The 4% and the real problems

I’ve been arguing that while inequality is a real problem, it is far down the list of serious problems faced by this country.  Certainly behind unemployment, poverty, and the war on drugs and probably behind global warming, access to health care, and government waste.  According to the Boston Globe, the public seems to agree:

THOUGH PRESIDENT Obama insists that income inequality is the “defining challenge of our time,” most Americans beg to differ.

“What do you think is the most important problem facing this country today?” asked Gallup in a nationwide survey this month. Dissatisfaction with the federal government — its incompetence, abuse, dysfunction, venality — topped the list, with 21 percent of respondents saying it was their key concern. The overall state of the economy was second, at 18 percent. Unemployment and health care were tied for third, with each cited by 16 percent as the nation’s most pressing problem.

How many shared Obama’s view that the gap between rich and poor is the issue that should concern us most? Four percent.

And even that 4% figure is misleading.  If you take away all the economics bloggers blathering on about inequality, it would probably be closer to 3%.

BTW, Free Exchange has a wonderful post by G.I. (Greg Ip?) on the myth that microeconomics is more scientific than macroeconomics.  The post also contains some excellent analysis of the minimum wage debate.  

The people who claim micro is more scientific are the same sorts of people who claim physics is more scientific than economics.  Economists can predict business cycles more accurately than applied physicists can predict the stuff we really care about, like weather and earthquakes and tsunamis.  The retort is that applied physicists are good at predicting stuff we don’t care about, like the orbits of Jupiter’s moons.

HT:  Joel Lidz

Update.  Some commenters pointed out that these problems are interrelated, at least to some extent. Here’s my response:

Sure, all problems are at least somewhat correlated and interrelated. But policies like a higher minimum wage that (supposedly) reduce inequality also (supposedly) increase unemployment. We can address inequality much more effectively by direct subsidies like the EITC, although I agree that ending the War on Drug Using Americans would indirectly reduce inequality. Think about the specific problem you are trying to address, and address that problem in a way that doesn’t worsen other problems.

Transferring money from Bill Gates to the middle class doesn’t reduce poverty, it makes it worse. If you had not taken the funds from Gates and given it to the middle class, he was planning on donating the money to reduce poverty in places like Africa.


Tags:

 
 
 

46 Responses to “The 4% and the real problems”

  1. Gravatar of Morgan Warstler Morgan Warstler
    29. January 2014 at 06:19

    That micro guys argue over marginal increases in MW is a yawner.

    NO ONE argues massive price cuts (like GI/CYB) don’t liquidate.

    100% of micro guys agree.

    And Ip should say so.

  2. Gravatar of Chris Ferguson Chris Ferguson
    29. January 2014 at 06:31

    Isn’t inequality correlated with unemployment and poverty ? The former could be a stand-in for the latter two (though a much more loaded stand-on). A decrease in unemployment and poverty would help solve inequality.

  3. Gravatar of Phil Phil
    29. January 2014 at 06:36

    half of the things you talk about being more important than inequality are drivers of or results of inequality. Throw in the war on drugs, and you’ve got 2/3.

  4. Gravatar of David R. Henderson David R. Henderson
    29. January 2014 at 06:45

    Morgan Warstler,
    What’s GI/CYB? I assume GI is guaranteed income. What is CYB?

  5. Gravatar of ssumner ssumner
    29. January 2014 at 06:53

    Chris and Phil, Sure, all problems are at least somewhat correlated and interrelated. But policies like a higher minimum wage that (supposedly) reduce inequality also (supposedly) increase unemployment. We can address inequality much more effectively by direct subsidies like the EITC, although I agree that ending the War on Drug Using Americans would indirectly reduce inequality. Think about the specific problem you are trying to address, and address that problem in a way that doesn’t worsen other problems.

    Transferring money from Bill Gates to the middle class doesn’t reduce poverty, it makes it worse. If you had not taken the funds from Gates and given it to the middle class, he was planning on donating it to reduce poverty in places like Africa.

