Does the dismal science deserve federal funding?

Here is today’s WSJ Op-Ed by Gary Becker and James Heckman.  In general I am sympathetic to government funding for science, but I’d like to tack on a few points to these arguments in particular.

1. There is not a peep about supply elasticity.  A lot of economics research comes pretty cheaply and presumably without direct government subsidy.  Isn’t the implied conclusion to invest a lot in data gathering and RCTs/field experiments, but not so much in large swathes of economics?  Should we not start by listing all those branches of economics which should not be eligible for subsidy?

2. Is there possibly a higher external benefit to directing the attention of economists to teaching or community service rather than research?  Somehow this argument ends up underplayed when economists discuss subsidies.  Or how about subsidies for economics bloggers?  Presumably there is lots of good economics research which remains underpublicized and underutilized.  Isn’t that often the relevant choke point, not lack of new research ideas and findings?

3. Economics research is already highly subsidized through our tax code and legal treatment of non-profits.  An argument for subsidy is not the same as an argument for further subsidy.

4. I find it easy to believe the subsidies for economists would bring higher returns than the worst uses of federal funds.  But surely larger subsidies for economists are not the highest return projects before us.  Isn’t it worth listing which projects would be even better than subsidies for economists (or at least acknowledging that they exist)?  How about reporting “Subsidies for economists are better than farm subsidies, but not as good as medical R&D subsidies or 347 other uses of the funds”?  Presumably the goal is to bring about the best outcome possible, not just to advocate further subsidies for economists, right?  Right?  Right?  After all, that is what the economic method is all about.



The greatest entitlement goes for warmaking. We spend more than the rest of the nations put together. But we spend less than them of social entitlement because it is not as glorious as war in the minds of establishment thinkers. The vain search for a better war is the underlying culprit that engenders government poverty via deficits.

That's not true and hasn't been for many years now. You clearly are ignorant of how much China spends on their military.

It's still a small percentage of the US, especially relative to population. Furthermore, unlike the US, China is surrounded by bases of potentially hostile competitors (that is, the US and its allies).

Jose, we spend over 65% of federal spending on the military (fed discretionary budget). I don't care how much China spends on the military because it is not a measure of force or effectiveness.

Bill, according to an article I read not to long ago from Planet Money, 22.6% of government spending in 2011 went to Defense. So, I'm going to air on the side that your making up statistics to prove a point.

Bill is not making them up. He is just cherry picking. He said discretionary.

The problem is that the fiscal problems in this country don't come from the discretionary side of the budget. And I don't say that as a person who thinks that the military budget as it stands now is money well spent.

Your first point is the only one worth considering, in my opinion.

Any discussion of subsidies whatsoever should be preceded by a discussion about where the limits ought to be. If you don't know what the line is, then you don't know when you've crossed it. Billions of dollars to take more high-res photos of Mars, anyone?

In general, one can always make some sort of vague argument in favor of the government's funding something. The problem is not starting down that path, the problem is knowing when to stop. Government money is a powerfully addictive substance that no well-meaning economist or scientist can resist. It's worse than heroin.

The question that they fail to raise it the political ramifications of taxpayer-funded, politician-directed economics research. Would the field of economics be better served by shifting from nonprofit, think-tank, and university research to government institutions? Would that better serve "the public good", or politicians self-interest?

I would like to see an analysis of the impact this using public choice economics, and preferably not a government funded public choice study of the effects of government funded studies.

I object to treating economics as a science. It's degrading to real sciences that make testable predictions, develop consensus, and learn from mistakes.

Evidence that government isn't spending enough on it. [/sarcasm]

HAHA!! +1


Economists make a multitude of testable predictions, and most of the important ones are eventually tested; witness the rapid growth of experimental economics over the past 15 years, which has largely been oriented towards theory-testing. Economists have developed wide consensus on a huge range of issues (although they clearly disagree on others, as do scientists in all disciplines from paleontology to theoretical physics), and build better models based on the shortcomings of previous ones revealed by the light of empirical evidence; witness the proliferation of behavioral models (e.g., social preferences, reference-dependent preferences) which generalize the neoclassical one and are increasingly used throughout the discipline.

I object to boldly opining on the state of a field in which one has no expertise.

Not being sarcastic, but what are some big, testable economic predictions of, say, the last 30 years that have been unambiguously verified.

Virtually all economists agree that minimum wages create least among poorly educated youth....maybe (sarcasm)
Seriously, with a few billion dollars of subsidy, we will finally be able to lay to rest this question.

The extraordinary low unemployment rate that has occurred over the past 12years proves tax cuts create jobs faster than Clinton's tax hikes..

