Book Review: The Reformation In Economics

Philip Pilkington published "The Reformation in Economics: A Deconstruction and Reconstruction of Economic Theory" in 2016. It is an ambitious book, outlining the structural flaws of mainstream economic theory. He discusses the potential replacement theory, but this reconstruction is somewhat overshadowed by the deconstruction.

Book Details

The Reformation in Economics was published in 2016 by Palgrave MacMillan. The book is 358 pages, and is written an academic style with endnotes in each chapter. 

It is divided into three parts, and 12 chapters (plus an appendix). The text is written at an academic level, and brings in subject matter from the humanities, such as philosophy. As a result, the writing style is quite different than my writing, or those of market commentators. 

Philip now works for the GMO investment management firm, and ran the "Fixing the Economists" blog.

The Deconstruction Is Impressive

Pilkington's deconstruction of mainstream economics is impressive; one might wonder if there is any theoretical area left that is not highly questionable.

He covers topics such as marginalism in microeconomics, equilibrium, theories of money and prices, methodology, and areas such as mathematics in economics. The book uses some basic mathematics to discuss these topics, although the book is structured so that the reader can skip the equations. However, the simplicity of the mathematics means that the latest excesses of mainstream theory can only be hinted at, which is an issue I discuss later in this review.

One point of note is that Pilkington's arguments are not driven by ideology. He notes that mainstream economics could easily be used to justify centralised control of all economic decisions; the "auctioneer" that sets prices in general equilibrium could easily be the State. At the end of Chapter 11, he writes that he is "averse to the idea that economics as it is currently taught is a mere reflection of the ideology of the ruling class." Rather,

Contemporary mainstream economics is less the ideology of the ruling class than it is the opiate for establishment intellectuals who find that their little models and ridiculously simplistic arguments get them invited to all the right parties.

Reconstruction Overshadowed By Deconstruction

The downside of devoting a large space of the book to deconstructing mainstream economics is that there is less space devoted to its replacement. I am not completely convinced by Pilkington's strategy in that regard, as I discuss in the next section.

I believe that I largely agree with Pilkington's views on how to approach economics. We have to accept that observed outcomes are the result of complicated interactions, and are contingent on a wide number of difficult to forecast events. Although this is a realistic world view, it has an obvious problem: it is impossible to give the precise-sounding forecasts that are the stock-in-trade of many economists. ("Brexit will lower U.K. GDP by x%!") For some reason, our ruling elites want to get such definitive answers, even if everyone knows the forecasts are inherently worthless.

Take the example Pilkington provides on Scottish oil/natural gas exports in his chapter on uncertainty and probability. He regressed export revenues on the oil prices, and he unsurprisingly gets a good fit. His argument that this exercise is useful as a mechanism to identify the magnitude of the exposure of Scottish exports to oil prices, a factor that could be used as an input into policy decisions. However, the regression model should not be viewed as predictive of much of anything. It just offers a new way to interpret data, but at the same time, one could find some other data which might tell a different story.

The book has a lengthy discussion of interest rate determination. It appears that my views on interest rate formation are closer to the mainstream than those in the book. (At least in the case of modern developed bond markets; not all interest rates markets behaved like these.) We need to distinguish between the level of the risk-free rate, and credit spreads, whereas the book's discussion lumps the two factors together. The willingness to lend to the private sector shows up in the credit spreads, while the risk-free rate is driven by the expectations about central bank policy. I may return to the interest rate discussion in a separate article.

Sure, Mainstream Economics is Inconsistent. So What?

As an initial disclaimer, I am only writing about economic theory here; there are many areas of "mainstream" economic study that are not covered by my criticisms (or those of the book, which overlap). A key example would be the work on defining and measuring national accounts data; there certainly is a certain amount of theory involved in such definitions, but there is nothing to predict the future movements of such aggregates (which is the sort of "theory" I am discussing here). Furthermore, there is a lot of empirical work that is relatively theory-agnostic. (It is hard to think of any theory that would justify the various quantitative easing event studies.)