  6. Gravatar of Morgan Warstler Morgan Warstler
    29. January 2014 at 07:20

    David, so glad you asked!

    http://www.morganwarstler.com/post/44789487956/guaranteed-income-choose-your-boss-the-market-based

    —-

    Actually Scott, “transferring” the money doesn’t work, BUT tax laws that valorize SMB owners, encoded smallness (which recognizes Big Govt. is an unmeasured negative externality of Big Business), shifts money within the the top 20%, creating more buyers of yoga lessons, sous chefs, dog walkers, etc.

    The bottom 2/3 (the C Power) who never make it into the top 20% of earners (A power = top 1/3 who do), would do best from themselves if they rotuinely sided with the A power, and not the B power (the oligarchs)…

    Just as China played US and USSR against one another.

    If you don’t even know how many players there are in the game, you will always lose.

    It’s pretty simple, the Dems ought to offer the Tea Party a deal guts oligarchs and gives all gains to Tea Party (SMB owners).

    More than anything else, this would improve the lot of the 80% who will work in the service side of the economy.

  7. Gravatar of mbka mbka
    29. January 2014 at 07:21

    Scott:

    “The people who claim micro is more scientific are the same sorts of people who claim physics is more scientific than economics. Economists can predict business cycles more accurately than applied physicists can predict the stuff we really care about, like weather and earthquakes and tsunamis. The retort is that applied physicists are good at predicting stuff we don’t care about, like the orbits of Jupiter’s moons.”

    Priceless.

  8. Gravatar of Kevin Erdmann Kevin Erdmann
    29. January 2014 at 07:41

    There is a wide range of income variance throughout societies and history that were acceptable to the societies that experienced them. Inequality could be half the scale that it is or twice the scale that it is, and the same supposed solutions would be proposed.

    http://idiosyncraticwhisk.blogspot.com/2014/01/policies-are-for-identifying-outsiders.html

  9. Gravatar of John Becker John Becker
    29. January 2014 at 08:22

    My idea for a direct subsidy is to get rid of everything else (tax and welfare wise) and have people making less than 75-125K get a bonus of $10,000 each year they work full-time. The cap changes each year so that people don’t structure their incomes to qualify. That’s what I came up with when I tried to think of a program that would help working class people without creating disincentives to work. To make the program most effective, make it super easy to get a job- no minimum wages, no anti-discrimination laws, employers can hire or fire for any reason that suits them, etc.

  10. Gravatar of MarkySparky MarkySparky
    29. January 2014 at 08:33

    Physicists can predict the results of subatomic particle measurements down to dozens of decimal places. Nobel prizes are given for moving that decimal place one more place to the left. There are not two physicists in the world with prestige exceeding zero that substantially disagree about the laws of Thermodynamics.

    Yet the content of this blog demonstrates quite regularly that even august economists is positions of great power disagree about basic things, like whether monetary policy *does anything*. This is trolling of the highest order to suggest that economics of any stripe is even remotely in the same league as physics, at least in terms of scientific rigor.

    There is, of course, a value judgement involved in assessing the relevance of each field to our lives. I would suggest that accurate prediction of business cycles is roughly as important to humanity as the gross observation of the regularity of sun/moon cycles. As for the orbits of solar system bodies, discrepancies of less than 1% error in Neptune’s orbit correctly predicted the existence of Pluto. I’ll be watching this space for a prediction about the price of 1/2″x3″ Grade 8 bolts at the Monument, CO Home Depot on June 4th, 2074…

    You’re better than this, Scott.

  11. Gravatar of Mark A. Sadowski Mark A. Sadowski
    29. January 2014 at 08:51

    “The people who claim micro is more scientific are the same sorts of people who claim physics is more scientific than economics.”

    A couple of observations:

    1) A lot of the impetus for such claims comes from the fact most people are very confused as to who is actually an economist. Lots of people working in finance claim to be economists, or to have expertise in economics, but when you actually check their transcripts there’s nothing there but a giant void. (“I never went to Med School, but I did stay at a Holiday Inn one night.”)

