Richard Denniss proves economics is a science!


I am not sure that this is what he intended, but in today’s AFR Richard has an article looking sceptically at whether economics is a science or just a bunch of opinions. He proves it is a science.

The article comes from a dispute he is having with Judith Sloan about tax and the rich. I don’t want to revisit the argument.  If you can lay hands on a copy of the AFR (or get behind its paywall) it is an interesting read. But the key feature of Richard’s article is that he approaches the dispute using the scientific method. He presents the different views and the underlying theoretical models. He isolates the key assumption that distinguishes between the models (I will let others judge if he got this right). He leaves us with the empirical question – what is the elasticity of tax avoidance for those with high incomes? This is an area that many economists have looked at so I suspect there are some good answers.

But despite using the scientific method to address the question, I suspect that Richard really wants the opposite conclusion. Thus he concludes “… if economists can disagree by three orders of magnitude about the likely impact of a policy, or they can’t admit when they are wrong, then they will add little more to the public debate than confusion and newspaper copy”.

Thus, Richard’s beef seems to be about scientists disagreeing, not about whether economics is a science. He proves the latter. As to the former, I would like Richard to show me a science where there are not areas with even more controversy! I don’t think there are any.

3 Responses to "Richard Denniss proves economics is a science!"
  1. Sorry Stephen. You do not understand scientific method. Richard has not proved economics is a science. It is certainly a discipline, but not yet a science. Observation, hypothesis, prediction – he fails on all counts as most economists do. Sorry to spoil the dream.

  2. I hate to break it to you Professor King, but all sorts of pseudoscientific fields ape the scientific method. It doesn’t stop them being pseudoscientific; indeed, the pretence of a scientific approach is what makes them pseudoscientific, as opposed to patent nonsense. The key feature of true science that Richard Denniss highlights in the article is the rejection of falsified hypotheses; pseudoscientific endeavours, in contrast, ignore the outstanding problems with their theories or resort to ad hoc rationalisations, ignore alternative explanations, and fail to progress as a body of knowledge as a result. In economics, can you ever really be forced, by the weight of evidence, to abandon a position, or do you always have room to question the evidence, reinterpret the evidence, and twist it to support your view, all the while being regarded as a competent economist (and maybe even getting a Nobel Prize for doing so)?

    By the way, I don’t know what Sloan’s argument is, but in the US, at least, the issue in the optimal top tax rate debate between Diamond et al and Mankiw et al is indeed the behavioural response, but it’s not (or at least not chiefly) the elasticity per se. It’s the thickness or thinness of the top tail of the income distribution, which also determines the significance of the behavioural response. And if “good answers” means sufficient to significantly narrow the debate among economists about what tax structure might be optimal, then no, there are no good answers, which just underscores my point.

%d bloggers like this: