Homogeneous Economicus
Conor Clarke points to Greg Mankiw’s table of propositions to which most economists agree. This list is from Mankiw’s economics textbook, so I assume that it was not casually constructed. Mankiw says of the list that:
If we could get the American public to endorse all these propositions, I am sure their leaders would quickly follow, and public policy would be much improved. That is why economics education is so important.
But there are several things that are striking to me about it.
First, it seems to me that, charitably, half (7 of 14) items on this are even falsifiable:
1. Tariffs and import quotas usually reduce general economic welfare. (93% of economists agree)
2. Cash payments increase the welfare of recipients to a greater degree than do transfers-in-kind of equal cash value. (84%)
3. A large federal budget deficit has an adverse effect on the economy. (83%)
4. A ceiling on rents reduces the quantity and quality of housing available. (93%)
5. Fiscal policy (e.g., tax cut and/or government expenditure increase) has a significant stimulative impact on a less than fully employed economy. (90%)
6. The gap between Social Security funds and expenditures will become unsustainably large within the next fifty years if current policies remain unchanged. (85%)
7. A minimum wage increases unemployment among young and unskilled workers. (79%)
In case you’re wondering, those are the falsifiable ones, assuming background technical agreement among all the people answering the survey about the meaning of pretty general terms like “welfare” and “adverse effect”.
The other seven statements do not seem remotely close to specific enough to be falsifiable, and are in fact mostly normative statements:
1. The United States should not restrict employers from outsourcing work to foreign countries. (90%)
2. The United States should eliminate agricultural subsidies. (85%)
3. Local and state governments should eliminate subsidies to professional sports franchises. (85%)
4. If the federal budget is to be balanced, it should be done over the business cycle rather than yearly. (85%)
5. The government should restructure the welfare system along the lines of a “negative income tax.” (79%)
6. Effluent taxes and marketable pollution permits represent a better approach to pollution control than imposition of pollution ceilings. (78%)
7. Flexible and floating exchange rates offer an effective international monetary arrangement. (90%)
Further, most of those that appear to be technically falsifiable, can’t, in practice, be falsified. Consider “Fiscal policy (e.g., tax cut and/or government expenditure increase) has a significant stimulative impact on a less than fully employed economy”. We are now about to engage on a trillion-dollar stimulus program, and no matter what happens to economic growth over the next several years, the lack of the counterfactual means economists will still be debating the causal impact of the stimulus even after the fact. A material number of respected economists will continue to believe this statement, while others will dispute it. That is, we will continue to have Keynesian and non-Keynesian economists, with waxing and waning fashionability, over decades. If a one trillion dollar stimulus won’t resolve this, what will?
Further, even those that seem to be more practically falsifiable, are directional statements, while the size of the effect is crucial for actual policy-making. How much does a minimum wage of what level depress unskilled employment? How large an adverse effect of what kind does a budget deficit of what size have on the economy? We would need quantification of these effects to answer practical questions. As per the current stimulus arguments, how much impact stimulus spending would have, (i.e., what is the multiplier?) is the essence of the debate. Many economists who agree to the stimulus statement on Mankiw’s list have entirely different opinions on the best stimulus policy, because they disagree about the size of the effect.
And further, even if we could get good quantification for some of the statements on Mankiw’s list for specific instances, we still wouldn’t have useful decision rules, because the real answer would always start with “It depends on…”. We would need a more comprehensive model of the economy to be able say “If variable X changed, given current values for this long list of other relevant factors, this would be the predicted change”. Of course, I’m describing large-scale econometric models, which, to put it mildly, are not super-reliable.
If Mankiw’s list is the best economics can do, it sure seems like a naked emperor moment to me. Where’s the beef?
My challenge would be simple: please list 14 useful, non-obvious predictive rules that economics provides that have survived rigorous, replicated falsification trials.
If you were to provide this challenge to physics or biology, it would be easy to come up with 1,400. Hence, human invention of aircraft, space travel, mobile phones, antibiotics, vaccines, MRI scans, the internal combustion engine and so forth. This – not the attempt to create pressure on public officials to support the policy preferences of most economics professors – is why actual science education is so important.
I got the first one. How about, “Our theories are most likely incomplete, inapplicable, or wrong.” Though it may be obvious, it’s definitely survived numerous attempts at falsification.
— JA · Feb 19, 04:17 PM · #
What is the differnce between this the following two staments that you lable falsifiable and not falsifiable?