The extraordinary robust financial sector that has significantly reduced poverty and inequality has proved deregulation of finance will trickle down wealth.

The government budget surpluses have proved tax revenue increases from growth and job creation that always follow tax rate cuts, more than offsetting revenue lost in rate cuts.

By defining government as the problem instead of defining government as the solution which was the case from 1935 to 1975, the economic growth and economic stability has greatly improved since 1980 compared by the constant government bailouts of banks and industries and corporations between 1935 and 1975.

Consensus has nothing to do with science, A majority of scientists can be wrong.


A majority of scientists can be wrong because the majority all collect data independently that is in error??

What evidence do you have the majority of scientists independently collect the same incorrect data??

"A majority of scientists can be wrong because the majority all collect data independently that is in error??"

Have you exhausted all possibilities of error? Faulty observation, incomplete observation, faulty deduction, faulty theoretical apparatus, faulty experimental setup ... and a myriad other factors.

"What evidence do you have the majority of scientists independently collect the same incorrect data??"

Your question is largely irrelevant. I hope you can figure out why.

Let me fix that for you.

Consensus has nothing to do with any profession, a majority of professionals can be wrong.

I second Chris D's objection

What is the purpose of your objection? It does not seem to pass any cost-benefit analyis that I can consider. Perhaps you ought to give economics a chance before you reject it. You might actually learn something.

I have a feeling if this debate were in an area other than education Tyler's views might be along the lines of: "Elites using their privileged position (i.e. access the WSJ) to rent seek."

I don't see you making much of a case that economics research passes any sort of cost/benefit test, especially when you think a good chunk is provided privately, without government subsidies.

The way I read Tyler's post, it was dripping in sarcasm and making exactly that point.

You read it correctly, in my opinion.

Re-read the last point that Tyler makes half a dozen times. Meditate on them for awhile. Why might Tyler feel compelled to repeat the word "right?" so many times? hmmmm. The possibilities are endless.

Tyler is is just so cryptic who could possible understand him?

So, you in favor of central planning? And who get the privilege of deciding which subsidies bring greater value to society (aka acting as the central planner)? Seems to me like your more than likely to end up with another Solyndra type company when the privileged central planners get to decide who get the subsidies. No? Your placing an awful lot of trust in the guys with all the power... more than I would be willing to.

The fact that a lot of economic research is already privately provided doesn't kill the idea that government subsidies are needed. Privately-created research might be focused on privately beneficial ideas, at the cost of overlooking potentially important research that is socially optimal. For example, there is a ton of academic finance research done by well-paid business school professors. So, one might say that we are already producing plenty of research privately and no government subsidy is needed. But where is most of this research focused? On asset pricing and corporate finance--areas that are focused on businesses, which is essentially where those professors' salaries come from. Practically no one spends much time on household finance, despite the huge need for research in this area. A government subsidy for household finance research would be greatly beneficial.

Making a list of all the branches of economics that do not need subsidies would be difficult because one would have to look at sub-fields and sub-sub-fields. You couldn't just say, "Finance doesn't need a subsidy," but you'd have to say, "Asset pricing and corporate finance don't need subsidies, but household finance does."

If we took a look at NSF grants for economics, I'm sure that we'd see nearly all of the money going to the top 50 research institutions. Given that the AVERAGE salary for professors at these institutions (not including their consultancy fees) places them firmly in the top 1% of wage earners, I see little need for increased government subsidies. These professors will produce nearly the same amount of research and rake in nearly the same amount of consultant fees even if we cut the current subsidies that exist.

Isn't the top 1% like $500,000? I doubt the average economist in a top 50 research institution makes that much without consultancy fees.

Yes, cutoff for 1% is about 500K. Professors at top 50 schools would still be top 5% or higher, though. According to the Tax Policy Center, the cutoffs are 200K for top 5%, 236K for top 4%, 290K for top 3%, 360K for top 2%, 506K for top 1%.

Correct, virtually no one is making that much from their university. Also, most social science professors, including economists, couldn't have gotten tenure at a large research university in the first place if they weren't PIs or co-PIs on large grants, and in some subfields, NSF is the only one regularly providing substantial funding.

government subsidies for economists? Sounds like a way to finally capture the economics profession the way the other sciences have been captured.

Yeah, it's terrible what NASA and military spending have done for science.

"... the other sciences have been captured."

What? All of them?

