Segment 1 — The Cold Open

In early 2017, two thousand unemployed Finns began receiving an unconditional monthly payment of 560 euros from the Social Insurance Institution of Finland, known as Kela. The designers had hoped the removal of means-testing and reporting requirements would free people to accept any work that appeared. Instead, the same payment structure that eliminated penalties for small earnings also reduced the net financial gain from moving into higher-paying or full-time roles. Consider a recipient who finds a part-time position paying a few hundred euros above the flat amount: under the old tapered system the benefit would have declined gradually, preserving some incremental reward for each additional hour, yet the new flat line meant every extra euro earned stayed fully with the recipient only after the original benefit had already been replaced entirely by wages. That arithmetic quietly reversed the incentive gradient once earnings crossed a modest threshold, leaving recipients with a stable but lower-powered reason to push further. The result was visible in administrative records as more part-time days worked and no measurable lift in full-time transitions, showing how a single design choice about payment shape could stabilize one level of attachment while dampening movement to the next.

Segment 2 — The Good Intention

The Finnish government launched the experiment to address long-term unemployment that had persisted after the 2008 financial crisis and the subsequent euro-zone slowdown. Policymakers at Kela and in the centre-right coalition believed that the existing unemployment-benefit system created high effective marginal tax rates whenever a recipient took even modest part-time work. An unconditional payment, they reasoned, would remove that disincentive, simplify administration, and let people test jobs without fear of losing all support. The approach drew on earlier Nordic discussions about basic income as a way to adapt welfare to a more fluid labour market while preserving dignity and reducing bureaucracy. Those discussions had emphasized that lengthy claim forms and abrupt benefit cliffs often discouraged small steps back into employment, especially when recipients weighed the paperwork burden against uncertain wage gains. By replacing the whole apparatus with a single monthly transfer that continued regardless of earnings, the designers expected individuals to weigh job offers on their own merits rather than on how the offer would interact with benefit withdrawal rules. They also anticipated lower administrative costs and less stigma, two outcomes that aligned with the coalition’s broader goal of streamlining public services without abandoning support for those out of work.

Segment 3 — The Implementation

Parliament approved a two-year randomised trial that ran from January 2017 through December 2018. A randomly selected group of 2,000 people aged 25 to 58 who had been receiving unemployment benefits at the end of 2016 were moved onto the flat 560-euro payment; a matched control group remained on the conventional system. Early administrative data showed that recipients reported less stress about paperwork and were quicker to accept short-term or part-time assignments. Proponents in the Ministry of Social Affairs highlighted these early signs as evidence that the simplified structure was working as intended. The random assignment itself was meant to isolate the effect of the payment rule from other labour-market variables, allowing later comparison of employment records between the two groups without the usual selection bias that plagues voluntary programs. Because the payment arrived automatically and without conditions, caseworkers spent less time verifying hours or recalculating entitlements, freeing capacity that had previously gone into monitoring compliance. At the same time, the trial’s limited scale meant it did not alter wage levels, vacancy postings, or regional job availability, so any employment changes would have to come from the changed incentives facing the recipients themselves.

Segment 4 — The Unintended Consequences

When researchers later compared the two groups using administrative records, the treatment group showed a clear increase in part-time employment days. At the same time, the difference in transitions to full-time work was statistically insignificant. The flat payment meant that any additional euro earned above the benefit level produced a smaller net gain than under the tapered unemployment system, lowering the return to intensive job search or to accepting longer hours. Recipients therefore had less financial reason to move from stable part-time roles into positions that demanded more commitment or carried greater risk of failure. To see why, imagine two otherwise identical workers: one still on the old system faces a benefit reduction that offsets only part of each new wage euro, while the other keeps the full 560 euros on top of whatever wages arrive, yet once wages exceed that fixed amount the marginal reward per extra hour is simply the wage itself, without the extra boost that a declining taper would have supplied. That difference in slope matters most precisely when a worker is deciding whether to add another shift or pursue a more demanding full-time opening. Second-order effects followed: some participants remained attached to low-intensity work that fit around the fixed payment, while employers in sectors that rely on part-time labour found a steadier pool of workers less motivated to seek full-time alternatives. The design had removed one barrier—the fear of losing benefits—yet left untouched others such as skill mismatches, regional job availability, and childcare constraints. Because the payment did not vary with local labour demand or individual circumstances, those who faced additional obstacles continued to weigh them against an unchanged return to effort. Over time the trial therefore illustrated how an unconditional transfer can anchor people at an intermediate level of participation without generating the stronger pull toward higher-intensity employment that a tapered or conditional structure might have retained.

Segment 5 — The Aftermath

After the trial concluded, the centre-right government did not extend the unconditional model nationwide. Subsequent administrations shifted focus toward activation measures that combined modest supplements with stronger job-search requirements. The Finnish experiment nevertheless influenced later European discussions, prompting designers in other pilots to test tapered or earnings-disregard versions rather than purely flat payments. No large-scale replication of the original flat structure has been attempted in Finland since. The decision to move away from the flat design reflected the evidence that part-time attachment had risen while full-time movement had not, leading policymakers to conclude that the incentive gradient needed recalibration rather than complete removal. Later pilots elsewhere incorporated earnings disregards that preserved some taper above a threshold, precisely to avoid the flattening of marginal returns observed in the Finnish records. Those adjustments themselves introduced new administrative layers, illustrating how attempts to correct one incentive problem can reintroduce complexity that the original unconditional model had sought to eliminate.

Segment 6 — The Lesson

Incentive structures always reveal their marginal effects once the payment rule is fixed. When a transfer is truly unconditional, it stabilises whatever level of work effort recipients find easiest to maintain, which may not be the level policymakers most want to encourage. Anyone designing income support today therefore needs to model not only the removal of penalties but also the altered rewards that remain once the new baseline is in place. The Finnish case shows that the arithmetic of marginal gain operates even when the payment itself is simple; changing the slope of that gain requires deliberate attention to what happens after the flat amount is secured. How might current proposals for expanded wage subsidies or earnings supplements change if we applied the same marginal-return test?

💬 Reply to this email — Patrick reads every one.

Share: X · LinkedIn · WhatsApp

Forwarded this email? Subscribe here — it's free.