Posts tagged with emh

I do not believe that financial markets make the economy more efficient. The analogy I use is Earth. If it were reduced to the size of a basketball, it would be smoother than a billiard ball. However, at a human scale, there are mountains and oceans we can exploit….

The guy talking about making markets more efficient is thinking of something like rolling rocks down a mountain to power useful work. This indeed makes the Earth smoother, wearing down mountains and filling in oceans. But … [that] bears no resemblance to what people really do. They’re more likely to build a hydroelectric dam that holds water back, that is it keeps the system farther from equilibrium, not moves it closer.

Aaron Brown, Red-Blooded Risk

(I rearranged his words a little bit.)

189 Plays • Download

Let’s reflect for 11 minutes on specific market, rather than on “free markets” in the abstract, for a real-world perspective on economic theory.

Simple fact that’s apparently obvious to everyone who trades muni bonds but not to me: 1 December is a common payout date, meaning that January & February usually have lower yields as there’s more money (tagged “for investing in bonds”) looking for its next home. So much for optimal selection of the best securities throughout all time or statistical arbitrage of the rates.

Not that there’s not an efficient-markets explanation for this or that Marshallian S&D is irrelevant. But there are some obvious kinks to it, right? You might thumbnail this as “institutions” if you have an economic-theory label gun in your holster.

  • Besides the regular cyclic dependence (I can easily imagine some theorist who doesn’t participate in the market assuming it must “obviously” be arbed away. Some papers on asset switching come to mind),
  • in this market we talk about fixed supply coming onto the market at a given time window, very different than a posted-offer (retail) or 
  • so your econ 101 S&D picture wouldn’t quite capture let’s say $AMZN’s decision of how to list its bonds. There’s some invisible demand curve
    (I drew this for an older post, not gonna re-draw an S&D curve.) that $AMZN is going to try to guess at (and they might hire some “banksters” to help them). Then they can delay or move forward that fixed vertical supply curve (they probably have pre decided how much they want to borrow based on their internal models and decision processes). So there’s a time element (try to list in Jan/Feb, if you “can” wait! Do or don’t list along with this other major issuance. Competing for the analysts’ attention. Etc.) and much less of a price element, since that’s pre-decided before listing.

    Sure, maybe there’ll be an OTC secondary market that changes the price afterwards. But now already we’ve split an adze through the “perfect efficiency” idea, simply by questioning whether it’s the primary auction or secondary trading we’re calling efficient. Now it sounds much more plausible that markets with more volume and smarter participants are going to price more “accurately” (which we haven’t defined and couldn’t, since defaults are one-off probabilities) and simply by questioning the axiom we’ve undermined the theorists’ go-to assumptions. If you wanted to model this market, you might start with something quite different to “Assume fully informed traders and no arbitrage condition”. You might start with, like, actual facts or data on trade records, look at legal documents, record phone calls, ….
  • And what about the ubiquitous government-versus-markets dichotomy? Oh, snap! This is private investors lending small (not federal) governments money to build highways. Hmm, I guess that “four legs good, two legs baaaaad” dichotomy is incoherent.
  • As far as politics goes, GARVEE’s must be a really important political issue (how to arrange for surface transport across the United States), but as it’s not a sex scandal or on either American party’s agenda to trash the other, I guess it’s unfit for discussion in the news.

Any market I look a little closer at, if you squint you can see the EMH in the outlines. But the details are more complicated and much more like a transaction you can imagine real people (who can afford lawyers) engaging in. Very head-to-head, can-we-make-a-deal, size-matters, quantity-over-price, get-it-done-rather-than-optimise-the-exact-details, ….

Equally as much as I might want to show this to overzealous, oversimple free-marketeers, I’d show it to the anti-capitalist zealots too. Does this sound like a monopoly of power by either large corporations or plutocrats? There’s a competitive playing field gunning for however much money is out there, you have to argue your case (marketing) for why the people with investors’ money (let’s say a pension fund) should trust you’ll pay them back, in general a lot of “power” floating around but sounds like they keep each other in check. It’s not like $AMZN can force people to subscribe to its bond issuance. And I can even imagine a legitimate, contributive role for high-powered lawyers and investment “banksters”. Do the computer programmers at $AMZN know how to market and list securities, track down and convince the people safeguarding the big sacks of money to lend it to them? Other than a prejudice toward large size, Doesn’t sound very plutocratic to me.

