Mainstream Economics as Ideology: An Interview with Rod Hill and Tony Myatt — Part II

Rod Hill and Tony Myatt are Professors of Economics at the University of New Brunswick in Saint John and Fredericton (respectively). Their new book, The Economics Anti-Textbook is available from Amazon. They also run a blog at www.economics-antitextbook.com.

Interview conducted by Philip Pilkington

Philip Pilkington: I think it was Joan Robinson who said something along the lines of “while we may have to teach a limited amount of material, we could at least teach that which is useful”. I’ve often encountered economics students who, frankly, seem to me to have a very tenuous grasp of the important aspects of economics. I recall one in particular who graduated from a very prestigious university not understanding what I meant when I said that I thought the chronic unemployment in Ireland was due to a lack of effective demand triggered by the bursting of the housing bubble.

In your experience do you find that students leave mainstream economics courses equipped to deal with real world issues? Maybe you could say something about this more generally.

Rod Hill: The person you describe – who was apparently unfamiliar with some basic concepts in Keynesian economics, like effective demand, or with ideas about asset price bubbles – reminds me of the distinction that Paul Krugman often makes in his blog between saltwater economics departments (where a student is likely to learn these things) and freshwater ones (where they are not so likely to learn them). I think both sets of schools might be broadly viewed as training students in ‘mainstream economics’, but clearly some would equip students better than others to understand real world issues.

This is another way of saying that ‘mainstream economics’ includes a pretty broad range of ideas and that the borders between it and heterodox economics are rather fuzzy. What we are really taking aim at in our book is the narrow range of ‘mainstream economics’ that is presented to undergraduate students, but I think your point is that the narrow range of ideas can also extend beyond that into graduate training, and I agree. This is a rather more pessimistic view than that expressed recently by Dani Rodrik, who seems more optimistic about what goes on at the graduate level, but then he’s in a saltwater school.

My own experience of students leaving mainstream courses is pretty limited, but I can say that I’ve been unhappily surprised during hiring interviews by quite a few fresh PhDs who don’t seem to know what the Keynesian fiscal multiplier is.

Tony Myatt: We could compare economics to other professions. In Canada, at least, someone can graduate from a university with an engineering degree, but they still have to pass a competency exam set by the national body of engineers in order to practice engineering. Perhaps the economics profession should implement a competency exam?

This raises the question: is economics really a practical science like engineering? We argue in our Anti-Textbook that it isn’t…despite the claims in all the mainstream texts to the contrary. Testing predictions isn’t all it’s cracked up to be and core hypotheses are incapable of being refuted. While disagreeing with much of what McCloskey says about methodology, I really like her characterization of economics as being the Art of Rhetoric. So, perhaps the gulf between Saltwater and Freshwater views that Rod mentions shouldn’t surprise us that much.

You cite an economics graduate who doesn’t understand that unemployment could be caused by lack of aggregate demand triggered by the bursting of the housing bubble. That could have a fairly mundane explanation. I’m in the early stages of research in writing the Macro Anti-Textbook. But from my reading of the mainstream principles textbooks so far, I would say that all of them contain an explanation of how wealth affects consumption spending, and all of them have chapters explaining how a decrease in aggregate demand can cause an increase in short-run unemployment. But there are differences between the texts that could underline or obscure that message. Some texts emphasize long-run results and fast speed of adjustment. They all emphasize inherent stability. They all emphasize that the fundamental cause of unemployment is wage stickiness – oh if only wages would fall quicker! And in all the books, by the time the student gets to the Philips curve chapters it seems that the only cause of unemployment is overly high expectations of inflation. So, the sensible stuff on how the bursting of an asset bubble could increase unemployment gets lost in the overall confusion by the time the student gets to the end of the course.

Macroeconomics has been in a real mess for quite a while. Yet there was – supposedly – a new consensus in macroeconomics by the early 1990s. A consensus which, in my opinion, was completely wrongheaded. In a way I’m delighted that the financial meltdown and subsequent Great Recession has blown this “new consensus” out of the water, both fresh and saltwater. It has stimulated interest in heterodox views. It has revealed deep schisms even among mainstream economists. I’m enjoying them.

