T’aint lark playin’ for Yorksha is it?

Last week, Gummo Trotsky wrote a feisty post, criticising the intervention of economists into the question of the impact of mobile phone usage on accident rates. More recently, commenter whyisitso on Andrew Norton’s blog asked economist Harry Clarke how he presumed to write about climate change.  

“I’ve taught students about climate change aned have written several papers on climate change – the earliest in 1991. My work is cited by leading economists in the area such as Robert Pindyck.”

OK Harry. I don’t doubt that for a layman you are well informed. But your comment brings to mind the wonderful (repeated) interview by Margaret Throsby yesterday with Michael Parkinson.

Parkinson relates the story of his ageing father’s comment about his own (ie Michael’s) fame and achievements – “Aye Lud. But t’aint lark playin’ for Yorksha is it?”

Tain’t like you’re a climatologist, is it Harry?

In both cases, I can see the frustration with economists presuming to stick their noses into areas that others (public health researchers, climate scientists) have been working on for years. In many cases, I think the annoyance also arises because such economists frequently don’t even devote their entire careers to that particular question. I’ve been blamed myself for doing just this with papers that range into public health, political science, and education. (When my last teacher quality study came out, the head of the Australian Education Union Pat Byrne – who I quite like on a personal level - told ABC TV that it proved “he just doesn’t live in the real world”.)

There’s certainly a kernel of truth in the critique of economics as imperialism. Economists often take too little account of previous research, and sometimes overstate their own contribution. But I do think that our empirical techniques have in many cases a lot to contribute to debates in other social sciences. To name three points, I think our training helps us handle very large datasets, think about causal inference more rigorously than other social sciences, and model incentives (Greg Mankiw has a less modest list, while Ed Lazear is the classic reference). And if you’re a non-economist social scientists who doesn’t like the contribution of economists to your favourite topic area, you’ll be pleased to know that the assumption of free disposal (of economics research) still holds.

Advertisements
This entry was posted in Economics Generally. Bookmark the permalink.

26 Responses to T’aint lark playin’ for Yorksha is it?

  1. derrida derider says:

    Of course, economists have long had the same complaints about physicists and engineers sticking their heads into their patch.

    It seems to me the only really valid complaint here is that the imperialists often don’t bother to get across existing knowledge and known methodological problems. So at best they often end up solemnly reinventing the wheel and at worst they remain ignorant of the problems that led to their approach being long-ago abandoned in the first place. That’s certainly the case with physicists and engineers dabbling in economics, and it’s probably the case in many places where economists stick their beaks.

    But other than that the colonised should welcome the colonisers – the more perspectives and expertise brought to bear on a question, the better.

    On climate change, that’s not economic imperialism at all. Climatologists’ expertise is in telling us what’s likely to happen, but straight economics gives the best guide to what we should do in response.

  2. Leon says:

    I think a similar phenomenon occurs with Mathematicians in the physical sciences and Philosophers in the humanities. Sometimes those with the most abstract, “meta-” knowledge feel they are better qualified or can pick up on the fallacies of those in conceptually “lower” disciplines. A lot of the time it seems they’re right, but who hasn’t met a philosophy student who feels they know everything.

  3. Verdurous says:

    Clearly, there is much to be gained from economists contributing to wide and diverse subjects. There is a pressing need for cross-fertilisation between disciplines. One must be cautious however. Derrida rightly highlights the potential for inadequate background knowledge of the subject area.

    (I would strongly disagree with derrida that straight economics should be the sole guide to how to respond to climate change. This is reductionist in the extreme. Adherence to “straight economics” is part of the problem that got us here! But of course it will contribute to solutions in a major way.)

    My suggestions for academic forays into fields hitherto unexplored:

    i) approach the area with humility and much preparation.
    ii) discuss the issue with experts in the field.
    iii) seek peer review from diverse disciplines.
    iv) publish in existing journals that cover the topic rather than creating a new micro-discipline.
    v) be open to others contributing to “your field” – for instance I think the pure and (especially) biological sciences have many insights that may contribute to economics, despite derrida’s quibble about “physicists and engineers dabbling in economics”.

    Definitions of economics are plentiful and very diverse. As I recall, Andrew has previously explained that economics is defined not by subject matter but through these features:

    i) researched by economists (a circular argument)
    ii) uses statistical techniques developed in the field (one could use this argument to alternatively call most papers “the epidemiology of….”)
    iii) primarly concerned with efficiency. (hmm.. maybe this one holds water)

    My take on this is that it is OK to call the topic “social science” ( or “health” or “anthropology” or whatever) if this is what is really about. Otherwise economics becomes all things to all people and loses any real meaning.

    But above all…humility and openness…for all adventurers on the seas of knowledge.

