Posts Tagged The Lucas Critique
Sometimes it seems like economist’s pet principles are applied selectively, in such a way that they attack ideas generally endorsed by the left end of the political spectrum. This isn’t to say economists themselves are ideologically inclined toward any opinion; merely, that key aspects of their own framework, and the way they present these aspects, lends itself to a more ‘right-friendly’ way of thinking.
In part, the issue is merely one of a disparity between how economists present issues to the public and how they speak to others in academia. Dani Rodrik noted this issue in his book The Globalisation Paradox. Here he describes a situation where a reporter asks an economist whether free trade is beneficial:
We can be fairly certain about the kind of response [the reporter] will get: “Oh yes, free trade is a great idea,” the economist will immediately say, possibly adding: “And those who are opposed to it either do not understand the principle of comparative advantage, or they represent the selfish interests of certain lobbies (such as labor unions).”
Rodrik then contrasts this with how such a question would be answered in the classroom:
Let [the student] pose the same question to the instructor: Is free trade good? I doubt that the question will be answered as quickly and succinctly this time around. The professor is in fact likely to be stymied and confused by the question. “What do you mean by ‘good’?” she may ask. “Good for whom?… As we will see later in this course, in most of our models free trade makes some groups better off and others worse off… But under certain conditions, and assuming we can tax the beneficiaries and compensate the losers, freer trade has the potential to increase everyone’s well-being…”
This adherence to basic, market-friendly principles over nuance can be found often in ‘pop’ economics: for example, economist Paul Krugman does it in his book Peddling Prosperity. The book is intended as an survey of nonsensical ideas from both the left and the right, remedying them both with a cold hard dose of facts, plus some basic economics. However, Krugman treats the left and right somewhat asymmetrically: with the right, he primarily opts for facts, whereas with the left, he uses economic principles
This is quite possibly because the right’s arguments, though they are taken to an extreme, have economic principles on their side, while the left’s do not. The ‘supply side’ economics that Krugman takes issue with is really just an extreme statement of the well known principle of deadweight loss, which suggests that taxes decrease output. If taxes reduce output by enough, then it logically follows that not only output, but overall revenues might fall if we raise taxes. Krugman would not question the principle, so he spends several chapters documenting evidence against the idea*.
Krugman then follows this up with a section berating the ‘strategic traders’, endorsed by Bill Clinton and others on the centre-left. Strategic trade suggested a role for government policy in promoting industry, because various clustering effects, economies of scale and positive feedback loops could mean that the initial wave of government investment could kick start an industry. As Krugman himself notes, such dynamic effects and ‘historical path dependence’ could render comparative advantage obsolete, since comparative advantage posits a more fundamental, innate reason a country produces a particular good, one that cannot be changed with policy (one that may be more applicable to agriculture).
Yet, in contrast with his section aimed at refuting the right, Krugman offers scant evidence suggesting government intervention doesn’t work. Instead, he effectively restates the theory of comparative advantage, coupled with a typical story to illustrate it. This is despite explicitly suggesting it might not be applicable in the previous chapter. When pushed, Krugman is prepared to fall back on his pro-market principles, even in areas where he knows they may not apply.
William Easterly does something similar in his book The Elusive Quest for Growth. The book is a survey of various policies than have purported to be panaceas for development, such as education, investment and population control. (As you can see, economists really love writing their “I’m an economist, here’s how it is” manifestos). Easterly finds every supposed development panacea wanting based on the available evidence, which is fine. However, occasionally he supplements his arguments with an excruciating example of ‘economic logic’ that always looks out of place.
For example, in the section on increasing availability of condoms, Easterly essentially makes the argument ‘how could people be lacking condoms? If they were, the free market would provide them!’ I am reminded of the joke about the economist who does not pick up a £10 note from the ground, because, if it were really there, somebody would already have picked it up. Easterly is a smart guy with a lot of concern for the poor, and I have a hard time believing he wouldn’t agree that a country might lack the institutions to deliver condoms, that people might lack the education to know why they’d need them, that it might conflict with their beliefs, etc. But the ease with which he can apply a pet economic principle is just too tempting, so he ignores these factors.