“Tariffs and import quotas usually reduce general economic welfare. (93% of economists agree)”
“The United States should not restrict employers from outsourcing work to foreign countries. (90%)”
These are both unspecific broad statments, but it seems like we have long term experiential data with tariffs and their effect but much less so with resticting outsorceing which has only been widely possible for a short time.
Couldn’t your list be classified as tested and untested theories, understanding of course that it is very had to rigorously test macroeconomic theories? That would mean that a lot of economist believe in untested theories and maybe they should be presented as such, but when you ar talking about macroeconomics don’t you have to act on your best guesses a lot of the time, becasue of the difficulties of macro economic experimention? So in that case the economicist who blieve these propositions are true are really saysing, we believe they are true to the best of our knowledge. And I don’t see how you can do macroeconomics without operating with that kind of conditional belief.
So I kind of think you a railing against reality here. You want the study of economics to be more perfect than they can be, and as such may be in danger of throwing the baby out with the bathwater.
— cw · Feb 19, 04:36 PM · #
“This – not the attempt to create pressure on public officials to support the policy preferences of most economics professors – is why actual science education is so important.”
For better or worse you’ve hit what is an insta-boner issue for me; education in math and science. A homily:
A few years ago I had a camera motor go down the day before a shoot. In all of New York City there are only two or three places that could do the sort of repair I needed, and only one of them open over the weekend, so Saturday afternoon found me in spare room of a row house in Queens.
My savior was a Russian who had emigrated to to New York some 15 years earlier. Back in Russia he was a physics professor, but had found a better life in the US as a camera mechanic; first working for one of the largest NY rental houses, then opening his own custom modification and repair shop. While he diagnosed and repaired my camera, we chatted about the dreadful state of math and science education in this country, and his observation stayed with me.
In a matter of fact tone and using value-neutral language he offered that Russian or Chinese students excel in math and science because there’s nothing else for them to do; no nintendo, no all star traveling soccer league, no trips to Disneyland (or Ft. Lauderdale) over Spring break, in short, none of the very pleasant diversions available students in the US. It was the first time I had heard anyone lay out a reason for our nation’s lagging math and science achievement that wasn’t rooted in nostalgic moralizing and actually made sense.
I don’t know the corrective might be. I am, (or at least used to be,) a pretty smart guy. None the less, it took me no less than two hours a day of rigorous study to pull a passing grade the physics class I referred to in another thread. Outside of my studio art classes and instrumental music studies, no other of my college courses required anywhere near that level of time investment. Even then, the art and music classes were upper level; the physics class was for underclassman.
Not too surprisingly, the physics class was not a popular. On a campus (not elite, but Pac10) with some 15K students, one section was sufficient. It started the year with about 100 students. By the end of the first quarter it was down to 50; by the end of the year it was down to 30. I don’t think this was because the students weren’t smart enough, or weren’t wiling to work hard. I think it’s because they had more interesting and amusing things vying for their time; and no sense that excelling in physics was going to improve their lot in an already luxerixous life.
Perhaps there might be a change in culture. I like “Cyberchase” very much; more importantly my daughter does, or did. She’s outgrown it. But the “real world/team challenge” tween program that seemed to be the next age-appropriate offering was lame and forgettable. Maybe something better will come along.
But for the time being, demonstrable scientific facts are open to debate. Creationism is taught in school, and a high school diploma can be obtained from perfectly respectable well-funded middle class schools without the student ever being obliged to fail algebra, must less master it.
— Tony Comstock · Feb 19, 05:01 PM · #
cw:
As I said, I was being charitable about the falsifiable list.
The big difference is that if you assume some specifiable meaning that was short-handed out of the survey but known to the respondents for “general economic welfare” (again, as per the post), one could, in principle test the proposition:
“Tariffs and import quotas usually reduce general economic welfare. (93% of economists agree)”
Though, also as per the post, this is a theoretical possibility, in that it’s hard to see how one would actually execute the required tests in the real world.
On the other hand, how could you test:
“The United States should not restrict employers from outsourcing work to foreign countries. (90%)”
If I ran such an experiment, the results could such that person A might find the effects of such restrictions great, and person B might find the exact same results awful. The word “should” is a dead giveaway that such a statement is not inter-subjectively testable.
I get your point that decisions have to be made in realms where we do not have lab-bench chemistry type certainty. I’m not railing agianst that. I’m railing against the false assertion of the level of certainty that the economics profession asserts about their beliefs, and their desire to lay the mantle of scientific knowledge over these assertions as a public rhetorical tactic. The reason that I don’t think this is simply emotional pique on my part is that I believe recognizing the uncertainty of this knowldge tends to lead to different kinds of decisions. Becasue these different decisions are based on a firmer recognition of how little we know, I believe that they will tend to be better.
