Common sense and the economistic worldview

It’s Sunday and thus time for another break from monetary economics.  Because of heavy grading early this week, I may not post anymore for 3 or 4 days.  But I have a couple interesting ideas for later in the week.

A New York Times article once reported that economists in academia tend to vote about 3 to 1 Democratic, whereas other academics vote about 7 to 1 Democratic.  Of course the general public tends to split about 50/50 between Democratic and Republican voters.  What should we make of this pattern?  A few years ago I began thinking about this issue and came up with a hypothesis that is based on the distinction between worldviews and values.  I think economists tend to be people with liberal values and a right wing worldview.  BTW, I haven’t had time to do much research in this area, so I claim no originality for the ideas here.  (Indeed, I believe that Bryan Caplan is far ahead of me in exploring the worldview issue that I discuss here.)  However in future posts on issues like political art, I think I might be able to make some original observations.

Before pursuing those other avenues of inquiry, we need to consider what terms like ‘worldview’ and ‘values’ actually mean.  For the purposes of this post, consider the term ‘worldview’ to represent one’s views about cause and effect, or what economists call positive questions.  Values relate to what is viewed as being morally right and wrong, or normative issues.  I don’t claim that there is any clear boundary between these two categories, but I hope they will prove useful anyway.

It seems to me that in economics (and to a lesser extent in fields like criminology) there is a distinction between what might be called a “common sense  worldview” and an “economistic worldview.”  Consider the following common sense worldviews:

1.    People don’t respond very strongly to economic incentives.  (I.e., higher prices don’t discourage consumption by very much, and higher taxes don’t reduce peoples’ work effort very much.)

2.    Imported goods, immigrant labor, and automation all tend to increase the unemployment rate.

3.    Most companies have a lot of control over prices.  (I.e. oil companies set prices, not “the market”.)

4.    Policy disputes over taxes and regulations are best thought of in terms of who gains and who loses.

5.    Experts are smarter than the crowd.

6.    Speculators make market prices more unstable.

7.    Price gouging hurts consumers.

I define the economistic worldview as essentially the mirror image of the preceding seven assertions:  incentives matter much more than one would expect; imports, immigrants, and automation do not raise the unemployment rate, prices are primarily determined by market forces, tax and regulatory policies often have little overall effect on income distribution, and a big effect on efficiency, the crowd (or market) is smarter than the expert, speculators tend to stabilize prices, and price gouging is socially beneficial.  As you can no doubt tell, I use the term “common sense” in a derogatory way, much as a snobbish 19th century lady might have used the term “common.”  BTW, I am avoiding macro issues for now, but if you’ve read my other posts you probably have some idea of where I’d put Keynesian economics.

It should be noted that not all economists have a completely economistic worldview.  This worldview is certainly most pronounced among right-wing economists, but even center-left economists (who are in the majority) tend to have a much more economistic worldview than non-economists.

[One interesting question is whether the best economists tend to have a more economistic worldview than the average economist.  I’d obviously like to think so, but is there any objective evidence?  I’d be tempted to point to the disproportionate number of Nobel prizes won by economists with University of Chicago ties, however that may be related to the fact that an economics prize only began to be offered around 1970, and thus the sample period falls mostly within the (economistic) era of neoliberal policy reforms.]

Now let’s assume that “ideologies” reflect values plus worldviews.  Thus the liberal worldview has many different ideologies.  Let’s also follow the standard practice of assuming that the term “left” applies to more socialistic versions of liberalism and the term “right” applies to more libertarian, or classical liberal, versions of liberalism.

Among liberals who are not economists, the common sense worldview is much more prevalent.  For instance, Orwell (1937) claimed that socialism was a common sense ideology:

“Everyone who uses his brain knows that Socialism, is a way out [of the Great Depression] . . . Indeed, from one point of view, Socialism is such an elementary common sense that I am sometimes amazed that it has not established itself already.”

The quotation nicely encapsulates two prevalent views of liberal non-economists; that smart people are liberal, and that common sense suggests the government should play a major role in the economy.  Further evidence in support of Orwell’s quotation comes from the current financial crisis sweeping the world, which has led most “common sense” pundits to call for increased regulation, or even a new “New Deal.”

So how can we explain the voting pattern of economists?  They seem very liberal compared to ordinary voters, but very conservative compared to other academics.  Suppose that an economistic worldview makes one vote more to the right than a common sense worldview.  That could explain why economists vote Republican more often than other academics.  Indeed I think this is a pretty standard explanation of their voting pattern.  Thinking like an economist makes one less receptive to socialist policies.

But economists also have a much more economistic worldview than the general public, so why don’t they also vote to the right of the general public?  Here’s where values come in.  Jonathan Haidt has an interesting essay on “What Makes People Vote Republican,” where he considers five different types of values; care/harm, fairness, hierarchy, pureness/sanctity, and loyalty.  Haidt says that liberals tend to put more emphasis on the first two, whereas conservatives tend to be more balanced in their value system.

Interesting, he says that one group of “conservatives,” known as libertarians actually have a very liberal value system.  I would add that in my view only pragmatic libertarians are liberals, I would put dogmatic libertarians (who believe taxation for redistribution is theft) into a different grouping.  It seems to me that the libertarian anomaly can help us understand the voting pattern of all economists.

Suppose that, for simplicity, we regard liberals as people with roughly libertarian, or at least consequentialist value systems.  Then it seems clear that many so-called “conservative” economists are best regarded as right wing liberals—people with liberal values and economistic worldviews.  Milton Friedman might be a good example, as although he was generally regarded as being very conservative, he also took very libertarian positions on drugs, prostitution, the draft, etc.  But I think a much more revealing liberal position taken by Friedman was his support for a guaranteed annual income for all Americans, as this was an economic policy proposal from one of America’s most conservative economists, which turned out to be far too liberal for the general public.  When thinking about “welfare” most people distinguish between the “deserving poor,” and those who are poor due to their own foolish choices (a perspective that is less relevant to utilitarians.)

If I am right that academics tend to be utilitarians who pay little attention to values such as loyalty, respect for authority, sanctity, social cohesion, nationalism, religion, patriotism, order, tradition, etc., then it is easy to explain the anomalous voting pattern of economists who happen to be academics.  They are people with roughly the same (egalitarian/utilitarian) value system as their more left wing colleagues, but a very different worldview.

Interestingly, most non-economists do not understand this distinction, and simply assume that economists who “defend the indefensible” must have a very different value system (as Larry Summers found out after he became president of Harvard.)  The economistic worldview is so counterintuitive that most people don’t really even know it exists.

[BTW, the phrase “defending the indefensible” refers to economists (sometimes even liberal economists) who argue against rent controls, against bans on child labor or sweatshops, for sending polluting industries to developing countries, against minimum wage laws, etc.]

Next Sunday I’ll explore this issue more fully by looking at why the economistic worldview is so rare, with some interesting hypotheses from Will Wilkinson and Friedrich Hayek.  I’ll also look at some intriguing findings from the field of experimental philosophy (no that’s not an oxymoron) and make a provocative argument that George Orwell’s Animal Farm might be one of the most deeply misguided books ever written.  Also a bit on the politics of writers from the former Soviet bloc.  In other words, it should be a bit more interesting than today’s post.  But at least you now know the analytical framework that I will use in my future posts on culture.



46 Responses to “Common sense and the economistic worldview”

  1. Gravatar of Greg Ransom Greg Ransom
    15. March 2009 at 20:06

    Friedman always claimed everything could be decided by looking at the data, and it turns out perhaps the clearest data producing “experiment” in social science was the guarantee minimum income experiment run in Seattle in the early 1970s — Friedman’s idea was a complete disaster.

    More than anything, it was this experiment which dumped Friedman guaranteed annual income idea into the dustbin of history.

    “I think a much more revealing liberal position taken by Friedman was his support for a guaranteed annual income for all Americans, as this was an economic policy proposal from one of America’s most conservative economists, which turned out to be far too liberal for the general public.”

  2. Gravatar of Winton Bates Winton Bates
    15. March 2009 at 20:55

    You list of “commonsense world views” overlaps strongly with the systematically biased beliefs of the public identified by Bryan Caplan: anti-market bias, antiforeign bias, make-work bias and pessimistic bias. Such views have also been appropriately described as pre-economic and DIY economics.

    I think Caplan goes too far in attributing such views to rational irrationality – an almost religious attachment to false beliefs – rather than to rational ignorance or just lack of understanding of economics. At the same time I think you are going too far in the opposite direction in labelling the mainstream views of economists as an “economistic worldview”.

    As I understand the term, economism involves attempting to explain everything in economic terms. Some economists tend to be economistic, but I don’t think the mainstream economic view (the opposite of what you describe as “the commonsense world view”) is economistic. For example, it is possible to maintain that incentives matter a huge amount in many areas of life without dismissing the role of informal institutions (arising for example from culture and moral traditions) in influencing many aspects of behaviour.

  3. Gravatar of ssumner ssumner
    16. March 2009 at 05:00

    Both good points.

    Greg, I agree with you that Friedman’s idea of welfare with “no strings attached” was probably a bad idea. Interesting, the conservative intuition about the “deserving poor” can also be defended on economistic grounds–that welfare for anyone (no strings attached) will strongly discourage work. If I am not mistaken, even the Nordic countries don’t have a no-strings-attached welfare system.

    I have always thought that the best welfare system would combine Singapore’s forced saving system (for unemployment comp), with eliminating the minimum wage, and then adding a sliding hourly wage subsidy for low income workers. Thus workers getting $3/hour might get a $6 subsidy, those making $7/hour might get a $4 subsidy, those making $11/hour might get a $2 subsidy, etc. It would somewhat discourage human capital formation, but at least people would get bigger “welfare payments” the more hours they worked, so in that dimension incentives would be improved. The hardest problem is disability, for which I have no good ideas.

    My point about Friedman is that in rejecting the concept of the undeserving poor, he was taking a very utilitarian approach. I do agree that his actual proposal was flawed. I think Hayek also favored some sort of welfare system, does anyone know what it was?

