1. "Econ should stop pretending to be value-neutral."
Call me crazy, but I think researchers should try to keep facts and values ("positive" and "normative" econ) separate. They won't completely succeed, but they ought to try their best. No matter what your values are, you'll be better able to implement them if you know the facts as much as possible. That doesn't mean economists shouldn't have values, it just means they should try not to let those values interfere with their assessment of the facts.
"But dude, being anti-ideological is an ideology too!"
Mmm, an insightful and trenchant observation. Now go take another bong hit and leave me alone.
2. "Econ should stop pretending to be a science. It's just a SOCIAL science."
What does it even mean to "pretend to be science"? Does history pretend to be science? Does anthropology? Does literature? Social science has different kinds of data from the lab sciences, but you do the best with what you have.
As for the idea that human beings can't be studied scientifically - well, you hear a lot of people say that, but it seems blatantly and obviously wrong. There are lots of models that predict human behavior very well in various areas. Not just in econ, either, but in operations research and other fields. So this critique is just incorrect.
3. "Econ has physics envy."
4. "Economists failed to forecast the financial crisis."
Hey, if you can't forecast, you can't forecast. We can't forecast earthquakes at all (except aftershocks). We can't forecast hurricanes very well. Maybe crises and big recessions are just really hard to predict.
Of course, the argument that econ ignored the importance of the financial system, leverage, etc. was totally valid, but much has changed since 2008 (see next section).
And I guess I'd like to see more academic macroeconomists express concern about the forecasting problem, but instead they've decided to just leave the job of forecasting entirely to the private sector (where tons of people are working on it).
5. "Econ focuses too much on equilibrium, which doesn't really exist."
I've got some news for the outside world. Economists have redefined the word "equilibrium" to mean "any solution of a system of equations." Every solution of every model is called an "equilibrium".
Now, some kinds of economic equilibria probably don't exist very much (see later section).
Critiques That Used To Be Valid But No Longer Are
1. "Econ ignores the financial system."
Not anymore, they don't!
Critiques That Are Less True Nowadays But Still Somewhat Valid
1. "Econ relies too much on theory, and not enough on data."
Empirics are on the rise. Theory is a smaller part of what econ does these days. I still see large numbers of wanky theory papers at seminars and such. But it seems like more and more of them are either A) job market candidates proving they're smart, or B) old folks having fun because they can. Nowadays, as Paul Romer says, "[I]n the new new equilibrium, empirical work is science; theory is entertainment."
But despite the decreasing flow of new theory, or at least prominent new theory, some econ literatures are still crammed with mutually contradictory models for which the scope conditions are neither known nor specified. And the stock of existing theories is still enormous. In some areas, especially in macro, economists really do have theories that make almost any prediction, with no real way to choose between them except priors and politics. And many economists still have very little problem using modeling assumptions that have already been taken to the data with discouraging results.
Also, as someone pointed out in the comments, econ models often sacrifice too much realism for the sake of tractability. If you have to publish theory papers, but it's too hard to get a result using solid assumptions, you go with silly assumptions. Hopefully the decreasing importance of model-making will decrease the pressure to pump out this kind of silliness.
2. "Econ uses too much math."
Go back and read some papers from the 70s and 80s. They are much more obsessed with pointless mathematical rigor and aping the style of math papers.
Of course, that doesn't mean that econ uses math for good purposes. There's a lot of what Paul Romer calls "mathiness" going around.
3. "Econ is a front for neoliberal/laissez-faire politics."
This was definitely somewhat true for a while. Especially in the Cold War, where economists became part of the ideological opposition to communism. Milton Friedman and Friedrich Hayek really were the public face of the profession for a while. But since then, the profession seems to have drifted in a leftward direction, in terms of economists' views, the profession's public face, and probably the implications of the results in econ papers.
There are still a few old Cold Warriors left, of course. There is still a lot of Koch money being funneled to certain econ departments and think tanks. There is probably still a subset of economists who were drawn to the profession because it seemed more conservative than other areas of academia. And there are still the Austrians, running around pretending to have something to do with the profession.
4. "Econ is sexist."
Yep, but it's clearly getting better, thanks to increased awareness and to the efforts of the AEA and other organizations.
5. "Econ colonizes other social sciences while ignoring their input."
My sense is that the Gary Becker project, of trying to understand social/cultural phenomena with standard econ models, is not doing so hot these days. You still see a lot of this on blogs and in pop econ books, but within the profession it seems generally recognized that things like norms and social preferences drive a lot of social phenomena. The problem with Becker stuff, as I see it, was that it focused too much on very standard, classic types of utility. Once you start bringing in new kinds of utility like norms and social preferences, those utility choices end up driving the results completely, and fancy econ techniques become only marginally useful. But it still sells pop books.
As for not taking input from the other social sciences, econ is still very siloed. Behaviorists have taken some input from psych. And poli sci methods and econ methods are converging so much that they are often indistinguishable at this point, in certain fields. But econ still ignores the existence of sociology entirely. Hopefully, as empirical methods become unified in sociology and econ, this will change.
6. "Econ is enslaved to the financial industry."
There has definitely been some corruption. Incentives matter, obviously. I don't think it was ever that widespread - most economists have nothing to do with finance - but there was a bit. But the AEA instituted a stricter ethics code in 2012, and there is more scrutiny of economists who do sponsored research. So the problem seems like it's being addressed.
7. "Economists assume Rational Expectations, which is B.S."
RE doesn't seem to be a very good assumption, except sometimes in the long run when you also have some lucky properties like time-invariant stochastic processes. It is very true that in the past, economists - especially in macro - were essentially forced to assume RE. Nowadays we are definitely seeing more deviation from that orthodoxy - Bayesian and non-Bayesian learning models, bounded rationality models, and other stuff. This critique is still mostly valid, though.
1. "Econ spends too much energy on macro."
Yep. Seems like macro is about a quarter of the econ profession. But given data limitations, a lot of that effort seems to go into making useless models. Macro is the glamour division of econ, but maybe more economists should avoid the glamour and get down in the muck where there are real conclusions to be had.