— Jim Manzi · Feb 19, 05:26 PM · #
To be fair, cw, “falsifiable” means something different. One experiment can kill a physics theory no matter how many similar kinds of experiments validate it. Nobody is asking that of social sciences.
— Sanjay · Feb 19, 05:38 PM · #
“The reason that I don’t think this is simply emotional pique on my part is that I believe recognizing the uncertainty of this knowldge tends to lead to different kinds of decisions. Becasue these different decisions are based on a firmer recognition of how little we know, I believe that they will tend to be better.”
and
“To be fair, cw, “falsifiable” means something different. One experiment can kill a physics theory no matter how many similar kinds of experiments validate it. Nobody is asking that of social sciences.”
I guess this where my emotional pique comes from. How can anyone understand these sorts of differences and make informed choices when they have virtually no exposure to math and science? Increasingly, ignorance is merely a difference of opinion.
Again, I can offer no solution. Arguing on the internet is more fun and less arduous than doing my job; so I rarely get around to it until the bills come due.
— Tony Comstock · Feb 19, 05:46 PM · #
Tony: Arguing on the internet is more fun and less arduous than doing my job.
It’s also more fun than studying for the Bar, which is in 5 days. Anybody wanna talk about law?
Devil’s advocate: wouldn’t admitting outright the uncertainty of our approach argue in favor of Krugman’s kitchen-sink korollary?
— JA · Feb 19, 05:54 PM · #
JA:
To a degree yes, hence my take on what to do: http://theamericanscene.com/2009/02/12/what-should-we-do-about-the-stimulus
Good luck on the Bar – but whay are you selling your soul? I thought you were a philosophy of science guy.
— Jim Manzi · Feb 19, 07:02 PM · #
How do you feel about String Theory – or theoretical sciences more generally?
I imagine that is how economists thing of themselves – as doing abstract theory building that may have insights that fall off the models.
And to be fair to them, a lot of Economic insights that fall off get absorbed into our mindsets so quickly that it’s tough to think of them as innovations like the car engine is one – it’s much different from when the police would set the price of bread in the 1730s town squares.
— Rortybomb · Feb 19, 07:29 PM · #
I actually looked up what falsifiable means and thought It meant that it was possible to prove something was false or not false, unlike the assertion that there is a god which is impossible to prove one way or another. Anyway, someone tell me what falsifiable means here.
And Jim, as a process person I agree with you 100% that assertions like “We know this to be true” can badly skew the process and results. SO the question for me is do economist really believe these statements to be absolutely true or do they believe them to be the current best theories, they way scientist talk about the big bang or whatever, as the commenter above mentioned. Maybe as laymen we don’t really understand how they are saying these things. But if they unconditionally believe these to be true then I’m on your side.
And my final point stands. You have to take action here n the best information you have. Economics are never going to meet scientific standards.
— cw · Feb 19, 08:38 PM · #
Jim, thanks.
I thought you were a philosophy of science guy.
Life’s too long for that. ;)
— JA · Feb 19, 08:47 PM · #
How about: “Would it be a good idea for the government to collectivize all agriculture and then force the new peasant collectives to build backyard steel furnaces?”
That was tried during my lifetime by the most populous country on Earth, and 30 million people died in the subsequent famine.
So, we don’t argue anymore about backyard steel furnaces in economic discussions. It’s boring because we all know the answer.
Instead, we hear a lot of arguments over whether it would be better for the top income tax rate to be 30% or 45% or whatever. That’s interesting to argue about because we don’t know the answer. But it makes economists look like they don’t know what they’re talking about because all the talk is about questions where the answer isn’t obvious.
— Steve Sailer · Feb 19, 10:01 PM · #
Jim,
I’m not sure how many of these you’d accept. Some may be obvious, or not satisfy your falsification requirements.
1) Increasing the price of a good or service leads to fewer sales.
2) Increasing the number of competitors in a marketplace leads to lower prices.
3) Decreasing the price of an input good will lead to a decrease in the price of an output good.
4) This effect is stronger when a market is more competitive.
5) Public policy that caps the price of a good or service will lead to shortages.
6) Public policy that interferes with markets leads to a greater incidence of black markets.
7) In the absence of barriers to entry, industries with high profitability attract new entrants.
8) Individuals migrate towards regions with greater earning potential.
— Ben · Feb 20, 11:39 PM · #
1) Increasing the price of a good or service leads to fewer sales.
Counterfactuals include Target and luxury (distinguishing) goods.
— JA · Feb 21, 04:04 PM · #