    Winton, Yes, I thought my list would partially or completely overlap with Caplan’s. I came up with the list years ago based on interacting with students, adults, and reading the press (and later discovered Caplan’s research.) I plan to read his book this summer, like everything else in my life it is now on the back burner. I didn’t even know what the term “economistic” means, if you are right about its meaning then I am using the term in a very idiosyncratic way—I should come up with a different term for the “economic way of thinking” as I agree that economists don’t think everything can be explained in terms of economic theory. Please view my use of the term here as reflecting the different (non-commonsense) approach to those 7 issues mentioned in the post. You are right that economics doesn’t explain everything, indeed one of those things it doesn’t explain is probably commonsense thinking (Wilkinson argues that evolutionary psychology can explain this type of thinking.)
    And I have always felt that a world where all activity was motivated by self interest would be a nightmare.

    I believe democracy is a very good system, so I will be interested in reading Caplan’s thoughts on that issue. On closer examination, problems that people attribute to democracy are often due to over centralization of institutions. (Thus New Hampshire town boards work better than the NYC school system, although both might be “democratic” in some sense.) Even though this post didn’t offer much beyond what Caplan has already done, I am confident that follow-up posts will strike off in new directions.

  4. Gravatar of Alex Golubev Alex Golubev
    16. March 2009 at 05:58

    Agree on the general public underestimating the importance of incentive systems and moral hazard. but to play Lenin’s advocate – doesn’t 10:1 leverage (at best) make it a bit of a “russian roullete” to leave things up to the invisible fist? The economy growth and contracts… this economic “breathing” makes me think that maybe one size fits it all policy is not the best. There a huge moral hazard in not having failure in capitalism, but we just need to learn to design safety systems with the CORRECT DISINCENTIVES. Airbags are great, but you can still get maimed in the car crash. But i think they’re a good idea to protect the non-DUI party. is that not fair? So my argument is that the safety nets are too safe.

  5. Gravatar of ssumner ssumner
    16. March 2009 at 17:37

    Alex, When I am arguing for a free market, the argument really has nothing to do with the current crisis. I know that sounds crazy at first glance, but think of it this way:

    Suppose there is some market failure that causes too much leverage (say all the backstops like FDIC and too big to fail) So you have a regulation limiting leverage, as you propose. That’s one regulation.

    Now what about the other 1.6 million regulations? I just made up that number, I have no idea how many government regulations there are in the U.S. But I am pretty sure there are a lot. And I am also pretty sure that less than 1 in 100 have anything to do with externalities or “market failure.” The vast majority of regulations are rules like:

    1. You need a license and 100 hours study to operate a hair salon.
    2. A stock prospectus must have 100 pages of legal gobble-digook.
    3. Or our IRS tax code (which has many thousands of regs.)

    That’s what I mean by moving to a free market economy. Cutting down on all that nonsense.

  6. Gravatar of Read This If You Want to Understand Me « The Emergent Fool Read This If You Want to Understand Me « The Emergent Fool
    16. March 2009 at 17:43

    […] 16, 2009 by kevindick Scott Sumner has a great post at his blog TheMoneyIllusion (highly recommended in general if you’re into monetary […]

  7. Gravatar of Alex Golubev Alex Golubev
    17. March 2009 at 06:24

    Don’t get me wrong. My arguments are more theoretical. I don’t doubt for a second that the new regulation coming out of this or ANY crisis will be magnitudes more complex, costly, idiotic, and ineffective, than what already exists. It will create many a job for regulators, tax accountants, and lawyers, but will do barely anything to prevent the next crisis of excesses. But i would argue that it is due to the incorrect incentives of our politicians and not the idea of more “socialist” policy. Yes, our system is one of the best out there, but it itself has too many self reinforcing processes of power and money. I believe, that the possibility of FAILURE is the greatest benefit of capitalism, because it gets rid of bad ideas of people who used to have good ideas and it needs to be reflected not only in our private markets but also be more pronounced in our public SOCIALIST policy. Yes, Socialism with a hint of creative destruction of capitalism… put on your cooking hat. I’m not refering solely to this crisis. i think any public programs like welfare ought to have a time constraint of 1-2 years and pay for a max of 2 children, while requiring some sort or technical training with a goal of forcing people to work. Same goes for medicare (get healthier or get cut off, unless completely chronic/genetic condition), and public school (study harder and improve your grades with respect to history of your own grades or pay for school). It’s not black and white – free markets or USSR. i’m calling for a smarter USSA. (Not the bailout zombie nation, where it’s always sunny and everybody always wins, that we’re becoming)

  8. Gravatar of Rafe Furst Rafe Furst
    17. March 2009 at 07:32

    “One interesting question is whether the best economists tend to have a more economistic worldview than the average economist.” I assume by “best” you mean most accurate in their predictions? If that’s the case, I think Tetlock should have some data on this.

    Regarding the veracity or falsity of the economistic worldview, I would offer the following conjecture: It’s true on the average, but breaks down on either tail. And since more and more the systems of interest are non-normal (c.f. network effects, increasing returns, power laws, chaotic attractors), viewing them through the Gaussian lens will be less and less useful. In other words, the economistic worldview will be less and less true as time goes on. BUT, so will the commonsense view. It’s a case of something being so wrong that not even it’s opposite is true.