    Take John Tamny of Real Clear Economics and Forbes for example. Until I started raising a big stink about it, his title was “Chief Economist” at H.C. Wainwright Economics, with an area of specialty in macroeconomics. (He’s now called an “Senior Economic Adviser”.) The problem was that the only course in economics Tamny has ever taken was a one semester survey of micro and macro as part of his MBA which spent a total of six weeks covering macroeconomics. This would be like somebody calling themself a physicist with a specialty in quantum mechanics after only taking physics 101.

    2) I hear the “economics is not a hard science, it’s a social science” taunt a lot. It’s not clear what they mean by “hard science” but it often seems to imply “quantitative”. And if that’s the criteria then economics is certainly more of a “hard science” than either biology or chemistry, and there’s virtually no difference between economics and physics.

    How do I know? Although I don’t have any graduate degrees in mathematics (I have completed 35 graduate hours in mathematics, and have multiple undergraduate degrees in mathematics, and am certified to teach mathematics at the secondary level), I have taught science and engineering track calculus at the university level. (I also have a minor in physics.) Thus I can tell you first hand how much mathematics that majors in various fields need to know, and mathematicians, physicists, economists and engineers are on a separate plane from other academic and professional fields in that regard.

    3) Anyone who thinks physicists approach their subject in a highly dispassionate and objective manner would have been cured of that delusion simply by reading the comment section of physics blogs (yes, they exist) during the Higgs-Boson discovery announcement. Economists, and macroeconomists in particular, hold no monopoly on academic pissing contests and immature behavior more generally.

  12. Gravatar of Morgan Warstler Morgan Warstler
    29. January 2014 at 08:55

    John, glad to have you aboard!

    http://www.morganwarstler.com/post/44789487956/guaranteed-income-choose-your-boss-the-market-based

  13. Gravatar of John Becker John Becker
    29. January 2014 at 08:58

    Mark Sandowski,

    I agree with you completely that physicists are highly biased towards their favorite theories. String theorists are especially guilty of this since it is impossible to test many of their predictions and the model is sort of “pretty.”

    Do you think that the mathematics in economics is useful? It seems to me that there are no equalities in people’s action. A transactions is always the result of inequality. In addition, if you can accomplish the same things through verbal reasoning, isn’t putting math into the mix a complete violation of Occam’s razor? As David Friedman said, the benefit of being a theoretical physics PhD is that I can do verbal economics without people accusing me of just being bad at math.

  14. Gravatar of Doug M Doug M
    29. January 2014 at 10:02

    A physicist at least has the humility to say that they can’t predict an earthquake, or the weather beyond next week. I seldom see that sort of humility out of economists.

    Quantification does not make an area of study a science.

    A science should produce testable hypotheses. And Macro generally comes up short.

    Physics produces its share of theories that, at least when they are put forth, appear to be un-testable. And, half of the physics community wants to excommunicate the string theorists for being “not even wrong.” But, most of the time the physicists have found something some way to test the crazier implications of their theories.

    I don’t see the economists coming up with good experimental models, and hence they continue to debate the whether the Keynesian multiplier exists nearly a century after it was proposed, even though it is just about the first thing taught to new students.

  15. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    29. January 2014 at 10:10

    Here are four economists who think more inequality will be good for preventing a future financial crisis;

    http://www.voxeu.org/article/inequality-and-household-debt-new-evidence

    ‘… the growth in household borrowing during the mid-2000s was driven in large part by credit supply expansions targeted at lower-income households.

    ‘However, to the extent that this expansion in the supply of credit to lower-income households is unlikely to continue (for example if it reflected a one-time securitisation of household debt), our results suggest that a continuation of recent trends toward rising inequality is likely to reduce access to credit for lower-income households.’

    Though I’m not sure they realize the implications of their conclusions.

  16. Gravatar of Nick Nick
    29. January 2014 at 10:17

    There is often a lot of confusion about what science is. It has pretty much degenerated at this point to if you use a lot of math, you are doing science, and if you don’t, you aren’t. But that is entirely incorrect.

