Why was economics so insular?

Noah Smith has a good piece on what seems like the never ending stream of popular articles in the UK slagging off economics (or economists). Here I outlined three potential reasons for this epidemic: people do not understand unconditional macro forecasts, politicians from the right do not like economists spoiling their pet schemes (e.g. Brexit), and many heterodox economists from the left wage endless war against the mainstream. All these complaints get airtime when the economy is bad.

The UK economy, right now, is perhaps in a worse state than at any time in the last eighty years. As John Lewis shows in this Bank blog, productivity growth has perhaps never been as bad as it is now: we have to go back to before 1800 to find anything comparable.

The natural reaction when the economy is bad is to criticise economists. That was what happened after the Global Financial Crisis, with some justification. But what is happening in the UK right now is mainly a result of first austerity and then Brexit. As I explained in detail in my earlier post, if we had followed the advice of mainstream economics austerity and Brexit would not have happened. [1] I have as yet not read a single critique of economics that has pointed that fact out, which if you think about it is extraordinary.

There is a little more to say about why economics is an easy target. Historically it has been very insular, and in this respect quite unlike other social sciences. I have already discussed the paper by Haldane and Turrell in the OXREP Rebuilding Macroeconomic Theory volume on Agent Based Models, but I did not have space to show an interesting chart from the introduction to that paper.

It tracks citations in papers to those in other disciplines. Until around 2000, there was no doubt which was the most insular discipline: economics. This is no surprise to me and I suspect most social scientists.

The paper does not explore the reasons why economics is so self-referential: their aim is simply to suggest that it needs to look to other disciplines to see what methods they use. Here I want to sketch why I think mainstream economics (and here the qualification mainstream is required) is so insular.

I once gave a lecture course on the methodology of economics, and in one lecture I used a large blackboard to describe how nearly all economics can be derived from the basic axioms of rational choice. For example the modern macroeconomics of consumption is just the choice between buying apples or pears transformed to the choice between consumption at different times. In that sense economic theory is like an immense tree, where every branch deductively builds on this core. Sometimes large branches grow by adding new elements, like asymmetric information, which then becomes part of the tree and can be used by other branches. This deductive tree of economic theory did not grow all by itself: its growth was and is influenced by the real world problems it wanted to address.

In using the idea of explaining decisions by optimising welfare under constraints economists have created a whole series of widely applicable tools. Economists naturally think about opportunity costs, adverse selection, moral hazard, incentives etc. There is something distinctive about thinking like an economist. To say, as Tom Clark does here, that sometimes this is just formalising common knowledge may be true (see also Cahal Moran here), but in many cases it is not. Try persuading someone who has invested in what is now a sub-optimal project about sunk costs.

This body of theory includes the neoclassical economics that heterodox economists and others love to hate, but it also includes game theory that has applications well beyond economics, and more. In my first year of studying economics I was told in some lectures that this whole endeavour was a huge ideologically driven misstep, but I began to see it differently after reading this famous 1963 paper by Arrow. It shows why (asymmetric) uncertainty in the health service means that the standard competitive model just cannot work for medical care. That may be obvious to us in the UK but it appears otherwise to many in the US. To be fair Clark also acknowledges that this economic theory has produced positive successes: he mentions auction theory but there are many more.

As to ideology, if you want an effective critique of neoliberalism you have to use economics (see, for example Colin Crouch’s book on neoliberalism or this by Dani Rodrik). So many critiques of economics use a kind of bastardised version that insists that workers are always paid their marginal products that the political right also employs. But monopoly and monopsony power are also part of the deductive tree. A paper I like to refer to in this context is by Piketty, Saez and Stantcheva (discussed here) which uses a simple bargaining model to show how cutting the top rate of tax can increase pre-tax CEO pay.

There is nothing like this deductive tree in other social sciences, and I think it at least partly explains why economics used to be so insular. As non-economists academics seemed to add little to building on this theory, there seemed little point in collaborating or even citing them. But, from the point of view of other disciplines, it was worse than that. Economics could also be imperialistic. Its methods, both theoretical and empirical, could be applied to other fields (with varying degrees of success): here is David Hendry applying his econometric methods to climate change, for example. So not only did economists not talk much to other social sciences, they trod on toes as well.

But although there may still be important branches to be added [2], the limitations of what you can do with a few axioms about rational choice have led in recent years to economics becoming much more empirical, and much less tied to this deductive theory. (See the article by Noah Smith which began this post. Unfortunately in my view an exception to this trend so far is macroeconomics.). We can see this in the citations data above, and the most obvious manifestation is behavioural economics. But a more immediate example of a data rather than theory based idea is the gravity model in international trade, which lies at the heart of why Brexit is such a bad idea. It is irony indeed that just at the point at which we have all these articles attacking economics, a large number of people who believe the UK is committing a large act of self harm are seeing the virtue of just one small part of what economists do.