So when you fire up your smart phone to find the nearest bookstore with a copy of Road to Serfdom, you don't trust the results? We have the science of orbital mechanics that predicts where the GPS satellites will be, the science of the propagation of electromagnetic waves through weakly conducting media (that's the ionosphere to you), the science of quantum mechanics that helps design the cesium clocks on the satellites, the science of general relativity that calculates the gravitational time dilation on the clocks, the solid state science that helps design the GPS receiver in your phone, the science of liquid crystals that lets you see the map on your phone .... yep, all captured by government. Different, of course, when exactly the same science allows someone in the USA to conbtrol a drone in Afghanistan.

1. Yes. Yes.
2. Yes. No. Bloggers are already subsidized by receiving a paycheck for teaching and research while they are blogging.
3. Yes.
4. Yes. Basic NPV problem with mutually exclusive projects.

Economists performing community service: Does this mean we would get to see Krugman, Cowen, Mankiw and all the rest in orange jumpsuits by the side of the highway?

Aren't most economists employed in universities, government and the FRB and other publicly-chartered institutions? They are already subsidized.

Contrary to "dismal science," they seem mostly concerned with telling their paymasters how government can transcend the physical laws of the universe.

Seems to me the appropriate area for investment is data collection. We screwed up the response to the recession because we didn't understand its dimensions. That's just one striking example of how we lose when we don't collect the right data timely. The data we collect today seems to fit 20th century economics, not the modern globalized world. The outgoing director of the Census suggested they needed to change what and how they collect data, yet Republican know-nothings want to scrap the American Community Survey.

Agree with Bill -- the return from collecting more and better data would be far superior to the returns from subsidizing more economists.

If you look at the data collected you find very, very detailed data on the old heavy manufacturing industries and virtually no details on the services sector and/or high tech sectors.

So, the detailed data collected on old industries done by FDR and his legacy of technocrats that was largely frozen at industry as of 1960-70 is enough, even though it doesn't tell us much about how manufacturing is done in Japan, Germany, China where successful US manufacturing draw a lot of their theories of operation, but what is need is the research not funded when the conservative "let the free market do it better" became the rule circa the 80s??

What happened in the 70s and 80s was exactly what is in point 1: the funding for data collection was tight, so the funding was limited to the parts of economics that it was agreed made sense: measuring the old pre-1970s industrial economy, plus some aspects of agriculture, and no new funding was put into the parts of economics dealing with the fringe topics.

That the "fringe" grew and grew, no funding has gone to data collection to define the service sector so anyone knows what data to collect, much less how to collect the unknown data to give details on the service sector, or on modern manufacturing.

And who should be funded to collect the data which requires deciding what data needs to be collected and how? Maybe physicists??? The physicists who came up with the models that have given us AAA rated bonds backed by bad mortgages, and flash crashes from microsecond trading? Has treating money like particle physics worked well?

The only way an economist can learn something not in a book is to collect data in the real world 90% of his time, or supervise his RA slaves who are collecting his data.

"A lot of economics research comes pretty cheaply and presumably without direct government subsidy. Isn’t the implied conclusion to invest a lot in data gathering and RCTs/field experiments, but not so much in large swathes of economics? Should we not start by listing all those branches of economics which should not be eligible for subsidy?"

This reflects the gross misunderstanding of science by economists. 90% of all the cost of science is in data collection. 9% is in science education, basically scientists being instructed in how to collect and analyze data. 1% is spent on theory.

In no branch of science is data collection not at the core of science.

In no branch of science is data collection done in lab experiments the core of the science - the data collection is done in the "real world". Data collection in the real world is expensive.

I defy you to identify one branch of economics that is irrelevant so that funding data collection in that branch is not worthwhile. If you argue that a branch of study is just a small micro part of economics, then that is an argument for erasing microeconomics from the curriculum, or at a minimum, writing the one book on microeconomics that is published once for as cheap as possible ($3 or $1) because it deserves no further research - research is data collection and analysis.

Economics is heavily subsidized: the ECB, IMF, CBO, CEA, NBER, GAO, Treasury, the Fed--virtually all of these are government entities.

Tyler, how would you measure "higher returns"?

This seems to be the fundamental issue of economics. How do we determine value without a market?

I'm very surprised to see you making objective value statements like that.

A great deal of research in pure mathematics requires only a pad of paper, a supply of pencils and a rubbish bin. Because the cost is low, the long term cost/benefit ratio is high. Economics research would have an even higher cost/benefit ratio, because economists never use a rubbish bin.

Oops :)

A great deal of research in pure mathematics requires only a pad of paper, a supply of pencils and a rubbish bin. Because the cost is low, the long term benefit/cost ratio is high. Economics research would have an even higher benefit/cost ratio, because economists never use a rubbish bin.

Your reading skills are excellent. Your grasp of logic is flawless. And your sense humor is above average.

