Obituaries

Partly Sunny with Showers 0° London Hi 5°C / Lo 2°C

Professor Leonid Hurwicz: Nobel Prize-winning economist

Thursday, 17 July 2008

Hurwicz invented a framework for the mathematical analysis of institutions, seen as abstract communication systems

AFP/Getty Images

Hurwicz invented a framework for the mathematical analysis of institutions, seen as abstract communication systems

Leonid Hurwicz was one of the finest economists of his generation. He made many important contributions to economic theory, the most significant of which was his invention of the field of mechanism design, an achievement which was recognised several decades later, in 2007, by the award of the Nobel Prize for Economics, which he shared with Eric Maskin and Roger Myerson.

Mechanism design is a framework for studying fundamental questionsabout social organisations. In the first half of the 20th century, there was a long-running debate among economists about whether socialism was compatible with economic efficiency. Oskar Lange and others argued that socialist institutions for rational resource allocation could be devised, while Ludwig von Mises and Friedrich Hayek denied that they were even theoretically feasible, for reasons to do with incentives and information. Hayek argued in a famous paper of 1945 that, since the information which is necessary for efficient economic planning is widely dispersed in society, no central planner could even in principle collect it, partly because nobody could be expected to reveal information truthfully if it would be used against their interests.

Hayek's argument, undoubtedly plausible as a critique of Soviet central planning, raised many questions. If economic institutions should be regarded as communication systems, in which people convey dispersed information to each other about desires and resources, how should one design the optimal communication system? Why should a system of free markets avoid the problem that people need incentives to reveal information truthfully?

Hurwicz invented a framework for the mathematical analysis of institutions, seen as abstract communication systems ("mechanisms"). In the formal theory, a mechanism is defined by a language and a set of rules for translating messages into decisions about, for example, production or distribution of goods. Mechanism design theory has great scope and elegance. A key notion was incentive-compatibility: a mechanism is said to be incentive-compatible if nobody has an incentive to violate the rules of the mechanism. Using these concepts, Hurwicz proved a fundamental theorem about private markets.

According to the classical theorems of welfare economics, it is socially optimal, given certain well-defined conditions, to leave production and consumption to be determined by competitive markets, and use redistributive taxation to deal with wealth inequalities. But these theorems do not take account of informational constraints. Hurwicz proved that no incentive-compatible mechanism based on private markets can be socially optimal.

Hurwicz's theorem naturally raised the question: which institution, in any given context, comes closest to being socially optimal? Mechanism design theory is ideal for answering such questions and has been applied to many disparate problems. For example, what is the best design for an auction, or a voting system? What rules should be used to regulate a natural monopoly? Which goods and services should be produced publicly and which privately? The theory has also been used to analyse the internal organisation of firms, the design of computer networks and the study of contracts, legal procedures and arbitration rules.

Leo Hurwicz was born in Moscow in 1917, two months before the October Revolution, of Polish-Jewish parents who had fled Warsaw to escape the Kaiser's armies. In 1919 the family fled back to Warsaw in a horse-drawn wagon, in fear of persecution by the Bolsheviks. Leo studied law at Warsaw University, graduating in 1938, but his main interests lay elsewhere – he had also studied piano, physics and economics.

In 1938 he began a PhD in economics at the LSE, but his visa expired after a year. Understandably reluctant to return to Poland, he sought refuge at the Institut des Hautes Etudes in Geneva, intending to study under von Mises. The day after he arrived in Switzerland, Hitler invaded Poland. Hurwicz's parents once more became refugees – again they fled from Warsaw to Russia, this time with Hurwicz's brother. His father was immediately arrested and interned in a labour camp, accused of "being capable of intellectual leadership", while his mother and brother were sent to a different part of the Gulag, in Siberia.

Hurwicz decided to emigrate to the United States. Unfortunately, the American authorities required a certificate of good behaviour from the British authorities, who refused to oblige, so Hurwicz applied his native charm and ingenuity. He persuaded a Swiss policeman to write to the London police to the effect that a certain Leonid Hurwicz was under suspicion – did they have any evidence of his bad character? The reply gave him what he needed.

Eventually Hurwicz made his way to the US, where his parents and brother were later able to join him. In the 1940s he lectured in meteorology at Chicago University and worked on econometrics, then a new field, at the Cowles Commission. He did not get on with Milton Friedman, who was the driving force at Cowles, and in 1951 he left to take a professorship at the University of Minnesota, where, aside from numerous visiting professorships, he was to remain for the rest of his life.

With Walter Heller, John F. Kennedy's chief economic adviser, Hurwicz built the Minnesota faculty into one of the best in the world, which it remains to this day. He had little interest in academic rank, but he always had time for ideas. Even when he was in his seventies a meeting with colleagues and visitors to discuss research, scheduled for brunch, might stretch into the evening until all but Hurwicz were exhausted. He was a polymath with a wide knowledge of many disciplines, but what marked him out, as his collaborator Kenneth Arrow said, were his intellectual caution, rigour and depth and his sense of the essence of a problem.

He was active in Democratic Party politics and served as a delegate for Eugene McCarthy at the Democratic National Convention in Chicago in 1968. A more successful political intervention had come in 1967, when a former colleague in the Minnesota economics faculty, Andreas Papandreou, was jailed by the Greek colonels when they overthrew the elected government, which had been headed by his father and in which he had been a senior minister. Hurwicz threw himself into a campaign to get the US government to press the colonels to release Papandreou. The campaign succeeded and Papandreou went on to become, in 1981, Greece's first socialist prime minister.

In 1990 Hurwicz was awarded the National Medal of Science by the first President Bush, one of only six economists to have been honoured in this way. Nevertheless, it seemed that the Nobel would elude him. He remarked that the award became less likely as the years passed and the people who knew his work gradually died off. Indeed, most younger economists, many of whom work all the time with ideas derived from Hurwicz's work, would have only a vague idea of where these ideas had originated.

When the prize finally arrived, he was, at 90 years of age, its oldest-ever recipient and only a few months from his death, though he was still writing papers and books until well into his eighties.

Robert Evans

Leonid Hurwicz, economist: born Moscow 21 August 1917; Research Associate, Cowles Commission for Research in Economics, University of Chicago 1944-46; Associate Professor of Economics, then Professor of Economics, Iowa State University 1946-49; Professor of Economics, Maths and Statistics, University of Illinois 1949-51; Professor of Economics, Maths and Statistics, University of Minnesota, 1951-1999 (Emeritus); Nobel Prize in Economic Sciences 2007; married 1944 Evelyn Jensen (two sons, two daughters); died Minneapolis, Minnesota 24 June 2008.

Interesting? Click here to explore further