  4. David Walker says:

    Since Roger Short’s comments referred to economic growth, climate change policy responses and (by clear implication) the effects of population on emission levels, I think there’s a fair argument that Harry was squarely on his own patch.

    This is the way of the world; economics is about responses to facts. That’s exactly why some people become economists rather than climatologists or experts in reproductive biology. Knowing that something is happening doesn’t usually tell you how to respond, but economic analysis can.

    Other disciplines will resent this. The most obvious fix is to suggest they learn some economics, the same way biologists have had to learn mathematics over the past couple of decades. If you want to play in the public policy field. you just need some economics.

    One of my great frustrations as a non-economist teenager at law school in the 1980s was that we were continually asked to address policy questions, in fields from penology to environmental law to taxation, that could only be answered with economic analysis. Most of my fellow students didn’t see this as a problem. I found that instructive in itself.

  5. Patrick says:

    I suspect that the real reason is that economics often offers poor support for the person in question’s preferences.

    Personally I like the influence of economics, even in law. But I suspect that if David Walker saw economics in every legal policy question then part of the answer was choice of vocation. Economics can inform legal reform, and can help understanding legal structures, and helps evaluate legal systems, but doesn’t necessarily tell you whether a given law is good or bad.

    Human rights laws are a classic example, although I actually think economics deserves greater weight in that debate than it gets (see eg South Africa).

  6. derrida derider says:

    Oh I’m not decrying all physicists and engineers who play in economics – historically the engineers in particular have made some big contributions. And yes, evolutionary economics – an important field – was basically created by wandering population biologists and ecologists.

    On response to climate change I can’t see what other field of science can illuminate the necessary tradeoffs for policy (with the possible exception of political science for policy feasibilty and sustainability). Certainly the few sociologists’ dabblings that I’ve seen in this sort of stuff is deeply unimpressive.

  7. Kymbos says:

    As a practicing economist, I’ve seen good and bad economics undertaken, but what worries me is when theoretically sound economics is implemented without a strong understanding of the context in which it’s applied. This is the approach which says “I have a solid theoretical knowledge of economics, and can thus apply it anywhere”. The result can be subtle effects with profound consequences, and poor advice.

    I think for economics to be useful, it needs to be backed by a detailed understanding of the market or context in which it is being applied. Failing that, it can be useless or dangerous.

  8. David Walker says:

    Just a clarification to my note above, prompted by Patrick. I have never seen economics in every legal issue. Rather, there were some issues in my legal education where we were specifically asked to make resource-allocation decisions. I remember a tutorial where one of the assigned questions was a discussion of the merits of a capital gains tax, an environmental law lecture where it was asserted that pay-to-pollute regimes were ineffective, and a discussion of the death penalty where most of the assigned reading was econometric studies (most of them, in retrospect, badly flawed).

    Sometimes people simply don’t realise they’ve stepped outside their own field of expertise.

    P.S. Gough Whitlam taught us human rights law for a little while. Economics didn’t figure. I didn’t miss it there, either.

  9. harry clarke says:

    I agree with the comments suggesting a complementary relation between economics and science. Then economics can guide you towards policy on the basis of goood scientific knowledge.

    But sometimes I think economists are just better-trained in quantitative reasoning and modelling than people in other fields and that a measure of imperialism is good for science in general.

    For example I have been studying the effects of smoking on the length of human life. Medical people have conducted long-term studies suggesting a loss of life of up to 10 years for long-term smokers. Recent studies by economists recognise that people who smoke tend to be risk lovers so that such people will drink too much and engage in other activities that reduce lifespan. Conditioning on these factors reduces the estimated loss of life due to smoking alone to about 4.4 years for US males.

    Maybe its a trite example but I often find that the training of economists often helps them to improve quantitative work in other fields. Economists should be modest but the core economics degree taught around the world is a useful start to various research-oriented careers that involve quantitative modelling of human behaviour.

    I try to listen to people working in related areas such as sociology and political science – and certainly we can all learn a lot from work in these fields – but is quantitative modelling by such social scientists on a par with comparable work done by economists? My intuition is – sometimes not.

  10. invig says:

    Yeah, you can handle datasets, but you shouldn’t be the ones asking the questions.

    You lot should be like technicians: on call for specialist jobs but otherwise stuck in a corner not causing trouble by mucking around with what you don’t understand.

    Cause how can you – understand I mean? With your head stuck in the world of stats, there’s no way you can keep abreast of the ‘real world’ of human interrelationships and macro effects. Its too complex and too fuzzy.

    Which is where I come in of course (or my work does anyway), but you econometricians will find out soon enough (when I ask for your help;)

  11. conrad says:

    HC,

    I thought all that stuff about risk taking and its influence on behavior was initially thought of by psychologists, not economists.

  12. Patrick says:

    I don’t know, Invig. Your comment is intuitively appealing, but I suspect that in practice the economists have often added the most value simply by thinking up the questions.

