The OECD and the problems of basic income

According to the OECD, basic income (BI) is not an effective tool for reducing poverty. However, the outcome would depend on the model chosen for implementing a BI system, as well as the changes made in other parts of social protection.


The Organisation for Economic Co-operation and Development (OECD) published in May a Policy Brief paper studying the feasibility of a basic income model in four OECD countries, one of which was Finland.

On June 16, Kela organized a seminar in which Herwig Immervoll, a senior economist at the OECD, discussed the findings of his study and analysed the strengths and weaknesses of a BI scheme. After the seminar, the national broadcasting company YLE reported: “Universal basic income might increase poverty and inequality”.

Apart from Finland, the OECD study includes France, Italy and the United Kingdom. The analysis was done with the help of the EUROMOD microsimulation model. In each country, the starting point for the analysis was to take all existing spending on social cash-transfers together and see what level of BI they would amount to. Eventually, the level of BI was set near the existing levels of guaranteed minimum-income benefits for single individuals in each country, adjusted so that it would not increase the public expenditures.

In Finland, this resulted a BI of 527 euros for the working age adults and 316 euros for children and youth under 18 years of age. Those entitled to old-age pensions within the current main statutory retirement age (in Finland over 65-year-olds) were excluded from the BI model.

In the BI model used in the OECD analysis, all existing working-age benefits (including social insurance benefits) apart from cash transfers for housing and disability would be abolished. Also, the zero-rate tax bands of income-tax schedules and equivalent tax-free allowances would be abolished, and all income-tax thresholds would be shifted downwards by a corresponding amount. BI would be made taxable under personal income taxation alongside other taxable incomes.


The OECD model would create many gainers and losers

The most important outcome of the OECD study is that the simulated BI model would strongly impact the income distribution in all studied countries. However, the effects vary greatly among the countries.

In all income groups, the BI model would create many gainers and losers. It would change the net income of most people in one way or another. It would lift some groups out of poverty and thrust others below the poverty line.

The simulated BI model would increase the income level of those small income groups who are currently not receiving any social benefits, or whose benefit level is very low. In turn, those receiving earnings-related benefits or several means-tested benefits would see a decline in their standard of living.

In Finland, those below 65-years-old receiving old-age pensions and single parents with low incomes would be among the losers of the model. The middle-income earners instead would generally benefit from the model.

The conclusion of the OECD is that particularly in countries with a comprehensive social protection BI is not an efficient tool for reducing poverty, since it does not target the benefits effectively. According to the OECD, a budget-neutral BI would not be distributionally neutral. High enough to be socially and politically meaningful and fiscally realistic, a BI would still require tax rises as well as reductions in existing benefits.

A very low basic income, instead, would have little other significance but increase poverty.


The outcomes of BI depend on reforms in taxation and social protection

How the findings of the OECD study are to be interpreted in the Finnish context?

Perhaps the most important issue that the research sheds light on is the fact that there are many institutional challenges in implementing a BI system, and those challenges differ among countries due to their different systems of social security and taxation.

As the OECD report (p. 5) notes, BI as an idea is simple, but the existing social protection systems are not. Therefore, there are grounds to argue that the same model of BI does not fit everywhere. If a reform such as BI were to be carried out, it needs to be adjusted to the existing institutions of social protection and taxation in each country separately. The parameters of the model should be adjusted so that it will not produce excessive changes in people’s incomes.

The greatest problems of the OECD’s microsimulation are that the income taxation is not changed to correspond with the BI model, and that the existing systems are demolished by the same means everywhere without examining the structures of social protection in each country separately. Due to this, BI seems to have unpredictable effects to income distribution.

The income distribution produced by a BI model can be influenced by adjusting the parameters of taxation and social security. In his presentation at the Kela seminar, Herwig Immervoll mentioned that tax reforms should be discussed in parallel with BI. Indirect taxes, such as environmental or value added taxes, have often been proposed as a complementary source for financing a BI scheme, combined with income taxation.

However, the OECD report does not mention these alternatives, and the premise seems to be that taxation in any form should not be increased.

In Finland, as well as in many other countries, some organisations and individuals have launched models of BI adjusted to the local context. Their objective has often (yet not always) been to not radically alter the income distribution or cause reductions in people’s after-tax incomes, especially in the lowest income groups. Microsimulation has been employed at least in the models of partial BI by the Green Party and the Left Alliance, and in the preliminary study for the national BI trial conducted by Kela.

In these models, BI is linked with a reform in income taxation that is designed so that radical changes in after-tax incomes will not occur in any income group. The aim is also to make the models budget neutral, that is, to cover the costs of BI by reforms in taxation and replacing the existing benefit systems. In these models, the old system will be abolished only in those parts where the level of benefits is lower than the BI.

One of the problems with the BI trial currently underway is that due to time constraints, the taxation reform proposed by the research team that designed the experiment was not included.


Will Finland implement a BI?

Though there exist BI models in Finland that would technically allow implementation of a BI system without radical changes in income distribution or public financing, the road of BI will probably be rocky even here.

The preparations of the BI experiment scheme revealed many institutional challenges in implementation of a BI model. The greatest obstacles for a BI are, however, ideological.

In Finland, BI has gained interest especially as a possibility to improve the incentives for paid work. The possibility to combine wages with social benefits more smoothly than today is an issue that no party opposes. Yet, many still find it morally wrong to give people money with no obligations. The opponents of BI fear that the “free money” would reduce people’s willingness to work and give a moral legitimacy to not apply for jobs.

If the only, or at least the most important function of BI is to improve work incentives, the great promises of BI may not be fulfilled after all. The preliminary studies for the BI trial revealed that BI models do not always unambiguously remove incentive traps, if parts of the old social security stay intact.

However, it seems likely that in Finland, as well as in other industrialised countries, the social security will be reformed in a direction that may contain some elements of BI, but not necessarily a ‘pure’ BI model.

If the political thinking emphasizing the labour supply and austerity in public economy prevail, the prospects for more generous BI models seem to be low. In the framework of current economic policies, the implementation of a BI would most probably mean at least demolishing large parts or other forms of social security.


BI as a social dividend?

The OECD report (p. 8) ends up recommending some kind of ’partial’ alternative of a BI model. One option mentioned is a possibility to introduce BI as a separate system from the existing social protection, whose function would be to share the benefits of globalisation and technological progress more equally.

This idea of ‘social dividend’ has often appeared in BI discussions. The state of Alaska is already giving an annual share of the permanent fund based on oil revenues to each citizen as a social dividend. There is similar thinking linked also to the idea of “helicopter money”, originally introduced by Milton Friedman, a cash transfer paid by the central bank to people’s accounts to stimulate consumer demand in economic downturns.

Considering BI as a social dividend would locate it in a new frame, where its function would not be to fix the problems of social security systems, but to distribute purchasing power also to those who lose their jobs or end up in low paid precarious jobs in the labour market turmoil caused by digitalization.

If BI were paid on top of other social benefits, its level could even be lower, or for instance connected to macro economy indicators. In that case, it could also be used to stimulate economies in downturns.


Johanna Perkiö
M.Soc.Sc., Doctoral Candidate
University of Tampere
email: johanna.perkio(at)


Read more:

Kangas et al. (2016) From idea to experiment. Report on universal basic income experiment in Finland. Kela Working papers 106.

OECD (2017), Basic income as a policy option: Can it add up?, Policy Brief on The Future of Work, OECD Publishing, Paris.




Sähköpostiosoitettasi ei julkaista. Pakolliset kentät on merkitty *