Having said all this, I think there is an unfortunate hangover from this insularity. As a discipline economics shows little interest in communicating its core knowledge to others [3]. This can be true both within academia and with the outside world. Within academia publishing in top economics journals still has far higher status than top journals in other disciplines. When it comes to policy and the public, there is a belief among many that when either requires our wisdom, they will seek out the best of us for advice. In part this epidemic of articles about the failings of economics reflects this communication failure. More importantly, both Brexit and Trump should be a wake up call that economists as a collective has to get better at communicating the core insights of economics.

[1] There are of course more underlying problems behind the UK productivity crisis beyond the negative shocks of austerity and Brexit. But economists overwhelmingly argue for more R&D spending and more public investment. In short if you want someone to blame for why the UK economy is currently in such a dire state, blame those who have ignored the advice of economists.

[2] Most of the good criticisms that I see of economics amount to requests to add to the tree. But economics is so rich that most things are possible. In part (but only in part) what is done follows the money: you will find it relatively easy to get money for work on free trade compared to work on rent seeking. To blame economists for that is just bad economics. As economists found out after the financial crisis, they had many tools to understand what had happened, but had just not applied them before the crisis.   

[3] I say as a discipline because I mean economists as a collective, not as individuals. There is no equivalent institutional infrastructure in economics to that built by the hard sciences. Of course many individual economists do their best, but there are also others who ignore the consensus to plug their own personal ideas or to further some political or ideological cause.

The fatal inconsistency within neoliberalism

The collapse of Carillion had some important implications for public sector outsourcing, as I discussed here, but I tend to agree with Will Hutton that the real lessons lie elsewhere. He writes

“While some public sector delivery is outstanding, notably in parts of the NHS, the general pattern is more patchy. It is for this reason that governments for decades have been contracting the private sector to deliver goods and services. Trying to extend that principle is not unreasonable if high-quality private sector partners step up to the mark: the problem is they are in such short supply.”

In other words the problem is not so much with outsourcing, if sensibly done by a government that does not automatically think private is best and a civil services able to write good contracts and effectively monitor quality. The problem is with the poor quality of so much British capitalism. Carillion is not just one bad apple, of course: this is a lesson that should have been learned from the financial crisis, and RBS in particular,

But how can that be, when the market ensures that only the most efficient firms survive? That idea, that the market ensures that only the most efficient prosper, is a central message of neoliberal ideology, and it has held UK and US governments under its sway since the time of Thatcher and Reagan. But that ideology contains a large and deep internal contradiction, which applies particularly to large firms like Carillion. To see what that contraction is, we need to talk about ordoliberalism and Ronald Coase.

Ordoliberalism is widely known as the German version of neoliberalism. It too celebrates the benefits of the market. It, like neoliberalism, ignores many of the failures of markets that Colin Crouch eloquently outlines and which economists spend a lot of time studying. But ordoliberalism does recognise one potential problem with their market ideal which neoliberalism ignores, and that is monopoly. Crouch makes a similar distinction in talking about market-neoliberals and corporate-neoliberals.

This distinction is important for reasons that go way beyond the textbook problem with monopoly: that prices are too high and output is too low. To see why it goes beyond that we need to turn to Ronald Coase. He was a British economist who subsequently worked at Chicago. His first major article was "The Nature of the Firm" written in 1937 which introduced the concept of transaction costs to explain the nature and limits of firms. The key point that Coase made relevant for our discussion is that, in theory, much of what firms do could instead be done by markets. For a non-economist this probably sounds strange, so I will briefly try and explain the idea.

In principle the activities of any firm could be duplicated by all its employees, including managers, becoming self employed. The worker who worked for a firm would become a self employed person who signed a contract with the self-employed manager. Furthermore, the self employed manager would not have to work with a fixed set of self employed workers, but for each task they could create a market for the task which any self employed person could bid to fulfil.

Coase asked why we have firms rather than this more market based setup. His answer is transactions costs. A typical worker employed in a firm has to do whatever (within reason) a manager asks them to do. This can vary, often at very short notice, depending on the needs of the firm. Replacing this with a large number of specific short term contracts would be very inefficient, because it takes time to write and read contracts (transactions costs). You could add search costs and many other costs that make the self employed contracting model normally inefficient.

