There’s only one way to test the EMH

Before beginning, I like to point out that I have a new essay over at The Economist.

Commenters often present me with market anomalies, which supposedly “prove” the efficient markets hypothesis is wrong.    I always respond that they’re just engaging in data mining.  They retort that no theory that can’t be disproved is worth anything.  But the EMH can be disproved.  The tests have been done, and it passes.  Sort of.

Finance professors have done many different tests of the EMH, and I’d guess 90% of the published tests (but only 5% of the actual tests) show that the EMH is wrong.  (Yes, I’m pulling these numbers out of thin air, but you get the point.)  They’ve found January effects, small stock effects and value stock effects.  They’ve found the market does better when P/E ratios are low.  They found the market does worse on rainy days (a study published in the AER!).  Of course you’d expect to find 5 anomalies for every 100 tests you do, and for the most part you only get published if you find an anomaly, and finance professors have a lot of computer power, so . . .

Here’s my analogy.  Suppose Stephen Wynn was concerned that some mysterious gamblers were getting away with cheating at one of his casinos.  He’d heard rumors, but had no proof, or even suspects.  You are statistician brought in to investigate.  You study 600 slot machines, and find 30 of them produced three cherries more often than you’d expect from mere chance.  You suggest that the anomalous slot machines be destroyed.  How should the casino owner react to that “investigation?”  I’m guessing Wynn wouldn’t be impressed.

I think you see the problem.  The investigator assumed he could figure out how the cheating was occurring.  He assumed that certain individuals had spotted slot machines with predictable tendencies, and that he (the investigator) could also find those machines.  But it’s really hard to walk into a casino and walk out with lots of money.  So how likely is it that an academic statistician is going to know where to look for cheating?  How likely is it that they would know how it’s done?  Not very, and that’s why their statistical search for anomalous machines will almost certainly be fruitless.  Oh, the data mining will unearth a few suspicious machines, no doubt about that, but only the number predicted by chance.

So how would one determine whether people were cheating?  Start by being humble—admit that you probably don’t know how it’s done.  If someone did have a secret formula, they wouldn’t put it in a book called “How to cheat at casinos” for all the world to see.  Any system that works is almost certain to be kept secret.  It’s like alchemy.  If anyone did have a formula to turn lead into gold, they would keep it secret; as the formula’s value would plunge to zero the instant it was released.

To find out whether cheating is occurring you need to look at whether the winnings of gamblers are serially correlated.  Are those who win once, more likely to win next time.  That’s the proper test, indeed the only practical test, of whether people are cheating the casino.  And it’s also the only test of the EMH.  Don’t look for “the system,” the secret way to beat the stock market.  Look for whether other people have found it.  Look to see whether people who did better than average one year, tended to do better than average the next year.  Don’t look for market anomalies—look for evidence that other people have found market anomalies.

Of course the study has been done.  I recall that Fama and French found that mutual returns were approximately serially uncorrelated, but not exactly.  There appears to be a slight serial correlation among the very best funds (top 3%), but not enough to give the average investor any advantage.

And that’s what I would have expected.  The EMH is approximately true; indeed it’s almost impossible for me to imagine any other model of financial markets.  But it’s not precisely true, again, just as you’d expect.  After all, if the EMH were perfectly true then no one would have any incentive to estimate fundamental values.  We know people are imperfect and hence that any real world human institution, including markets, will be at least slightly imperfect.

A smart person like Eugene Fama should have been able to come up with both the EMH, and its limits, by just sitting in a room and thinking.  Much as David Hume got the QTM by imagining what would happen if everyone in England woke up one morning with twice as much gold in their purses.  Or Fisher’s theory of inflation and nominal interest rates.  Or Cassel’s purchasing power parity.  Or Friedman/Phelps’ natural rate hypothesis.  Or Muth and rational expectations.  Certain ideas are simply logical, and that’s why I have no doubt that despite all those economists on the left arguing the EMH has been discredited, it will still be taught in every top econ/finance grad program 100 years from now, whereas fiscal stimulus will be long gone from macro textbooks.

PS.  Why will fiscal stimulus be gone?  Because even Krugman admits it only makes sense at the zero bound.  And we are rushing headlong into a world of all electronic money–probably within 50 years.  There is no zero lower bound with electronic money, and hence the Taylor Rule is all you need.  Old Keynesian economics will vanish, leaving only new Keynesianism.


Tags:

 
 
 

83 Responses to “There’s only one way to test the EMH”

  1. Gravatar of JimP JimP
    28. September 2010 at 06:07

    http://blogs.ft.com/money-supply/2010/09/28/posen-the-case-for-doing-more/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed:+ft/money-supply+(Money+Supply)

  2. Gravatar of ssumner ssumner
    28. September 2010 at 06:40

    Thanks JimP, That’s encouraging.

  3. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    28. September 2010 at 06:40

    Ben Bernanke says better financial market models are needed:
    “Another issue that clearly needs more attention is the formation and propagation of asset price bubbles. Scholars did a great deal of work on bubbles after the collapse of the dot-com bubble a decade ago, much of it quite interesting, but the profession seems still quite far from consensus and from being able to provide useful advice to policymakers. Much of the literature at this point addresses how bubbles persist and expand in circumstances where we would generally think they should not, such as when all agents know of the existence of a bubble or when sophisticated arbitrageurs operate in a market. As it was put by my former colleague, Markus Brunnermeier, a scholar affiliated with the Bendheim center who has done important research on bubbles, “We do not have many convincing models that explain when and why bubbles start.” I would add that we also don’t know very much about how bubbles stop either, and better understanding this process–and its implications for the household, business, and financial sectors–would be very helpful in the design of monetary and regulatory policies.”

  4. Gravatar of Josh Josh
    28. September 2010 at 06:50

    Scott,

    You wrote, “Why will fiscal stimulus be gone? Because even Krugman admits it only makes sense at the zero bound. And we are rushing headlong into a world of all electronic money-probably within 50 years. There is no zero lower bound with electronic money, and hence the Taylor Rule is all you need. Old Keynesian economics will vanish, leaving only new Keynesianism.”

    Why does the zero bound disappear?

  5. Gravatar of mlb mlb
    28. September 2010 at 06:56

    If you believe EMH then you must also believe that macro investors who have consistently beaten the market have a better understanding of the economy & markets than the sum of all “conventional wisdom.” I can think of about 10-20 such investors (many of whom are not household names). The problem then is that by my estimate, 80% of these investors would completely disagree with your policy advice. I am inclined to believe your form of EMH, but I am also inclined to want to elevate to power the people who have proven they understand the economy best – and the only way to do that is to consistently beat the market.

  6. Gravatar of ssumner ssumner
    28. September 2010 at 07:04

    123, The paradox of all-anti-EMH models is that as soon as we can model a bubble, it will no long happen. Ignorance of what’s going on is a sine qua non of bubbles. So I doubt we’ll ever be able to design regulatory policies around bubble theories. We had just had a huge bubble in 2000, and so everyone was on the lookout for bubbles. Even so the regulators missed the housing bubble. If they missed that one, I doubt they’ll ever be able to spot bubbles better than the investment community.