Anyway. I listened to this and got the feeling I have many times on learning just some basic obvious stuff about real markets. Like wow, grand economic theory is missing details that are obvious to actual market participants, mired in overgeneralisations and simplifications, and the theorist who gets all tooth-and-claw about their holy assumptions needs to get more exposure to the real world.

Just from the merest actual facts about this market regime I’m already thrust into a “middle ground” where, sure, the price system, self-interest, and competition seem like they’re going to be pretty good and robust-over-time and so on. But certainly not “optimal”! And certainly not full Intrade-worshipper style, where price corresponds to exact probabilities and markets are a crystal ball | prediction engine. So the extremists on both ends lose, on this story.

Amazing what you can learn about the world when you actually observe it before writing the theory.

"To generalize is to be an idiot. To particularize alone is a distinction of merit." —William Blake

Hat tip @munilass.

PS Obviously I don’t know anything about munis or corporate bonds. All of my “you“‘s and “I“‘s above can be interpreted in a strict sense as “Now that I’ve learned the very first thing about this real market, how does plausible do various abstract economic-theory ideas sound afterwards?” If you work in these markets and I said something wrong, please correct me.

The seeds of my dissent from economic orthodoxy were pretty much sown for me by my 1st professor on the 1st day of my 1st economics class.

This prof had gone to a great personal trouble to begin our exposure to the dismal science with a very down-to-earth and super-important lesson. She went so far as to spend her own cash on some things from the store, of varying cost, and gave us all at the beginning of the class random items. Some people got candy, some got socks, one or two got things of greater value.

This was a masterful teaching stroke, by someone who cared deeply about her subject and teaching it to newbies: she would have us all participate in voluntary trade within the classroom and end up than we started. Gains from trade—the fundamental point about economics—are really “the only thing we know about welfare”. Sure, some people start off with more—more wealth, more smarts, better looks, genes that will make them grow taller so they can reach the mayonnaise jars from the back—but hey, at least we can make all of them better off and not hurt anyone by allowing them to trade freely.


We each reported, on a scale of 1 to 10, how satisfied we were with the Stuff we had been randomly given at the beginning of the class, and the prof wrote these scores down on the board. Then we were asked to stand up, walk about the room, and see if anyone would voluntarily exchange Stuff with us. Multiple transactions were allowed, even encouraged—and after a few minutes of cluelessly blitzing with each other, the trading day was closed and we resumed our seats.

The prof asked our scores again, fully expecting that ∀i in the class, utility before utility after.

But one girl reported a lower score.

Instead of taking this as evidence against her belief that transactions are always mutually beneficial—a cornerstone of normative economic theory—the prof instead scolded the girl. "Well, what’d ya do that for?!”

By the way, this was not a prof who prepended test questions with the phrase “According to the theory we learned in class,” which means I still dispute that I got that one about the lobstermongers right! (Since it asked about “What would happen” not “what the theory says would happen”.)

At the time I thought the outburst a bit rude and over the years to come I remembered the episode. (well, obviously) I still think of it as a microcosm of certain intellectual misdeeds by economists. The framework is too important to hold onto; if anyone undermines then you get angry and yell at them! It’s a plausibility war, after all.

Not too far off from real comments by economists: But if you took away the mutually-beneficial assumption, then you’d have no theory at all! (Regardless of whether nullset is the only true theory we have.)

The assumptions about what goes on in transactions are so appealing that even when you see them violated in front of your eyes, they’re still so implausible and—hey—what about all this stuff I learned about indifference curves? If I saw so many graphs with them not overlapping or going backwards, then that has to be the truth, because maths!

Nevermind that people don’t always know what they want, or maybe it’s contradictory or impossible, and even in well-defined classroom experiments they may just, um, do it wrong.

Happy Independence Day. Here’s to hoping you don’t use the independence to shoot yourself in the foot.

Free Money

  • @isomorphisms: Higgs Boson Particle to be observed on/before 31 Dec 2012 = 68.0% @Intrade
  • @isomorphisms: Higgs 2013 = 80%, Higgs 2014 = 85%, Higgs 2015 = 80.1%. Oops #EMH #arb
  • There's $2 three years from now sitting on the table for whoever wants to pick it up.

[T]he Efficient Markets Hypothesis [is] in contention for one of the strongest hypotheses in the whole of the social sciences.

Strictly speaking the EMH is false, but in spirit it is profoundly true. Besides, science concerns seeking the best hypothesis, and until a flawed hypothesis is replaced by a better hypothesis, criticism is of limited value.

Martin Sewell

Um … the criticism might not be useful to science, but knowing that the EMH is false is certainly useful to market participants.