You can see exactly these tensions, between the advantage of the market versus internalising activities within a firm, played out when a firm decides to outsource any of its activities. But these tensions have a darker side. To what extent have firms internalised the market so that they can exploit customers (traditional monopoly), exploit workers (if one large firm is the only potential employer for many workers i.e. monopsony), or as a vehicle for managers to exploit shareholders.

There are two paths you can travel once you recognise all this. The first, and more ordoliberal, is to distrust monopoly of any kind, and be highly skeptical of large firms and their creation through mergers. The second is to assume that firms are always right, and that large firms exist simply because they are more efficient at doing what they do than the same firm broken up. Given the darker side mentioned above, there is no logical reason to take the second path. Skepticism about large firms is the right path to take. The second path, where you presume that there are always good reasons why large firms exist, makes no logical sense, unless your goal is for whatever reason to defend these large firms. It is the path that neoliberalism took.

Will Davies has a fascinating paper of how ideas about the virtues of competition, and the need to break up monopolies, changed within the University of Chicago from the 1930s until the 1980s under the influence, among others, of the work of Ronald Coase. It shows how many of the leading actors during that period chose the second path. Economics is rich enough that, if they so wish, an economist can always invent reasons why monopolies might be efficient, and so earn lucrative fees for so doing.

Thus neoliberalism that might have started as extolling the virtues of the market ends up as “a consultancy racket, in which corporations purchase complex models to justify their chosen strategy”, to quote Will Davies more recently. This is what I meant when I wrote back in 2016.

“It is important that those who use the term neoliberalism today recognise this contradiction. It does not mean that using the term neoliberalism to describe the dominant ideology is wrong, but it is a mistake to assume the ideology has not be moulded/adapted/distorted by those in whose interest it works. These changes have made it intellectually weak at the same time as making it politically strong.”

There are important mistakes that those opposing neoliberalism can make if they do not see this point. They involve opposing anything your opponent appears to favour. Thus, for example, because neoliberals preach the virtues of the market but in practice are normally simply pro-business, this does not mean that you should always oppose markets: you can be pro-market and anti-neoliberal. More generally, just because neoliberals can use ideas from economics to argue their case does not mean there is something wrong with economics (as opposed to some economists). Louis Zingales, from Chicago, exemplifies both points, and is not afraid to make the distinction between pro-market and pro-business in public.

This is all related to the distinction between ideas and interests. While it is right to recognise that interests are to some extent based on ideas, it is also important to see that ideas can change in time to reflect interests. How an ideology based on seeing markets through rose colored spectacles turned into an ideology justifying crony capitalism is at least part of the story of neoliberalism. These compromises with power allowed the ideology to become dominant, but at the same time it also made it rather easy for it to be subverted by different kinds of plutocracies, such as those represented by Brexit and Trump.

Why does economics get so much stick?

Because the advice of economists is so hopeless, you may say. Well think about the following thought experiment. After the financial crisis. suppose people had done the opposite of what the majority of economists said they should do. We do not need to imagine over Brexit, because most of the 52% who voted for Brexit chose to ignore, or more likely did not hear, the advice of 90+% of economists that Brexit would make them worse off. For those who work that belief was quickly shattered as their real wages fell as a direct result of Brexit.

Immediately after the financial crisis interest rates would not have been cut and austerity would have started in 2009, not 2010. Banks would have gone bust because economists said we needed to bail them out. In which case the Great Recession would have become the second Great Depression. Because the majority of economists did not support austerity you would have had continuing cuts in spending during this new depression.

So comparing this thought experiment with reality, we can see that economists have prevented a rerun of the 1930s depression, and if their majority advice had been taken we would have had a stronger recovery and the UK would not have left the EU. Sounds pretty good to me. But, as I’m sure you are now saying, what about the financial crisis the economists failed to warn of?

That was a mistake, but what are the consequences? Do you really think that if most economists had warned about how fragile the sector was anything would have happened? Banks would have continued to lend because they were making money and they had a guaranteed bail out from the state. Their campaign contributions would have weighed far more heavily in politicians’ minds than warnings from economists. So yes, not warning about the financial crisis was a mistake, but it would not have changed anything if the mistake had not happened. Economists are often told to stop being naive about politics, but the same needs to be said to their critics.

Despite such a strong record in macroeconomics since the crisis, why does economics get so much stick? I think there are three reasons. The first is simple: when the economy goes wrong, economists are easy to blame, particularly because of those forecasts that never predict downturns. In reality virtually no academic macroeconomists are involved in forecasting because they know that kind of unconditional forecasting is a mugs game [1], and furthermore most economists are not macroeconomists, but for some that kind of detail is irrelevant. (There are also plenty of highly successful pieces of microeconomics, but most critics act as if economics was just macroeconomics.)