    Josh, There would be no zero bound because you can have negative interest rates on electronic money balances.

    mlb, Actually, the EMH says that those investors who have consistently beaten the market are mostly lucky, although I would argue that a very few may have some insights the consensus lacks. But it’s almost impossible to prove that one way or another. Even the best investors (Buffett, etc) are often wrong.

  7. Gravatar of Joe Joe
    28. September 2010 at 07:38

    Professor Sumner,

    Your argument concerning things like low P/E, low P/Book, High Dividend Yield, is very good. Essentially, its easy to go back and notice all of this, but can you actually use this insight to beat the market decade after decade.

    But, have you read Buffet’s famous Columbia speech, the “The Super Investors of Graham and Doddville” http://www.tilsonfunds.com/superinvestors.html

    When it comes Graham’s followers, its not a bunch of dudes cheating the system. They’re all old and have been consistently beating the markets for decades using variations on a basic outline.

    Take Walter Schloss, a guy in his 90s who directly learned under Graham, I quote…

    “Although he stopped running others’ money in 2003-by his account, he averaged a 16% total return after fees during five decades as a stand-alone investment manager, versus 10% for the S&P 500”

    Thats an outstanding track record, how can that be anything but skill?

    Joe

  8. Gravatar of Silas Barta Silas Barta
    28. September 2010 at 07:44

    Josh, There would be no zero bound because you can have negative interest rates on electronic money balances.

    Which is why people would adapt and hold their savings some other way than in an electronic account that can arbitrarily decide to deduct from them when it thinks the economy god needs them to spend some more.

  9. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    28. September 2010 at 07:46

    Scott, you said:
    “So I doubt we’ll ever be able to design regulatory policies around bubble theories. We had just had a huge bubble in 2000, and so everyone was on the lookout for bubbles. Even so the regulators missed the housing bubble. If they missed that one, I doubt they’ll ever be able to spot bubbles better than the investment community.”

    I agree. In fact, regulators are busy designing new regulations that will create bubbles. Housing bubble was supported by Basel regulations that subsidized triple-A junk. Guess what? There is a new Basel III project, with a slightly different set of subsidies, but this time it will have even broader reach as it has support of all global policymakers.

  10. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    28. September 2010 at 07:48

    Scott, you said:
    “The paradox of all-anti-EMH models is that as soon as we can model a bubble, it will no long happen. Ignorance of what’s going on is a sine qua non of bubbles.”

    There are models of bubbles where all agents know of the existence of a bubble, yet bubble grows.

  11. Gravatar of Josh Josh
    28. September 2010 at 07:50

    Scott,

    I realize that it is possible to have negative interest rates on electronic balances. My question is regarding whether it is likely. For example, it is possible to have negative interest rates right now. Promise to remove any currency from circulation in December whose serial number ends in a random number or simply outlaw currency and lower the rate below zero. Why don’t we do this? Is it solely because it is politically unpopular? Would it not be politically unpopular to have negative interest rates with electronic balances?

    I don’t want to speculate as to the political nature of the problem, so let’s consider the economics.

    Even in a world without cash, individuals would still have accounts (like, say, money market mutual funds) that earn a return and provide transactions services. In other words, there is still a demand for money, but the return on transactions balances is no longer zero. The marginal rate of substitution between commodities stocks and transactions balances would be equal to the ratio of storage costs to the return on money. In other words, the relative costs of commodity stocks would fall relative to transactions balances. Wouldn’t this increase the demand for gold, silver, and other durable, storable commodities to be hoarded just the same as currency would have been?

  12. Gravatar of Indy Indy
    28. September 2010 at 08:31

    I’m not so sure about the electronic money no-zero-bound hypothesis. Maybe not “zero”, but at some point of negative interest you would make it profitable to do contango inventory storage arbitrage. Spread this around in hedged futures contracts for the proportional amount of all goods in the CPI-basket for which these contracts are available, and you’ve got a little profitable industry that will put some kind of limit underneath electronic-money accounts.

    Or have I made an error in logic somewhere? Or is the argument that, in general, the kind of theoretical negative-interest rates we’re talking about are too low to matter?

  13. Gravatar of Benjamin Cole Benjamin Cole
    28. September 2010 at 08:43

    Another fascinating post by Scott Sumner.

    Regarding EMH–the fact the EMH is roughly and largely true does not mean that individual investors cannot outperform the market.

    There is luck.

    There is also the practice of constantly searching the market for brief-lived anomalies to exploit. Keep most of your powder dry, then invest in “odd” moments or niches, or in situations in which legal complexities inhibit EMH.

    With the emergence of ETFs, I think momentum investing, by sector, may prove worthy. I ran some money this way a while, and it worked. It may no longer–and that makes my point.

    Anomalies get wiped out by EM. You have to seize the anomalies quickly.

  14. Gravatar of Nick Rowe Nick Rowe
    28. September 2010 at 09:03

    Indy: Buying inventory to store is investment. If you can do that costlessly, then the natural real rate of interest cannot fall below zero (with prices measured in that inventory good). The problem is that we cannot push the market rate below the natural rate. And the danger is that if the market rate is above the natural rate, nobody invests in real goods. Inventory investment is a solution to the problem, not a cause of the problem.

  15. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    28. September 2010 at 09:07

    Scott, Fama/French study you link to does not use serial correlation. Instead, it compares a simulated distribution of multiyear returns of no-skill mutual fund industry, and compares to the actual distribution of mutual fund returns.

    Serial correlation is not a good approach, as some profitable strategies have small losses 9 years out of ten, and have huge gains 1 year out of ten. On the other hand, some bad strategies have the opposite profile of returns, and their small gains are serially correlated.

  16. Gravatar of Philo Philo
    28. September 2010 at 09:52

    “Look to see whether people who did better than average one year, tended to do better than average the next year.” But there are *a lot* of people, and at most very few have found the secret to success in the stock market. So the serially correlated results of these few will be drowned in the ocean of uncorrelated results achieved by ordinary people.

    “There appears to be a slight serial correlation among the very best funds (top 3%) . . . .” I am worried that, just by chance, there should, over any finite time-period, be positive correlation for a few funds (counterbalanced by negative correlation for others). But the successful positive-correlation funds will thrive, while the unsuccessful ones will quickly fold up and so will not appear in the data set.

    “Certain ideas [in economics] are simply logical . . . .” Are you coming out as an apriorist, à la Mises?

  17. Gravatar of Josh Josh
    28. September 2010 at 09:53

    Nick,

    What if they are not investing in productive assets? The point that I made above is this:

    Even in a world without cash, individuals would still have accounts (like, say, money market mutual funds) that earn a return and provide transactions services. In other words, there is still a demand for money, but the return on transactions balances is no longer zero. The marginal rate of substitution between commodities stocks and transactions balances would be equal to the ratio of storage costs to the return on money. In other words, the relative costs of commodity stocks would fall relative to transactions balances. Wouldn’t this increase the demand for gold, silver, and other durable, storable commodities to be hoarded just the same as currency would have been?

    Am I wrong? Shouldn’t this have the same effect as an increase in currency demand? I have no idea if I am correct, but it would seem to me that the answer isn’t as simple as “without cash, the zero bound goes away.” (And this ignores political feasibility.)

  18. Gravatar of Joel Joel
    28. September 2010 at 10:22

    http://arxiv.org/PS_cache/arxiv/pdf/1002/1002.2284v2.pdf

    The link above proves that, as the title says, “Markets are efficient if and only if P = NP.”

