Bill Woolsey on micro vs. macro

Here’s Bill Woolsey:

I sometimes have the impression that microeconomists smugly look down upon macroeconomists.    Microeconomics is scientific and empirical.   Macroeconomics is a mess of conflicting theories that apparently cannot be distinguished empirically.

But then…

What about the debate regarding the minimum wage?

As soon as microeconomics becomes politically relevant and controversial, suddenly the basic model (supply and demand) is subject to dispute and the empirical evidence becomes doubtful.

This reminds me of debates I occasionally get into about whether economics or physics is more “scientific,” or more “successful.”  To me it’s a nonsensical question, as one is comparing apples and oranges.  Physics can be used to predict planetary motions more accurately than economics can predict GDP and inflation.  But economics can predict GDP and inflation more accurately than applied physicists can predict earthquakes or long range weather conditions—something we care about far more than planetary motions.

Comparisons only make sense when two distinct fields are trying to explain the same thing.  Thus you might want to examine whether economists or sociologists are better able to explain crime.

Speaking of crime, I love this Alex Tabarrok comment:

Australia has great natural beauty. The British should have left the convicts behind and moved everyone else.

I agree with the entire post.  He forgot to mention that Australia’s government is just as small as the US government, but they lack all sorts of problems that American progressives claim are caused by the fact that the US government is smaller than European governments.  I hope to visit Queensland this summer.



38 Responses to “Bill Woolsey on micro vs. macro”

  1. Gravatar of Il Gattopardo Il Gattopardo
    19. March 2013 at 06:56

    Alex wrote the post about Australia. More important, following Bryan Caplan’s insights, I hope the Micros leave Accountants and Macros behind and continue to move forward.

  2. Gravatar of J J
    19. March 2013 at 06:58

    A corollary is that if microeconomists want to be critical of macro, then they must come up with better theories and models to explain macro phenomena. There is no such thing as no monetary policy, so microeconomists must come up with a better way to determine the right monetary policy if they want to complain that macro is missing all sorts of important micro factors. They can’t simply say “those factors matter and macro doesn’t include them, so macro and its predictions are worthless.”

  3. Gravatar of J J
    19. March 2013 at 07:00

    This is the second time in the past week that I have seen an Alex post attributed to Tyler. Apparently, nobody realizes that Marginal Revolution is a collaboration.

  4. Gravatar of Grim23 Grim23
    19. March 2013 at 07:32

    Problems with Australia:

    – Ridiculously complicated employment law
    – Ridiculously complicated tax system (not as bad as the US)
    – Lots of red tape and archiac regulation, especially at the state level
    – Giant nanny state; slow speed limits, sneaky hidden speed cameras, fireworks are illegal, riding a bicycle without a helmet is illegal, cigarettes are in plain packaging, toughest anti-smoking regulations in the world, you can’t serve alcohol unless you have a “responsible service of alcohol” license, restrictive liquor licensing laws, jaywalking is an offence, in WA street drinking is illegal everywhere (even for passengers in a car)

  5. Gravatar of Doug M Doug M
    19. March 2013 at 07:35

    At least the physicists have accepted that some things are fundamentally unpredictable (radioactive decay) and some things are fundamentally immeasurable (Heisenbergian uncertanty). I am not so certain that the Economists are as in tune with what is unkowable.

    Chaos theorists will tell you that physicists will never have good long-rage wheter prediction, and economists will never have long good long rage forcasts of any economic or market variables.

  6. Gravatar of ssumner ssumner
    19. March 2013 at 08:00

    Gattopardo and J, I have a bad habit of doing that.

    Grim23, When I was in Australia it seemed less of a nanny state than the US. Gambling was legal back in 1991, prostitution was legal. You were allowed to take more risks. (Less litigious?) Perhaps it’s changed.

    Doug, No, physicists have not accepted that some things are random, only one of the two versions of quantum mechanics says things are random.

  7. Gravatar of J J
    19. March 2013 at 08:05


    Actually, often in economics it’s easier to predict long-run relationships than short-run relationships (see: models of the nominal exchange rate).

  8. Gravatar of TallDave TallDave
    19. March 2013 at 09:18


    I think the other version of QM was ruled out a while back by Bell’s inequality (i.e. no hidden variables as Einstein had postulated). As far as I know pretty much everyone accepts that QM is probabilistic rather than deterministic (i.e. the Copenhagen interpretation).

    I like to think of macro and micro like relativity and QM: generally fine on their own, but problematic where they converge.

  9. Gravatar of adam adam
    19. March 2013 at 09:33

    Actually, Physicists have accepted randomness for a long time in QM. Also the uncertainty principle isn’t about randomness but about fundamental ability to measure certain variables.

    Also, complaining that physics can’t predict earthquakes or weather movements is a bit of a non sequitur; these aren’t physics so much as geoscience and…climate science or whatever you want to call it. You might as well complain that physicists don’t understand the brain (it’s made of physical matter! biophysics!) or that economists should be able to explain collective ant behavior.

  10. Gravatar of Suvy Suvy
    19. March 2013 at 09:40

    I think the biggest mistake in macroeconomics is trying to build the macro by aggregating the micro(microfoundations). I think that’s where macro went astray. In the macroeconomy, you have a financial sector that plays a key role in the formation of capital and financial assets, which play a central role in determining prices, output, inflation, and unemployment. You’ve got all sorts of different markets that interact with each other and you also have a financial sector that interacts with all of those markets as well. The more markets you have, the faster and faster the complexity of the system grows. Ergo, the complexity blows up, which means that the nonlinearity blows up(more complexity means more nonlinearity due to greater second order effects, third order effects, etc). If the system is nonlinear, then then the sum of the parts will not be the sum of the whole.

    Another key point that is missed in macro is that a long-run equilibrium is assumed. Not only is a long-run equilibrium assumed, but none of the models that are used ever have time as a dependent variable. The problem is that the behaviors of people over time shift as time shifts(in math, this is called path dependence). So it’s quite surprising to me when there are very few macro models that actually have time as a variable.

  11. Gravatar of nickik nickik
    19. March 2013 at 09:51


    I disagree 100%. Macro has gone wrong when they stopped carring about individuall action, when the stopped carring about price theory.

    Keynes was all aggregation with no knowlage or understanding of why anybody did what. The monetary disequilibrium is the view that strongly comes out of microfundations.

    The changes in relative prices, the problem with the price-level aggregat, the insentives firms/people face in terms of changing real schocks and there effects on money demand. These for me are the microfundations you really need to make this a good story that actually explains whats going on.

  12. Gravatar of ssumner ssumner
    19. March 2013 at 10:00

    TallDave, They most certainly do not all accept the Copenhagan interpretation—I did a post on this a while back, and plenty of physicists accept the MWI, indeed it has recently increased in popularity.

    Adam, But physics fans brag that it can predict planetary motion! By your logic that doesn’t count, as it’s astronomy, not physics.

    And see my previous response to Talldave, you are wrong about randomness being universally accepted among physicists.

    Suvy, I agree that it’s a mistake to try to ground all of macro in price theory–but you’d like to use price theory as much as possible. I think Keynesians went too far to the other extreme.

  13. Gravatar of Laurent Laurent
    19. March 2013 at 10:10

    Every year, physicists increase de precision and accuracy of weather forecasts. In fact, this improvement is nearly linear.

    Can the same be said about inflation and gdp predictions?

  14. Gravatar of Bill Woolsey on micro vs. macro | Fifth Estate Bill Woolsey on micro vs. macro | Fifth Estate
    19. March 2013 at 10:39

    […] See full story on […]

  15. Gravatar of Suvy Suvy
    19. March 2013 at 11:17

    Prof. Sumner,

    “Suvy, I agree that it’s a mistake to try to ground all of macro in price theory-but you’d like to use price theory as much as possible. I think Keynesians went too far to the other extreme.”

    I think that’s a fair analysis. I’d tend to agree.

  16. Gravatar of J J
    19. March 2013 at 11:21

    I think the Keynesians (and other non-micro macroeconomists) did something vital. They produced stylized macro facts. For example, they brought to light the importance of price stickiness. Had we only worked from a micro perspective, we might never have realized how important price stickiness is in explaining macro phenomena. Now that we are aware of its importance, we can properly microfound it.

  17. Gravatar of adam adam
    19. March 2013 at 12:46

    I think you don’t understand maybe don’t understand QM, nor the Bell inequalities? They say that there are no hidden variables in QM, right? And it isn’t dependent on your interpretation of QM?

    The only out that you actually have here with respect to the Bell inequalities and MWH is that things are deterministic but appear random because we have no way of knowing which world we move into. But then things are deterministic but totally unpredictable ie effectively random. Which is what Doug and I were trying to say in the first place.