The second reason is politics. Carlyle in 1849 called economics the dismal science because economists did not support his idea of reintroducing slavery. Ever since then economics has annoyed politicians and their supporters of various colours by pointing out the problems with various political programmes or schemes.

Politics is also at the heart of the third reason for criticism: politicians and ideologies of the right use the aspects of economics that suits their cause. Want to promote markets? Just take the idea from economics that an ideal market is an optimal way of exchanging goods, and ignore all the ways that real markets deviate from this ideal (ways which, incidentally, a great many economists spend a lot of their time studying). Some heterodox economists of the left, rather than use mainstream economics to point out how the right plays fast and loose with economic ideas, prefer to suggest that mainstream economics is much closer to the right wing caricature than it is in reality. It is why, as Noah Smith observes, so much of this criticism can be found in the pages of the Guardian.

This misrepresentation of mainstream economics is either deliberate or reflects ignorance. Ignorance about the fact that a lot of economics has become more empirical and therefore more eclectic in its use of theory over the last few decades, perhaps in part because of the influence of behavioural economics. Ignorance that even in macroeconomics, where ideological influences can be strong, there is more consensus around New Keynesian economics than some mainstream Keynesian economists imagine. (See my survey with André Moreira of post graduate teaching at the top schools here.) Nowadays you will find that in most areas of economics (alas not yet macro so much) there is nothing limiting the analysis to selfish individualistic behaviour. The idea that economics is like a religion is absurd.

But sometimes it is hard not to believe that popular criticisms of economics choose to ignore how far economics deviates from the neoliberal characterture. There is no excuse for ignoring that, for example, the best arguments against health care being left to the market can be found in a paper by Nobel prize winning economist Kenneth Arrow written decades ago. As the recent book by Colin Crouch suggests, the best critiques of neoliberalism come from within economics.

Another ridiculous charge against economics is that economics has a natural bias against state intervention. Indeed it is possible to argue the opposite. In my own field it is typical to assume the existence of a benevolent policy maker, who maximises social welfare. It is essentially just a useful analytical device, but you could argue if you wished to that this device biases those that use it to favour state intervention.

Judging by recent conversations I have had, many heterodox economists attack the mainstream because it uses the distinction between positive (value free) and normative economics. An example of positive economics would be me saying a temporary cut in government spending when interest rates are stuck at their lower bound reduces output. A normative statement would be that austerity is unfair. Heterodox economists like Sheila Dow seem to suggest that everything is value laden, and the positive/normative distinction allows economists to avoid being “morally implicated in the advice they give.”

I think this criticism is either trivial (yes, of course there may be normative reasons for choosing particular research topics) or dangerous. It is dangerous if it suggests that economists should be encouraged to base their analysis on assumptions that reflect their values. Economics, even though it is a social science, should conform to the scientific method: it should be as much like a science as medicine. Indeed I think it would greatly improve the public debate if both economists and their critics realised that economics, even though it is a unique and inexact science, is more like medicine than any of the hard sciences.

Dow writes “Getting policy-makers or the general public onside over a particular argument is therefore, critically, a matter of persuasion rather than demonstrable proof (since that proof is impossible).” But surely the best way of trying to persuade a policymaker not to impose austerity is to say that most models, including the consensus theoretical model, and nearly all the evidence suggests austerity will reduce output. In contrast it is far too easy to persuade a politician of things they want to hear. We do not want politicians to pick advice only if it is given by ‘one of us’ (by those who share their values), or as a result of the rhetorical skills of the academic.

The danger in encouraging plurality is that you make it much easier for politicians to select the advice they like, because there is almost certain to be a school of thought that gives the ‘right’ answers from the politicians point of view. The point is obvious once you make the comparison to medicine. Don’t like the idea of vaccination? Pick an expert from the anti-vaccination medical school. The lesson of the last seven years, in the UK in particular, is that we want mainstream economists to have more influence on politicians and the public, and not to dilute this influence through a plurality of schools of thought.

All this does not mean that economists are beyond criticism. As my last post pointed out, I have fundamental criticisms about current macroeconomic methodology. An important point to note about the microfoundations methodology is that it excludes economists who are not prepared to sign up to what is currently considered (by macroeconomists) acceptable microeconomics, or who do not think microfoundations is where you have to start in doing macro. But this critique has nothing to do with values. The mistake macroeconomics made in the 1980s was not in their desire to look for microfoundations, but in deciding that models that had internally consistent microfoundations were the only admissible models.