It’s astonishing how quickly things are changing. Where there was a “consensus” there is now vociferous debate. Whereas in the past this debate would go through the journals, with long time lags, now the debate is instant and online. The blogs are becoming where it’s at. So, you mustn’t underestimate what you’re doing here Philip.

PP: If the blogs are doing anything – and I think they are – they are pulling back students who became completely disillusioned with economics (or who never did it because they thought it was nonsense) and who now see that they need to understand it in a very pressing way if they are to partake in any way of fixing what has gone so wrong. Actually, this ties into what I was just going to ask.

Tony, you say that you think many very important (I would say: chronically important) things get lost in a sea of context. This, I think, is what many students find when they go into an economics classroom with a relatively critical mind. It’s very hard to pick out what’s what and – in my experience at least – you are confronted with a closed system that appears to work perfectly with no problems and this grates with what you see all around you in the real world (especially now – I did economics during the bubble era and even then it was obvious to me).

What do you think about this? Is mainstream economics giving students the impression of a closed system in which nothing can happen that is not predetermined – and that everything that is predetermined is always already happening? I guess that’s a pretty oblique question and I apologise, but I think it’s worth asking.

Tony Myatt: Yes, I think that’s an interesting way to put it. A system that is inherently stable, with a unique equilibrium, determined by tastes and technology – this is a very closed system. While the standard consensus DSGE model (dynamic stochastic general equilibrium model) allows for random shocks, these only influence the short-run movements around the long-run equilibrium. Stability guarantees that unemployment will revert to its natural rate. Of course, there are still a few microeconomic problems in this kind of world. There are the usual problems of externalities and inequality, but these are portrayed as political choices or political failures rather than failures of the political-economic system. To get a more open system we could introduce true uncertainty, hysteresis, multiple equilibria, and strategic interactions. But to account for the asset bubbles I think we also need a good healthy dose of Minsky, and a recognition of the role of power and corruption.

Rod Hill: We did manage to get a hint of some of this into The Economics Anti-Textbook in a way which I hope is not too abstract for the undergraduate audience that we are primarily targeting. One of the authors we cite is Alan Kirman, who expands on some of these themes (multiple equilibria, no guarantees of convergence to equilibria, and how mainstream economics ignores this) in a recent interview in the Erasmas Journal for Philosophy and Economics.

As Kirman points out, Walrasian microeconomics is hardly a suitable foundation for macroeconomics, but (as he puts it) macroeconomists “simply said that we will have to simplify things until we get to a situation where we do have uniqueness and stability.” So what’s constructed is a closed system that appears to work perfectly, as you put it, because it’s internally consistent. But it doesn’t necessarily have much contact with reality.

PP: Another point that is noted in the book is that textbooks rarely engage the student with empirical material. Could you say something about this and perhaps indicate how you set out to deal with it in your book?

Rod Hill: The microeconomics principles texts make some real effort to link the theories they present with evidence or observations, much more so than higher-level texts that deal only with mathematical models of imaginary situations. The principles texts do this to convince the student of the practical relevance of the theories they are learning. However, the point we try to make is that important empirical evidence is often either absent or incomplete.

We spend some time in the book indentifying and filling those gaps. In that way, the readers get not only the missing information, but also some sense of how the rhetoric of the texts works. A couple of brief examples might help to illustrate what I mean.

Every beginning microeconomics student struggles through the theory of the firm’s production and costs. While trying to follow what’s going on in all the diagrams, the student might not notice that all of the examples are of simple imaginary firms making a single product by varying mysterious homogeneous inputs called ‘labour’ and ‘capital’. Little or no empirical evidence is given about the behaviour of firms’ actual costs.

It’s long been known that the general story being told about costs is fiction. The textbooks say that eventually firms’ costs of producing another unit of output in a given time period (their ‘marginal costs’) rise. In the short run, this is attributed to the diminishing marginal productivity of the inputs the firm can vary; in the long run, it’s said to be due to decreasing returns to scale and rising average costs.

In reality, marginal costs for a great many production processes seem to be generally either constant or decreasing. The texts ignore the evidence and they do this for an important reason.