    That suggests to me that the practical differences between the actual market and an efficient one, just like the practical differences between problems whose solution can be verified in polynomial time and whose solution can be arrived at in polynomial time, are important.

    While it’s true that a literal traveling salesperson can make a living traveling some slightly non-optimal route, some times inefficiency will have important consequences.

    I look forward to a future where economists learn the EMH in the way that physicists now study Newtonian dynamics: verifiably false, but a useful heuristic while it approximately holds true, and a worthwhile foundation on which to build an understanding of more advanced theories.

  19. Gravatar of Morgan Warstler Morgan Warstler
    28. September 2010 at 10:44

    Um, when all money is electronic:

    1. knowing the fixed amount of it to the dollar, euro, yuan is doable.
    2. who holds each unit of each is knowable.
    3. any meaningful currency is readily usable with any seller.
    4. exchange rates head towards friction-less.
    5. counterfeiting disappears.

    And suddenly there is NO LONGER MACRO.

    “Printing money” instantly devalues all other known units against other forms of currency…. and as such QE is seen EXACTLY for what it is a tax on noble savers….

    Forget negative interest, forget printing money, forget fiscal stimulus.

    Governments become very competitive based solely on policy…. losers adapt quickly or get gutted, and the whole world bends towards anarcho-capitalism.

    I actually think we’ll see a currency spawned based on human hours (my hour is worth 2.3X of tom’s, etc.) and it will crush other currencies in its wake.

  20. Gravatar of ssumner ssumner
    28. September 2010 at 11:37

    Joe, You said;

    “When it comes Graham’s followers, its not a bunch of dudes cheating the system.”

    You misunderstood my analogy. I am comparing bettors who beat the system at a casino to investors who are smarter than the market. So using my analogy, you are claiming these smart investors are like gamblers who beat the system.

    You said;

    “Take Walter Schloss, a guy in his 90s who directly learned under Graham, I quote…

    “Although he stopped running others’ money in 2003-by his account, he averaged a 16% total return after fees during five decades as a stand-alone investment manager, versus 10% for the S&P 500″³

    Thats an outstanding track record, how can that be anything but skill?”

    This is a common mistake. It is very likely they were just lucky. Take some averge guy who wins the Ohio state lottery. There are ten million people in Ohio, so what are the odds that John Smith of Dayton, Ohio would have won the lottery? 10,000,000 to 1. But someone had to win. In investing, most people will be close to average. But when there are a large number of investors, you’d expect some to do extremely well, and some to do extremely poorly (which is just as hard, by the way.) Your mistake is to single out the person you already know was successful, and ask “how likely is it that it was just luck.”

    Silas, Fine, but that doesn’t change my argument that there would be no liquidity trap.

    123, I doubt regulators could create a bubble, even if they tried. If I’m a regulator, how do I convince Bank of America to invest in MBSs that will drive its stock down close to zero?

    123, I’m not sure I buy rational bubble theories, as least as applied to the real world.

    Josh, The serial number idea is a complete non-starter, for all sorts of reasons. It just wouldn’t be practical. Negative interest on base money is certainly possible, Sweden already did a small program on reserves. The Fed could no longer say “we can’t cut interest rates any further,” and hence I think there would be a lot of pressure on them to go below zero. It would also reflect changes in the way economics is taught. When I was young, no one could imagine IOR, now it’s common. The textbooks will no longer be able to say there is a zero bound, so a whole generation of economists will be taught monetary economics w/o the zero bound. Eventually they’ll start working at the Fed. When I proposed negative rates on reserves many commenters said a central bank would never do it. Six months later the Swedes did it, and Alan Blinder just proposed it for the US. Believe me, it’s coming.

    Josh, If people start hoarding gold, and stop hoarding cash, then you are out of the liquidity trap–and prices start rising. That’s what you want. The point is to stop the hoarding of cash.

    Indy, Everyone seems to be missing the point. The whole point of negative IOR is to cause people to NOT to want to hold money.

    Benjamin, Yes, someone has to make the market efficient, and presumably they earn a competitive rate of return.

    Nick, Yes, that’s right.

    123, Thanks, that’s a very good point. But isn’t that similar in spirit? They are seeing whether the successful funds actually knew something, or just had a fluke year where they were really profitable. Or have I misunderstood the article?

    Philo, Regarding Mises and a priori logic, I’d say no, because I also believe in theories that are totally illogical, like AS/AD. If the data didn’t support the AS?AS/AD theory, no one would have developed such an affront to common sense.

    Joel, I think the Newtonian analogy is a good one, but as of now we are still waiting for our Einstein. Seriously, it differs from Newtonian physics in several ways. It is probably less of an approximation to reality than Newtonian physics, at least under local conditions, but also harder to improve upon. Einstein didn’t have to worry about the fact that discovering relativity might cause it to become false. Indeed quantum measurement is the better analogy.

    Morgan, I think we’ll get electronic money long before we get flexible wages and prices.

  21. Gravatar of JimP JimP
    28. September 2010 at 11:38

    And here is a way for the Fed to buy bonds without having to buy many at all – expectorations do the work for them.

    http://ftalphaville.ft.com/blog/2010/09/28/355076/so-qe-asy-it-hurts/

  22. Gravatar of Josh Josh
    28. September 2010 at 12:05

    Scott,

    Your wrote:

    If people start hoarding gold, and stop hoarding cash, then you are out of the liquidity trap-and prices start rising. That’s what you want. The point is to stop the hoarding of cash.

    Yes, but we are talking about a cashless economy. Nobody is hoarding cash.

    Suppose that financial innovation progresses to a point at which there are no reserves and there are no cash balances. Individuals hold money in the form of money market mutual funds or some other such asset. Negative nominal interest rates imply that the value of the balances of the money market mutual fund will decline. Who is going to continue to maintain these balances? There exists an incentive to find and hold commodities with low (or zero) storage costs.

    What’s more, can monetary disequilibrium exist in the absence of base money?

  23. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    28. September 2010 at 12:39

    Scott, you said:
    “I doubt regulators could create a bubble, even if they tried”
    They do this all the time whenever a currency peg is established.

    you said:
    “Thanks, that’s a very good point. But isn’t that similar in spirit? They are seeing whether the successful funds actually knew something, or just had a fluke year where they were really profitable. Or have I misunderstood the article?”

    They were trying to determine if the results of the best funds can be explained by pure luck vs. skill. But they were only examining total returns for the period – it does not matter at all if the skill or luck based outperformance is consistent or was generated in a single year.

  24. Gravatar of Silas Barta Silas Barta
    28. September 2010 at 12:39

    @ssumner: ,i>Silas, Fine, but that doesn’t change my argument that there would be no liquidity trap. … If people start hoarding gold, and stop hoarding cash, then you are out of the liquidity trap-and prices start rising. That’s what you want. The point is to stop the hoarding of cash.

    Wait, so it’s a liquidity trap if people hoard their savings in cash and stop buying things, but not if they hoard it in commodities that functionally take the place of (slight less liquid) cash and … also stop buying things?

    Your position seems to be, “That’s not a liquidity trap because I choose not to call that asset money!” Huh?