    “But physics fans brag that it can predict planetary motion! By your logic that doesn’t count, as it’s astronomy, not physics.”

    That, of course, is called astrophysics. If it is even considered a separate field (and not a subfield), it is because it branched off of physics; and I doubt anyone would consider geoscience to be an “offshoot” of physics. There’s a reason many departments are called “physics and astronomy”.

  18. Gravatar of TravisV TravisV
    19. March 2013 at 13:02

    Marcus Nunes has an excellent new post on Greenspan vs. Bernanke:

    “Scott also says he thinks Bernanke is better than Greenspan. That got me confused. Is he talking about the person, the economist or the Fed chairman? Maybe Bernanke is more pleasant and polite. Surely the academic standing of Bernanke is higher. But to the extent that the Fed Chairman is influential in getting FOMC policy made, as Fed Chairman Greenspan was much better than Bernanke. If they were awarded bonus by results, Bernanke would have gotten none!

    Everyone knows the academic qualities of Bernanke. But even as an academic inferior, Greenspan developed a reputation not only as the world´s preeminent central banker, but also as the best economic forecaster. His knowledge of economic statistics and ability to dissect and connect official statistics, market/industry data, and so called anecdotal evidence was legendary. I would think those are the qualities/characteristics that make up a good central banker.”

  19. Gravatar of TravisV TravisV
    19. March 2013 at 13:12

    By the way, please see the graphs here of historical NGDP in Germany:

    Could someone please explain how Germany has been able to achieve such a robust rate of real GDP growth while the rest of Europe and the U.S. have been so sluggish? Should real GDP growth in Germany be slow as well? Hard for me to understand……

  20. Gravatar of TravisV TravisV
    19. March 2013 at 14:27

    What is the right way to interpret this data?–s-p-stocks-at-lowest-valuation-since-1980-13669.html

    “Stocks are close to the least expensive ever versus government bonds, using a valuation method favored by former Fed Chairman Alan Greenspan that compares earnings with interest payments. S&P 500 companies currently generate profit equal to 6.5 percent of their share prices, about 4.5 percentage points more than yields on 10-year Treasuries. The average spread in the past 10 years was about 2.5 percentage points, data compiled by Bloomberg show.”

    My initial take: expectations for real earnings growth from U.S. companies are still fairly low.

  21. Gravatar of Neal Neal
    19. March 2013 at 17:20

    To compare two fields, come up with a common standard, that you can apply to any scientific field. This standard should be a pure number, so you’re not comparing field-specific measurements, and it should naturally arise from the scientific process.

    Obviously, I have something in mind: precision of prediction. A field’s success can be measured by how precisely its ppredictions match up to its observations.

    I think it’s pretty clear how physics and economics stack up. This isn’t to diss economics — it’s a much more difficult field for several reasons. That is why ot is more primitive than physics.

  22. Gravatar of Josh Josh
    19. March 2013 at 18:32

    Apologies in advance on the behalf of Australia for Queenslanders. A visit to Melbourne or Sydney might be a good idea, just for perspective if nothing else.

    Grim23 is correct about those nanny state policies. Australian’s LOVE health and safety. Tax and regulation not too bad though, if not ideal. But we’ve still got some growth happening and unemployment a point lower as of last week. They’re are worse places to be at the moment, even if I do have to wear my helmet on my bike ride to work.

  23. Gravatar of Suvy Suvy
    19. March 2013 at 19:25


    Actually, Keynes did deal with the way that investment is financed and the way capital creation is financed. He actually understood the role that balance sheets played–Keynes was a speculator and ran a old version of what is now a hedge fund. What passes off as “Keynesianism” today isn’t actually Keynes’ economics. For example, Keynes’ ideas were actually first used to control inflation during World War II. Yet the common critique against Keynes’ ideas(mainly monetarist) is that Keynesian policies lead to the inflation of the 1970s(they did, but the policies weren’t built on Keynes’ ideas).

    “Keynes was all aggregation with no knowlage or understanding of why anybody did what.”

    This is completely false. There are several of Keynes’ papers where he specifically talks about the behavior of individuals. A few that you should look at are:
    1. The General Theory of Employment(not The General Theory, his book) which was a paper written in 1937
    2. Alternative Theories of the Rate of Interest
    3. The Ex-Ante Theory of the Rate of Interest

    Keynes’ ideas are actually built on how individuals like households, banks, financial institutions, businesses, etc make decisions based on an uncertain future. Keynes was just saying that there is too much complexity in the economy for the classical (and neoclassical) ideas to work. He specifically compares the classical ideas to Benthamite calculus that requires the ability to know the probabilities and distributions of everything that may happen in the future.