    Science is one method by which we can choose to believe in certain universal truths instead of others despite the specific and contingent knowledge from our sensory experience. Specifically, it is the method that formulates a hypothesis, designs and experiment that could disprove that hypothesis, performs the experiment, and reviews the experimental data and develops a conclusion.

    If a field isn’t designing and then running experiments, it isn’t doing science! Without experimentation, all you are doing is writing narratives, even if you use lots of interesting mathematics. Those might be very convincing narratives, but narratives are a very different path to universal truth than the scientific method is.

    But since the truths that the scientific method reveals are so powerful, so robust, and so reproducible (and often even useful!), it should come as no surprise that everybody wants to claim that what they do is “science.”

  17. Gravatar of john t john t
    29. January 2014 at 10:26

    here’s a challenge nassim taleb posed noah smith on twitter re macroeconomics being science: name a single result in econ that is robust to assumptions.
    best noah came up with was okun’s law. once country from 47 to 02 and its more a rule of thumb than a law.

  18. Gravatar of Mark A. Sadowski Mark A. Sadowski
    29. January 2014 at 10:33

    John Becker,
    “Do you think that the mathematics in economics is useful?”

    You’re definitely asking the wrong person, as I’m enormously biased. If there was no mathematics in economics it wouldn’t hold any interest for me at all.

    Try asking the same question about physics. Economics without math would be like physics without math, and just as totally useless. It would be the province of armchair philosophers.

  19. Gravatar of Alex Godofsky Alex Godofsky
    29. January 2014 at 10:33

    Physicists have a pretty good track record on predicting the behavior of doped semiconductors, which I think quite a few people care about quite a lot.

  20. Gravatar of john t john t
    29. January 2014 at 10:50

    isn’t engineering applied physics? seems quite useful

  21. Gravatar of Tom Brown Tom Brown
    29. January 2014 at 11:19

    Mark,

    “Take John Tamny of Real Clear Economics and Forbes for example.”

    I’m happy to hear you raised a stink about him. I have exactly 0 hours of economic schooling, and yet even I could see that one of his Forbes pieces was truly ridiculous. I’m sure Tamny can run circles around me in a lot of ways, but I was really shocked to see that particular article in Forbes. It’s more a comment on Forbes perhaps that they printed it.

    Re: science and math: sometimes I think that in econ we may be partly fooling ourselves because there are so many numbers floating around in the field. They’re there because they are readily available: humans love to quantify things and make records of it. Here’s my thought experiment on this: Can you explain how the economics of a population of chimps could possibly be any MORE complex than our own economy? Yet in many respects it appears to be more difficult to study a population of chimps (there’s not as many numbers readily available)… but at the root of it there are decisions being made about resource usage by those chimps. My gut tells me the chimp situation can’t possibly be more complex than our own, and in all likelihood it’s actually simpler. Their brains are smaller! How could it not be a simpler problem? I’d love to get your feedback on that.

  22. Gravatar of Tom Brown Tom Brown
    29. January 2014 at 11:24

    … also, legally, it’s probably easier to run experiments on chimp populations than human populations, so that ought to make chimp econ easier as well.

  23. Gravatar of LK Beland LK Beland
    29. January 2014 at 11:50

    “Economists can predict business cycles more accurately than applied physicists can predict the stuff we really care about, like weather”

    Practically all applied physicists use similar frameworks and hypotheses to model the weather. Most agree that with more computational power, we could extract more predictions of great interest to the public. We are getting better at it every year.

    The same cannot be said for macroeconomics. Almost every founding principle is up to debate. Even with computational power 1000000 times that of today, I doubt that the macroeconomics community would be better at predicting business cycles than today.

  24. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    29. January 2014 at 12:07

    ‘…legally, it’s probably easier to run experiments on chimp populations than human populations….’

    Not according to Barack Obama. You just need to issue an executive order.

  25. Gravatar of ssumner ssumner
    29. January 2014 at 12:11

    Thanks mbka.

    Kevin, Sounds right.

    John Becker, I like the idea of a subsidy per hour worked, a reverse payroll tax.

    markySparky, You are comparing apples and oranges. When looking at problems of equal complexity both economics and physics do equally well (or poorly).