The Kelly Rule for optimal bet sizing says to leverage your bet in geometric proportion to your confidence. So a small reduction in confidence means a big reduction in (optimal) leverage.

I could also wheedle about how obvious modifications of the EMH are more correct than strong EMH, but I won’t. My main thought on this quote is that Sewell has effectively redefined “truth” so that his viewpoint is unfalsifiable. (What example could I cite to demonstrate that EMH is not “profoundly true in spirit”?) On the other hand, it would be naive of me to think that esthetics and organa are not driving the engine of any kind of theory—be it economics, physics, or Foucault.

Anyway … despite the fervour of the abstract (from which I’m quoting), Martin Sewell’s History of the Efficient Markets Hypothesis is an informative read. 


market makers are normally paranoid that the other side of the trade “knows” what is going on and that it is a pickoff.

as a market maker, you think “what can i do against this that makes sense?” so you look for a vertical or 1×2, for example. or if there is nothing, then “how much i am going to bleed out of you in order to make [this] worth the risk?”

nuclear phynance user apine


Most people think that day-traders don’t add economic value. The NuclearPhynance traders’ phorum might change their minds. Those [economists] whose idea of finance = prices converging efficiently upon a "true" value due to a kind of value-investor who also sells short—might see in apine’s words a different story. Financial markets can be more like a discussion between people who don’t necessarily know what’s going on, but try to figure that out based on what everyone else is saying.

For those who don’t understand what “liquidity" means to a trader: it means a market maker who is willing to get scraped up in a scary environment. Liquidity sometimes means a market maker (sometimes: a high-frequency trader) who is willing to play on a muddy pitch and not even charge much for the trouble.

A boxer expends much energy avoiding punches, even if they’re only feints.


[A] strong case can be made for the [random walk hypothesis]… This market view is supported by the fact that the vast majority of mutual funds fail to beat the broader market year after year, and history shows us that the ten best-performing funds in any one year will drop to the bottom of the pack in the following two to four years…. Simply put, there is no way to consistently beat the market.

Needless to say, this view of things does not sit well with Wall Street, which preaches that … relying on expertise are the keys to investing (and their business model!)….

[A]lthough the random walk theory paints a strong case against mutual funds, it is not entirely bullet-proof. Investors consistently fall prey to fear, envy, overconfidence, faddism, and other recognizably human imperfections that make markets not only inefficient but predictably inefficient…. If the DOW goes up one week, it is more likely to go up the next week. In the long run all of these moves smooth themselves out, but in the short run, predicting and trading these constant adjustments can actually make for quite a profitable proposition.

Agustin Silvani, Beat the Forex Dealer

What I’m hearing—not for the first time by a trader writing a book—is the implication that the way to consistently make money as a trader is to make the market more efficient, more stable.

If I were running a company based on this asset, I would be thanking the trader who stabilized my business decision. Is that what happens when these guys run longs and shorts all year long?

The [foreign exchange] market is by far the largest and most liquid market in the world, with daily FX turnover estimated at around $2 trillion. If this seems like a lot to you, it is because it is. Compare FX volumes to the tiny $50 billion traded at the NYSE or the $800 billion traded in government debt and you get an idea of the size of the market.
Agustin Silvani, Beat the Forex Dealer

μ dt + σ dWt, my #$$

The simplest model of a stock price movement is that the log of the price moves in a direction, plus some noisy drift (like adding a Gaussian W𝓽 at every timestep).

Agustin Silvani gives a counterexample: Federal Open Market Committee meetings precede a volatile market episode, meaning long-term changes in μ and short-term spikes in σ come after an FOMC announcement.

"Stop hunting" by dealers and other smart-money players can sometimes shake up the price pattern ONLY during a super-short period. This occurs most in the least regulated market, FX, because dealers will pull the rug out from under their retail clients’ feet. The dealers can see their clients’ stops and will just blatantly cheat the retail clients (according to Silvani), because they only need to retain a big client’s business. (Hence it’s more profitable to cheat the small fry — there will always be more.)

Efficient markets, my #$$

Not only does this violate the Black-Scholes model (a freak σ^7 comes in and then disappears), it also violates the Strongly Efficient Market Hypothesis, which says that market price—at any instant—reflects all available information and are the best measure of the "true" value of an asset. You would obviously get a more accurate valuation from taking a one-hour average than from relying on any instantaneous price, in the above graph.

According to Silvani’s story, the dealers are very efficient at bilking worse-informed or less-experienced participants, but don’t use this as a Prediction Market!