Another example is where Easterly asserts that population growth cannot be a problem, because “an additional person is a potential profit opportunity for a person that hires him or her” and as a result “the real wage will adjust until the demand for workers equals the supply.” It’s quite clear things don’t function this smoothly in labour markets even in developed countries; for theoretical reasons as to why, Easterly need look no further than John Maynard Keynes; failing that, modern work on labour market frictions might prove sufficient. Again we see a neat but overly simplistic principle applied when even the economist themselves surely knows better.
So it is not uncommon for economists to prefer their more ‘free market’ principles over nuance when writing for a popular audience**. But is this problem only limited to popular economics? Economists seem to think so; to them, the issue is primarily one of communication, and knowing the limits of your models. This is fine as it goes. However, there are reasons to believe this bias extends into the murky depths of academia.
In my opinion, there is one major culprit of selective application in economics, and it is one that cannot be explained by economists simplifying their work for public consumption: the Lucas Critique. The Lucas Critique suggests that adjusting policy based on observed empirical relationships from the past will alter the conditions under which these observations were generated, hence rendering the relationship obsolete.
Unfortunately, in practice, Lucas’ version of the critique seems to have been used to beat ‘Keynesians’ over the head, rather than being universally applied as a tool to further understanding. To illustrate this, here are some areas I think Lucas critique-style thinking could be applied, but hasn’t:
- Milton Friedman’s methodology. If a ‘black box’ theory corroborates well with past evidence but we aren’t entirely sure the internal mechanics are accurate, there’s no reason to believe the corroboration will hold, or to know how the mechanics of the system will change, if we change policy.
- Nominal GDP Targeting (NGDPT). This hasn’t caught on much on the left (in my opinion, for primarily ideological reasons: it’s anti-Keynesian, it partly absolves the private sector of responsibility for recessions). But it doesn’t seem to have occurred to proponents of NGDPT that we must ask if the relationship between inflation, RGDP and NGDP will break down if we try to exploit it for policy purposes. This is despite the fact that we are talking about precisely the same variables as the Phillips Curve, the primary theory to which the Lucas Critique was initially applied.
- The supposed “deep parameters” of human behaviour on which Lucas suggests we construct economic models, such as technology and preferences. For a neoclassical economist, you are born with a set of preferences and you die with them, while in many models technology is a vaguely defined exogenous parameter. Yet a single example can show that both of these things can change with policy: government investment, which is at the root of a large number of technological break throughs. These break throughs have often resulted in new products, creating preferences that otherwise wouldn’t have existed. A model with fixed, exogenous parameters for technology and preferences is therefore hugely fallible to policy changes.
The fact that the critique hasn’t been applied to these examples leads me to believe it’s often only used to preserve existing economic theory. In fact, the critique itself is really just a narrow version of the more general principle of reflexivity, noted by many before. Reflexivity is an ever-present problem that suggests an evolving relationship between policy and theory, not a principle that means we can fall back on economist’s preferred methods.
Is the Lucas Critique the only culprit? Well, I’ve found economists are generally critical of the assumptions and mechanics of heterodox models, despite appealing to Friedmanite arguments when questioned about their own. I’ve also found economists (okay, one) appeal to how businessmen really behave when defending their theories despite not paying much credence to alternative theories based on the same principle, such as cost-plus pricing. So maybe economists need to air out their theories and principles a bit, rather than simply applying them where it suits them.
Economist’s simple stories often capture some truths, which is why they will defend them to the death. But too often this becomes a matter of protecting a core set of beliefs, and being unwilling to apply them in new ways or even abandon them altogether. So economists end up deferring to their framework when it isn’t appropriate, or only interpreting it in their preferred way, particularly when they communicate their ideas to the public. The result can be that misleading conclusions about the economy remain prominent, even when economist’s own frameworks, interpreted completely, don’t necessarily imply them. Perhaps if economists were more willing to open up their theories, which can sometimes feel like something of a black box, these misinterpretations would be exposed.