    Mark. Very good points.

    Doug, You said;

    “A physicist at least has the humility to say that they can’t predict an earthquake, or the weather beyond next week. I seldom see that sort of humility out of economists.”

    Very good point. I don’t agree with your other points. Economics has come up with extremely counterintuitive testable hypotheses that have been confirmed. For instance, we predict that you should not let people with a “proven track record of success” manage your stock portfolio. I’ll bet most physicists think you should let them do so.

    Patrick, Good point.

    Nick, The original definition of science was something like “any field of inquiry aimed at better understanding the world around us.” Later people added math, experiments, refutability, and other concepts unrelated to the original definition. Now the term means different things to different people. So people simply talk past each other.

    Alex, Of course they do, however economists can tell you whether North or South Korea has a better economic policy regime, which is something else that people care about quite a bit.

    LK, It’s possible that weather forecasting will improve faster than economic forecasting. After all, economic theory predicts that economists will be unable to predict asset prices, but physics theory does not predict that physicists will be unable to predict weather. But the EMH is itself a very useful theory.

  26. Gravatar of Mark A. Sadowski Mark A. Sadowski
    29. January 2014 at 12:16

    Tom Brown,
    Economists like Vernon Smith already conduct economic experiments using human subjects, so it’s not unheard of:

    https://en.wikipedia.org/wiki/Experimental_economics

    The problem with studying the economics of chimpanzees is that their economic behavior may not be the same as that of humans. See this for example:

    http://www.sciencedaily.com/releases/2007/10/071005104104.htm

  27. Gravatar of Tom Brown Tom Brown
    29. January 2014 at 12:25

    Mark, Thanks for the reference and your thoughts. I fully expect that chimp econ is different than human econ: I just mean in terms of getting a better handle on one “branch” (or “species” Ha!) of econ.

    I suppose you could argue that because humans quantify and keep records, it truly does make human econ easier: we’re better at communicating and recording what’s going on in our heads. But I have serious doubts that this effect isn’t more than offset by the complexity of what’s going on in our heads, and the feedback effect I already discussed.

  28. Gravatar of Tom Brown Tom Brown
    29. January 2014 at 12:36

    BTW, that was an interesting econ experiment on the chimps. Thanks.

  29. Gravatar of Kevin Erdmann Kevin Erdmann
    29. January 2014 at 12:40

    “Economics has come up with extremely counterintuitive testable hypotheses that have been confirmed. For instance, we predict that you should not let people with a “proven track record of success” manage your stock portfolio.”

    Uh oh. I think you might have chosen poorly. There’s a ceteris paribus problem there. If someone had a proven track record of success with $100 million in assets, and you and $10 billion worth of other clients want a share of that success, you’re basically pulling the price of the fund into equilibrium through diseconomies of scale.

    Both of these links are somewhat lengthy, but:

    Here’s a study that showed consistent patterns of performance in truly active funds:
    http://www.cfapubs.org/doi/abs/10.2469/faj.v69.n4.7

    I wrote about it here:
    http://idiosyncraticwhisk.blogspot.com/2013/11/risk-premiums-reputation-and-actively.html

    But that’s the problem with comparing economics and physics. Physicists don’t usually have to control for the agency of their subjects.

  30. Gravatar of Tom Brown Tom Brown
    29. January 2014 at 12:57

    Reductionist fantasy thought experiment: for the sake of argument assume this truly is a mechanical universe and that re-creating the workings of a human brain turns out to be relatively easy (a lot easier than recreating the workings of a global weather system for example) and this is done through the usual reductionist approach: a full simulation is built up through simulating the activity of each neuron. An this simulation is indistinguishable from a real human. Assume further that large groups of humans can be simulated in a simulated environment (“The Matrix”, so to speak).

    Ignoring ethical considerations, would this be a valuable tool for studying economics and macro in particular? Why or why not?

  31. Gravatar of Lorenzo from Oz Lorenzo from Oz
    29. January 2014 at 13:44

    If you want to reduce household/family inequality, reverse feminism. High income women marrying high income men (i.e. “assortative mating” when women have similar income opportunities to men) has driven up household/family inequality since 1970.