  25. Gravatar of Benjamin Cole Benjamin Cole
    28. September 2010 at 13:03

    OT, but saw a great column title in the NYT: “Profiles in Timidity.”

    Defines the Fed.

  26. Gravatar of William William
    28. September 2010 at 13:07

    Re: People who manage to make money in the stock market, Armen Alchian has people pair off in coin-flipping tournaments to illustrate the point.

    Assuming one toss per second for each eight-hour day, at the end of ten years there would still be, on the average, about a hundred-odd pairs; and if the players assign the game to their heirs, a dozen or so will still be playing at the end of a thousand years! The implications are obvious. Suppose that some business had been operating for one hundred years. Should one rule out luck and chance as the essence of the factors producing the long-term survival of the enterprise? No inference whatever can be drawn until the number of original participants is known; and even then one must know the size, risk, and frequency of each commitment. One can see from the Borél illustration the danger in concluding that there are too many firms with long lives in the real world to admit an important role to chance. On the contrary, one might insist that there are actually too few!

  27. Gravatar of William William
    28. September 2010 at 13:19

    (following up) … which, incidentally, is the point of comparing actively-managed funds to index funds. Even if the EMH says that it’s not possible to make money in the stock market, we know that it is possible to lose money by constantly buying and selling different stocks. The existence of a few managers who beat the market is evidence against the EMH. If it were just chance, people should be doing better!

  28. Gravatar of William William
    28. September 2010 at 13:20

    Ugh… sorry, correction. “The existence of a few managers who beat the market is evidence against the EMH” should read “the existence of just a few managers who beat the market is evidence *for* the EMH.”

  29. Gravatar of Indy Indy
    28. September 2010 at 13:24

    Hmm… I don’t think I’m expressing myself well. I understand the idea of wanting to diminish the demand for money, the point about storage costs, etc. Here, let me propose a hypothetical:

    Let us say that I set up a new financial institution, IndyFed. The purpose of IndyFed is to create and manage an alternative, competitive, form of money (easily imaginable as a kind of foreign currency, I suppose, which would help establish something closer to a “risk-free” rating, so we can hand-wave away the risk-premium).

    Anyway, what IndyFed does is this. It sells 2% negative interest (in physical quantity terms) CPI-basket promissory notes, redeemable on demand. It sells them at today’s market price plus a small transaction charge, a charge also imposed when you redeem them for the US-cash value of the market price of the diminished-basket.

    And this is perfectly similar to what you would do if you were storing wealth in a foreign currency you trust to be well-managed as to maintaining its professed target trend of real purchasing power (kind of like Canada?), except that, in the corporate setting, you trade-off the traditional “risk-free” bonus for the new “strictly liable under contract law without the sovereign’s prerogative of debasement” bonus.

    IndyFed kind of operates like a bank therefore and has all the fundamental maturity transformation and asset-management problems that banks have – but let’s assume it is as prudently managed as possible, unlike, say, almost the entire global industry over the last few years.

    Of course, it’s not a perfect replacement for “money”, in terms of everyday transactions, (though if popular it could easily become prominently written in private contracts in the form-of-payment terms, as many contracts abroad are indeed written in foreign hard-currencies). But it’s close as least in it’s capacity to offer a cash-like savings mechanism that is “real” instead of nominal, as ordinary bank deposits would be.

    I’m not convinced by the Boolean logic of everything being strictly money or not-money. I think it’s more a matter of degree as to how “money-ish” something is depending on how the markets actually treat and value the instruments – similar to the money-ish liquidity coefficients in Barnett’s Divisia metric. Yen are not useful as money in the US, but they are very money-ish.

    So, in this world, it’s true, negative interest rates on electronic money would diminish people’s propensity to hold US electronic-notes. But so what? It doesn’t change their propensity to consume or create additional demand – it would just shift the preferred form of their electronic “liquid” savings. People would seek a savings safe-harbor in alternative money.

    Of course, IndyFed has to use the dollars it receives from its sales, and it buys some intelligent distribution of actual physical CPI-baskets stored as efficiently as possible, in addition to futures of various maturity. I’m sure some quant wizkids can figure this out without breaking the world financial system, well … actually … hmm …

    I could be wrong – but my hunch is that this management-buying would goose AD a little but not much – though I’ll concede I’m not sure and I’d have to see some models, I suppose.

    But let’s say our quants can do a decent job. In this scenario, it still seems to me that you will indeed hit an interest-rate floor beyond which you can’t effectively influence macroeconomic aggregates because of the existence of this safe form-of-savings money competitor.

    The establishment of such a thing, even as a subsidized toy prediction market on Intrade, would probably be a very useful thing – as the exchange rate between something that actually accomplishes the trend the Fed is trying to establish, and actual cash probably reveals all kind of important information the actual Fed could use for targeting.

    It’d be useful like a NGDP futures market would be useful. I bet a good graduate-level paper could be written about how to properly exploit such information.

    Or maybe it’d get an F. I’m probably missing something. Monetary Economics is indeed highly counter-intuitive.

  30. Gravatar of Benjamin Cole Benjamin Cole
    28. September 2010 at 14:47

    http://online.wsj.com/article/SB10001424052748703694204575518222145769804.html

    The above WSJ article seems to suggest that the Fed is contemplating Scott Sumner’s $100 billion a month QE plan (yes, Sumner’s plan includes other elements).

    This is great news.

    All monetary bulls: This is your hour. Make your viewpoints known in blogs, letters to the editor, letters to Fed officials, whoever.

    The buzz around Sumner’s idea is…well, buzzing.

  31. Gravatar of Morgan Warstler Morgan Warstler
    28. September 2010 at 15:06

    Karl, just knocks it out of the park today:

    http://market-ticker.org/akcs-www?singlepost=2189736

    “This fiscal year (2010) we have approximately $13 trillion outstanding in debt (including “intergovernmental borrowing”, that is, Social Security and Medicare.)

    Our total debt service is projected (it’s not quite over!) to be 4.63% of the budget, or about $165 billion. That’s approximately 7% of revenues, incidentally.

    That’s an effective interest rate of about 1.27%.

    Yes, 1.27%.

    Now what happens if we take no more debt at all but rates normalize to 5%?

    That would be $672 billion, or about $500 billion more than it is today. Incidentally, that’s fifty-two percent of all (personal and corporate) income taxes, up from today’s thirteen percent.”

  32. Gravatar of Jim Picerno Jim Picerno
    28. September 2010 at 15:40

    Scott, great post. But a few points. One, it’s important to clarify that variation in p/e (or dividend yield, or something similar) has shown some linkage with predicting return over the medium/long term. Relatively high (low) div yld, for instance, tends to a good predictor of relatively high (low) return 10 years out. Some EMH critics say such a relationship is a sign of market inefficiency, since you can “beat” the market by buying when yields are high and selling when yields are low. But that’s not necessarily a refutation of EMH.

    It’s hardly a sign of inefficiency to discover that expected returns vary. It would be inefficient if returns were static. At the end of 2008, when it seemed like the economy was on the verge of collapse, expected return for equities soared, as implied by a surge in the trailing dividend yield. That’s surely rational, since the only way investors can rationalize buying stocks in, say, Dec. 2008 is that the market’s offering better terms.

  33. Gravatar of Benjamin Cole Benjamin Cole
    28. September 2010 at 16:06

    Jeez, Bill Gross/PIMCO, says more QE is a given….

    Foreign Stocks Offer Better Investment Than US: Gross
    Published: Tuesday, 28 Sep 2010 | 2:45 PM ET T

    The weak dollar will mean meager returns on US stocks and bonds for years, so US investors need to look at overseas markets for better yield, Pimco’s Bill Gross told CNBC Tuesday.

    “The developing world grows at a much faster rate, so investors should be looking outside the United States,” he said. “Especially if the dollar is declining and reducing your standard of living and purchasing power.”

    Among those areas where he expects better growth””and returns: China, other Asian countries, Australia and Canada.

    “Non-dollar currencies, in combination with higher growth, is really the receipe for investment success,” he said.

    Gross said he expects the Fed to begin another round of easing soon, which he called “a last gasp.”

    The Fed, he said, will begin “buying hundreds of billions of Treasurys with the hope that lower interest rates will stimulate the animal spirits, forcing investors to buy stocks” because bonds offer such low yields.

    Gross said that if the Fed is successful in stimulating economic growth over the next few years, stocks will be the main beneficiary….

    QE is coming? I hope so.

    Time for a secular bull market in equities and property? I think so.

  34. Gravatar of Wimivo Wimivo
    28. September 2010 at 16:12

    A book written by Leonard Mlodnow (he’s been in the news lately) called The Drunkard’s Walk deals with highly successful portfolio managers. In it he analyzes the success of Bill Miller’s Legg Mason Value Trust Fund, which “was the most celebrated fund manager on Wall Street because his fund outperformed the broad market for 15 years straight”. In a related article, he points out that “if all the comparable fund managers over the past 40 years had been doing nothing but flipping coins, the chances are 75% that one of them would have matched or exceeded Mr. Miller’s streak.”
    (http://online.wsj.com/article/SB10001424052970204556804574261942466979118.html)

  35. Gravatar of Lorenzo from Oz Lorenzo from Oz
    28. September 2010 at 17:24

    Scott, enjoyed your Economist piece. You even got the requisite clueless comment. It is worth noting that you have extremely high quality comments on your blog: a sign you are doing something very right. (Such as following the suggested model set out here.)

    Silas: So it’s a liquidity trap if people hoard their savings in cash and stop buying things, but not if they hoard it in commodities that functionally take the place of (slight less liquid) cash and … also stop buying things? Q: what did they do with their cash? You need to grab Henry Hazlitt’s Economics in One Lesson and read (or re-read) the bit about looking at the next step in the causal process. I take Scott’s point to be that if they hoard gold, they are spending their cash.

  36. Gravatar of Greg Ransom Greg Ransom
    28. September 2010 at 17:56

    Even Popper abandoned this dogma …

    “They retort that no theory that can’t be disproved is worth anything. “

  37. Gravatar of Greg Ransom Greg Ransom
    28. September 2010 at 17:58

    Ask these folks if there is any evidence they would accept refuting their theory of worthwhile theories …

    “They retort that no theory that can’t be disproved is worth anything. “

  38. Gravatar of jean jean
    29. September 2010 at 03:40

    @Joël, ssumner:
    I think that the conclusion of the paper on P=NP and the efficient market hypothesis is that the latter should be reformulated as follows:
    It is possible to find market anomalies from past prices analysis, but the cost of computation to find them before other people is greater than the benefit you could get by using this finding.

  39. Gravatar of DanC DanC
    29. September 2010 at 06:19

    Very good post, a little bit wordy, but most excellent.

  40. Gravatar of ssumner ssumner
    29. September 2010 at 08:35

    JimP, Interesting article, and the link to the WSJ is also interesting.

    Josh, I was assuming the continued existence of base money, in the form of electronic reserve balances. I’m not sure how a true moneyless economy would work.

    123, I don’t see how a currency peg is a bubble. To make the peg stick, they must adjust the money supply so that the pegged rate becomes the equilibrium rate.

    123, I think I now understand your point about the total return. Tell me if this is correct: Extremely high returns could occur if some funds took huge risks. But in that case you’d also see extremely low returns. If you have an unusual number of extremely high return, but no unusual number of extremely low returns, then the explanation can’t be high risk behavior, rather it seems the high returns show real stock picking talent. Is that the basic idea?

    Silas, Yes, that’s what I’m saying, and as far as I know that’s what all economists are saying when they talk about liquidity traps. Why is that a surprise?

    Benjamin, Yes, a good title.

    William, That’s a good point.

    Indy, I don’t understand why you think this near-money alternative supports the liquidity trap idea. If people tried to switch from cash to your near money, then prices would rise (assuming the supply of cash was unchanged.)

    Money can be defined as any medium of account, an asset in which other prices are quoted. I find the base to be the most useful medium of account, because it is controlled by the Fed. There may be close substitutes for the base (though I doubt it during most times.) But that doesn’t change anything fundamental in the analysis. If some alternative asset is a perfect substitute, then the most useful definition of the relevant money supply is the base plus that perfect substitute.

    Benjamin, Yes, I was also struck by the $100 billion/month figure, which I also proposed. Of course if that’s all they do, I’d prefer a much bigger figure.

    Morgan, Yes, I’m also a bit worried about fiscal policy.

    Jim, Those are good observations about the P/E issue. I am agnostic on that issue. I tend to believe in the EMH, but also understand why Shiller thinks this ratio is so valuable–it does look awfully important when you fit past data.

    Wimivo, That’s a good point.

    Lorenzo, That’s an interesting post. In another life I’d actually learn how to use computers, and make my blog much more reader friendly.

    Greg, That’s a clever point.

    Jean, That sounds right, but I don’t know enough about the P=NP stuff to rely on anything more than my intuition.

    Thanks DanC. Keep in mind I have two types of posts; a little bit wordy and way too wordy.

  41. Gravatar of Silas Barta Silas Barta
    29. September 2010 at 08:57

    @ssumner: I’m not sure which one that was a reply to, but I think you’re confirming that “If everyone rejects this electronic money [for the most part], stores all savings in commodities, and even uses them for most of their trade, then it doesn’t count as a liquidity trap when people don’t spend those preferred ‘money-like goods’, even if no one’s lending because the real return on holding those goods is positive.”

    My confusion is that it seems to be erasing a supposed problem because of a simple naming convention.

  42. Gravatar of Andy Harless Andy Harless
    29. September 2010 at 09:38

    Indy:

    my hunch is that this management-buying would goose AD a little but not much

    I don’t get that hunch at all. To the extent that IndyFed is engaging in an enterprise intended to be profitable, it is buying stuff, either directly or indirectly. (Buying futures is roughly equivalent to buying physicals in that it gives arbitrageurs and incentive to store commodities.) Buying stuff is what AD means. To the extent that IndyFed hedges the risk of a lower inflation rate (which it should fully hedge, to the extent that its business model is to provide a service rather than to engage in speculation), it will increase AD by the entire amount of its deposits.

    Now if you want to make the argument that not all AD is equivalent, and that attempts at negative interest rates will result in excessive production of storable commodities as a store of value to replace money, you can make that argument. But it’s not really an argument about money and near-money; it’s about commodities and other real stores of value. IndyFed is only an intermediary and not necessary for the argument.