The big problem with most criticisms of economics you see in the media is not that economics is beyond criticism: as the paragraph above suggests it in many cases should be criticised, and there are plenty more interesting criticisms of economics available. The problem is that these more important criticisms are not those you find in the pages of the Guardian. The typical criticisms you see in the press are just not very good, and I fear reflect either ignorance or ideological antipathy.

[1] A lot of the criticisms of forecasters are themselves spurious. Someone who writes “economists should not need to pretend that we can predict things that do not really matter to several decimal places” are themselves pretending that there are any serious forecasters who do pretend this.

How Neoliberals weaponise the concept of an ideal market

This new book by Colin Crouch will perplex many on the left who simply believe neoliberalism has to be overthrown. Indeed the author starts his book by talking about the Grenfell Tower disaster, which he along with many others believe epitomises the failings of neoliberalism. Yet he writes that the book

“is not a contribution to the demonology of neoliberalism, but an attempt at a nuanced account. Only in that way can we assess its capacity for reform.”

Such an account can of course also be justified on the basis of intellectual curiosity, but in addition the author sees some positive aspects of the ideology: He summarises these as

“the discipline of price and calculation [recognising efficiency and opportunity cost}; helping us appreciate the limitations of democratic government; facilitating trade and reducing barriers to it; and facilitating links among people [reducing national divisions].”

So what exactly is neoliberalism? He defines it as

“a political strategy that seeks to make as much of our lives as possible conform to the economist’s ideal of a free market”

The problems and deficiencies of this strategy come when the conditions required for the free market to be ideal do not hold, and the author’s long discussion of these problems would be useful for any economics undergraduate.

One of these conditions for an ideal market is competition: a free market is an ideal from a social point of view if (alongside many other conditions) each good is produced by a very large number of producers. The author recognises, for obvious reasons, that most neoliberals (as opposed, perhaps, to ordoliberals) tend not to go around wanting to break up monopolies and reduce monopoly power. As a result, he distinguishes between market-neoliberals who might, and corporate-neoliberals who would not. He talks about past competition (that may have resulted in monopoly) and current competition. As Luigi Zingales describes it rather well here, business tends to be in favour of a competitive market before it enters it, but once it has a dominant position in that market it is happy to put up barriers to further competition.

The author goes on to discuss conflicts between corporate and market neoliberalism, and much else besides. I think it is a great book, free from unnecessary jargon that you often find elsewhere. It got me thinking about the concept of neoliberalism again as you can see below. Whether that is a good thing or not, I would encourage you to read the book. The author also of course discusses whether he thinks neoliberalism can save itself. For his answer to that question you will have to read the book.

Now, for what it is worth, are some of the thoughts the book inspired. They go back to the distinction between market-neoliberals and corporate-neoliberals. It seems a little odd to define an ideology as the evangelisation of the free market, and then go on to say most neoliberals happen to exclude a crucial component of that free market (competition) when it suits them. I am quite prepared to believe that some of the people who first wrote about neoliberalism many years ago (and perhaps one or two today) could be described as what the author calls market-neoliberals, but as I have suggested in the past I think neoliberalism has evolved (or if you like been distorted) by ‘big money’ or capital to become a tool for self justification.

As a result, I would tend to suggest a slightly different definition that seems to work quite well today. The definition would be: 

neoliberalism is a political strategy promoting the interests of big money that utilises the economist’s ideal of a free market to promote and extend market activity and remove all ‘interference’ in the market that conflicts with these interests.

This replaces a definition based on following an idea (the author’s market neoliberalism), by one of interests promoting an idea so long as it suits those interests.

This alternative definition seems to fit two cases I have used in the past to question more conventional ideas. Large banks benefit hugely from an implicit subsidy provided by the state (being bailed out when things go wrong), but neoliberals do not worry too much about this form of state interference in the market (whereas economists do). Regulations on the other hand they do complain about. It is a very selective focus on market interference.

The second is executive pay. This is always justified by neoliberals as being something determined by the free market, when obviously it is not. Yet if you pretend that there is a market in executives and salaries etc are set by that market and not the remuneration committees of firms, then you are being a good neoliberal by defending these salaries. This example is interesting because it involves defending one part of ‘big money’ (CEOs or some workers in finance) at the expense of another (shareholders). It is why I do not talk about the interests of capital in my definition. 

Is this alternative definition simply negating the power of ideas and going back to good old interests? Only in part. Interests utilise an idea because the idea is a powerful persuasive tool. There is an obvious lesson for the left here. Because neoliberals promote the concept of an ideal market only when it suits them, so opposing neoliberalism does not necessarily mean opposing the concept of an ideal market. The left should utiliise the same concept to oppose monopoly power, for example. The idea of a free market is too powerful an idea to cede to the other side.