The rising costs of the textbook firm limit its size. At some point, the competitive firm looks at the market price (over which it has no control) and decides that it’s not worth producing and selling another unit of output: its marginal cost of producing it will exceed the market price. But if its marginal costs were constant or decreasing, it could just go on expanding output. However, then the firm could become large relative to the size of the market and it would no longer be a little price-taking firm and the theory of perfect competition would not be applicable. Because the theory of perfect competition is the centrepiece of these texts, the evidence about actual costs has to go.

We identify a number of other key ideas where the texts set out some theoretical ideas but fail to discuss evidence in any serious way. For example, the texts assert that there is an equity-efficiency trade-off, but we show that there is plenty of evidence that contradicts this. This is an omission with important political consequences: the texts invite millions of students to conclude that reducing social and economic inequalities and strengthening social insurance is significantly more costly than it actually is.

Aside from ignoring inconvenient evidence, texts can also present incomplete evidence about a question. We point out, for example, that discussions of rent controls in the texts are simplistic and ignore features of many actual rent control systems that mitigate the negative effects of controls. While some texts now do a better job of discussing the minimum wage by acknowledging that other models are possible (such as monopsony, in which employers have power over the wage), most still use only the perfectly competitive model of the labour market to examine the policy. The further points about the minimum wage that Dean Baker raises in a recent article are something I’ve never seen in a textbook, even though it’s not technical or hard to understand.

What the texts are doing in these kinds of examples is making a single point over and over: government interference in market outcomes causes inefficiency, which is bad. Again, the political implications are obvious should large numbers of young people absorb that point of view.

Tony Myatt: My favourite example is the incidence of taxation. The texts assume (of course) perfectly competitive markets, and derive the result that the incidence of a tax (say the sales tax) is determined by the relative elasticities of supply and demand. It is not determined by who legally has to remit the money to the government. The texts then “illustrate” the theorem in a wide variety of markets, which invariably include cigarettes and gasoline. The illustrations are purely diagrammatical.

Logically the texts should show that the model’s predictions are corroborated by the evidence. They should go out and measure the elasticities of demand and supply for say cigarettes and gasoline, and generate their predictions about the incidence of taxation in those markets. Then they should go and find detailed empirical studies on the effects of taxation in those markets and compare their predictions with the detailed empirical evidence.

But this is never done. And there is a very good reason why it is never done: it is impossible to do it! No-one can measure the elasticity of supply when the supply curve doesn’t exist! There is no supply curve of cigarettes or gasoline because they are sold in oligopolistic markets. You can only define and derive supply curves when all firms are so small they have no influence on market price. This explains why

no text

presents any corroborating empirical evidence on the ability of the competitive model to predict the incidence of taxation.

The moral of that story is not to confuse an “illustration” with empirical corroboration.

My second favourite example is about minimum wages. For example, perfect competition predicts that any small increase in minimum wages (no matter how small) must produce a decrease in employment. On the other hand, non-competitive markets predict that a small increase in the minimum wage might actually increase employment – though a large increase in the minimum wage would decrease employment. What does the evidence say? Actually, the evidence is all over the place. Sometimes minimum wage increases seem to increase employment; sometimes they seem to decrease employment. This mixed evidence is much easier to explain using non-competitive models.

PP: This touches on something very interesting. In mainstream economics academics generally distinguish between ‘positive economics’ – that is, describing the ‘real world’ – and ‘normative economics’ – that is, describing how things ‘should be’. The former is supposedly the realm of ‘science’, the latter the realm of ‘policy’ and ideology’.

But if, as both your points imply, the texts assume perfect competition and bury monopoly and oligopoly in much of their models, then the texts are always making not ‘positive’, but ‘normative’ judgments. They are saying that the world should be free from monopoly and oligopoly (and government) when in reality it is not. Could you say something about this? How do academics that recognise the positive/normative distinction gloss over this? I assume they do so in their own minds as well as in their texts and classrooms…

Tony Myatt: Mainstream textbook writers recognize that much of the real world is non-competitive. But they justify the assumption of perfect competition on positive grounds. They say it is simpler than alternative models and most crucially that it is a good approximation for more complicated market structures. This follows on from their methodology that realism of assumptions is not important. What is important is predictive power. That’s why it is so crucially important for us to show that perfect competition is NOT justified on the basis of its predictive power.