**In fairness to Krugman and Easterly, these books were written a while ago, and I’m sure they have updated their positions since then. I only wish to show that economists use this tactic, not that any one economist endorses any particular position.
I’ve previously spoken about how many great insights supposedly ‘discovered’ by economists – classical and modern – had really been known for a long time, but had been ignored or perverted before they were put in terms neoclassical economists approved of. The more I learn, the more it seems that this is the case with a vast amount of critical ‘insights’ on which macroeconomists pride themselves.
The fact is that 1970s ‘anti-Keynesian revolution’ was really just a restatement of things already stated by Keynes and Phillips – who were incidentally two of the main targets of the revolution, but both completely misinterpreted.
The psychological time-preferences of an individual require two distinct sets of decisions to carry them out completely. The first is concerned with that aspect of time-preference which I have called the propensity to consume, which, operating under the influence of the various motives set forth in Book III, determines for each individual how much he will reserve in some form of command over future consumption.
But this decision having been made, there is a further decision which awaits him, namely, in what form he will hold the command over future consumption which he has reserved, whether out of his current income or from previous savings.
Furthermore, the Lucas Critique itself – the idea that macroeconomic aggregates cannot be used as a guide to future policy because a change in policy will change behaviour and therefore relationships between variables – was also stated by Keynes:
There is first of all the central question of methodology—the logic of applying the method of multiple correlation to unanalysed economic material, which we know to be non-homogeneous through time. If we are dealing with the action of numerically measurable, independent forces, adequately analyzed so that we were dealing with independent atomic factors and between them completely comprehensive, acting with fluctuating relative strength on material constant and homogeneous through time, we might be able to use the method of multiple correlation with some confidence for disentangling the laws of their action…in fact we know that every one of these conditions is far from being satisfied by the economic material under investigation.
Even more ironically, it was stated by the primary target of Lucas’ criticisms, the much misunderstood Edmund Phillips:
In my view it cannot be too strongly stated that in attempting to control economic fluctuations we do not have two separate problems of estimating the system and controlling it, we have a single problem of jointly controlling and learning about the system, that is, a problem of learning control or adaptive control.
One might argue that Lucas deserves credit for formalising this point, but in reality I think Phillip’s one sentence formulation is better – it emphasises continual change and vigilance in recognising the feedback loop between policy and the real world. In contrast, it seems the Lucas Critique has become little more than a tool with which to cling to outdated methodology despite empirical falsification.
This is why I am frustrated when people like Krugman say that they “don’t care” what Keynes or others “really meant,” and people like Scott Sumner and Robert Lucas pay barely any attention at all to the history of thought. Ignoring the history of thought just means you are condemned to rediscover the same insights over and over – often, it seems, in a far less enlightened way than they were originally stated.
As we expand debt in the process of want creation, we come necessarily to depend on this expansion. An interruption in the increase in debt means an actual reduction in demand for goods.
In fairness to Keen, I wouldn’t paint him with the same brush as the above – he readily acknowledges that this insight was noted by Schumpeter and Minsky before him.
The doublethink with which mainstream economists are able to ’embrace’ new economic thinking and simultaneously shout down any attempt at, well, new economic thinking, is quite incredible. For example, see the infamous Krugman/Keen debate, where Krugman behaved startlingly similarly to his usual opponents on the right, even despite his own essay readily acknowledging the failure of economics in the crisis. Mark Thoma also had a similar reaction to Keen, but he then went on to write an essay praising new economic thinking.
In fairness to Thoma, his essay appears to acknowledge that economists like him are set in their ways and cannot embrace change on the scale required, but of course this in itself isn’t encouraging. Why can’t they change their ways? Why carry on if you strongly suspect your paradigm is flawed? I am reminded of a Richard Dawkin’s documentary, where he spoke of a scientist who had been working on a theory for the best part of his career. A new scientist arrived at his place of work, and falsified the theory – the older scientist, however, thanked him. This type of attitude would be helpful in economics.
