Should we reform economics?

This is one of a series of columns that were produced for Moneyweb Investor in which Stuart Theobald explores the intersection of philosophy of science and finance. This followed an earlier series for Business Day Investors Monthly on the same theme.

Much coverage was given in the international press earlier this month to a student-led critique of the “neoclassical” and mathematical approach of traditional economics teaching. An open letter penned by a multinational collection of student societies declared the teaching of economics in crisis and called for theoretical “pluralism”. Some commentators jumped on the letter to push their own pet hates such as the heavy reliance of economics on mathematics or the fact that economics training seems largely designed to service the recruitment needs of financial firms. The letter itself was more sensible, making the point that although maths and statistics training were crucial, economics students would benefit from studying some of the historic controversies in the discipline as well as understanding some of the qualitative tools developed in others.

Well, let me defend traditional economics teaching. Such a defence is not a claim about what students “should” learn if you want them to become better rounded people. I am a big fan of the “PPE” (philosophy, politics and economics) that Oxford pioneered. The University of Cape Town now offers an excellent version of it. I encourage everyone to engage in the controversies, but that does not amount to a call to reform economics. The reforms economics needs are about the fundamentals: its concept of probability and the epistemic foundations of models. But that is a matter for a different column.

Most commentators seem to see traditional economics teaching as a single edifice of agreed content. This is both right and wrong. There are major disputes in economics and there always have been. In Britain, the post-war theoretical battle between Cambridge (led by John Maynard Keynes) and the London School of Economics (Friedrich Hayek) set the scene for what continues to be deep ideological differences at the highest level of policy debate. In the United States, the University of Chicago’s profound influence on the discipline was motivated by operationalism and the emergence of computers together with the collection of large amounts of data on markets for the first time. Yet the Chicago school’s Milton Friedman spent his career in a fraught and profound theoretical battle with Paul Samuelson over regulation and markets.

On the other hand, there is underlying agreement within the traditional discipline. It is committed to a broadly positivist philosophy, a belief that it should be engaged in making positive truth claims about how the world works. It is also committed to the use of theoretical modelling as its primary methodological device. One of the general challenges that economics tries to address is how to manage uncertainty in the world. The positivist answer is to use statistics to analyse data to determine probabilities. Separately, economic modelling has become very mathematical, with causal relationships expressed in equations rather than other methods of representation such as scale models. At the London School of Economics you can still see the hydraulic model Bill Phillips (he of the Phillips Curve) built in 1950 to show how monetary and fiscal variables affected the economy. These days we prefer mathematical representation rather than scale models because they are more precise and less reliant on the irrelevant features of the physical system you put together. (Phillips’ model used water prone to leak. A mathematical representation of the same model that used nine simultaneous equations didn’t have this problem.) It is this use of statistics to capture probabilities and maths to build models that has given economics its mathematical character today.

Much of traditional economics teaching is the accumulated “orthodoxy” that decades of argument have led to. But this orthodoxy implicitly includes much of the historic controversy. While commentators consider Friedman the arch disciple of neoclassical economics, it is in fact Samuelson’s textbook that has been the biggest seller and used for decades in teaching. That textbook includes the traditional Walrasian equilibrium theory but also the aggregate expenditure models of Keynes. The historic controversy is not made explicit.

But so what? What if economics did draw out the controversy? Consider a discipline that does. Philosophy teaching is heavily reliant on history. While there is great benefit in teaching Plato and Kant, and great intellectual battles like that between John Rawls and Robert Nozick, it has made philosophy departments detached from the real world. The skills philosophy departments teach – formal logic, analysis, applied ethics and argumentative writing – are hugely important to society. I think there is a case to be made that philosophy departments should split into “applied philosophy” and “history of philosophy”. Economics has already made that change; many universities have separate departments for history of economics and economics. This allows students to focus on what they want to learn: the tools or the controversies, especially when those include major sub-disciplines like Marxism. It is also about efficiency: you want, in the three years you have them, to get students to become masters of the tools, which requires focus. A focused training in the use of the tools, which economics provides, is valuable. That is not to say other things aren’t valuable too, but that is no reason to change economics.

Theobald has studied economics and philosophy, including at the London School of Economics, but sadly never did a PPE.