    Think, for example, of the revolution effected when a greeny economist thought ‘hey, I wonder how much that super-wonderful idea might cost?’ Or when an socially concerned economist thought to ask ‘does deterrence work? Where’s the evidence?‘ That last question is one that economists are perhaps particularly likely to ask.

    Lawyers are also prone to ask that, which is probably part of why Economics and Law proved such a popular discipline.

    NB: to clarify, I used human rights because, like Invig’s comment, it is an intuitively appealing example. In practice, I think that a greater dose of economics in human rights literature might a) lead to better protection of substantive human rights and b) lead to less concern for mushy social rights.

  13. invig says:

    Patrick,

    Thats the point, and the difference between conceptual theory-making and data-correlation.

    However, the problem with economics (as opposed to econometrics) is not that it does not prompt the ASKING of questions, but rather in the process it utilises already-flawed theory. As has been said elsewhere, much of economics since the turn of last century is bunkum (due to an unfortunate assumption of closed system equilibrium, when – due to the sun – no economic system on earth is closed).

    So, there needs to be a better process of theory construction, one not constrained by the mistakes of the past. Once theory has been produced, using the mind IN COMBINATION with conceptual tools that ACCENTUATE its abilities, the statistical tools and big data sets can be brought in to test it.

    NOT the other way around!!!!!!!!!!

    (and each of those exclamation marks was delivered individually and emphatically)

  14. christine says:

    Patrick & David: Having studied a bit of international human rights law and not worked in the area at all, I’d agree with you that a bit more economics would be nice there. So, if you want to be part of the solution (and help me out as well!), I’ve got a question for you: I’m teaching a course on ‘The Economics of Human Security’, as part of a team with two political science people next year (ie not terribly much explicit int law). Have you got any brilliant ideas on what students in the area should be learning?

    On the broader point that it’s helpful to know about the topic you’re commenting, I think that’s right in general. It does bug me when outsiders say “aha economists (or whatever) have missed this really important point, so it’s all nonsense!” when the economists are fully aware of the point, have written about it for 30 years and have pretty much solved it and moved on. There are some times, though, when knowing not very much about previous approaches is an advantage. Only every now and then, but you can get a complete regime change if someone from outside a discipline comes in and rethinks something crucial. Got to keep a bit of an open mind about these things, I think – having smart people working on a particular topic is a good thing, regardless of their discipline. If their ideas are good, the discipline can pick them up and run with them (as economics has with some biology, and vice versa, eg), and if not, they can criticise them on substantive grounds or ignore them.

    My guess is Gummo and whysitso are worried that economists/ econometricians are more likely to get a hearing in the corridors of power, even if they are wrong. (And Gummo thinks the econometrician is more likely to get the astrologist with the cute bum??? Any empirical evidence for that? I’d have thought it pretty unlikely, myself.)

  15. cba says:

    “As has been said elsewhere, much of economics since the turn of last century is bunkum (due to an unfortunate assumption of closed system equilibrium, when – due to the sun – no economic system on earth is closed)”

    nice straw man. ridiculous statements like that don’t make you credible, invig.

  16. conrad says:

    invig, here is Box’s comment, which should give you a different perspective on the usefulness of historical model and theory development and why people do it.

    “All models are wrong, some are useful”

  17. Meh says:

    As an engineer, economists are amusing. The best response to their “imperialism” is just to keep mocking them. Occasionally they actually will then engage with known problems in their work and correct them.

    The best bit is when economists claim “statistical superiority.” Given that most of them don’t even understand the significance tests they use, it’s sort depressing to see them go out and recommend policy on the basis of their implementation of those tests. There is a reason why most of the heavy lifting in industrial applications of economics (e.g. market simluations, hedging and the like) is carried out by physicists and engineers, not economists.

    The second best bit is when people like Mankiw claim that the publications by various economists in the Annals of Statistics carries through to the majority of public policy economics. Would that it were true! The reality is that most of public policy economics is done with back of the envelope curves that should thoroughly embarass the profession.

    Finally, the most amusing bit (as opposed to the best bit) is when economists try to apply their statistical knowledge to situations where the numbers just don’t have the precision or accuracy to be meaningful. You can see this with most of the work on “the economics of the firm.” What is it about economists, that makes them constitutionally incapable of analysing whether a model actually corresponds with some closeness to reality or not? Is it the “core economics degree” or is the PhD work? Inquiring minds would love to know.

  18. Patrick says:

    And Gummo thinks the econometrician is more likely to get the astrologist with the cute bum??? Any empirical evidence for that? I’d have thought it pretty unlikely, myself.

    It’s all relative…relative to a boring lefty, maybe they are 😉

    Christine, I don’t have any bright ideas, but South Africa post-apartheid is an example of human rights with too little economics. What is particular about it was that human rights were used as a means to a largely economic goal.