    By the way, Krugman does not understand what Keynes said. Krugman probably just read bits and pieces of The General Theory and claims that Keynes said various things that he supports. However, anyone that has read a good many of Keynes’ works would know that IS/LM is not Keynes’ model. In fact, it was a model developed by John Hicks who even admitted later that it wasn’t Keynes’s model and a poor interpretation of Keynes. In all of the 3 papers that I stated above, IS/LM clearly contradicts what Keynes is saying.

  24. Gravatar of Suvy Suvy
    19. March 2013 at 19:34


    By the way, these are the papers of Keynes that I was referring to:
    1. The General Theory of Employment
    2. Alternative Theories of the Rate of Interest
    3. The Ex-Ante Theory of the Rate of Interest

    In paper (2), this was a quote by Keynes:
    “The alternative theory held, I gather, by Prof. Ohlin and his group of Swedish economists, by Mr. Robertson and Mr. Hicks,and probably by many others, makes it to depend, put briefly,on the demand and supply of credit or, alternatively (meaning the same thing), of loans, at different rates of interest. Some of the writers (as will be seen from the quotations given below) believe that my theory is on the whole the same as theirs and mainly amounts to expressing it in a somewhat different way. Nevertheless the theories are, I believe, radically opposed to one another.”

  25. Gravatar of mbk mbk
    19. March 2013 at 20:58

    When I was in Western Australia recently, it seemed a lot freer (now) than those parts of thew US I knew (in the 90’s: CA, HI). It seemed like what I thought CA must have been in the 60s but with modern safety standards. BUT: Prices are atrociously high, retail, restaurants, accomodation, you name it. And that’s not just for foreigners but I believe Australians are complaining too. There was something in The Economist on that recently. And anecdotally, there just isn’t enough labor / manpower around for services.

  26. Gravatar of Rien Huizer Rien Huizer
    20. March 2013 at 00:17


    What you say about Australia may be factually true but lots of people will not agree that thst kind of regulation is a bad thing. And WA people drink copiously, wherever they are. There may be a rule but as so many, not enforced. Nanny states have rules and enforce them. Australian rules is different.

  27. Gravatar of TallDave TallDave
    20. March 2013 at 05:22

    TallDave, They most certainly do not all accept the Copenhagan interpretation””I did a post on this a while back, and plenty of physicists accept the MWI, indeed it has recently increased in popularity.

    Scott — MWI is a Copenhagen derivative, because it doesn’t have hidden variables. Hidden variable theories have almost no proponents today, i.e. virtually everyone accepts that QM is probabilistic (at least in our world-line).

    MWI is distinguished by two qualities: it assumes realism,[18][19] which it assigns to the wavefunction, and it has the minimal formal structure possible, rejecting any hidden variables, quantum potential, any form of a collapse postulate (i.e., Copenhagenism) or mental postulates (such as the many-minds interpretation makes).

    MWI may be deterministic across worlds, but since we live in one world-line the practical effect is to accept probabilistic outcomes.

  28. Gravatar of TallDave TallDave
    20. March 2013 at 05:40

    Of course there’s also superdeterminism, but that seems to run into epistemological problems about what science could even be said be doing in such a construct.

    It’s also amusing to point out that, at least in classical (non-QM) terms, physics should not only be able to predict the weather, but also the economy! See:

    Of course, we lack the ability to model on that scale, even assuming the outcomes aren’t ultimately probabilistic.

  29. Gravatar of Ritwik Ritwik
    20. March 2013 at 10:39

    “..economics can predict GDP and inflation more accurately than applied physicists can predict earthquakes or long range weather conditions..”

    Umm, no.

  30. Gravatar of TravisV TravisV
    20. March 2013 at 11:14


    The Federal Reserve put out its latest monetary policy statement today and the news is no news. Zero change. And yet the range of core inflation forecast for 2013 was revised downward. So was the range of overall inflation. And the real growth rate for 2013 was also revised downward, as was the real growth outlook for 2014.

