Minna Ylikännö

Research team leader, Social Insurance Institution of Finland



Olli Kangas

Professor of Practice, University of Turku



The evaluation of the Finnish BI Experiment reveals the problems in the current conditional labour market policies

The first results of the Finnish Basic Income (BI) experiment were published at the beginning of February 2019. According to the results, the benefit mimicking basic income did not have any positive employment effects when compared to the control group, i.e. those not included in the experiment. There are several possible reasons for this result.

Firstly, we must keep in mind the specific characteristics of the target group of the experiment. Both the treatment group and the control group were drawn from the unemployed job seekers receiving unemployment benefits from the Social Insurance Institution of Finland (Kela) in November 2016. In order to understand the specificity of the group, we have to take a cursory look at the Finnish unemployment system.

An unemployed individual who is a member of a voluntary unemployment fund and meets the employment condition of 26 weeks during the previous 28 months is eligible for earnings-related unemployment benefits. At the average wage level, the benefit is about 65 percent of previous earnings. The benefit can be paid up to 300-500 weekdays depending on the work history and the age of the claimant.

An unemployed individual who is not entitled to the earnings-related unemployment benefit (the right to the benefit has expired due to long-term unemployment or the unemployed individual does not yet fulfill the employment condition) can qualify for unemployment benefits paid by Kela [1]. These Kela’s ‘basic security’ benefits (basic unemployment allowance and labour market subsidy) are not income-related. In principle, the labour market subsidy is income-tested and the duration is unlimited.

At the end of the year 2017, 369,058 people received unemployment benefits in Finland. Out of them, 196,452 (53.2%) received labour market subsidy and 31,460 (8.5%) basic unemployment benefits. Hence a minority, i.e. 138,949 (37.6%), were entitled to an earnings-related daily allowance from unemployment funds.

When consulting the above percentages, we can argue that the basic income experiment covered the population group that represents a majority of unemployed job seekers in Finland. However, despite this, the experiment was targeted to the long-term unemployed or those unemployed with a very short working history. Maybe the experiment would have yielded different results if it had included those unemployed individuals receiving earnings-related unemployment benefits, not to mention low-income earners, free-lancers and micro-entrepreneurs [2].

Secondly, at the beginning of 2018, the center-right government implemented a new labour market policy measure, namely ‘the activation model’. The aim of the reform was to increase labour market participation by sanctioning the unemployed by benefit cuts if specific activation criteria are not met [3]. While the activation model contaminated the control group, the results may be distorted.

A third possible explanation is that people do not react on monetary incentives as strongly as is usually supposed e.g. in the economic literature. But once again, we must keep in mind the specific characteristics of the target group. It may be that this group would react more strongly on services than on income transfers.

The fourth explanation to the “zero” result may relate to the possibility that the effects of a basic income counteract each other. In the treatment group, the strong monetary incentive to accept jobs (€560 net a month) might have increased employment, while the less stringent conditionality might have made the BI receivers ‘lazier’. Thus, these opposite effects may have repealed each other resulting in an overall zero outcome.

Too strong an emphasis on labour market behaviour?

The proponents of basic income disagree with the above view of laziness. On the contrary, they argue that by empowering people and giving them the freedom to choose for themselves the outcomes are more positive than what is achieved through sanctioning the unemployed. As desired outcomes of basic income, they emphasize not only the increased activity in the labour market but also increased the wellbeing of the citizens.

In fact, the proponents of basic income hardly consider the activation to employment as the main aim of BI, but rather the wellbeing of its receivers. For the supporters of BI, the economic activity e.g. in the labour markets is not a sin non qua for a functional society even if it was widely considered to be the prerequisite for it, at least if we aim to obtain the present welfare state model that is financed via income tax.

When the Finnish government decided to launch the basic income experiment, its main interest was in the employment effects of BI. The government obviously wanted to see if BI is good for employment. Thus, the government’s goal was rather limited and only when discussed in the parliamentary committees, the interest was laid in the wider wellbeing aspects of BI. However, the main interest still was – and still is – in the labour market behaviour of the BI recipients.

The preliminary results for the first year do not display any employment effects, making the opponents of the BI claim that the BI does not solve a low labour supply. However, we can turn the focus upside down. The zero result is an interesting one. According to the proponents of mainstream labour market policies in Finland, we should have witnessed a clear decrease in the labour supply among the basic income recipients compared to the control group. However, we did not observe anything like that. Neither was the employment rate better in the control group that remained in the current, conditional and punitive benefits system.

Hence, in regard to the long-term unemployed (which was the case for most of the recipients in Finland’s basic income experiment), the mechanism improving their labour market status is much more complex and perhaps more attention should be paid in the wellbeing effects of the experiment than on the employment effects. This is done in many other experiments outside Finland – in the evaluation of the experiments the focus in on the sine non qua for activity, namely wellbeing.

What should then the political decision makers do based on the results? Shall they abandon basic income, if it does not increase employment, even if it improves people’s well-being? Or shall they abandon basic income, while strengthening the conditionality of the existing system in order to more effectively “activate” labour, even if it decreases people’s wellbeing? Considering the recent developments in Finland’s labour market policies, one should perhaps bet on the latter option.




[1] If the unemployed person violates the Unemployment Security Act, he/she may lose the right to the unemployment benefits and he/she may to apply for last resort social assistance of in need for financial support.

[2] Explanations why the unemployed were the target group of the experiment are given at https://tutkimusblogi.kela.fi/arkisto/3316.

[3] The activity is monitored in periods of 65 days of payment. The activity requirement are met if an unemployed person, over the course of a single payment period, has been in salaried employment for at least 18 hours, or have earned an income of at least EUR 245.64 from self-employment, or if he/she has participated in five days of employment-promoting services or some other employment-promoting activity arranged by the Employment Services.

About Guest Contributor

Guest has written 183 articles.