Watching economists react to Keen’s work, and the work of others, I feel there are a few major barriers to economists accepting new economic thinking; once these are addressed they, hopefully, will find it easier.
Identifying Neoclassical Economics
Neoclassicism is seen by some economists as either a non-existent school of thought; a swear word used by their opponents, or as a long outdated paradigm which has been abandoned in favour of sticky wage/price models and other developments in the DSGE framework. So the first step towards engaging sympathetic neoclassicals is convincing them that their school exists.
I have commented on this briefly before, but I think that Christian Arnsperger & Yanis Varoufakis’ essay on this subject is excellent. It identifies neoclassicism as a methodology, rather than as the ‘rational self maximising, perfectly informed’ agent criticism that economists are so able to brush off with appeals to higher level work. That neoclassical economics uses methodological core of individualism, instrumentalism (preference satisfaction) and equilibration is hard to dispute, and so is an important starting point different schools to engage each other in both directions.
The Lucas Critique
(1) It has given grounds for economists to revert to their old mantra of ‘that’s OK in practice, but does it work in theory?‘ Krugman’s first post on Keen mentions that there is ‘a lot of implicit theorising’ in his paper – in other words, there are no microfoundations. Krugman uses this as grounds to dismiss the overwhelming empirical relationship between private debt and other economic variables.
(2) In practice, application of the Lucas Critique has basically amounted to the use of rational expectations and representative agents, rather than any deep change in economic modelling. There is a great discussion of the flaws in these approaches here, but that’s not necessarily relevant – what matters is that the LC is only applied sparingly.
(3) There is no empirical evidence to support it’s application – that is, it doesn’t appear to be useful when developing new theories or policies.
(4) The most important criticism of Lucas’ paper is that he suggests we model based on the ‘deep parameters’ of human behaviour. As anyone with even a passing familiarity with anthropology and history should know, these parameters simply don’t exist. You can find people throughout history behaving in any number of ways, both as societies and as individuals – even the most basic instincts, such as the need for sustenance and reproduction, have been overcome by environment (abstinent monks, lent, self sacrifice). The fact is that, for economists, ‘deep parameters of human behaviour’ seems to mean nothing more than the individualist, instrumentalist core outlined by Arsnperger & Varoufakis. This is as vulnerable to the Lucas Critique as any other theory or methodology.
So what are we left with? In essence, a suggestion that using a model for policy might have unintended consequences. This is true, and unfortunate, but it’s the reality of the social sciences, and has been known for a long time.
The ‘It Doesn’t Matter’ Mentality
Economists sometimes acknowledge that a model is flawed, but assert that the real world still behaves as if it corroborates with their models. This mentality can be found in one of my textbooks:
…[the student] rightly assumes that few firms can have any detailed knowledge of marginal revenue or marginal cost. However, it should be remembered that marginal analysis does not pretend to describe how firms maximise profits or revenue. It simply tells us what the output and price must be if they do succeed in maximising these items, whether by luck or judgement.
And also in Nick Rowe, defending exogenous money on the grounds that:
So the central bank must stop them creating loans and deposits out of thin air. The central bank will raise its rate of interest by whatever it takes to stop banks creating loans and deposits out of thin air. It is exactly as if the banks were reserve-constrained and couldn’t create money out of thin air.
This is one of those positions that I find it hard to articulate a response to. Of course it matters that we get the mechanics of a system right, otherwise we simply don’t have a model of the system – we’ve got something else! This is what I’ve been trying to get at with useful assumptions – useful ones simply eliminate a complication, whereas ones that are essentially hypotheses about how agents behave can be falsified in their own right. Economists seem to enjoy clinging to the ‘hypothesis’ variety of assumption, and this needs to stop.
There are some other important traps economists fall into – three of which I mentioned in my post on how to unlearn economics. From the perspective of accepting new economic theories, however, these three (and maybe the third one in the aforementioned post) are the most important – if they are not addressed, heterodox and mainstream economists will continue to talk past each other.