2. "Econ is obscurantist."
Definitely. This is a problem plaguing much of academia. In math and physics there are natural barriers to entry, because getting stuff right, though possible is just really hard, so only really smart people can enter those fields. In lab sciences, huge fixed costs are the barrier, with a fairly high degree of smarts also required. But in social sciences and humanities, there are fewer natural barriers, so the practitioners have to create them in order to increase their monopoly rents.
One way to do this is hyper-specialization. You see a lot of this in history, for example. Another way is to make up a ton of bullshit neologisms and spend all day arguing over what they mean, which requires defining them in terms of other neologisms, ad infinitum. This seems to constitute most of "critical theory". But a third method is to artificially increase the level of intelligence needed to do work in your field. This seems to be the approach taken by econ, where people who will spend their careers digging up natural experiments to test simple linear models are nevertheless forced to learn what a sigma algebra is. It also might be the reason for the use of purely mathematical methods to study every possible problem, even when those methods may not help or illuminate. It might even be one reason for the turgid, impenetrable style of many econ papers...
3. "Economists are arrogant."
Well, when you get paid more than almost any other academics, constantly sought after for policy advice, and treated as a sage on every conceivable topic from how to find a good restaurant to how to find a good spouse, you do tend to get a little full of yourself! :-)
4. "Economists believe their theories too much."
"Models are just tools for understanding specific situations," says every economist who believes his own models are The Truth And The Way.
5. "Welfare economics is silly."
Yeah, pretty much. Add realistic non-standard preferences and everything changes wildly. Add time-inconsistency and it becomes utterly hopeless.
Critiques Of Whose Validity I Am Not Really Sure
1. "Econ makes people bad, selfish, and immoral."
There's some evidence to this effect, but it doesn't seem conclusive.
2. "Economists need to get out and have more contact with the real world."
I hear people say this, but I really don't know how much contact the average economist has with the real world. I know some who seem to have essentially none, and some who seem to have a lot.
3. "Economies are complex adaptive systems, hence you will never be able to understand them."
4. "Econ 101 is filling kids' heads with bad ideas."
I'm not sure about this, since Econ 101 must be taught very differently at different schools. The only thing I really know that is constant across schools is the identity of the most popular 101 textbooks. And Mankiw and Krugman's textbooks are pretty good, I think, although they're almost all theory and very little about empirical evidence. That's something I'd like to see change at the 101 level, for sure.
5. "General equlibrium/Walrasian equilibrium sucks."
I'm not sure. I haven't seen a lot of general equilibrium models that look like they've had a high degree of success in explaining observed phenomena. But it's a big, big world out there, and there are LOTS of general equilibrium models, and people studying how these models fit the data.
Anyway, off the top of my head, there's my list. Anyone have any more you'd like me to add? Leave them in the comments!
3. "Economists believe their theories too much."ReplyDelete
everybody believes their theories too much
would you consider an assertion like "the early pioneers of neoclassical econ (jevons, menger, walras etc) informed by their philosophical views shaped the path of their their research interests" false?ReplyDelete
if so, in which category of the ones you've presented above would you classify such an assertion?
Don't know, to be honest. I don't know much about those early days of economic thought.Delete
"Economists ignore criticism from other social sciences, while themselves colonializing these social sciences with their theories". Personally I would put that under the "somewhat true but less than it used to be" headline. I got the impression that there is a lot more openness today than say 25 years ago, and the wave of economic imperialism seems to have also ebbed a bit.ReplyDelete
Ahh, good one. I'll add it.Delete
You forgot "the dismal science".ReplyDelete
Seriously though, all the value neutrality is on the side of the liberals, and it kills kids.
From the perspective of of a neuroscientist, I'm mad econ spends plenty of effort on derivative markets, yet it apparently takes neuroscientist lead teams to get a simple RCT on cash transfers to poor parents in the US.
I'm also super mad that the policy framework we have means we can't address neurotoxic impact of known poisons, like lead in drinking water. That's not the fault of economists, but it is true that as a member of a society, if you see something wrong that could be remedied, you at least have a duty to tell people in policy making roles that they could be acting. Economists have a duty to highlight, if not advocate for, cheap public health interventions. I see a few who do well, but vastly more on the Koch brother funded side of highlighting how to make the wealthy moreso.
all the value neutrality is on the side of the liberalsDelete
This is basically "Econ is a front for neoliberal/laissez-faire politics", right?
I'm mad econ spends plenty of effort on derivative markets, yet it apparently takes neuroscientist lead teams to get a simple RCT on cash transfers to poor parents in the US.
Are you sure? I know of quite a number of field experiments giving cash transfers to households in poor countries. Are you sure there aren't any that do the same with poor American households? I haven't checked the literature to see if this has been done...have you?
Economists have a duty to highlight, if not advocate for, cheap public health interventions.
I think economists probably just don't know about public health stuff.
I think this opinion just comes from a misapprehension about where economists' research money comes from, what economists do research on, and what things they advocate for. I'm surprised Noah didn't fully correct this.Delete
The money first: economists serve many markets and one of the markets most willing to pay is finance. We might want to change the amount of resources controlled by people in finance - and there are plenty of economists who argue for that - but the economists are not sucking that money away from the neuroscientists. Another source of the money is students who want jobs - often these jobs are in finance or the private sector which again drives schools to hire people who do research in these areas. This is more a problem with the economy rather than economists, a problem that some economists work on.
What we do research on: One of the fasting growing areas of economics has been RCTs trying to understand how best to help the poor. In fact, the research of Kremer et al. (rightly or wrongly) has lead to massive deworming programmes. Economists in development look at public health a lot. So do the economists in health, mental health too. See the work of Richard Layard for example, who has used his research to aggressively push to improve treatment of mental health issues in the UK. Or see Banerjee and Duflo who, with the Poverty Action Lab at MIT, have worked tirelessly to find policies that can help lift people out of poverty. Or see .... the list is endless. Should more economics be focused on these issues, I expect the answer is yes, but again this is a problem with the funding. A lot more of medicine should be focused on public health and illnesses associated with poverty rather than on cancer or even worse, cosmetic surgery - but cancer is a first world problem so it gets more of the money.
Work on cancer regularly reveals fundamental biological truths that help bioscientists. And the last cosmetic surgeon I met mostly focused his career on operation smile type work.Delete
But really, there are good and bad actors in all fields. My basic critique of econ is about the high level of harm done by the bad actors, and the ethos of the field. I think the proper role of econ is to tell us how to be effective altruists. Most academic economists seem to think the role is to tell people how to be effective at whatever we choose. But a subset of the finance guys think the role of econ is to tell people how idiotic they are to aspire to altuism (or alternatively, to tell the powerful how to exploit those with such weaknesses).
F. Economists use a strange ordinal numbering system that sometimes has two 2s.ReplyDelete
bad critique: Economics is useless as all these highly paid economists couldn't see X coming. (X = financial crisis, recession, oil price changes, stock market crash, etc.) See also: seismologists. Wait..ReplyDelete
I thought this was included in "Econ ignores the financial system"...Delete
I always use Earthquake prediction in arguments about Econ but I wonder how valid it is. A seismologist can very accurately predict the general locations of Earthquakes and even their potential strengths and natures. The problem is that the boundary conditions on their model predictions SEEM useless because human lifetimes are so short compared to geologic timescales.Delete
I think the "Economics is useless because is couldn't predict X" critique is functionally the same as "Economies are complex adaptive systems, hence you will never be able to understand them."
The real analogy to me is weather prediction. Economic forecasting and weather forecasting are pretty analogous and the real issue is effective calculability at specific resolutions. There are some agent-based models used in civil engineering that are very good at predicting human behavior in for things like crowd control or traffic prediction.
Can human behavior be estimated in enough detail individually and in aggregate for a model to be computationally feasible is yet to be shown. It works for individual market prediction and design for things like commodities trading and designing markets for power. Who knows.
The simplifying assumptions of modern macro make it useless for prediction for the sake of tractability.
So really this seems like a *shrug* and not a bad critique although it is framed in a negative way.
3. "Econ is a front for neoliberal/laissez-faire politics."ReplyDelete
This was definitely somewhat true for a while. Especially in the Cold War, where economists became part of the ideological opposition to communism.
Please note that this is still very true in the Eurozone, which is dominated by Germany, where the Econ opinion is dominated by neo-liberal clap-trap from employer financed think-tanks, which dominate TV strudio debates, for example.
If not it is the neo-classical arguments which are heard. And that is about it.
You have heard of the hostility towards Krugman, eg, with his views in Germany. Wehreas Krugman is pretty mainstream, he is seen as "dangerous" in Germany, (and by implication in the EU)
Yep, I have heard of that. But in Germany, do they use academic econ as a front for desired policies? Or do govt. officials simply assert that they know the right thing to do?Delete
"Call me crazy, but I think researchers should try to keep facts and values ("positive" and "normative" econ) separate. They won't completely succeed, but they ought to try their best. No matter what your values are, you'll be better able to implement them if you know the facts as much as possible. That doesn't mean economists shouldn't have values, it just means they should try not to let those values interfere with their assessment of the facts."ReplyDelete
The problem here, I believe, is that pretending not to have bias is not an effective way to overcome bias. Recognizing and trying to make your potencial bias clear to yourself and to your readers makes the whole process more honest and objective. Stopping pretending to be neutral is a necessary condition to not letting values interfere.
And then there's an immense difference from objectivity to neutrality.Delete
The first is about being honest and rigorous in your research, which I believe everyone in any area of knowledge would consider to something to look for.
The second is about the idea of an impartial, rational, unbiased person doing science immune to instincts, feelings and values. This is pure, and poor, ideology indeed. And it is real and strong and dangerous to democracy and freedom as economists and policy makers believe that.
I don't think it really has to do with pretending or not pretending. You just have to try to be as unbiased as you can. People aren't really aware of their own biases, so this effort will not be a complete success...but by the same token, that also prevents people from admitting their biases, too.Delete
The second is about the idea of an impartial, rational, unbiased person doing science immune to instincts, feelings and values. This is pure, and poor, ideology indeed.
See, I don't get that at all. How is this any different from an idealized form of the thing you said was good in the previous sentence???
1. Just as "the 1st step is admitting you have a problem" works for an alcoholic who needs to get sober, so it is for the social scientist who needs to overcome his/her biases. Denial gets you nowhere.Delete
2. I wish to understand whatever I'm studying as good as I can, so I carry my research as logically and technically rigorous as I am capable of. That's being objective, and that's universal, no matter if you are a homo erectus trying to replicate your colleague's fire, a marxist trying to undestand class strugle or an economist studying the impact of development policies.
Taking your question, believing in "neutrality" is about believing that objectivity (logical and technical rigour) is enough to guarantee the absence of bias. In a very narrow way it is, but in social sciences the non-neutral character of research is much deeper than that. Describing a result, choosing a model, a way of modelling, a research question, a research theme, that's all non-neutral, and not in a trivial way, as it impacts how much attention and money are to be given to one or other social object, and how it is being conceptualized and treated.
Recognizing that your way of definining things and undestanding the world around is deeply determined by the society you live in and where you fit on that society might not have a relevant impact on how techically correct your paper will be, but will help you understand your limitations. Similarly, recognizing that whatever your paper says does have an actual (even if marginal) impact in how other people perceive the world and act upon it, should give you some responsibility and help you mold more consciously your interests, and not be molded by others or by your original biases.
1.Just as "the 1st step is admitting you have a problem" works for an alcoholic who needs to get sober, so it is for the social scientist who needs to overcome his/her biases. Denial gets you nowhere.Delete
How do you know economists deny their biases?
And remember, we can get some info about how much bias is leaking into economists' work: http://noahpinionblog.blogspot.jp/2014/12/economists-arent-that-ideological.html
2. Still trying to figure out what this means. Do you mean that economists should consider how their research will be applied and used before they do the research? (And if that's what you mean, how do we know they don't consider it?)
1. Surely there is an enormous heterogeneity about economists' attitute towards their biases, but it seems clear to me that among social scientists we are the ones who care the least about them.Delete
As for ideology, Economics is a status-quo oriented discipline, it is no surprise that economists tend to be near the center, and most who see themselves far from it end up adapting. For its dominance on social sciences, it is almost tautological to say that economics would be at the center. But being near the center does not mean being neutral, only means that your "ideology" is the dominant one, and that the right-left wing political spectrum is constrained precisely by your (economists and those who think alike) ideology's rules. Calling this "neutrality" is to deny the sociological nature of economists power over policy makers and public debate.
One might argue if this is good or bad, but economics profession, and the way it reasons, and the things it focus on, is a very specific way to view the world, which at this specific point in time, at some specific countries, dominates others in capacity to please the status quo and to influence it.
2. I believe economists should have clear that their non-partisan technically perfect paper is still, as any social scientists' work ever, a political document representing some specific way to see things and having some underlying moral and philosophical assumptions that escape scientific reasoning.
As in the first pont regarding bias, I believe a lot of economists have this sensibility, I just think that the way the whole teaching and professional environment dismisses those discussions is deleterious to our capacity to understand what we do. I would agree that the problem is exagerated, or at least poorly understood, by a lot of critics, but it is a real one.
"Yeah, but in Econ 101 / 1st degree, they still learn to maximize profit and ignore all this stuff you said economists don'r ignore anymore. That is what most people remember"
You mean, Econ 101 is stupid?Delete
I'm not sure if it is, because it must be taught differently at every school, right?
Quick question about your comment about the "anti-idealogy is an idealogy view". It's not clear to me; are you saying it's a true and useful statement that simply isn't relevant here, a banal truism, just wrong, "not even wrong", or something altogether different? I get the feeling you mean the first or second, but may be projecting my own views onto you and am curious to hear your explicit thoughts.ReplyDelete
Sure, thinking that separating facts from values is good is itself a value judgment. That's obvious. But also very beside the point.Delete
A philosopher of science such as John Dupré would ask whether the measure of inflation that we use can be determined "objectively", or is inescapably value-laden in important ways. His piece on fact/value can be found on-line.ReplyDelete
It's the use that incorporates values, not the calculation. In fact, official statistics calculate a bunch of different measures of inflation.Delete
So which one do we choose to use? How do we explain that there are different measures if variance in underlying values is not why economists created these different measures?Delete
I don't understand your second question.Delete
As for how do we choose one to use, it depends on what we're going to use it for, and why.
If I remember Dupre, he is asking: Is it simply an objective matter to determine how much and which types of goods are weighed in the basket of goods we use to measure inflation? If there is not one objective way to do this right, then how do we choose which way to measure inflation? People may determine their inflation measures in terms of underlying values and class-biased interests. The facts about inflation may turn out to be shot through with values. The fact/value separation may not be as easy to maintain as we had hoped.Delete
Here's how Dupre puts it:Delete
With inflation, normativity comes in a little later. The
primary problem, as has long been familiar to economists, though it
often appears to surprise others, is that there is no unequivocal way of
measuring this economic property. It would be easy enough if everything
changed in price by identical percentages, but of course that does
not happen. How should we balance a rise in the prjce of staple goods
say, against a fall in the price of air travel? The immediiJtely' obvious reply
is that we should weight different items in proportion to th amount
spent on them. The problem, then, is that not all goods ar.e equally consumed
by all people or even by all groups of people. It is quite commonly
the case that luxury goods fall in price while basic necessities
rise. It might be that these cancel out under the suggested weighting, so
that there is no measured inflation. But for those too poor to afford luxury
goods, there has manifestly been an increase in the price level.
How, then, does one decide how such an index should be constructed?
The unavoidable answer, it seems to me, is that it depends on
the purposes for which it is to be constructed. There are many very practical
such purposes. People on pensions, for instance, may have their incomes
adjusted to account for changes in the level of inflation. For such
purposes, the goal might reasonably be to maintain the value of the pension,
in which case the ideal would be to enable typical pensioners to continue
to afford the goods that they had previously consumed. Of course,
no pensioner is absolutely typical, but a case might be made for addressing
particularly the case of pensioners dependent solely on the pension. For
such ends, it would clearly be desirable to have specific indices designed
for specific groups. But the goals might be quite different, calling for different
measures. For example, and perhaps more plausibly, one such goal
might be to save the taxpayer money.
i share this concern. noah says that which inflation measures we use depends on what we are going to use them for and why. so how do we chose without making a value judgment about which measure is best for the given purpose?Delete
Wow you guys are dumb. Inflation is just a statistic -- in some contexts it may be interesting and important, in others not so much. For some questions we want core inflation, or trimmed inflation, or PCE inflation, or CPI inflation. Damn, why are you people so insistent on not making sense?Delete
The problem is the "we". "We" may not be unified. One group may want inflation to be measured one way; another group another way. And the two measures may be each internally consistent and valid--even though they generate very different numbers.Delete
Disagree on the theory vs. empirics one. The pendulum has swung way more in the empirics direction than you allow for. Hardly see any theory any more. Even theory-lite (just simple models) is getting rare. Yes, macro is somewhat of an exception but even there it's not really "theory" (name one big theoretical idea in macro from the past ten... no wait, twenty, years. And if you say "zero lower bound" you get slapped upside the head), more like tweaking the tweaks that tweak the theory.It's all the same (mostly wrong) model over and over and over again. Maybe another exception is in behavioral although that's pretty much fizzled by now too.ReplyDelete
1. I'm using "theory" and "modeling" interchangeably, yeah I know there are few real insights nowadays...mostly, macro people are going back and mining the literature for insights that they overlooked when they came out, like John Geanakoplos' leverage stuff. Seems that way to me, anyhow.Delete
2. Interesting to hear that the pendulum has swung more toward empirics than I allow for, outside macro. I mostly see macro and finance stuff these days - since grad school I haven't had frequent contact with other areas of econ.
3. But what about micro theory? That's still a huge field.
4. Also, theory (or "model-making", if thou wilt) still seems to be very important for the job market: https://afinetheorem.wordpress.com/2014/01/27/is-theory-really-dead-2013-14-job-market-stars-in-economics/
I know a ton of people who went on the job market with theory papers. Seems like it was mainly used as an IQ test.
And for the record I don't think that's a good thing. Empirics without theory is just data mining, which is what a lot of what goes on now really is. How about we leave that to the sociologists?ReplyDelete
Well, presumably a lot of the empirics being done now is of the natural-experiment variety, and natural experiments (well, good ones, anyway) don't exactly grow on trees. So that would significantly curb the data-mining problem.Delete
" (well, good ones, anyway) " <-----Delete
Yep, but crappy natural experiments can be observed and checked, unlike true data-mining which is invisible.Delete
Some more for you, Noah.ReplyDelete
Welfare economics. I don’t mean this or that proposition, I mean the whole thing. Utility and utility maximization. As we both know, it’s indefensible on close inspection. And it has horrible effects throughout applied econ. I’ve gotten to the point where I can barely read a paper that has “optimal” in the title.
The lamppost problem. Economics has it real bad. Problems become hot because econ has tools for them, not because they’re the most important ones to solve. (This is one of the critiques of RCT’s.) Assumptions are made so that models will be tractable and the tools work, even if they assume the can opener, to mix metaphors. (My bete noire is convexity of production and preference sets, but you could fix on others.)
Econ 101 is a huge problem! Agreed, it’s taught differently all over to some extent, but take a look at the textbooks. Now you could argue that the problem with the texts is the profession, since publishers collect reviews and force authors to heel. If the reviewers weren’t so blinkered maybe we’d have better texts, but that’s all guessing. What I know is that the textbooks are terrible, and this is not only what most college students will read, but it also, subtly, influences the way their professors think about their discipline.
Finally, it helps to distinguish between economics as what economists do and how it’s perceived from the outside. If you read a whole lot of working papers as they flash by on the web you get one impression. (Not so bad in fact.) But the weight of prestige and publicity falls on just a portion of the work and the practitioners, and it’s not the best sample. One reason for this is the intermingling of professional and ideological selection pressures. It’s not like the think tanks, private money, business press etc. are on another planet from the pristine universities. It’s not all the same, but it’s not divided by a Chinese wall either. (I assume that selection is stochastic, so a few counterexamples don’t prove anything.)
Welfare economics. I don’t mean this or that proposition, I mean the whole thing. Utility and utility maximization. As we both know, it’s indefensible on close inspection. And it has horrible effects throughout applied econ. I’ve gotten to the point where I can barely read a paper that has “optimal” in the title.Delete
Welfare econ is indeed crapalicious (I'll add that), but utility maximization is very predictive in many cases, so I'd say utility is "real", or at least often is.
The lamppost problem. Economics has it real bad. Problems become hot because econ has tools for them, not because they’re the most important ones to solve.
Sometimes yes (I'll add this), but usually I think important policy problems do get attention, like minimum wage or health care or stuff like that. Econ does a huge amount of policy-relevant work.
Econ 101 is a huge problem! Agreed, it’s taught differently all over to some extent, but take a look at the textbooks.
Personally, I think the theory sections of those are OK, though they give too much sense that economists are sure about these things. But they need a lot more about evidence. Just talking about Mankiw and Krugman's books.
It’s not like the think tanks, private money, business press etc. are on another planet from the pristine universities.
Good point. There's a ton of hackonomists out there. I try not to be one of them... ;-)
Properly defining utility, what it means, and how it came to mean what it means, and what would it mean if we did not use the utility concept is a good example of why we need Critical Theory and deconstructive methodology in Economics. The foundations need to get sorted out.Delete
Utility and utility maximization is indefensible? Utility is real? What would it mean? This is just representation of preferences. Which axioms of preferences do you think are wrong?Delete
Clearly Peter Dorman knows very little about economics.Delete
One Remaining Anono-Troll, you need a girlfriend so bad.Delete
Here are 3 criticisms I'd be curious to see you address...ReplyDelete
1. Economics hasn't improved its predictive accuracy over time, or hasn't improved it as much as fields that are clearly scientific. Meteorologists have gotten a lot better at predicting the weather in recent decades. Have economists made similar strides?
2. The "rise of empirics" within economics is still relatively superficial. Consider an analogy. Mathematics is a clear case of a field that is not a science. My knowledge that "2+2=4" is a priori, it isn't based on observation. I don't make math "more empirical" by pointing out that 2+2=4 has lots of real world applications... 2 apples plus 2 apples equals 4 apples; 2 neckties plus 2 neckties equals 4 neckties; etc. But how is this different from what economists do? (Compare the complaint that string theory is not a science, it's just math.)
3. Economics relies too much on folk psychological concepts (e.g., preferences) that shouldn't figure in a mature science. (The Alex Rosenberg objection.) Again, an analogy. There is no science of Tuesdays--no laws of nature about Tuesdays, no causal mechanisms that essentially depend on Tuesdays, etc. Of course, lots of things happen on Tuesdays--earthquakes, mosquito bites, power outages, etc.--and you can study those things scientifically. But the concept of "Tuesday" won't be part of your toolbox for studying them. The objection then says that our folk psychological concepts are like Tuesdays, they shouldn't be part of your toolbox for serious scientific inquiry. Physics only really started as a science when we rejected our folk physical theory (a theory you might think of Aristotle as outlining); biology only really started as a science when we rejected our folk biological theory (invoking essences and so on); similarly, the objection goes, the social sciences will only really become sciences when they reject folk psychology as a foundation.