    Inequality among individuals has barely changed:
    http://politicalcalculations.blogspot.com.au/2014/01/in-which-were-vindicated-again.html#

    Equality before the law matters. Anything else is better looked at in terms of poverty and opportunities.

  32. Gravatar of errorr errorr
    29. January 2014 at 14:35

    I would say macro isn’t that much different from statistical mechanics or other applied physics like weather prediction.

    The problems are on different levels. To construct a model in either physics or mechanics requires both data to define the initial variables and then basic rules on how the individual variables need to evolve.

    Physics has the problem that ascertaining initial variables in any system is by definition impossible, even good estimates are questionalble.

    Economics has much the opposite problem. Initialization would be easier as the data is inherently discoverable, but the rules of interaction are much harder to determine. Micro foundations in Macro were part of the attempt to better define how a system evolves through interactions among its constituants. However, I would guess utility functions are oversimplified. Rational expectations is so alluring because it simplifies this problem to a level of seeming irrelevance even though there are too many instances where this assumption is ridiculous. Although I would concede that some day there could exist a utility function that estimates the probability space of various human economic choices.

    There is an irony in Scott’s comment that we can locate a moon in a specific place. Physics can estimate a moon in space fairly accurately in a short time frame. However, the system is inherently chaotic and this estimates are limited by the accuracy of knowing the initial conditions. The beginning of chaos theory started in realization that the 3 body problem has no inherent equilibrium. I find it ironic that Scott used the example of locating a planetary body in space to show what physics was good at predicting when that is the precise example I would use to show the inherent limitations of physics in predicting things.

    They discovered the Higgs Boson only when the data exceeded the arbitrary threshold of 5 standard deviations that a committee decided was required to be “certain” of there claim. That isn’t discovery as Much as being extremely likely.

  33. Gravatar of MarkySparky MarkySparky
    29. January 2014 at 15:46

    “markySparky, You are comparing apples and oranges. When looking at problems of equal complexity both economics and physics do equally well (or poorly).”

    That caveat of “equal complexity” is a real minefield… What problem in economics has the equivalent complexity of weather prediction, and to what level of accuracy? I’m honestly curious.

  34. Gravatar of Vivian Darkbloom Vivian Darkbloom
    29. January 2014 at 16:20

    “When looking at problems of equal complexity both economics and physics do equally well (or poorly).”

    I think this echoes something that Scott has written before and I would agree (not being a physicist or economist) that problems of economics are as difficult or more difficult to solve than those of physics. In fact, I would say that economics is likely much more difficult when it comes to “solving problems”.

    But, I’m curious as to what “doing equally well” means in this context. If a physicist were to accidentally blow up the world or destroy the economy in trying to solve a difficult problem, I would not exactly say that they are “doing well”.

    The problem is not that economics is not difficult, it is that economists often do not themselves recognize the difficulties and therefore the limits of their own knowledge or more likely are simply loathe to admit it. There seems to be a real disconnect between the message that “this is really very, very difficult, almost impossible stuff” and “I’m pretty sure I know the answer to this, or at least I know it better than you do”. I’m not sure you can have that both ways.

    It is in this respect that perhaps economists are doing rather poorly compared with, say, physicists who seem as a group to be pretty upfront about what they know and don’t know. Besides, because most economic problems are so closely associated with politics, it is not a very winning political strategy to admit your proposed economic policy is just a shot in the dark. When politics enters areas of traditional “science”, e.g. climate change, one can see the effect that has.

  35. Gravatar of Major_Freedom Major_Freedom
    29. January 2014 at 17:01

    Economists can predict business cycles more accurately than applied physicists can predict the stuff we really care about, like weather and earthquakes and tsunamis.

    Pardon my language, as I am trying to say what deserves the most courteous way it can be said, but that comment is BS.

    The track record of economists predicting business cycles is abysmal. Climate scientists are far more reliable and dependable than an economists predicting the economy.