  9. Gravatar of chrismealy chrismealy
    17. March 2009 at 08:31

    That so many economists are not Republicans is a relief. It shows that deep down they don’t take the rational agent/hedonistic sociopath view of the world as seriously as their models do. Now if only they could fix their models …

  10. Gravatar of John John
    17. March 2009 at 08:57

    You seem to assume that it’s economics that makes economists relatively conservative, and not that their own conservatism makes people economists.

  11. Gravatar of ssumner ssumner
    17. March 2009 at 08:58

    chrismealy, I am not sure that the rationality question is what separates Democratic and Republican economists. I believe that it may have more to due with mundane topics like estimates of elasticities. Republican economists who support supply side tax cuts, for instance, seem to think that work, savings, and investment will respond strongly to those cuts. Democratic economists are more skeptical.
    I agree with you that many Republican economists are too in love with the “hedonistic sociopath” model, I am just not sure whether this worldview is what separates the two camps on policy questions. On the other hand, I think the rationality assumption is generally a good one for purposes of public policy discussion–although I do have an open mind regarding behavioral econ.

    Rafe, I’m afraid that I do not agree with your assumption that predictive prowess is the test of a good economist. A good economist is someone who can convince his peers to see the world in a new way. Prediction may play a role in achieving that success (although I would emphasize conditional prediction, not unconditional) but so might many other factors. These include plausibility (in terms of accepted theory, or based on introspection), as well as theoretical elegance. A useful theory might also be one that provides a COHERENT explanation of complex historical events that formerly seemed unintelligable. I am a methodological pluralist.

    Regarding your “increassing non-normal” hypothesis, I am not convinced that the economy is changing in the fundemental way you describe. On the other hand, you probably know much more about micro than I do, so I’ll defer to your judgment for now. But in macro, something I do know quite a bit about, the economy seems to operate pretty much as in did in the 1930s. When you have a monetary policy stance that generates falling nominal GDP, all sorts of bad things happen, from unemployment to financial crisis.

    Alex, The way you put things seems very reasonable. But I am not comfortable with the USA/USSR framing. We need a far more free market economy than even the USA has–something closer to Denmark, which in an earlier post I argued (based on Heritage Institute data) was the most free market economy on earth. We don’t need to find a happy medium between two countries that both have far too much regulation.

  12. Gravatar of ssumner ssumner
    17. March 2009 at 09:08

    Regarding my previous post, I should have said his/her, not “his”, as I didn’t mean to exclude female economists.

    John posted right before I finished the comment. My response to John is that economists vote 3-1 Democratic, so they don’t go into economics because they are conservative. I think what you meant is that they have an economistic worldview (which I argue is combined with liberal vlaues.) OK, maybe they do go into economics because of this worldview, but Sunday I will discuss evidence that the common sense worldview isn’t just intuitively appealing, it is also wrong, and economists have good scientific evidence to show it is wrong. I have little doubt that the economistic worldview on questions like incentives is at least valid as far as left-leaning economists push it, (and I think much farther.)

    It seems to me that I am getting some new visitors. Please don’t be put off my my disagreement, I think you guys are putting forth very good arguments. Welcome.

  13. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    17. March 2009 at 10:09

    Good timing, what with all the hullabaloo about the AIG ‘bonuses’ that were just paid. A lot of people are exposing themselves as common sensers, including our new President.

  14. Gravatar of Values vs. Worldviews « Blackadder’s Lair Values vs. Worldviews « Blackadder’s Lair
    17. March 2009 at 10:18

    […] vs. Worldviews Here are some excerpts from another interesting post from Scott Sumner: A New York Times article once reported that economists in academia tend to vote about 3 to 1 […]

  15. Gravatar of Values vs. Worldviews « Vox Nova Values vs. Worldviews « Vox Nova
    17. March 2009 at 10:20

    […] vs. Worldviews Here are some excerpts from another interesting post from Scott Sumner: A New York Times article once reported that economists in academia tend to vote about 3 to 1 […]

  16. Gravatar of Alex C Alex C
    17. March 2009 at 11:54

    How is price gouging socially beneficial? Not trying to be snarky, just curious what economic theory this is evoking. As I understand it price gouging is by definition bad since it’s a derogatory term used only by a buyer when he feels a price is too high.

    Also, don’t economists believe in bubble-and-bust cycles driven by speculation? That’s the sense in which common sense thinks *excessive* speculation is destabilizing, even though moderate speculation is beneficial, so on that point wouldn’t these two straw men agree?

  17. Gravatar of Jon Jon
    17. March 2009 at 13:05

    “Good timing, what with all the hullabaloo about the AIG ‘bonuses’ that were just paid. A lot of people are exposing themselves as common sensers, including our new President.”

    The focus ~200m in bonus payments is actually pretty shocking. Talk about missing the forest for the trees in dissecting the fate of 0.1% of the money AIG has gotten. In the middle of the second page of the WSJ’s article on the subject, you get to the line that says these were retention payments. Very, very standard stuff. 200m in retention payments on a multibillion dollar payroll in bankruptcy? If anything its conservative.