1. Economics hasn't improved its predictive accuracy over timeDelete
For what? Recessions, no. But that's not nearly the only thing economists might want to predict.
2. Not sure what this one means.
3. Hadn't seen that one before. Aren't preferences important? They're the key to Shapley-Roth matching, for instance, or Random Utility Discrete Choice Models.
We have seen a lot of big data type exercises. What is missing though thorough historical analysis to properly explain the data (and with documentary evidence being supported by the data, and not only the reverse). This is a complicated, and largely qualitative exercise which should under no circumstances be framed by a model.Delete
Continuing to make the same stupid comment doesn't make it true, Henry.Delete
How about the criticism that economics has misused and overapplied ideas about equilibrium? From Schumpeter we have the idea that out of equilibrium states are necessary for dynamic efficiency and wealth creation. This is not taught in Econ I, as far as I know. In Econ I students are often taught that markets can make adjustments to and achieve equilibrium before the supply and demand curves have shifted, but this basic lesson is taught dogmatically, i.e. without any argument that the time needed for adjustment to equilibrium is short enough that parameters will not have changed in the meantime. Markets are justified in terms of their ability to achieve equilibrium states, but this seems to be dogma. I think Meghnad Desai has isolated the equilibrium assumptions of the profession for targeted critique in his last book.ReplyDelete
Sure, I'll add this! I don't think it's right, though. Economists have redefined "equilibrium" to mean "any solution of a system of equations".Delete
It's true that Walrasian/price-taking/general equilibrium doesn't seem to produce many usable results.
"Economists have redefined "equilibrium" to mean "any solution of a system of equations".Delete
Does that have any meaning or relevance at all?
It's just a word...Delete
Econ has an extremely hard problem to analyze and considering the challenge, it's actually been quite successful. The small problem is that majority of its successes are in the past. Current econ research, to a large degree, either refuses to confront the data or just practices sophisticated curve fitting, focused on signalling mathematical ability with no value whatsoever.ReplyDelete
In some sense, it's understandable: the limits of the available data sets are pretty severe. However, the current focus is to pretend that we can increase the power of our tests by disaggregating the data sets. This might obviously help for the time scales for which the disaggregated data sets are relevant. However, majority of serious econ questions do not fall into this category.
Econ spends too much energy on macro.ReplyDelete
The problem is not that too many economists are working on macro - the problem is that they are accomplishing too little.
It's hard, man!Delete
Granted, until the crisis they thought they were accomplishing a lot when in fact they weren't. I think some of them realize their mistake now, some don't...
It's hard, man!Delete
So is quantum mechanics.
It's the fact that it is hard that makes it a worthy topic.
Raghuram Rajan is living proof that understanding what is going on is not impossible. Maybe we just need fewer, smarter, more open minded economists.
Quantum mechanics has great data. Macro does not. There are almost no natural experiments, and certainly no actual experiments. There are a million causal factors, none of which can be controlled or isolated. What's more, history only happens once, so we're often limited to time-series econometrics, which is basically a giant piece of shite. That means that even if you get stuff right, you'll probably never know you got it right.Delete
That's what I mean by "hard". Not mathematically hard.
I believe I have something to add to the debate, if I may:ReplyDelete
I enrolled in a post-grad economics course earlier this year. The first subjects were Math and Statistics. Both sucked for different reasons, but I'd like to focus on the Math classes. The teacher was this rather obnoxious French microeconomist who would bombard us twelve hours a week with demonstrations. I had plenty of reasons to hate him, and among them was his disdain for empirical work. It wasn't intellectual enough for the bloody wanker, I suppose. Of course, there's no need for him to test his theories himself, but he just didn't care. Great mathematician, poor scientist. And a shitty person in general. But I digress. I'm sorry for ranting.
On the other hand, my macro teacher would present us various models about a subject, such as inflation, and present the methods he used to test them and choose which made better predictions. Now that was science!
In the end, I dropped out of it: as much as I enjoyed the macro and econometric courses, I dedicated nearly all my time studying microeconomics, beating my head against the wall to understand Farkas' Lemma, Separation of Convex Sets and other delightfully difficult math for which there might well be a purpose, but struck me as intellectual signalling, as the gentleman/lady above said.
forgive me, I forgot to ask: how prevalent are economists who share this attitude with my beloved Frenchman?Delete
how prevalent are economists who share this attitudeDelete
I think John Cochrane is an economist who is sometimes blinded by the beauty of his equations.
Cochrane is pretty empirical. He likes equations that are simple and easy to work with, not super-formal math. But simplicity is another kind of beauty that may also be seductive...remember what Einstein said...Delete
I think that his "neo-fisherian" stuff is an example of Cochrane getting lost in his equations.Delete
What Absalon thinks about economics is about as important and relevant as what Ned Beatty thinks about my shoe choices.Delete
I was wondering when my one remaining anono-troll would show up and offer his scintillating insights.Delete
What Absalon thinks about economic ..Delete
John? Is that you?
"I was wondering when my one remaining anono-troll would show up and offer his scintillating insights."Delete
Also, why didn't anono-trolls didn't make the bestiary?
I should have put them in... :-/Delete
Yeah, I think I'm down to one. He always comes on and says the same line, "X doesn't know shit about economics." He always sounds boring and butthurt.
Unless there are 2 identical guys like that. Wow, now that's a scary thought.
I would add that economist should learn moral philosophy especially if they wish to influence policy. Agree about trying to be value neutral. I think learning moral philosophy helps with that. Otherwise the economist thinks she is value neutral when in fact she is not value neutral. Hope this makes sense.ReplyDelete
That does make sense.Delete
"What does it even mean to "pretend to be science"?"ReplyDelete
"Intelligent Design" (as opposed to evolution by natural selection) comes to mind.