Most of economics is normative, in so far as we’re discussing what the government ought to do to increase overall wellbeing. Analysis of how governments actually make decisions, and what objectives they actually try to fulfill, while “positive”, is normally left to political scientists.

Economists recognise that governments have two main jobs: promoting equity and promoting efficiency. Promoting equity is recognized as the normative bit. But most texts don’t emphasize that promoting efficiency also has a normative basis. Mainstream texts argue that efficiency is obviously a “good thing” because this potentially allows everyone to have more stuff. But it is just an assumption that more stuff will make us happier. Empirical tests of this assumption suggest that the evidence in favour of it is pretty weak.

When it comes to industrial policy, the texts get themselves into a right mess. They go through all their efficiency arguments to suggest that perfect competition is ideal, and monopoly and oligopoly involve “waste”. But these arguments are based on static analysis. As soon as we bring in technological innovation, static efficiency becomes irrelevant.

PP: Are you saying that monopoly/oligopoly might be more efficient? I believe I’ve heard this from both the right (Schumpeter) and the left (Galbraith), perhaps you could outline the argument and speculate why the mainstream economists ward it off.

Tony Myatt: When discussing efficiency the textbooks focus on “static efficiency.” That is, how to maximize wellbeing with given resources and technology and tastes. They go through an awful lot of trouble to show that – with this proviso and that proviso – that perfectly competitive markets are ideal for maximizing static efficiency.

But I mentioned earlier that the one thing that capitalism has been undeniably very good at is technological change and innovation. Indeed, this is probably the most important feature of the modern world. Now, in which market structures does this innovation occur? Both theory and evidence suggests that it is NOT in perfectly competitive markets. On theoretical grounds, firms in these markets are small and lean, just covering their costs. They lack resources to invest heavily in research and development. Empirically, William Baumol in his book The Free Market Innovation Machine reckons that oligopoly is the best market structure from the point of view of innovation. And as Schumpeter argues, if the lure of monopoly profits are the reward for successful innovation, and if that monopoly power is then eroded by new innovators, then it’s hard to see the “static inefficiency” of monopoly as being a serious problem. It’s dwarfed by the dynamic benefits.

So, where does this leave us? It leaves us realizing that most of what the microeconomic textbooks teach concerning efficiency is a tale told by an idiot, just sound and fury signifying nothing. It’s completely irrelevant.

Mainstream economists certainly don’t want to admit this. They’d have to find something else to teach.

Rod Hill: And this raises again the distinction between what we call ‘textbook economics’, the version of what’s taught to undergraduates, that has hardly changed since I was a student in the mid-1970s, and what some (even mainstream) economists are actually doing these days. Lots of people have been working on models of economic growth, some of them using Schumpeterian ideas, for example.

PP: That’s really interesting. But I guess we should wind this up. Are you guys hopeful that the next generation will be more aware of the flaws in the textbook version of economics? Do you think they’ll be more willing to engage with alternative ideas? And what do you see for the future of economics as an academic discipline if such a sea change fails to take place?

Rod Hill: It would be easy to be pessimistic. Plutocratic power seems deeply entrenched, as does the propaganda system that supports it. The ‘mainstream’ textbook version of economics is part of that system, ignoring critical perspectives and encouraging a Panglossian view of the world. If that’s all that economists were to continue to offer, they would ultimately condemn themselves to irrelevance as the disconnect between the world and the textbook grows unbearably large.

That’s one reason I’m actually rather optimistic about the long term. The unsatisfactory performance of the political and economic system is too hard to miss and encourages economics students to consider alternative ideas. Unlike in my student days in pre-history, before the internet existed, those alternative ideas are now freely available and easy to find. Good alternatives to the mainstream texts also exist, such as the Economics in Context texts published by M.E. Sharpe, for instructors who are tired of the standard fare.

As well, within the economics profession itself, there is lots of interesting work going on. The growth of behavioural economics is just one example. Another is the increase in work being done on economic inequality, for instance as seen in the rising prominence of those like Emmanuel Saez and Thomas Piketty who are examining the explosion of ‘high incomes’. Every year, the traditional mainstream principles texts look ever more outdated. I hope Tony and I have been able to help in a small way to hasten their demise.

Tony Myatt: Nice one Rod. Let’s hope you’re right.