  18. Gravatar of ssumner ssumner
    17. March 2009 at 13:26

    Alex, Price gouging refers to the practice of setting prices at the market clearing level, when there has been a sudden supply disruption or increase in demand. It is beneficial for several reasons.

    1. It allocates goods to those with the greatest willingness to pay, and thus those who get the greatest benefit from using the product.
    2. It encourages suppliers to rapidly bring new supplies to the market with the sudden scarcity. If price gouging does not occur, then there is a supply/demand imbalance (shortage) which can persist for quite some time.

    Speculators tend to stabilize prices by buying something when the price is low (and thus driving up its price) and selling when the price is high (and thus driving down the price.) Speculation gets a bad name for several reasons:

    1. Many people don’t understand the useful service provided by speculators (which is similar to the price gouging example–except that goods are reallocated over time, not location.)
    2. Sometimes speculators make mistakes and buy high and sell low. This destabilizes markets (such as housing circa 2006). Of course they lose money when doing this, so we can assume it is not the norm.

    Jon, I agree. BTW, I’m probably dense, but I couldn’t tell whether Patrick was being sacrastic about me, or Obama. I should say that there is nothing wrong with a common sense attitude toward economic questions. I have a common sense attitude toward quantum mechanics, in other words I’m clueless about it. I have no idea why people who haven’t spent years studying economics should be expected to understand it. It’s a really hard, really counterintuitive subject (although not as hard as quantum mechanics.) Many people in real life ask me to explain my views, but it is very hard because there are so many hidden assumptions behind each step in the reasoning process–so I am never sure if I am actually getting my ideas across. Most of the readers of the blog, on the other hand, seem to have a pretty good knowledge of economics.

  19. Gravatar of Jon Jon
    17. March 2009 at 14:17

    “Jon, I agree. BTW, I’m probably dense, but I couldn’t tell whether Patrick was being sacrastic about me, or Obama. ”

    I couldn’t tell either to be honest…

  20. Gravatar of Marti Marti
    17. March 2009 at 14:31

    “Regarding my previous post, I should have said his/her, not “his”, as I didn’t mean to exclude female economists.”

    Um, no. No you should not. If you do not want to follow the centuries old English custom of using “mankind” and the non-specific “his” to refer to all people, then just choose a gender and stick with it. In place of generating a mentally internal audible sound, the symbol “/” generates a very distracting external eyeroll.

    Marti in Virginia

  21. Gravatar of Alex C Alex C
    17. March 2009 at 15:27

    Thanks for the clear explanation!

  22. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    17. March 2009 at 16:18

    Believe me, when I’m being sarcastic you won’t be in any doubt. I was just pointing out that there is a good example of the point being made in this post; the AIG stink. Jon understood the point I thought.

    Alex C., a good tutorial on the price gouging point is Russ Roberts’ recent short novel, The Price of Everything:

  23. Gravatar of ssumner ssumner
    17. March 2009 at 17:32

    Patrick, Yes, when I re-read it I see you weren’t being sarcastic. But at least Jon was also in doubt, so I don’t feel quite so dense. The Roberts book looks like another one I should read this summer.

    Marti, I sympathize with your view, but being a right wing liberal I feel I have to please both cultural liberals and cultural conservatives who read my Sunday posts. So maybe that’s why I’m a bit too PC. Next Sunday’s post will be much more interesting (I hope.)

  24. Gravatar of Melchi M. Michel Melchi M. Michel
    17. March 2009 at 18:53

    As I read through your list, the thing that struck me (like Alex C) was the bit about speculators (#6). This makes it seem as though the effects of speculation on the market are clear cut (to economists). While I am a lay person regarding economics, I thought this was a bit unclear. You yourself point to the impact that speculation may have had in destabilizing the housing market. As a scientist, I wonder if you could point me to experimental or carefully controlled correlational data pointing to stabilizing effects of speculation. While I understand your explanation to Alex C., it seems as though one could easily come up with a similarly cogent argument for the destabilization effects of speculation. For instance, speculators are not necessarily interested in the ‘inherent value’ (for lack of a better term) of an asset; they are only trying to make money. That means that I, as a rational speculator, can realize that housing is rediculously overpriced, that we’re in a bubble, and that the floor will soon drop from under the market, but nonetheless invest in housing, betting that the bubble’s peak has not yet arrived. In any event, the assumption that speculators are, on balance, rational seems to argue against the success of the gambling industry.

  25. Gravatar of Sören Höglund Sören Höglund
    18. March 2009 at 00:14

    Pardon my “common sense” (I stumbled over via Andrew Sullivan’s blog) spit-take, but as someone who’s never really studied economics, but has a reasonably fair grounding in science, I find this fairly shocking:

    “Rafe, I’m afraid that I do not agree with your assumption that predictive prowess is the test of a good economist. A good economist is someone who can convince his peers to see the world in a new way. Prediction may play a role in achieving that success (although I would emphasize conditional prediction, not unconditional) but so might many other factors. These include plausibility (in terms of accepted theory, or based on introspection), as well as theoretical elegance. A useful theory might also be one that provides a COHERENT explanation of complex historical events that formerly seemed unintelligable. I am a methodological pluralist.”