    In any case, the interest rate could be made sufficiently negative that the incentive to create real stores of value (to satisfy either direct investor demand or that of intermediaries like IndyFed) would be high enough to bring AD up to the full-employment level (however one might choose to define that level). If you want to say that the excessive production of store-of-value commodities would be even more inefficient than having idle resources, you can make that argument. But again, it’s not really about near-money: it’s an argument that the availability of actual money, as a store of value, is a good thing because it prevents this inefficiently high production of real stores of value.

    And it strikes me as a pretty weak argument, anyhow. If the real stores of value are good stores of value, then by definition they will have value in the future. It is useful to produce them, because they will necessarily be useful in the future, or else they wouldn’t be valuable in the future and therefore wouldn’t be good stores of value. So, from a social welfare point of view, why not produce them instead of leaving workers unnecessarily idle?

  43. Gravatar of Silas Barta Silas Barta
    29. September 2010 at 09:59

    @Andy_Harless:

    And it strikes me as a pretty weak argument, anyhow. If the real stores of value are good stores of value, then by definition they will have value in the future. It is useful to produce them, because they will necessarily be useful in the future, or else they wouldn’t be valuable in the future and therefore wouldn’t be good stores of value. So, from a social welfare point of view, why not produce them instead of leaving workers unnecessarily idle?

    1) Because the workers *aren’t* producing them — they’re just holding onto existing commodities.

    2) If you object that, “But all this saving will *lead to* production of these commodities, which *is* good”, then you’re rejecting the usual argument against the gold standard, which is that spending all this effort to mine more gold (or otherwise produce these commodities) is wasteful, because it’s only going to be used as money, when we could just use paper money.

    So, we’ve come full circle!

    1) You shouldn’t use the gold standard because it requires constant mining of new gold that carries real costs with no corresponding benefit.
    2) So let’s use electronic currency that the government can arbitrarily decide to take from you when it decides you’re not spending it frequently enough.
    3) So people who don’t like the idea of their savings being arbirarily leeched away will have to store their savings in commodities.
    4) This means that when people save, they’ll do it by mining gold that’s intended only for use as money, which is GREAT!
    5) *falls out of chair*

  44. Gravatar of scott sumner scott sumner
    29. September 2010 at 18:06

    Silas, No, it’s not just about language, it’s about economics. I assume everyone agrees there is no such thing as a liquidity trap when inflation and NGDP growth are very high. My point is that if no one wants to use money, then inflation will be very high. Inflation is defined as a fall in the value of money. When the demand for money falls, its value will tend to fall.

  45. Gravatar of Silas Barta Silas Barta
    29. September 2010 at 19:22

    Yes, but in the liquidity-trap-antidote scenario, inflation negates (ends?) the liquidity trap because people are spending money. But this obviously isn’t true when the value of money falls because people *aren’t* using it! So *that* kind of inflation needn’t mean the end of a liquidity trap — just the opposite.

  46. Gravatar of Andy Harless Andy Harless
    29. September 2010 at 20:55

    Silas,

    Considering the amount of volatility in the real price of gold, it’s not a good store of value, and in any case, there isn’t enough gold to make up a large fraction of the value people want to store. And the primary objection to the gold standard is not the mining costs but the uncertain supply and demand, which result in a volatile price level (and worse, if prices are sticky). When gold was the primary unit of account, it functioned as a good store of value for individuals, but it was never fundamentally a good store of value; it was only sort of a “stable bubble” enforced by the stickiness of prices.

    As long as the monetary authority controls the unit of account (which should be the case as long as fiat money is legal tender), no individual commodity is a good store of value (because any bubbles will be unstable), and a portfolio of commodities is a good store of value only because the commodities are useful (because the fundamental values depend on usefulness, and any bubbles will be unstable).

    And in your response to Scott, I don’t see how it makes sense to talk about “when the value of money falls because people aren’t using it.” It is not the ultimate act of ceasing to use money that causes the value to fall; it is the process of preparing to stop using it. You don’t just throw away your money while others are still accepting it; you get rid of it by spending it. And that’s what ends the liquidity trap.

  47. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    30. September 2010 at 03:17

    Scott, you said:
    “I don’t see how a currency peg is a bubble. To make the peg stick, they must adjust the money supply so that the pegged rate becomes the equilibrium rate.”
    Pegged currencies are a bubble for some time before the peg crashes.

    You said:
    “I think I now understand your point about the total return. Tell me if this is correct: Extremely high returns could occur if some funds took huge risks. But in that case you’d also see extremely low returns. If you have an unusual number of extremely high return, but no unusual number of extremely low returns, then the explanation can’t be high risk behavior, rather it seems the high returns show real stock picking talent. Is that the basic idea?”
    No. In their simulation they recreate a universe of zero skill mutual funds that have the same distribution of risk characteristics as the real world has. Then they compare the number of very high return funds in the actual world and in the simulation, and find some evidence that some of the high performance funds have skill, because in the real world the number of high return funds is higher. This has nothing to do with the annual distribution of returns, and this has nothing to do with the ratio of very high/ very low return funds. In fact, they find that some funds have negative skill.

  48. Gravatar of scott sumner scott sumner
    30. September 2010 at 05:43

    123, I disagree with the notion that pegged currencies violate the EMH. I don’t see any evidence for that. There’s a reason George Soros is so famous–if pegged rates were a bubble we’d all be as rich as Soros.

    I would think that the finding of “negative skill” is a huge red flag, and a big point in support of the EMH. Surely no one is trying to beat the market on the downside. Why wouldn’t the negative skill people just reverse their strategy? Probably because they don’t know they have negative skill. And since negative and positive skill are mirror images, I think its also very unlikely that those shown to have positive skill, actually know what they are doing.

    I would also note that given the complexity of real world portfolios, the various mutual funds probably have unidentified characteristics that were not put into the F&F model. If so, that could explain the abnormal returns.

    F&F may not have worried about that, as they already got the finding they were looking for.

  49. Gravatar of Silas Barta Silas Barta
    30. September 2010 at 06:01

    @Andy_Harless:

    You don’t just throw away your money while others are still accepting it; you get rid of it by spending it. And that’s what ends the liquidity trap.

    Until people stop accepting the hot potato …

    And the primary objection to the gold standard is not the mining costs but the uncertain supply and demand, which result in a volatile price level (and worse, if prices are sticky).

    Maybe not *your* primary objection, but I never hear a consistent case, and that’s definitely what a lot of people think.

    and in any case, there isn’t enough gold to make up a large fraction of the value people want to store.

    Oh wait, it looks like it is a big concern.

    As long as the monetary authority controls the unit of account (which should be the case as long as fiat money is legal tender), no individual commodity is a good store of value (because any bubbles will be unstable), and a portfolio of commodities is a good store of value only because the commodities are useful (because the fundamental values depend on usefulness, and any bubbles will be unstable).

    That still shows how and why people would switch to a different money, including non-negative interest currencies.

