Compensating the losers of reform: the forgotten rule of radical centrism, with EU and Spanish cases

Every serious reform has losers. Removing a subsidy, closing coal mines, liberalising a profession or reorganising the labour market can make a country richer and still leave particular workers, firms and towns worse off. Radical centrism has a rule for this, and it is the one most often forgotten: compensate the losers, for real, as part of the reform.

The rule sits in the radical centre constitution we finished today, and it comes from Matthew Miller. His The Two Percent Solution (2003) proposed paying urban teachers much more in exchange for making it easier to dismiss poor performers. He called it institutional bribery. The people whose security a reform threatens get paid, openly, to accept it.

The test economists use, and its gap

Mainstream welfare economics judges reforms with the Kaldor–Hicks criterion. A change counts as an improvement if the winners gain enough that they could compensate the losers and still come out ahead. The key word is "could". As the standard summary of the criterion explains, the compensation never has to be paid, so some people can end up worse off under a reform the test approves. The best-known objection asks why a compensation that only exists in theory should carry any moral or political weight.

Amartya Sen's method gives the objection teeth. Sen asks us to compare real, available alternatives by the outcomes they actually produce, and to rank them through open public reasoning. In that comparison the losers are real people with a status quo they can lose, and they get a voice. A reform that improves the national average while a mining valley collapses has to be defended in public against the alternative where the valley is supported.

What Europe has learned

Europe has run this experiment several times, and the results are fairly clear.

The EU's 2005 agricultural reform replaced subsidies tied to production with a payment tied to the land a farmer keeps in good condition. A 2013 paper by Hans Peter Grüner for the European Commission's economics directorate treats it as a model. The reform included compensation for those who lost from it. A survey of farmers in Lower Saxony found about half satisfied with their transfer, and Grüner reports that there was basically no lobbying to return to the old subsidy system. He draws a practical lesson from the case: compensatory payments for reforms have to be reliable, and a phase-out may be more credible than an open-ended promise.

Germany's labour market reforms of the early 2000s show the other path. Grüner credits them with reducing unemployment by encouraging the unemployed to look for work. They also came with a long period of stagnating real wages, and the reform package included no compensation for that. The governing Social Democrats lost several regional elections in 2004 and 2005, lost the 2005 federal election, and fell to 23% of the vote in 2009, from 38.5% in 2002. The reforms survived. The party that made them paid the bill.

Spain's coal transition sits closer to the first case. On 24 October 2018 the ministry, the trade unions and the employers' associations signed a framework agreement for a just transition in coal mining, combining closures with social measures for miners, training, environmental restoration and aid to revive the mining areas. According to the Ministry for the Ecological Transition, by November 2025 it had channelled more than €975 million in aid to mining areas since 2019, linked to more than €2,100 million of total investment. The figures for workers are specific: 330 early retirements and 36 incentivised departures, at a cost of €175.5 million. Municipal projects funded by the recovery plan created 300 direct jobs, and closure and restoration aid another 153.

The EU's compensation toolkit

The Union has since built standing instruments for the same job. The Just Transition Fund has €19.32 billion for 2021–2027 to diversify the economies of the regions most exposed to the move to climate neutrality, with reskilling, support for small firms and clean energy. The Social Climate Fund, worth €86.7 billion over 2026–2032, starts two years before the new carbon price on buildings and road transport in 2028, with money for greener homes and transport and temporary income support for vulnerable households. The European Globalisation Adjustment Fund covers workers laid off in large restructurings. In July 2026 the Commission proposed €2.7 million from it, 85% of a €3.2 million package, for 671 workers dismissed by 20 automotive suppliers in Galicia after supply chain disruptions and the shift to electric vehicle platforms. Galicia pays the remaining 15%. The package covers training, job search help and a return-to-work bonus of €200 a month for up to six months.

How the radical centre would write the rule

Put these cases together and a position emerges that each side can recognise. From the left: the people and places that lose from change are owed real money and real support, paid by the wider public that gains. From the right and the market tradition: the reform itself, whether it removes a distortionary subsidy, closes uncompetitive mines or opens a protected sector, should go ahead, because the gains are real. And the idea that fits neither camp is that compensation is a tool of political design. It buys durable consent, and it works when it follows a few disciplines.

The compensation is costed and written into the reform from the first draft, so it cannot be traded away later. It goes to identifiable losers, by place and by job. It is reliable, because, as Grüner's farm case shows, a promise that gets renegotiated poisons the next reform. And it has an end date, so it buys a transition and stops short of creating a new permanent subsidy.

Spain has plenty of reforms waiting where this rule matters: housing supply in cities where existing owners fear for their property values, the energy transition beyond coal, the regulated professions, the pension system. In each case the radical centre's first question is the same as Miller's. Who loses, by how much, and what will we pay them to say yes?