    You can also look at ‘the economics’ of human rights action – eg Amnesty over the last decade is an interesting case study there.

    I don’t know how broad the scope of that subject is, but you could also consider ‘human security’ as an economic problem, and contrast various practical models (tribalism, monarchy, autarchy, clerarchy, fascism, democracy) and theoretical models (eg Nozick and Marx).

    Alternatively, and in contrast to the South African case, you could study the use of economic means to human rights ends – sanctions, divestments, the WTO, and even in some cases, subsidies (look at some of the European rhetoric behind subsidies/tariffs).

    Finally, I am already sure from your comments on this blog that you will cover the basic value of time spent on cost-benefit analysis.

  19. invig says:

    cba: I’ve got credibility to burn champ. Address the point or go away.

    conrad: I know about models, i have an engineering degree. “Wrong” is a misleading word that presumes a single dimension of correctness.

    Meh: word 🙂

  20. Where’s the empirical evidence that chickens cross roads to get to the other side or that elephants paint their testicles red to hid in apple trees?

    To the casual observer the comments here present an interesting, if self-selected, sample showing an inverse correlation between taking economics seriously and the ability to see the funny side of economists inflated claims for their discipline and professional training.

  21. My pet peeve are cases were an ‘economic’ model is proposed to examine an aspect of human behaviour, the model then ‘explains’ a tiny portion of the behaviour (very low R-Sq), but the author never makes this clear to the lay reader and then builds a castle of speculation on the assumption that he (and it is always a he) has found the real, true, hard-headed explanation. Posner’s Public Intellectuals is a case in point.

  22. Andrew Leigh says:

    I’m enjoying the exchange, but I would ask a few people to drop the unnecessary put-downs. That goes double if you’re commenting anonymously; pseudonyms shouldn’t be a cover for rudeness.

  23. cba says:

    Invig (because you asked so nicely 🙂 ), Meh, Gummo Trotsky and Geoff Robinson:

    Just as there is good and bad scientific work, there is good and bad economic research. Bad economics is like bad science – there’s no shortage. A good economists, just like a good scientist, is a skeptic. Skeptical of policy prescriptions made by non-economists, skeptical of the work of other economists and skeptical of his own work. Do you seriously think that even the average phd economist isn’t aware of the limitations of his or her work? Those flawed models have their use, if only for rhetoric. Those controversial empirical studies are controversial not because they are right but because they challenge an existing view point or draw attention to a new subject. Just like any other field of endeavour certain paradigms dominate, but new lines of inquiry open up if they bear fruit.

    Do you think that a medical researcher or engineer that makes policy prescriptions based on lab work should be immune from the criticism that the lab conditions might omit critical behavioural factors encountered in the field? (Steve Levitt’s car seat study springs to mind, not that I agree with his findings). In the rough and tumble world of policy, everyone gets a turn at the pulpit. The scientist that expresses moral indignation when her view is challenged by a mere economist should probably have rethought stepping out into the mudpit in the first place. It’s a place that most PhD-trained economists are wary to tread (our generous host a notable exception)

  24. invig says:

    I’m sure that logic is the best that can be found for persisting with economics as it stands.

    I just hope you will be prepared to throw away the vast majority of your beliefs when you are required to do so.

    I do not say ‘if’ because we need answers that economics (as is) continues to fail to provide, and I am an eternal optimist.

  25. christine says:

    Patrick, thanks for the comments. SA is a good example to look to, and the economics tie ins you suggest are nice. Must say though I do in fact understand the theory of opp cost of time, I try to ignore the practical implications when reading/commenting on blogs 🙂

    Gummo: ” … inverse correlation between taking economics seriously and the ability to see the funny side of economists inflated claims for their discipline and professional training.”

    Er … I think all the serious economists here are recognising what I think was a serious point you made about expertise, and are taking that seriously, and also recognising your more lighthearted points, and taking them lightheartedly (which, if I can be allowed to call myself a ‘serious’ economist, I was doing when I asked about empirical evidence on the pick-up ability of the econometrician vs the biologist. Sorry if it wasn’t clear).

    On relative social eptness of PhDs/profs: I once had a prof who was a Bayesian econometrician, who tole me that a huge advantage of being a Bayesian econometrician was that their conferences were so much fun relative to those of the other boring econometricians and economists out there. Based on the particular individual, I must say I find that really hard to believe, but haven’t hung around Bayesian conferences at all myself so shouldn’t be too sceptical. Anyone want to highjack the thread towards which group of academics is relatively more interesting/fun?

  26. Damien Eldridge says:

    Christine,

    Clearly micro theorists and game theorists are the most intersting and fun. I mean, it must be called game theory for a reason right? And their semninars never get bogged down into tedious debates over some boring detail of some data set!!! 😉

Comments are closed.