When I met James Mirrlees, perhaps the word's greatest tax theorist

Tax is just a matter of taking the most from those with the greatest capacity to pay, right? Or is it a matter of taking the most from the people least able to resist? In Australia we attempt to do both, pretty successfully, due in large part to work of Sir James Mirrlees, the most distinguished British economist of his generation who passed away at his home outside Cambridge last Wednesday aged 82.

In 1996, he shared the Nobel Memorial Prize in Economics for his work on “asymmetric information”. It had long been known that the party on one side of a transaction often knows more than other. A car owner selling it second hand knows more about it than the potential buyer, for example. George Akerlof shared the 2001 Nobel for his legendary paper examining the implications entitled “The Market for Lemons^[1]”.

Mirrlees’ contribution was to formalise the problem and apply it to the design of tax systems^[2]. He saw it as a problem where individuals know whether they are willing to work at a certain rate of tax but the designer of the tax system does not. And his modelling allowed the income tax system to be progressive, as Australia’s is, rather than flat, as had previously had to be assumed in academic work.

And he did more. His work helped the design of rules for auctions such as those used in Australia to allocate mobile communications spectrum. The award of the 2007 Nobel to Leonid Hurwicz, Eric Maskin and Roger Myerson for “mechanism design^[3]” owes much to his work.

His groundbreaking contributions to the theory of moral hazard (which examines the way in which institutions such as banks and individuals take greater risks when they are insured) and the so-called principal-agent problem (where the people who run the company have different interests to the people who own the company) formed the basis of Bengt Holmstrom’s 2016 Nobel.

This work has implications for how to design incentive schemes for everyone from CEOs to school children to journalists. His contribution here was fairly technical, pointing out that the existing formal models of the problem suffered from two previously unrecognised mathematical problems.

His important paper on moral hazzard and the principal-agent problem was left unpublished until 1999^[4]. Apparently he had ensured that those who needed to know about it did, and allowed it to as good as vanish in the days before the widespread availability of the internet, leaving anyone who wanted to read it to seek out a physical copy from a library.

As a budding young student in 1996 I did just that, on a trip to Oxford.

Having secured a copy of the paper I was eager to read it, and was part way through it on the train ride back to London when a voice behind me asked “what do you think?” And there he was – the great man – who proceeded to chat with me for the rest of the trip about the intellectual history and technical details of his paper.

For a budding economics nerd it was the moment of a lifetime. And it was a measure of the scholarly seriousness, intellectual generosity, and humility of Mirrlees that he took the time to talk to some random Australian kid.

Mirrlees worked on some of the most profound social questions of our time – how to balance equity and efficiency, and how to provide incentives – and he did it with precision and rigour.

He was the model of a great economic theorist. He will be missed. But his work will live on in the design of Australian and other institutions and will be on reading lists for decades. If the Turnbull government had proceeded with its tax white paper, he would have been a reference point, as he was for the Henry Tax Review for whom he acted as a consultant. He’ll be helping us out for generations to come.

References

1. ^{^} The Market for Lemons (www.sas.upenn.edu)
2. ^{^} apply it to the design of tax systems (www.jstor.org)
3. ^{^} Leonid Hurwicz, Eric Maskin and Roger Myerson for “mechanism design (www.nobelprize.org)
4. ^{^} was left unpublished until 1999 (www.researchgate.net)