    How is predictive prowess not *essential* in judging if someone’s a good economist or not? You can construct the most elegant hypothesis imaginable using post-hoc analysis, but if it has no predictive value it’s at best useless and at worst dangerous when people use it to make decisions. People do base public policy on these things…

  26. Gravatar of ssumner ssumner
    18. March 2009 at 04:01

    Melchi and Soren, Both great questions. Welcome.

    Melchi, I would guess that the best data in support of speculation revolves around products with season supply or demand. I would guess you could find studies showing that speculators buy up fuel oil when it is cheap (summer?) and sell it when it is expensive (winter.) I’m not sure about this particular case, but I am pretty sure that you could find many of this sort of case involving agricultural commodities, etc. In the stock market, speculators quickly bid up the prices of stocks to reflect new information. Almost all of these effects occur within seconds, and immediately move stock prices to the new (perceived) equilibrium based on the new perception of fundamentals.
    My sense it that the general public only tends to notice speculation when it (ex post) goes wrong, when it moves prices away from their fundamental equilibrium. And this certainly happens a lot, but that’s because prices are so hard to predict. If they were easy, it would be easy to get rich. Oil speculators buying last summer is another example of (ex post) destabilizing speculation. And those speculators lost a lot of money. I believe that Milton Friedman wrote the classic defense of speculation–try googling it.

    Your hypothesis that they might try to profit by buying in an already overheated market, and selling when it was even more overheated, sounds appealing at first glance, but is probably not much of a problem in reality. On problem is that in markets like housing, the asset is very illiquid. So even though there is a bit of price momentum, the iliquidity of houses makes it hard to profit from that momentum. But I think it would be easier to address your question by abstracting from liquidity, and looking at the same idea in the highly liquid stock market. Could speculators have done the same during the tech bubble?

    I suppose it is possible, but I would make two observations.
    If they bought in an overheated market and sold at an even higher price, wouldn’t they make the bubble happen more quickly, but peak out at a lower price? More importantly, I think it would be very foolish for a speculator to follow such a strategy, as stocks are incredibly hard to predict. Thus if you buy at an already overheated price, you are making a job that is difficult in the best of times, even more difficult. If the speculator really can predict somewhat, why not buy at a time where the asset is below its fundemental value, that makes it more likely that you will be successful in market timing. I think that most speculators that tried to time the market peak would end up losing money.

    I know these arguments are not very rigorous, and you could find rigorous theoretical models (by behavioral economists) that support you intuition, but I think most economists still believe that speculation is stabilizing more often than not, and I think they are right. I think the common sense view is a cognitive illusion–appealing at first glance, but less so on closer inspection. Sunday I will talk more about cognitive illusions in economics.

    Soren, I have class now, but will answer your very interesting question around midday.

  27. Gravatar of ssumner ssumner
    18. March 2009 at 08:18

    I’m still not happy with my answer to melchi, so I will do an entire post on why I think the efficient markets hypothesis is unfairly maligned, hopefull by Thursday or Friday. I also realize that I need to do an tire post on Soren’s question, so rather than give a slapdash answer at work, I will do a whole new post (hopefully tonight), in response to the methodological issues raised by Soren.

  28. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    18. March 2009 at 09:29

    Soren’s question seems to be Milton Friedman’s position.

  29. Gravatar of Alex Golubev Alex Golubev
    18. March 2009 at 10:07

    i think the choice of words is key. speculators do not make prices MORE stable. Rather, speculators make prices stable MORE OFTEN than they make them unstable, but when they make them unstable it might bring the whole financial system down. But yeah, i’ve been looking forward to a post on this topic. Predictability is one priority of any field of study. another priority is changing/improving human behavior. There’s an argument there that economics can create a better system after the fact without predictability of the future system’s weaknesses. I think that’s the goal of economics. it’s the rest of the folks including myself that choose to speculate on the future. almsot by definition economics cannot predict much, because agents will change their behavior once something is learned, but the argument is that the new system is better at allocating resources. once again, this is where i’d love to hear thoughts on irrationality of democracy.

  30. Gravatar of Melchi M. Michel Melchi M. Michel
    18. March 2009 at 10:24

    After reading your response (and rereading your response to Alex C), it strikes me that your point #6 may have more to do with differences in definitions of speculation than in differences between common and economic worldviews. What I (and possibly other laypersons) consider speculation seems to correspond to what you might call meta-speculation or second-order speculation. In particular, it seems that you would consider anyone who is neither a producer nor a consumer to be a speculator. My definition is a bit more narrow. I wouldn’t generally classify a distributor, for example, as a speculator because his valuation of a commodity is strongly bound to its (estimated) fundamental value. For me, speculators are agents that trade on trends, in a sort of Keynesian beauty contest manner, without much concern for the fundamental value of the underlying commodities.

