Development

Prior to his move to Oxford, following the suggestion of Amartya Sen, in the late 1960s Mirrlees joined the India Project run by Paul Rosenstein-Rodan for the MIT Center for International Studies, spending time at MIT (during which he met Bob Solow and Paul Samuelson) and then in India. He applied his attention to the issue of development. The end result was the work with Ian Little and the ‘Little-Mirrlees’ method of cost-benefit analysis for developing countries. This was nothing less than a general framework applying appropriate shadow prices to public sector activities to enable efficient allocation of resources by the public sector. Distributional issues were included in choosing the shadow price of labour, as were distortions such as currency controls. Prior to Little-Mirrlees, the main approach had been to employ market prices to undertake cost-benefit analysis in developing countries: whilst there are limitations to using market prices to measure opportunity cost and marginal benefits in developed countries due to distributional and market failures, in developing countries these limitations become huge and market prices can be highly misleading (see Mirrlees 1969a; Little and Mirrlees 1969, 1972, 1974). Again, this is an instance of Mirrlees wanting to do things properly rather than taking the easy option. Also, we see the expert economic adviser looking beyond and behind the market to design policy that can increase welfare and promote economic efficiency. Ian Little also shared Mirrlees’s utilitarian perspective (see Little 2002).

Mirrlees’s interest in development has remained, more recently focusing on the issue of taxation (see Mirrlees 2011a). Mirrlees is very much aware of the problem of collecting taxation in developing countries: he believes that governments should use ‘all the kinds of taxes that really work in lower income countries’ (ibid.). Here, of course, his experiences in India were highly instructive to him: a sales tax is the most practical to raise, but was in effect an ‘urban’ tax with little chance of being levied in rural areas. In order to make a sales tax progressive you need to add subsidies. An income tax is harder to make work: he is of the opinion that when he was in India, it was ‘totally ineffective’, although nowadays that is not true in India and China:

With the urban sector growing rapidly in many countries, only a few people paying tax should be quite a temporary phenomenon. For example, a lot of people now pay tax in People’s Republic of China. There will always be many people in rural areas that you should be able to get into the tax system. But it actually becomes very inefficient including large numbers of low income tax payers as you end up only collecting a small amount of tax and it usually costs almost as much to collect the tax. So I come back to thinking that a sales tax is a pretty good way of doing it (ibid.).