaited results from the two-year basic income experiment have been published. The results from the experiment, which was conducted in Finland between 2017 and 2018, cast serious doubt on popular beliefs on basic income and employment. In short, carrots do not seem to matter, but sticks are welcomed.

Basic income has been presented as an alternative to workfare policies which require claimants to participate in active labour market programmes, such as training and subsidized work. Those who do not participate may face sanctions, such as the termination of their benefits. Basic income comes with the idea that unemployed people know what is best for them, and that without obligations and sanctions they will be more equipped to navigate the labour market. As people are able to keep the money they receive from basic income even if they get a job, the programme comes with significant benefits.

The aim of the Finnish experiment was to test if the carrot works better than the stick in encouraging unemployed citizens to find new job offers and to seek income from entrepreneurial activities. In essence, the experiment analysed three interlinked mechanisms in the Finnish public social services system.

Currently in Finland, unemployed individuals are required to participate in active labour market policy measures. Those who refuse to participate lose their unemployment benefit for two months and may also face a reduction in social assistance, the last-tier income support system in the country. This is the first mechanism that was studied.

In the experiment, the basic income recipients were not obliged to participate in such measures and faced no sanctions.

Secondly, as elsewhere, all Finnish minimum income benefits are means-tested. This means that income acquired from work is deducted from government-sponsored benefits.

Those who received basic income during the experiment were able to keep all the extra income from work without having to risk losing one euro of the basic income amount.

Thirdly, the recipients of basic income did not need to report their incomes to the unemployment office or to participate in active labour market policy measures. This reduced bureaucracy and insecurity caused by fluctuating benefit levels.

In the experiment, a sum of 560 euros per month was paid to a randomly selected group of 2,000 unemployed Finns on a flat rate unemployment benefit. The basic income replaced the existing unemployment benefits and was paid even if the participants took up jobs.

The rest of the unemployed in Finland who receive a flat rate unemployment benefit, formed the control group. Since the group getting the ‘treatment’, i.e. the 560 euros a month net sum, was similar to the control group in all relevant background characteristics, the experiment mimicked studies in natural sciences and medicine. The idea was that if there were any differences between the treatment group and the control group after the experiment, we would be able to establish a causal loop.

Disappointing employment results

On May 6 2020, nearly 1,5 years after the end of experiment, the final results were released.  The verdict was that there was very little difference in employment or earned income between the groups. Over the first year of the experiment, there was no difference between the groups. The second year, the basic income group worked six days more, but the difference cannot be attributed to basic income alone. A new sanction regime (activation model) was implemented in 2019, which invalidated the research setting.

According to Kari Hämäläinen, Chief Researcher at the VATT Institute for Economic Research, the effects of the second year of the experiment cannot be separated from the effects of the activation model.  In the case of very low paying jobs, the control group actually worked as much as the basic income group. The research group concluded that the huge carrot included in the basic income model did not actually work in increasing employment.

More surprisingly, the results from the final experiment report revealed that many members of the basic income group did not want to abandon employment services and active labour market policy measures. They were in no way obliged to register at employment office but a great majority of them decided to do so. The share of people among the basic income group who were registered in the employment office was at most only 17% lower than among the control group. The research group concluded that the majority of the basic income group did not have a problem with the conditionality of employment services.

The final report also discussed the survey results showing that the basic income group had clearly higher subjective wellbeing across a large variety of measures. The survey was conducted at the end of the study period. There was no baseline survey to analyse changes during the experiment. Therefore, as the research group concluded, it is not possible to determine if the positive results can be attributed to basic income.

After the preliminary results of the Finnish experiment showed no positive employment effects some basic income advocates remarked that there were no negative effects either. However, the results of the Finnish experiment need to be discussed in the context where the experiment was conducted.

The centre-right government implemented the experiment to see if improving financial incentives would lead to increased employment. The aim was not to study non-conditional cash transfer would reduce employment as a consequence of guaranteed basic income. For that purpose, other population groups such as freelancers and low-paid workers should have been included in the trial. The government was also not primarily interested in improving the wellbeing of unemployment benefit claimants. This explains the absence of baseline survey which invalidated the findings on broader effects of basic income. This is a major weakness since wellbeing effects are at the core of the global basic income debate.

The Finnish basic income experiment generated huge international attention. It was the first statutory nation-wide randomised field experiment on basic income. However, contrary to a common misunderstanding, the experiment did not study universal basic income, but partial basic income targeted to young and long-term unemployed. The government of Finland cannot be blamed for the fact that the international discussion around the experiment was completely blown out of proportions.

There are still important take-home messages. The experiment demonstrates that the problems that young and long-term unemployed individuals experience in finding work do not relate to bureaucracy or financial incentives. Similarly, the results show that financial incentives for employment and reduction of benevolent bureaucracy should not be used as selling points for basic income. The results also call into question the punitive workfare policies based on the deterrent effect of unemployment services.

This notion gives room to develop employment services and cash benefits on the basis of mutual trust between the claimants and the policy makers.


Read more:

Universal ba­sic in­come ex­per­i­ment in Finland: new research re­veals sig­ni­fic­ant di­versity in par­ti­cipants’ ex­per­i­ence

Heikki Hiil­amo: "Dis­ap­point­ing res­ults from the Finnish ba­sic in­come ex­per­i­ment"

University of Helsinki experts on the basic income experiment


Researcher's View articles are science and research-related statements from University of Helsinki researchers.