This disclaimer is not just a bit of boilerplate: I would guess that most self-identified "mainstream" researchers work in such areas. As a result, such economists would have good reasons to be amused with raw denunciations of "mainstream" economics on the grounds I discuss here (or in the book): the criticisms do not apply to their work, yet it is lumped in with the more questionable theories.

My suspicion is that the majority of "mainstream" economists may use many of the embedded assumptions that Pilkington objects to, but they are not greatly troubled by the internal consistency of mainstream economics as a whole. Instead, they are focused on whatever issues they view as most interesting. It is highly possible that the need to use particular mathematical formalisms in published papers is just part of the publishing game, and not taken too seriously (even if they cannot admit it in public). In any event, if their publishing record is within the mainstream tradition, they have a strong incentive to not see that tradition thrown in the dumpster.

As a recent example, I would point to some of the online reactions I saw to one of my articles criticising the mathematics in mainstream economic theory. One response was essentially: yes, the mathematics makes no sense, but everyone knows that already. From the perspective of a pure or applied mathematician, such a response would be jaw-dropping. But if we accept the possibility that mainstream theory is just a cobbled together set of assumptions that few people actually believe, such an attitude makes perfect sense.

Furthermore, although this lack of consistency appears to be an unwelcome state of affairs, I should not be particularly perturbed by it. I used to be in the theoretical wing of control systems engineering, which put me relatively close to pure mathematicians. I could have easily raged against internally inconsistent mathematical models used in control systems engineering. However, if one were a purist on that score, we would have had to shut down the entire aeronautical industry: aircraft control systems are the triumph of engineering pragmatism over theoretical purity. I see no particular reason that economic systems would be any easier to analyse than aircraft.

This dynamic suggests that there are only a couple of strategies for developing and presenting sensible economic theory.

• Ignore mainstream theory. The obvious defect with this strategy is that it will be ignored by the wide audience, which instead will obsess about things like the size of the Federal Reserve's balance sheet.
• Confine debates with mainstream views to topics where there are concrete issues to discuss, and only worry about the issues at hand. Since we need to assume that mainstream economists do not truly care about the internal consistency of their theories, there is no value in dragging that in to the debate. Even if this does not end the stranglehold of the mainstream on credential gathering, it raises the profile of heterodox economics in the wider community.

My concern with The Reformation in Economics is that it takes the theoretical coherence of mainstream economics too seriously, with the embedded assumption that mainstream economists also care about such matters.* The experience of the past decades of such critiques shows that this assumption is faulty, as the book itself ably documents. Meanwhile, the attack on mainstream economics is so aggressive that I doubt that many self-identified mainstream economists would read to the end. (My own writings suffer from this defect.) This means that the book ends up preaching to the converted.

I am not greatly excited by the criticism of older mainstream economics, such as Paul Samuelson's textbook (which is discussed in the book). I can see how older post-Keynesian professors might be fixated by such topics, and it certainly gives a human interest angle to the story. However, if you read a contemporary Dynamic Stochastic General Equilibrium (DSGE) text, Samuelson's textbook is unlikely to be cited. Meanwhile, most numerate central bank watchers have at least some familiarity with DSGE macro -- even if they think it is a load of nonsense. It is hard to relate the criticisms within the book to the contents of the modern papers. It may be that there are underlying errors that spring from faulty assumptions put in place decades ago, but without discussing the modern work directly, this is not obvious.

Concluding Remarks

Although the bulk of this review ended up discussing my differences with regards to strategy, I broadly agree with almost all of the conclusions of the book. If one is new to economics, the book provides a good introduction to the issues around economic theory (albeit at a more complex reading level than my articles). I would highly recommend focusing on Pilkington's discussion for how economics should evolve, and less on the series of accidents that led to its current dismal state.

Footnote:

* I focus here on internal consistency, and not empirical predictions. This is because mainstream economics in practice has structured itself so that its models can explain any observed outcome. For example, if negative real rates do not induce higher growth and inflation, all that tells us is that the natural rate of interest suddenly became negative. It is a self-defeating strategy to argue on the basis of empirical predictions against a non-falsifiable theory. All that is left is to discuss whether the framework is internally consistent.

(c) Brian Romanchuk 2017