Haha true, if you keep saying you're "science", that definitely counts as pretending.Delete
"Econ relies too much on theory, and not enough on data."ReplyDelete
To which one could respond:
"The theory-dependence of observation, by rejecting the role of observation as a theory-neutral arbiter among theories, provides another source of incommensurability. Methodological incommensurability ... denies that there are universal methods for making inferences from the data. The theory-dependence of observation means that even if there were agreed methods of inference and interpretation, incommensurability could still arise since scientists might disagree on the nature of the observational data themselves."
Yeah, that right: I went there! From here, which I found here. And yes, I might be suffering from this!
Angus Deaton has actually to say quite a lot on this (he's quite a bit the foucault of development economics statistics): http://www.lse.ac.uk/newsAndMedia/videoAndAudio/channels/publicLecturesAndEvents/player.aspx?id=2756Delete
And yes: as all definitions used to be able to measure anything are by definition phrased in a language, there is a linguistic side to this. Also, the definition many economists use to measure capital is the asset side of the balance sheet definition, leaving out the liability side (which defines a lot of the distribution of non-wage income), using such a limited, literally one sided definition of capital (a metod which was popularized by this apologetic of USA style neoclassical economics, John Bates Clark, you should read him) is loaded with values as it disables a thorough analysis of the distribution of income. Same for all those DSGE models which leave government production of for instance education out of the Euler equation - a choice, which is linked to their definition of consumption (which leaves out consumption of government production). The point: choices are made. Other choices could have been made - even within the paradigm of 'broad' economics, including business economics and the national accounts (which do estimate government production and consumption of government services). Hey, why do you think these Chicago boys leave government production out of the Euler equation!
@Merijn, I'm not sure I follow all that. I was just trying to sound clever by providing a quote (which I don't fully understand) which I think means:Delete
"...you need theory in order to make sense of facts. There is no such thing as theory-neutral facts."
which I do understand (from my 2nd link BTW). Now whether it's true or not, I don't know. The fact that I can't even tell what you think lets you know where I'm at. ;^)
1. "Econ ignores the financial system."ReplyDelete
In Noah's post I don't know why he just talks about the catch-up post 2008. Kiyotaki-Moore 1997 Credit Cycles model has finance right at the heart of it and it was very highly cited even before the crisis.
Then there was a huge money and banking literature looking at bank runs, debt crises, interbank lending and so forth. It had been left out of the heart of graduate macro programmes but this stuff was not being published in the weeds but in "top 5" journals.
"This seems to constitute most of "critical theory"."ReplyDelete
It's this sort of arrogance that gives economists a bad name. Unless it fits in to a neo-classical model things do not exist. What is a better way than Critical Theory to demolish this approach?
Almost all social scientists and philosophers know something about Critical Theory and are aware of its importance. The exception is economics. Once critical theory touches economics, the subject is revealed for what it is (essentially fiction); there are strong motivations to exclude it from the discipline.
Btw, do you know the first thing about Critical Theory?
please forgive my cheapshot, but I think this strip explains the problem:Delete
It's not that Critical Theory or post-modern thought have no value; rather, the abuse of jargon and the contrived language make both nigh impenetrable and, at times, hide a lack of content. Mr. Sokla's "Fashionable Nonsense" displays some examples of how it is used to conceal rather than illuminate.
But come to think of it, those super formal mathecon papers suffer from a similar flaw. The only difference is that they abuse and contrive mathematical language.
Second anon wins. XKCD link getting bumped to main post!Delete
It's just a SOCIAL science."ReplyDelete
It's neither a science nor particularly social.
I used to think that economists believed their theories too much. Then I read Graeber and Diamond. In many ways, anthropologists like them are the evil beard mirror verse opposites of economics. They generalize from theories to try and predict economic outcomes just like econ. Looking at the alternative and realizing how close-minded economics could be gives me a lot more appreciation for the humility of the discipline.ReplyDelete
Economists may have been paying more attention to finance lately, but has this caused an improvement in our understanding of the relationship between the financial sector and macroeconomics? Looking forward, how should we investigate this relationship going forward? Should adding financial frictions to DSGE models be the main tactic? What else can/should we do?ReplyDelete
There has been a lot of empirical work as well - much of the almost-useless time-series variety (finance crashes when the rest of the economy does, but hard to separate cause and effect), but also a lot on linkages between financial firms and non-financial firms.Delete
Whether it is a good or a bad critique, I don't think you get this one at all:
1. "Econ should stop pretending to be value-neutral."
Call me crazy, but I think researchers should try to keep facts and values ("positive" and "normative" econ) separate. They won't completely succeed, but they ought to try their best. No matter what your values are, you'll be better able to implement them if you know the facts as much as possible. That doesn't mean economists shouldn't have values, it just means they should try not to let those values interfere with their assessment of the facts.
Nobody would disagree that researchers should try to keep facts and values separate.
And yet, I bet if you asked any economist - orthodox or heterodox, freshwater, saltwater or dry-as-a-bone - he would say that's precisely what they do. "I keep facts and value separate," they would say. "It's the other guys who mix things up."
That's how you find austerians claiming that deficits and debts are the path to the Dark Side and Keynesians claiming it's the solution.
So, what gives?
Clearly, that problem is not a preserve of mainstream economists. Many a vociferous critic is not immune to the Very Serious Person syndrome.
Well, Zebub-san, remember that you might be thinking of economists in the public sphere, while I'm thinking more of academics.Delete
Sure, most people would like to believe that they're more objective than the people who disagree with them. But does "admitting" ideology mean embracing it? For many it does. See Russ Roberts, for instance. http://www.forbes.com/sites/modeledbehavior/2015/06/13/empirical-and-ideological-skepticism/
"Well, Zebub-san, remember that you might be thinking of economists in the public sphere, while I'm thinking more of academics."
I'm not sure I understand that distinction. The economists who say something in the public sphere, aren't they also academics? Without mentioning anybody in particular, the top of the econo-commentariat is composed of hot-shot professors, who do - or once did - research, teach, give lectures, write books, and publish papers. Or am I mistaken?
From where I stand, it's the same people wearing the two hats.
Do you mean to say the hat determines what they say? Something in public, something else in scholarly research?