  50. Gravatar of Freestate Freestate
    30. September 2010 at 06:52

    Scott, I would take issue with the idea that the only way to test the EMH is to test for serial correlation. In particular, I am assuming that the biggest inefficiency is overvaluation – call it a bubble if you will. If momentum drive inefficiencies exist at the MARKET level, then my strategy to earn excess returns would be to not invest when valuations are extreme and to invest when they are not extreme. For example, I would not be invested when Shiller p/e’s are above 20. This will lead to long periods of under-performance followed by long periods of over-performance. Over the very long term it will lead to market out-performance. It won’t lead to a string of serially correlated superior performances (i.e. winning in every consecutive year).

  51. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    30. September 2010 at 08:17

    Scott, you said:
    “I disagree with the notion that pegged currencies violate the EMH. I don’t see any evidence for that. There’s a reason George Soros is so famous-if pegged rates were a bubble we’d all be as rich as Soros.”
    It’s hard to get rich from currency bubbles. If a peg has 2% annual probability of blowup, yet interest rate differentials are pricing in only a 0.1% probability of blowup, serious malinvestment will develop. But a speculator might wait 50 years until the bubble bursts, and with incomplete markets, it might not be possible to make a bet at all.

    You said:
    “I would also note that given the complexity of real world portfolios, the various mutual funds probably have unidentified characteristics that were not put into the F&F model. If so, that could explain the abnormal returns.
    F&F may not have worried about that, as they already got the finding they were looking for.”

    This is certainly a possibility, but one of the main directions of research Fama & French are doing is risk factors, and they are #1 “experts”. There is also a danger in the opposite direction – F&F assume that size and book factors are real risk factors and not an indicators of market inefficiency, so they already assume a part of their conclusion.

    I have also checked – according to gross return simulation, more funds have positive skill than negative skill.

    There are two explanations for negative skill funds – their returns is a combination of negative skill and luck, and it is hard to determine which is which. Negative skill funds might also have an advantage in marketing, as in “my neighbor is rich, he invested in all those well-performing dot com funds”.

  52. Gravatar of scott sumner scott sumner
    30. September 2010 at 18:33

    Freestate, That’s an anomaly that’s worked in the past. But what makes you think it will work in the future?

    123, Well Soros didn’t have to wait 50 years. My serious response would be that you are arguing that currency markets are a tiny bit inefficient, which is hard to distinguish from the case where they are efficient.

    I accept you argument on the negative skill funds, if they really are less common than the positive skill funds. So we’re back to F&F’s result, which I view as saying the EMH is roughly true, but not exactly.

    I don’t know enough about finance to comment on your risk factor discussion.

  53. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    1. October 2010 at 02:44

    Scott,
    currency bubbles and associated 2 percent interest rate inefficiencies are not tiny, if you measure long-term investment project demand.

    Positive skill is more common when measured according to gross returns. Negative skill is more common when measured according to net returns.

  54. Gravatar of ssumner ssumner
    1. October 2010 at 18:15

    123, It also seems to me that you’d have trouble proving it was a bubble, and not a market inefficiency caused by government intervention.

  55. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    2. October 2010 at 04:40

    If currency markets were perfectly efficient, interest rates would rise under the peg, and the investment bubble would be prevented.

    But if Soros doesn’t come and short the currency, interest rates stay at levels that are too low, and investment bubble grows.

    What the government did was dangerous, but it was the job of markets to price in the danger, and markets failed.

  56. Gravatar of ssumner ssumner
    2. October 2010 at 08:22

    123, That would be difficult to prove. You’d have to prove the markets knew the exchange rate was out of line. That’s much harder than you think–it depends on much more than CA deficits of deviations from PPP.

  57. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    2. October 2010 at 13:29

    Scott, it is very easy to observe the mismatch between the volatility of fundamentals and the volatility of exchange rates and interest rates. One obvious example is Greece. One year ago we had a bubble in Greek government bonds.

  58. Gravatar of ssumner ssumner
    2. October 2010 at 17:56

    123, Did you sell them short? Can they be sold short?

  59. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    3. October 2010 at 05:31

    I did not short Greek bonds because it is hard to do this for retail investors. But for some institutional investors with flexible mandates shorting is a possibility.

    I was advising an owner of large bond portfolio, and just after Lehman I strongly recommended not to invest in Portuguese government bonds, one of the reasons was that eurozone breakup risk was underpriced. In November 2008 I turned bearish on Greece in my public comments, because investors started the process of risk reassessment. In January 2009, I explained that Greek government bubble is bursting:

    http://translate.google.com/translate?js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&sl=lt&tl=en&u=http://www.nematomaranka.lt/2010/01/kredito-krize-tesiasi-graikijos.html&act=url

  60. Gravatar of ssumner ssumner
    3. October 2010 at 11:51

    123, Why didn’t the wholesale investors short greek bonds, if it was obvious there were going to default?

  61. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    3. October 2010 at 12:16

    Scott, even now it is not obvious that Greece will default. But it was obvious that the risk of default was underestimated. Some investors did short Greek bonds, especially by using credit default swaps.

    Sometime ago you have linked to Friedman – Mundell debate on exchange rates and currency zones. In 2007 the price of Greek bonds was telling us that it is 100% obvious that Friedman is an idiot. In my view it is a strong sign of market inefficiency. Of course efficient market’s estimate of Friedman’s intellect should fluctuate a bit, but not by that much.

  62. Gravatar of Richard W Richard W
    3. October 2010 at 13:02

    Buying the CDS for the underlying bond is effectively shorting them. The underpricing of the sovereign bonds of the euro periphery was really a story of the market misunderstanding the nature of the monetary union. The yields of the periphery started to diverge from the euro core when it dawned on investors that the eurozone was more a fixed exchange rate regime than union. Even now there is great uncertainty. Will the rest of the members allow a periphery nation to default? Can a member leave? Would the others force a member out?

  63. Gravatar of Scott Sumner Scott Sumner
    4. October 2010 at 11:34

    123, If it was “obvious” then why didn’t people see it? And how do you know it was obvious? It seems to me that your argument rests on this logical fallacy:

    1. I believe X will occur because of reason Y.
    2. My neighbor doesn’t think Y implies X will occur.
    3. X actually occurs.
    4. This shows that my neighbor is a moron.

    Richard, Good point. I think one reason people underestimated the risk was because the Greek government was lying about their finances.

  64. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    4. October 2010 at 12:19

    Scott, my argument is different. Two Nobel laureates in string theory (M&F) had opposing opinions about the number of dimensions in the universe. One day Hansonian betting market indicates that there is 100% probability that M is right. Next day the betting market indicates that there is 100% probability that F is right. On the third day this again market indicates that M is 100% right. It is obvious that the betting market did not efficiently process information about string theories, and it did not efficiently process routine information produced by hadron collider.

  65. Gravatar of Scott Sumner Scott Sumner
    5. October 2010 at 06:23

    123, What if the market only had 2 or 3 traders, and was very illiquid? Perhaps the bid asked spread was 100%. Then arbs couldn’t make that market efficient.

  66. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    5. October 2010 at 12:39

    Scott, this was not applicable to Greece – there were many actual and potential traders, narrow bid/ask spreads. And yet prices were set by Mundellians as if Friedmanites did not exist at all.

  67. Gravatar of ssumner ssumner
    7. October 2010 at 06:01

    123, Again, That’s something you are simply assuming. There is no data that would prove that was true ex ante.

