Why Sweden's Welfare State Needed Reform

Abstract

According to Professor Walter Korpi the Swedish economy would be in better shape if leading politicians had paid less attention to the advice from an influential group of Swedish academic economists allegedly hostile to the welfare state. Their claim that high taxes and extensive social security programmes have strong adverse effects on aggregate economic performance rested on a politically biased reading of comparative international data. In purporting to show that Sweden had fallen behind other comparable countries (the 'sclerosis diagnosis'), they set the stage for cutbacks in the welfare state, and a major tax reform involving substantially lower marginal tax rates.' Apart from creating inequality at a negligible gain in efficiency, the very same reforms may also explain why Sweden ran into a macroeconomic tailspin in the early I 990s. Rather than an instructive case study of the ultimate consequences of a long-term obsession with income equality and social protection, the exceptional economic crisis can be traced to severe macroeconomic policy blunders. The views of Professor Korpi are quite familiar for anyone that has followed the Swedish debate over the last decade. Supporters of the welfare state applaud Professor Korpi for revealing the ideological bias underlying the evidence presented in support of the sclerosis diagnosis. Some Swedish economists rather thank of him as a modern day Don Quixote, eager to pick a fight with imaginary villains. Being a Swede, and a professor of economics, I can hardly claim the status of an impartial observer. I certainly believe that Harberger triangles convey information, and I am convinced that most of the recently implemented reforms make sense. Nevertheless, I believe that Professor Korpi raises important issues. Many of the accusations made against the welfare state, in Sweden and elsewhere, rest on evidence which does not survive normal academic scrutiny. In their capacity as policy advisors, academic economists far from always live up to the standards of the seminar room. Speculation and sweeping generalisations are poor substitutes for facts and empirical investigations. Like Professor Korpi, I doubt that very important insights on the harmfulness of the welfare state can be gained from comparative aggregate data. Over the period I970-90, Swedish economic growth was below the OECD-average. At the same time the public sector was very large. However,