    Science cannot predict the economy. Humans learn in a priori unpredictable ways, because we can’t learn what we will learn now, before we learn it. And since knowledge affects our actions, it means our actions, and thus business cycles, are scientifically unpredictable.

  36. Gravatar of ssumner ssumner
    29. January 2014 at 17:52

    Kevin, You said;

    “Uh oh. I think you might have chosen poorly. There’s a ceteris paribus problem there. If someone had a proven track record of success with $100 million in assets, and you and $10 billion worth of other clients want a share of that success, you’re basically pulling the price of the fund into equilibrium through diseconomies of scale.”

    It’s even worse than that. Success in stock picking is almost all luck, so there is little serial correlation. So people are better off with index funds. That’s extremely counterintuitive. Do you know any other profession where past performance is not a predictor of future success?

    Tom, I honestly don’t know. It seems to me that the complexity of the human brain isn’t the real problem. It doesn’t explain why we disagree about the minimum wage, for instance. We disagree because we can’t decide whether labor markets are competitive or monopsonistic.

    Lornenzo, I was thinking about the same thing. Could this explain the high equality and upward mobility in Utah (which presumably still has lots of traditional values?)

    errorr, Interesting. BTW, there is a lot of similarity between currency crises and earthquakes. Currencies are held fixed for long periods of time, imbalances build up, then a sudden sharp devaluation. Continental plates try to slide past each other but friction stops it until enough pressure builds up. Then a sudden adjustment.

    Markysparky, There are no perfect analogies because there is nothing like the expectations issue in physics. But business cycle analysis is a bit like climate forecasts. We know GDP and global temperatures are likely to grow over time, but have trouble predicting year to year fluctuations. But we have at least a tiny bit of ability to predict year to year fluctuations (huge oil shocks in macro, Mt Pinatubo in climate.)

    Vivian, I mostly agree, but physicists are less impartial than you might think. But yes, economists are worse.

  37. Gravatar of Saturos Saturos
    29. January 2014 at 18:37

    YES, THE HIERARCHY OF SCIENCES IS REAL: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0010068

    Scott, I hope you don’t have such a narrow view of “prediction” as to think that the only physics predictions that are of use to us are the weather and climate? That might be their “macro”: but their “micro” underlies basically all engineering, which requires predicting, conditionally and unconditionally, the behaviour of systems. Physics is the largest and most successful science; as reductionism leads to the unification of science that unification is expressed under the umbrella of physics.

  38. Gravatar of Saturos Saturos
    29. January 2014 at 18:39

    The point of Noah’s posts is to show that, although in some ways economists may be said to have a harder job, they are not in fact doing their job (as an academic community) as well as physicists do theirs. And perhaps the amount of controversy in blog comments isn’t the best metric of that.

  39. Gravatar of Lawrence D’Anna Lawrence D'Anna
    29. January 2014 at 18:57

    Being able to make useful predictions isn’t what makes a science scientific. How many useful predictions does palaeontology or astronomy or QCD make?

    Here’s the definition I like: A science is something that can produce extremely convincing, universal, empirical evidence. Meaning:

    extremely convincing: It can convince you to adopt subjective probabilities very close to 0 or 1.

    universal: It should be able to convince anyone who is reasonably rational and is starting with reasonable priors.

    evidence: It’s a valid argument, not an exploit of bias or irrationality.

    Economics does not have this quality, at least not compared physics. Economics is full of people claiming to know things with high degrees of certainty and other people claiming to know the exact opposite with the same certainty. Usually it is very difficult for an outsider to tell who, if anyone, is right. Economics is far, *far* less scientific than physics.

  40. Gravatar of Lorenzo from Oz Lorenzo from Oz
    29. January 2014 at 19:17

    Lawrence D’Ana: how much is that about economics and how much economists? Which is to say, economics is a much more morally loaded discipline than physics.

    Scott: I would call it a good working hypothesis. High levels of congruent preferences/clear signals/trust might also be a factor.

  41. Gravatar of Jason Jason
    29. January 2014 at 22:15

    “Economists can predict business cycles more accurately than applied physicists can predict the stuff we really care about, like weather and earthquakes and tsunamis.”