  68. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    8. October 2010 at 02:24

    Here is just one random paper about relative efficiency of credit markets:

    “With the exception of LBO news, stock market seems to reveal information about negative credit events before the CDS market.”

    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1497537

  69. Gravatar of ssumner ssumner
    8. October 2010 at 14:58

    123, Again, the key question is “ex ante.” That’s an ex post study. It’s an anomaly study.

  70. Gravatar of 123 123
    10. October 2010 at 03:53

    Yes, this is anomaly study. But after this and other similar studies, lots of funds have implemented credit-equity arbitrage strategies. This tells us that the study wasn’t a fluke, there was a real inefficiency in the credit market. This inefficiency was quite easy to arbitrage, as it was related to the events that already have happened or will happen in the very near future. It is much harder to correct mispricing that results when German and Greek bonds reflect different probabilities of Greek default, because in this case inefficiency is related to unknown future events. Supporters of Mundell were purchasing Greek bonds five years ago, supporters of Friedman were purchasing German bonds five years ago, and there was little arbitrage between these two opinions.

  71. Gravatar of ssumner ssumner
    10. October 2010 at 04:59

    123, I read the evidence in exactly the opposite way you do. The fact that all the funds are now taking advantage of the pattern suggest it will go away. And any anomaly that goes away after being discovered is simply a product of data mining, not a true anomaly. To show market inefficiency, you need to show patterns that continue to persist, EVEN WHEN GENERALLY KNOWN TO MOST MARKET PARTICIPANTS.

  72. Gravatar of 123 123
    12. October 2010 at 12:15

    Scott, it is a tiny minority of funds that are engaging in credit-equity arbitrage, and their actions will cause it to disappear, even though most equity funds will continue to ignore daily credit market developments, and most credit funds will continue to ignore daily equity market price changes.

    I think that most people are using a different definition of EMH than you are using here. There is a strong version of EMH, that says private information held by some market participants should be fully reflected in prices. There is a semi-strong version, it says that publicly available information (such as in housing bubble articles in Economist magazine, and in academic debates between Friedman and Mundell) should be fully reflected in prices. Your strange definition says that information should be reflected only if it is generally known to most market participants. Your definition says that markets where prices are distorted by noise traders who ignore some publicly available information could be efficient, but this is not what EMH usually means.

  73. Gravatar of ssumner ssumner
    15. October 2010 at 15:10

    123, No, I meant publically available information. But it can’t be found by looking backward. You can’t go back and say “look, everyone should have been able to see X, and we now know that X affected market prices, yet it wasn’t incorporated into the prices at the time.” By that definition the whole idea of the EMH would be absurd. I’m asking whether the anti-EMH view has any policy implications. I’m not interested in sure things looking backward, I want to know what investments you KNOW right now will out perform the S&P 500 in future years. That’s what I am skeptical about. It’s easy to go back and say “X was obvious, and it was public, and it wans’t included in the market prices.” Clearly it wasn’t obviously relevant to the market at the time. That’s why the F&F test is the only meaningful test of the EMH. Don’t search the past for anomalies, search the past for evidence that smart people are able to consistently find anomalies.

  74. Gravatar of 123 123
    18. October 2010 at 03:45

    Scott, I’m not saying “look, everyone should have been able to see X, and we now know that X affected market prices, yet it wasn’t incorporated into the prices at the time.” I’m saying “look, according to the markets, the risk of X was zero, scientific community was divided on the matter, yet only the views of some of the scientists were incorporated into the prices at the time.”

    Anti-EMH view has policy implications. If markets are efficient, the harm of credit market regulations can be measured by the net present value of the subsidy involved. If markets are not efficient, regulations can nudge markets and cause enormous increases in inefficiency.

  75. Gravatar of ssumner ssumner
    18. October 2010 at 18:15

    123, I don’t see how you can assume markets think there is zero risk.

    I should have said the current evidence against the EMH doesn’t give the anti-EMH any value. No one has shown regulators are smarter than markets. Find that evidence and I will listen.

    Is there a typo in your last word?

  76. Gravatar of 123 123
    19. October 2010 at 01:39

    Scott, there is no typo in my last word. I am concerned that new Basel regulations will nudge credit markets to create new credit bubbles.

    In 2005 markets thought that there is zero probability that Friedman was right in the Friedman-Mundell debate, as Greek-German bond spread has virtually disappeared, only tiny difference related to the relative size of the bond markets remained.

  77. Gravatar of Scott Sumner Scott Sumner
    21. October 2010 at 05:01

    123, Ex post, they did underestiamte the probability of default. But those forecasts may have been rational at the time. This is just another anomaly.

  78. Gravatar of 123 123
    23. October 2010 at 03:04

    Scott, I’m saying that ex-ante they underinvested in collecting information about Greece. Even though every year billions are wasted by mutual funds that are trying to predict relative performance of various stocks, it doesn’t mean that that there is a similar overinvestment in collecting information about other issues.

  79. Gravatar of Scott Sumner Scott Sumner
    24. October 2010 at 07:01

    123, I just can’t imagine anyway of showing that. Suppose Greece had not gotten into financial problems. Would you be able to convince people that traders underinvested in info gathering? I doubt it. Which tells me that’s it only appears to be a mistake because ex post Greece did get into trouble.

  80. Gravatar of 123 123
    27. October 2010 at 11:26

    Scott, it depends. In 2006 (when according to markets there was no risk of Greek financial problems) I would have succeeded in persuading Milton Friedman and other US based Eurozone skeptics that traders are underinvesting in the analysis of their thoughts.

    If I had raised the issue with actual Greek bond investors, they would have laughed at me, in a polite European company it was a gross violation of political correctness to talk about Eurozone breakup risk, people would have said that I am a Hitler or a Borat.

  81. Gravatar of ssumner ssumner
    27. October 2010 at 17:50

    123, I think I’ve reached the end of the line here. I just don’t know enough about the ex ante risk of greek default to make a sweeping judgement about the EMH. I will keep an open mind.

  82. Gravatar of Doc Merlin Doc Merlin
    27. October 2010 at 20:02

    @123
    ‘Scott, it depends. In 2006 (when according to markets there was no risk of Greek financial problems) I would have succeeded in persuading Milton Friedman and other US based Eurozone skeptics that traders are underinvesting in the analysis of their thoughts.’

    1. Basel II bank regulation gave preferential treatment for sovereign debt, so it wasn’t a completely free market.

    2. EMH doesn’t say that the market will be perfect at predictions, it just says that the market is best on average. So, the market can miss a lot of stuff, but on average it should be better than non-market prognosticators.

  83. Gravatar of 123 123
    28. October 2010 at 07:06

    Doc Merlin,
    EMH says markets are investing an optimal amount of money in information gathering activities in order to achieve the optimal amount of reduction of the bias in market forecasts.
    Yes, Basel II means that it is a supervised market, but it is fascinating that Basel II was able to reduce the amount of information gathering activities so drastically. It was as if bank regulators said you are allowed to no longer worry about the sovereign default stuff, and banks got lazy (they didn’t have to, information gathering by the banks was not prohibited, just no longer mandatory), and there were not enough non-bank market participants who continued to worry about such things.

    It was the same with the AAA subprime, the magic of regulators and Moodys was so strong, that only ten investors in the whole world were critically checking the assumptions embedded in those securities.

Leave a Reply