  31. Gravatar of Melchi M. Michel Melchi M. Michel
    18. March 2009 at 10:47

    I had missed Soren’s point above, but I agree wholeheartedly with it. While I don’t know what makes a ‘good economist,’ I can guarantee you that If I came up with an abstruse black-box model that reduced the uncertainty in market predictions by, say 50% over the best existing models, that people who care about money would quickly adopt that model over the most elegant of existing models.

  32. Gravatar of ssumner ssumner
    18. March 2009 at 11:28

    Melchi, Anyone who follows Keynes’ “beauty contest” approach is likely to lose money. I think speculators do mostly trade on fundamentals. The problem is that the fundamentals are often extremely hard to discern. But those treating markets as a casino are foolish if they think that can predict trends. I don’t know of anyone who can reliably out-predict markets by even one percent, much less 50%. If someone finds such a get-rich-quick black box please let me know. But I completely agree with you that if they could find such a model, people would take the model over a more elegant model–I would never deny that prediction is a very important test of models.) I will have a much more serious answer for you in a few days in the form of a post on efficient markets. My argument will not focus on the unresolvable debate about whether markets are efficient, but rather the much more tractable debate about the policy implications of the EMH vs. anti-EMH models, which I think might surprise some people.

    There is also a lot of misunderstanding about the impact of speculation on market prices. I often disagree with Krugman, but he was exactly right when he ridiculed arguments that the high oil prices in 2007 were caused by speculation. In order for that to be true, inventories would have had to go up, but there was no evidence of this occurring. If two people trade futures contracts in the oil market, one taking a long position and one taking a short postition, it has no effect on the spot price. The spot price is only influenced when oil actually goes in or out of storage (or a producer holds back on production–which is the same thing), but that is exactly the activity that you said you didn’t consider true speculation.

    Regarding Alex’s comment. I think people grossly overestimate the disruptive effects of speculation. Housing speculation didn’t bring down our financial system–Fed policy which sharply reduced nominal GDP brought down that system. So even if some speculation is destablizing, I don’t think it does all that much harm, and surely speculation is more often than not stabilizing, or speculators couldn’t earn a living. If we are going to fix our financial system so that this doesn’t happen again, the first thing we have to do is correctly diagnose the problem–which is what I am trying to do with this blog. We als misdiagnosed the Enron fiasco, which led to sarbanees Oxley, which did absolutelely no good in this crisis. So I implore people to think more seriously about a forward-looking NGDP target. If we had had one in 2004, the whole crisis might never have happened, or at least been much milder.

    Patrick, Yes, it is Friedman’s position, but I think that position is no longer taken very seriously by philosophers. I’ll have more to say on that in my next post. So I agree with Friedman on speculation, but not on methodology (and I doubt that Friedman really believed what he said about methodology.)

  33. Gravatar of Melchi M. Michel Melchi M. Michel
    18. March 2009 at 12:05

    Thanks for the response. I look forward to your post.

  34. Gravatar of Jon Jon
    18. March 2009 at 12:08

    “I don’t know of anyone who can reliably out-predict markets by even one percent, much less 50%.”

    The right body of literature here is from computer learning and computational finance. Arbitrary strategies depend on speed. The market gaps close very rapidly. Profits come from many small trades. A friend of mine runs such a fund, the transaction volume is 1000x times the fund-balance.

  35. Gravatar of Jon Jon
    18. March 2009 at 15:17

    Obviously the word I was looking to write was arbitrage

  36. Gravatar of ssumner ssumner
    18. March 2009 at 17:42

    Thanks Jon, I’ll defer to your expertise on arbitrage. BTW, I just posted a long response to Soren’s question. Either tomorrow or Friday I’ll do one on the EMH, for which I am anxious to see how people react. So please check back later Melchi, as I am interested in how skeptics view my argument.

  37. Gravatar of Alex Golubev Alex Golubev
    19. March 2009 at 06:46

    “Housing speculation didn’t bring down our financial system-Fed policy which sharply reduced nominal GDP brought down that system” – 1. Your argument seems to be that the organization that was put together to prevent panics (which are also speculative phenomena) is responsible for not preventing a panic and not the panicing agents themselves? I don’t think it’s fair to blame the safety mechanism and not the agents themselves. 2. Whether there is a Fed or not, the resources were MISallocated by speculative agents. it doesn’t matter whether it brought the system down or not. Too much money and credit was tied to housing that should have been for “optimum resource allocation”. We can argue about what ORA is i guess, too. I believe it’s RISK-ADJUSTED IRR and speculative agents underestimated risk (“house prices never go down”). I don’t think one can deny that speculators get caught in positive feedbacks quite often.

  38. Gravatar of ssumner ssumner
    19. March 2009 at 17:45

    Alex, I agree that the sort of problems we had in late 2007 and the first half of 2008 (when nominal GDP was growing and unemployment rose only modesly) were due to misallocation of credit. I also believe that the recent sharp fall in nominal GDP is the Fed’s fault. Yes, the Fed was set up to deal with financial crises, but it long since has evolved into the instutition that controls the path of nominal spending (which only the central bank can do in a fiat money regime.) In my view it has failed at that job.