    1. Ha! No they can’t.

    2. Those are earth sciences, not physics. Sure, there is some physics involved, but it’s a bit like saying economists can’t predict revolutions or elections.

  42. Gravatar of Benny Lava Benny Lava
    30. January 2014 at 07:57

    Talk about insecure. Comparing economics to physics is a complete joke. The Nobel in physics recently went to a guy who invented a process to make the lossless fiber optic cables that currently make high speed internet possible. I guess people don’t care about that because economics brought us…what again? What tangible benefit has economics brought anyone? Nothing. That is the retort. That people are ignorant of the gains of physics but econ brings us nothing. As useful to society as women’s studies.

  43. Gravatar of Floccina Floccina
    30. January 2014 at 11:49

    @john t

    isn’t engineering applied physics? seems quite useful

    Isn’t accounting applied economics? Seems quite useful.

    Every economist believes that if you raise the minimum wage to $50/hour it will increase unemployment. I would go as far as to say that all economists would agree that if the delta in total wages paid after a MW increase is greater that total profits in an economy that it will increase unemployment.

  44. Gravatar of J Mann J Mann
    31. January 2014 at 05:49

    My late three cents:

    1) The “is X a science” question breaks down to your definition of science.

    1.1) Popperian definitions are very popular, but IIRC Friedman and IIUC Sumner are more interested in whether a discipline can make useful predictions.

    1.2) Given that words are nothing more than the skins worn by living thoughts and all, I think both defitions have their place. Certainly, when you see an op ed by two doofuses arguing that the Fed chair should not have a governing theory because economics is not a science, you’ve gone too far. (And my guess is that about 1 in 3 people who claim economics doesn’t count because its not a science still believe that papers showing that the minimum wage doesn’t lower employment provide useful information for predicting the future, not that it’s the output of a random number generator).

    2) Scott’s point is that there are levels of complexity at which both physics and economics can make useful predictions. Certainly, I think that Randal Monroe’s predictions about how many lasers it would take to shift the moon’s orbit are suggestive of the real answer, although it wouldn’t suprise me if we tried it and found out it was wrong.

  45. Gravatar of ssumner ssumner
    31. January 2014 at 15:27

    Saturos, The question of “success” is complex. What is the counterfactual? What counts as economics? Is the vast difference between North and South Korea due to economics or physics? That seems like an important issue. Is India’s poverty due to bad economics? That seems like an important issue?

    Lawrence, I don’t like that definition of science. I like the original definition, something like “a field that tries to explain the world around us.”

    When people claim physics is better than economics it sounds to me like someone claiming accounting is better than poetry because it is more precise or useful or something like that. You can’t compare accounting and poetry and you can’t compare economics and physics, they do radically different things.

    BTW, I am very critical of economics in many ways (especially monetary economics.) I think economists caused the Great Recession. So this isn’t a sort of knee jerk defense. I have no problem with the abstract idea of economics being a trivial unimportant field. The world is what it is. It’s that’s true, I’m fine with that. I just don’t like people adopting these cliches without thinking.
    Maybe chemistry is the most important field—the green revolution, pharma and biotech are all related to biochemistry. Why isn’t that more important than physics?

    Jason, If physicists don’t call those fields applied physics, is it because they are embarrassed? Aren’t the laws of physics supposed to explain those things. Indeed why shouldn’t the laws of physics explain stock market crashes? Stock markets are composed of atoms that should be following the laws of physics. They’d say it’s too complicated. But economics is also complicated! You can’t use the same criterion for accuracy as with the orbit of the moon around the Earth, a simple system.

    J Mann, Good comment, but actually I don’t think something needs to make useful predictions to be a science. I regard history as a science.

    Econ can’t go much beyond conditional predictions.

  46. Gravatar of Tom Brown Tom Brown
    1. February 2014 at 15:24

    Saturos, you and I are on the same page regarding physics and science and reductionism. Only you express it a lot better than I do. Thanks.

Leave a Reply