I mean, that would be pretty much enough to say I rest my case.
The Ozimak article is interesting, sure. Particularly because Roberts - candidly and probably unwisely - admits to "reading the evidence" according to his ideology.
If you ask me, the question is how many other economists - of different persuasions - are a lot more discreet in their admissions.
Yup, that was me, B.L.
Sure, the people in the public sphere are academics, but they're not representative of the average academic. They are a selected subset. And what are they selected for? Their desire to engage the public, and to get in rhetorical battles! That's going to make them more ideological than the average academic.Delete
I am tempted to conclude from this exchange that the real point is that the "average economist", when compared to Roberts, is a lot more discreet in their admissions. Twitter is a bitch.
By the bye, my friends call me Tengu
Is it because you have a long nose?Delete
I am glad that you are willing to consider the point I made above about equilibrium/disequilibrium. I am relying here on Schumpeter, Janos Kornai, Meghnad Desai and Henryk Grossman all of whom were/are interested in disequilibrium dynamics.ReplyDelete
I really don't get your point about prognostication. It's not like meteorologists tell us that there are not going to be any more hurricanes or tornadoes. But economists such as Paul Samuelson and Robert Lucas have told us that the business cycle has been conquered soon before deep downturns.
There is a problem here; it should be waved away.
It should NOT be waved away.ReplyDelete
I was a bit late to this, but it always seemed to me the big weakness of econ was not paying enough attention to the Asian Development Model. After all:ReplyDelete
- the ADM has pulled more people out of poverty than anything else since WWII.
- the early "strongman" leaders who championed the ADM in their home countries were generally not trained in Western economics but were applying their own patriotic ideas about national development, and also copying Japan who had perhaps copied Germany.
- Attempts by Western economists to cover the ADM have perhaps been a bit lacking, e.g. Krugman's belittling of Singapore's capital accumulation strategy (was that in 1994)
- limitations of the ADM (e.g. not all of MITI's industry bets turned out to be good) seemed to get more attention than the overall factor of its success
- while some aspects of the ADM everyone would agree with, such as fiscal responsibility and education, a lot of elements are unconventional by Western standards: financial repression, exchange rate manipulation and competitive practices, for example.
- based on their contribution to human welfare, studies of the ADM should have made up more of this list for example: http://pubs.aeaweb.org/doi/pdfplus/10.1257/aer.101.1.1
Ah good, I'll add "free trade" in general in the next update...Delete
The items in the lists cover critiques that are commonly made. An interesting, separate section would cover "critiques that have validity, but almost no one is making them."ReplyDelete
Yeah. I added the "obscurantist" one, which so far I've never heard anyone else but me say. But I have a few more...Delete
"But econ still ignores the existence of sociology entirely. Hopefully, as empirical methods become unified in sociology and econ, this will change."ReplyDelete
And here is one enormous problem. Why should those methods become unified? Different methods (various forms of quantitative analysis, as well as various forms of qualitative analysis) have utility for different kinds of questions or empirical explorations. Getting at "practice" is incredibly difficult, and perhaps impossible, quantitatively--but ethnographic and anthropological methods are pretty good at getting at what practice is and its importance in social action. (How much serious ethnography and anthropology have you read? You try this sometimes when you write about Japan, although any half-competent ethnographer or anthropologist would rightly rip you to shreds.) There there should be one method, around which sociology should be "unifying," is an assertion--grounded in your own assumptions and lack of reflection on what "methods" are. (And this is the real physics envy, by the way. If you knew anything about chemistry, you'd realize that variation in methods is no bad thing.)
Why should those methods become unified? Different methods (various forms of quantitative analysis, as well as various forms of qualitative analysis) have utility for different kinds of questions or empirical explorations.Delete
Sure, not all methods will be unified. But some will be. And then maybe economists and sociologists will cite each other more, and the line between the disciplines will blur a bit more.
How much serious ethnography and anthropology have you read? You try this sometimes when you write about Japan, although any half-competent ethnographer or anthropologist would rightly rip you to shreds.
Ooooh, aggression. Rawr. :-)
And this is the real physics envy, by the way. If you knew anything about chemistry, you'd realize that variation in methods is no bad thing.
Aww, now you're just being st00pid, I know plenty about chemistry.
Feeble attempts at wit, Noah. Aggression? Not really, although that suggests something about your own priors. And chemistry? You know "plenty"? Let's just say I'm dubious. You've revealed plenty of hubris before.Delete
I think the first bad critique "Econ should stop pretending to be value-neutral." has been misunderstood a little; it's not that value-neutrality is seen as a bad thing (well some postmodernists might, but ignoring that), but rather that econ is not value-neutral, because the way its set up (in its assumptions) contain certain biases and ideological viewpoints (all social science to an extent does - the phenomenon being looked at can't be easily divorced from social, cultural and political backgrounds in the way that studying the movement of planets can). It's more true of some sections than others, but it's there and the argument is that being more self-aware about it would lead to better understandings, rather than acting as if such value judgments don't happen and don't guide thinking.ReplyDelete
This is not a reply, but I seem unable to just post a comment.
Anyway, as someone who blathers a lot about economies being complex adaptive systems, I would say that only rarely do such people claim that therefore it "cannot be understood." More frequently the CAD argument leads to related arguments, such as that rational expectaitons rarely hold because of all tha complexity. That is certainly something I have frequently argued, as have some others in the complexity ecomomics gang.
BTW, I have seen this argument more frequently made by climate change skeptics who argue we can ignore models predicting global warming because we know the climate system is chaotic, which it is. But that is also not a valid argument, in that chaotic systems are bounded and one may be able to predict general trends of the bounds, even as shorter term forecasting (much longer than a week) is basically impossible due to the butterfly effect.
The flow of water in a river is chaotic but I can predict with a high degree of confidence that the trend will be that the water will go down hill.Delete
You just did post a regular comment.Delete
Got it, Noah. Things just looking different on my system, but I figured out how to post, even though I went through looking like I was replying.Delete