(I can't spend all my time complaining about other commentators)

Wasn't that joke originally about philosophers? But I like the modification.

I guess if this spreads, it'll take a few lost generations before populations and cultures realize the value of research. There should be some sharing of the fact that all we see around us is a built up environment made by accumulation of smart ideas. We Lose the breadth of smarts only at our peril.

The article was not arguing for more government funding for economists, but rather the preservation of current funding. Seems like support for basic research in science (including social sciences like economics) is one of the basic roles of government. Sure basic research is and should happening in the private sector too, but that's been a smaller fraction of research than the public sector. And private sector research in economics is often a different kind of research in which the client/funder looms larger. Basic research is our seed corn. I would be worried about what those supply elasticities are, especially quality adjusted.

Economists have designed the auction mechanisms for selling spectrum and the additional revenue from better mechanism design could well be about the entire historical spending on the NSF on economics. Maybe that's a reason to channel funding into sub-fields that relate to government buying and selling but I thought the idea with basic research is it is hard to tell which sub fields are going to be useful so you mostly fund the best work in each area.

It actually doesn't matter much if these are the highest return projects before us. If government is going to fund basic research then what matters if the projects are positive NPV and that the funds exist to fund the projects. Basic economics research could plausibly be positive NPV (I argue most likely strongly) and the government can easily fund it.

I find this discussion unfocused and pretty uninformed. As far as I can gather, what's being debated could be NSF style grants to "pure" economics, if there be such a thing, i.e., theory with no immediate applications. That's a mere pittance of what the fed govt spends on economic research. First, federal subsidies to higher education benefit economics just as any other academic discipline, and no one is proposing to stop those subsidies. Second, the fed govt employs thousands of economists, micro and macro and econometricians. Every single agency is full of them. I bet the second most frequent occupation in Washington, D.C., after lawyers, is economists. That's an important piece of evidence that the market supports economists -- the govt uses them extensively. Third, every agency in the fed govt has funding for studies and analysis that is spent on academic and independent economic studies. Again, the market at work. They could buy sociologists, political scientists, statisticians, what have you, but they spend a huge amount on contracting for economists' services. Pretty much all inhouse and contracted economic studies are of the "applied" kind. This will continue as long as agencies receive funding from Congress. So, I don't know what the fuss is all about. Becker and Heckman give examples of "useful" research -- but all of their examples, as far as I can tell, concern the kind of applied research that agencies always have and always will spend money on through their own studies and analysis appropriations. Both Becker and Heckman have made considerable contributions to economic theory as well as applications, but their arguments for continued subsidies to "pure" economics fail to convince me -- and, believe me, as a bureaucrat I've spent millions of federal dollars on outside contracts, hired a slew of economists to work in government, and been on the receiving side of many hundreds of thousands of federal dollars in grants for applied research to outside organizations. If federal funding for research in pure economic theory is discontinued, only a small pittance of funding would be saved, and no one (except a small number of economists) would ever notice. So, I repeat, I have no idea what this debate is about -- it strikes me a totally off base.

The NSF policy was under debate because they tried to kill NSF funding for political science research and there was talk of coming for economics next. There was also talk of ending the census's American Community Survey of which most the expense is in non-economists gathering the data but many economists privately benefit from using.

I imagine that there are vastly more accountants and soldiers than economists in DC. The American Economic Association has about 10,000 members and the IRS employs over 100,000 people (though that's all over) and almost 30,000 people work in the Pentagon.

Or, you know, we could subsidize economics bloggers. We're just putting it out there.

The signalling joke was amusing.


You're looking at the consequences of change in subsidies without why subsidies would increase.

If in equilibrium the increase in government subsidies for economic research was due to a commitment to better economic policy, then there would be a benefit. Otherwise I don't see how there could possibly be a benefit to more subsidies for economic research.

At best it would re-locate funds from other activities to growing trees for the papers used by economists and do something about global warming.

Or never reason from what a price change could bring :p

As a free marketer, I'm disgusted by Gary Becker this time (I usually love him).

Yep, if we can't beat big government, let's stick our nose in the trough.

Here's an antidote:

From my comment 8/14/2012

Still available from Liberty Fund:

The Logic of Liberty by Michael Polanyi

These kinds of ideas raise again:
What are the functions of governments?

Is "Economics" a "science?" If so, what makes it so? Or, are we simply viewing human conduct from a particular perspective?

Even in Medicine, which is the Art of applying "science," we have objective norms for investigations, research, and testing conclusions and conjectures. Where is that to be found in the field of "Economics?"

Besides, there is no government money; there are only levies (taxes) that reduce the funds left to those who produce or serve.

Comments for this post are closed