    So if projected inflation and projected real growth are being revised downward, why aren’t new policy initiatives being announced? Or another way to look at it, if back in December the Fed liked policies that it thought would produce the outcomes that it foresaw in December why doesn’t it now want policies that will achieve those same outcomes? Recall that both inflation and real output were revised downward. That is, by definition, a demand problem. A supply shock would send growth and inflation in opposite directions.

  31. Gravatar of Lorenzo from Oz Lorenzo from Oz
    20. March 2013 at 12:17

    Australians famously regard the state as a giant utility and have a utilitarian political culture. (Think of us as the country where the Benthamite-Chartist stream in British history won.)

    There is also a long history of focusing on risk management. Droughts, floods, isolation, the “tyranny of distance” all encouraged such an outlook. The “Deakinite” system of trade protection, wage arbitration, state paternalism, white Australia and imperial benevolence which dominated public policy from a few years after Federation to the early 1980s was a giant risk management system. It did not work terribly well, Australian per capita income fell in relative standing from the highest in the world to around 20th or so.

    White Australia was abandoned in 1967-1972. Change occurred in economy policy because policy-makers became convinced it was reducing Australia’s ability to manage risk, not improve it. Subsequent events have proved the reformers correct. Despite somewhat hysterical claims in some quarters, the reform program was never about abolishing the welfare state, but making it sustainable.

    But that utilitarian/risk management concern has become very focused on health and safety issues pertinent to urban life–Australia is a very highly urbanised society (89% life in urban areas).

  32. Gravatar of Lorenzo from Oz Lorenzo from Oz
    20. March 2013 at 12:18

    That should be “89% of Australians live in urban areas”.

  33. Gravatar of ssumner ssumner
    20. March 2013 at 19:19

    Laurent, I don’t forecast inflation, but I believe our long run forecasts are more accurate now than in the 1970s.

    J, Price stickiness was well understood long before the Keynesians.

    Adam, Yes, it’s possible that things are actually deterministic, but appear random.

    And very convenient that physics fields are only considered “physics” when they deal with real easy problems and hence get accurate forecasts. Maybe economists should do the same!

    Neal, That’s just silly. The EMH says asset prices are random. It would be a horrible black mark for economic theory if we WERE able to predict prices! You are comparing apples and oranges.

    Josh, I’ve already been to those two cities.

    mbk, Australia was quite cheap in 1991.

    Lorenzo, Thanks for that info.

  34. Gravatar of Geoff Geoff
    20. March 2013 at 19:34

    Everything macro is micro caused.

    Nothing micro is macro caused.

    You do the math.

  35. Gravatar of flow5 flow5
    21. March 2013 at 04:03

    Scientists collect data from which to make forecasts. Economists throw data away & capitulate.

  36. Gravatar of TallDave TallDave
    21. March 2013 at 06:54

    Adam, Yes, it’s possible that things are actually deterministic, but appear random.

    Well, again, that would require hidden variables (i.e. “something, that if we knew it, would allow us to predict outcomes that appear random”), which goes back to the original point several of us made: almost everyone agrees hidden variables have been ruled out (this was the original Copenhagen interpretation, defined in opposition to the Einstein-championed view that there must be hidden variables; later MWI came along as a way to deal with the inelegant concept of wavefunction collapse).

    I think it’s fair to say that resolving the EPR paradox with nonlocal effects (“spooky action at a distance”) rather than hidden variables was a very counter-intuitive result, one that a lot of people never liked, but unfortunately it’s been tested out to 242 standard deviations so today virtually everyone accepts it.

  37. Gravatar of ssumner ssumner
    23. March 2013 at 07:28

    TallDave, I believe if you go back a few weeks and read the comment section from a QM post I did, you will find that you are mistaken. There is a lot of dispute about the Cpenhagen interpretation. Highly knowledgeable commenters who are physicists told me that an increasing number of physicists accept the MWI. I have no reason to doubt that.

    I just read a book on QM that insisted the multiverse is deterministic. I’m not saying that’s definitely right, but it’s certainly one valid point of view.

  38. Gravatar of TallDave TallDave
    24. March 2013 at 19:10

    Scott — yes, I agree MWI is quite mainstream, I’m just pointing out MWI accepts random outcomes in our universe.

    As far as I know there is no theory of MWI that would allow you to predict (in theory), say, which atom in a hunk of cesium-137 is going to decay first. MWI just says there are universes for all the possible outcomes — that is, one could find a universe where any of them decayed first, but one could never say beforehand “this is the universe in which this particular atom will decay first.”

Leave a Reply