THE concept of a universal basic income (UBI), an unconditional cash payment to all citizens, has in recent years captured the imagination of a wide spectrum of people, from leftist activists to libertarian Silicon Valley techies. Proponents see a neat solution to poverty and the challenges of automation; detractors argue it would remove the incentive to work. Trials of UBI have been launched, or are about to be, in several countries. Most are publicly funded, although Y Combinator, a Silicon Valley startup accelerator, is starting a privately funded experiment in America.

Finland was one of the first movers. In January 2017 it began a trial for 2,000 people, each receiving €560 ($680) a month. That drew legions of foreign journalists and camera crews. This week, however, international media attention abruptly centred on the ending of the experiment in December 2018. Headlines suggested that it had been “scrapped” or had “failed”. The truth is more nuanced.

The trial was always due to finish after two years, although Kela, Finland’s national welfare body, which was responsible for the experiment, had hoped to expand it (it was denied funding in January). The scheme was also more limited than the hype suggested; it was not a truly universal benefit, because all the recipients were chosen from among the unemployed. And the trial is not ending because of failure. Indeed, Kela has refused to publish any results until it is finished, for privacy reasons and to avoid biasing outcomes. The government simply has other priorities. In particular, it has decided to adopt Danish-style active labour-market policies.

More important, the UBI trial was always as much about the principle of policy experimentation as it was about the outcome. As Heikki Hiilamo of Helsinki University points out, Finland has tested policies before, such as a “full employment” trial which sought to provide a salaried job for every unemployed person in the small town of Paltamo. And the country is still keen on novelty. After the UBI trial, the government is planning to test a universal credit system.

In its desire to try new policies on a small scale, Finland is no different from many other countries. A study in five Dutch municipalities, billed by some as a basic-income trial, is in fact mainly focused on testing various options for unemployment benefits, dividing participants into three groups. (To be fair, one of these was supposed to receive something much more like a UBI before changes imposed by the national government.)

For all the hype around UBI, surprisingly few are using the most rigorous research approach—a randomised control trial. Y Combinator’s plan in America, as well as an experiment in Kenya run by GiveDirectly, a non-profit organisation, are exceptions. In the Kenyan project, 300 villages were assigned to four groups. In one, villagers get a UBI for 12 years; in the second, for two; in the third, they get a lump sum; the fourth is a control group. But all these trials are just getting started, and none has released results yet.

Meanwhile, other basic-income trials are going ahead. On April 24th the provincial government of Ontario said that it had successfully finished the enrolment phase of its UBI study, which has more than 4,000 participants in four towns. And in Scotland, four local authorities, including Glasgow and Edinburgh, are busy hashing out pilot projects to implement a basic income. Whatever the results of Finland’s trial, a definitive answer on how UBI actually affects citizens’ long-term behaviour is still many years away.