I had been wondering if any contributors to this debate would respond to my comments about the near-exclusion of certain social groups from economics, and finally Thomas Palley did so – but only to dismiss them as irrelevant compared to the social exclusion of heterodox white male economists. I do agree with him that economics is important because of its influence on policy and in society more widely through its effect on practices in business and financial markets. This is why I regret the social exclusiveness of the subject; even though there are many fine economists from all identity groups who do research economic outcomes for women and members of minority groups in society, their perspective simply can’t be the same as that of a researcher from one of those groups.

To a degree, yes. At any rate, I wish some amongst them would welcome the undoubted opening up of the mainstream to a wider range of ideas and approaches. The absence of new papers on Keynesian macroeconomics in the AER doesn’t prove anything to the contrary. As Thomas says, there’s no scientific basis for rejecting all Keynesian descriptions of the economy – or for accepting them – although there is a sound evidential basis for some beliefs about how to run macroeconomic policy, such as the institutional status and aims of the central bank and the long-term role of money. When you’re talking about any model of the whole economy simple enough to warrant a label like ‘Keynesian’ (or anything else) you’re in the realm of belief and not science. These can be extremely useful in many circumstances. Reaching for Keynesian metaphors to try to understand events was a sensible reaction to the post-2000 period when deflation looked a greater possibility than at any time since the 1930s.

Yet economics does have a good claim to a scientific basis. Like a number of the non-experimental natural sciences such as geology and evolutionary biology, it tries to pick out the most important variables driving change, sets them in a model which ensures internal consistency, and confronts the model with the evidence. Models may be ephemeral, with one that suits one era not working in another. They are certainly falsifiable.

I’m completely with Paul Krugman in thinking that the ‘orthodox’ approach of modelling via maximisation problems and asking what the equilibrium solution would be is incredibly fruitful. Most of my work is in competition economics, and in new technology markets at that. I’m not so daft as to believe that the businesses I look at succeed in maximising anything, nor that their industries are in a perfectly competitive equilibrium, or equilibrium of any kind. But the mainstream approach tells me more than the alternatives about where, as a competition regulator, I should be interested in intervention. Few heterodox economists have offered alternative tools but where they have – such as Paul Ormerod’s application of agent-based complexity models – I’m delighted to use them too. I’m certainly not rejecting heterodoxy on the grounds of ideology, although I do find it lacking in traction for applied work.