  39. Gravatar of Lorenzo (from downunder) Lorenzo (from downunder)
    29. March 2009 at 00:02

    Without disagreeing with your discussion about the economistic world view, there is data on academics beyond just economists and non-economists. I have mounted an argument from that on the differences in ideological and partisan identification.

  40. Gravatar of ssumner ssumner
    29. March 2009 at 10:06

    Thanks Lorenzo, That data is slightly different, as you say, but roughly comparable to the 3-1 and 7-1 ratios I cited.

  41. Gravatar of Arare Litus Arare Litus
    5. April 2009 at 05:50

    “then adding a sliding hourly wage subsidy for low income workers. Thus workers getting $3/hour might get a $6 subsidy, those making $7/hour might get a $4 subsidy, those making $11/hour might get a $2 subsidy, etc.”

    Regarding sliding wage subsidy – isn’t this asking companies to transfer payment of their workers to the public? Canada has a system that is somewhat analogous, unemployment benefits: companies game the system by rotating layoffs such that unemployment benefits pay for workers for a sizable fraction of their income. Workers pay into unemployment, but I believe that for many industries the amount drawn out is much larger than the workers within that industry pay in – so this insurance program is more of a transfer program in many cases.

  42. Gravatar of ssumner ssumner
    5. April 2009 at 16:00

    Arare, Under my hourly subsidy system unemployment wouldn’t game the system. That’s the advantage. I don’t doubt there are flaws, but fewer than our current welfare state has.

  43. Gravatar of Arare Litus Arare Litus
    5. April 2009 at 20:32

    Mr. Sumner,

    I did not mean to suggest that unemployment would be gamed, but that companies would simply underpay workers as the public would pick up the tab (and, yes, I do realize you have a sliding scale) so that this sliding wage would be misused.

    Work that is not cost effective would therefore be subsidized by the taxpayer. I meant only to make a rough analogy with the current situation in Canada, were *effective* we have the situation of a sliding wage scale, implemented crudely via gaming the unemployment system – “companies” are paying (companies + the taxpayer) employees at a rate that does not reflect the actual benefit of the labour.

    I agree is likely better than the current welfare system, I heartily agree. At least *something* productive is occurring, skills are being transferred, and people are working (an intrinsic good for many). But that is setting the bar low, it seems like it would take a real effort to design a welfare system that was worse than our current one.

  44. Gravatar of ssumner ssumner
    7. April 2009 at 03:14

    Arare, Companies do not determine wages, markets do. If companies had that ability, why would McDonald’s often pay starting employees wages far above the minimum wage? Or why would companies pay workers in Switzerland 100 times more than Bangledesh? The answer is that wages are determined by productivity. Low skilled workers would still prefer more wages to less, and will seek out the job that is best for them–where they are most productive. So I am not at all concerned about the issue you raise. My proposal would generate full employment, and the wage level would then be determined in each industry by productivity levels, not by company decisions. It only seems like companies have an ability to determine wages because there are cases where the labor market is not free, and companies have monopsony power.

  45. Gravatar of Arare Litus Arare Litus
    7. April 2009 at 05:13

    Mr. Sumner,

    “So I am not at all concerned about the issue you raise.”

    At all?

    Good points, which I of course agree with, but having the sliding scale removes the freedom from the market. We are really arguing implementation details, but that is key for structures insulated from market forces: as implementations are locked in once set, any problems they have can have very bad effects.

    I imagine political forces would turn the sliding scale into a “minimum wage” bottom, and companies would contribute as little as they have to towards the wage the is set by company + taxpayer. The wage a company agrees to pay is set by productivity, yes, but the wage that is “supplemented” by society is blind to such realities.

    To use an unfair analogue, one could argue that the Soviet Union had a (very) poor implementation of this system: where the state picked up the floor, and the additional wages supplied for performance were weak and highly politically determined (fame, additional perks, special stores w/out massive lines, etc.). Yes, this is clearly an extremely poorly implemented version, but indicates – nicely as it is so extreme – what is a possible danger and inherent weakness of the system of public supported wages.

    I bicker at this idea, as I tend to like it – I *want* the NIT to work, I *want* this modification to work, but I don’t have faith (perhaps because my bias is so aligned with the idea).


  46. Gravatar of ssumner ssumner
    8. April 2009 at 16:45

    Arare, Well I can’t argue with anything you say here. I think I wrongly assumed that you didn’t understand my original idea. A wage subsidy is of course the mirror image of a wage tax (in one respect.) While a tax reduces hours worked, a subsidy increases hours worked. Because the subsidy increases employment to the full employment level, it does bring in low skilled workers who were not previously working, and their wages would be fairly low. But they would still equal productivity. Just as companies don’t absorb labor taxes (which are passed on as lower wages) companies don’t benefit from labor subsidies.
    The main problem with a wage subsidy is that it discourages human capital formation, as your after subsidy wage rate rises by less than your before subsidy wage rate (due to the phase out provision.) But that is true of taxes as well. The problem of politics is always there, but I am just describing what I see as the best welfare state (in case we every get a government that is actually concerned about economic efficiency, like Singapore and Denmark.)

Leave a Reply