That’s quite a good formulation for bringing out what looks like a tautological aspect to the EMH.

I don’t know anybody – trader, investor, manufacturer – who makes their investment and production systems assuming that “financial markets are the best possible guide to the value of economic assets” well, quite – this also shows the convoluted nature of the thing, perhaps analogous to the idea that everyone is a price-taker. As though the market takes its course and the participants are just enganged in hovering around the underlying equilibrium it dictates. I’ve never heard a clear formulation of the Invisible Hand which didn’t seem trivial either, for that matter.

NB I don’t think I’m disagreeing with you here. Just remarking on the EMH in its proffered formulation.

Let’s bracket out outcomes, results and consequences; in fact, let’s bracket out alternative courses of action and assume that our (imaginary) perfect market is the only game in town. We’ve got zero transaction costs and infinite knowledge, and we’ve got self-interested participants each of whom has an interest in maximising income and minimising outgoings.

My observation was that, in that (imaginary) setup, the tendency will be for everyone trying to sell to be driven down to their own cost price – with the result that nobody has an edge and nobody makes a profit.

What’s wrong with that picture?

More generally, you can define the word “perfect” however you want, there’s no committee in charge of the English language. In mathematics, a perfect number is a positive integer which is the sum of its proper positive divisors. This as far as I can tell has nothing to do with how engineers or economists use the word perfect. If you want to define a perfectly efficient market as one that functions in a particular way, there’s nothing to stop you. In terms of clarity of English it might not be the best idea, but that’s not stopped many economists in the past. Or engineers.

Similarly, I don’t see how it can be in the definition of perfect efficiency for a market that any participant makes “enough money”.

It isn’t. It’s a result of the definition of a perfectly efficient market, as defined at least in the Econ 300 microeconomics course I took. As the definition of a perfectly efficient market includes that there are no transaction costs and infinite knowledge, the consequence is that if participants in the market get less utility from participating than they would doing something else, they stop participating in the market and go and do that something else. Also, if a non-participant in that market can increase their utility by starting to participate, they promptly do so (no transaction costs and infinite knowledge, of course). As I stated before, I don’t believe that any real world markets are perfectly efficient, the model simply has its uses for understanding real world markets.

Incidentally, economists talk about utility, not happiness. If someone doesn’t value being as happy as much as some other goals, eg acting for the greater glory of God, then it is at least theoretically possible for them to knowingly chose to act in ways that reduce their happiness. The point of using the word “utility” is to avoid arguments about what ends people aim at.

Dsquared, I’ve read the article you pointed me to – thanks. Some points:
1. The existance of liquidity rebate traders doesn’t strike me as affecting the truth or otherwise of the EMH one way or another. If the market centers want to pay people for posting orders then that strikes me as the same in concept as paying for an interior decorator to make your office look fantastic so you can lure people to meetings. Neither may be a wise use of your money, but I don’t see how someone could devise a trading scheme to beat that.

2. Predatory Algos and Program Traders – why do people keep writing institutional algorithms that are open to such manipulation? The article itself ends by recommending not doing that.

3. The AMM by learning hidden information, aren’t they increasing the efficiency of the financial market? I still don’t believe in the strong form of EMH, in fact my reading around this has started to convince me of the relative merits of the adapative markets hypothesis, but the AMM strikes me as improving the efficiency of the financial markets relative to whatever they were before (I’m using efficiency here in something like the engineering sense of output/input, not in the sense of perfect efficiency).

IANAE - boy, am I NAE - but it seems to me that you’re still using ‘perfect’ in two different senses, and covering the gap by using two different senses of ‘normative’. You said:

“in a market that clears perfectly the marginal participant should be making enough money to compensate for the time they spend given their other opportunities, and any extra pains from their occupation”

But a perfectly efficient engine isn’t one that consistently gets me to work on time – it’s a description of how the mechanism functions, not what outcomes it delivers. Similarly, I don’t see how it can be in the definition of perfect efficiency for a market that any participant makes “enough money”. It seems more likely that perfect efficiency would mean that nobody has any kind of edge, since there are no imperfections or pockets of localised knowledge to exploit – whether they’re happy about that state of affairs is secondary, surely.

The EMH says that financial markets are the best possible guide to the value of economic assets and therefore to decisions about investment and production.

This is not strictly so. EMH merely says the price of an individual security at any point in time reflects all known public information about that security. Therefore, at market level, an individual, on average, cannot outperform the market, because that individual could not possibly have all the information that has made up these market prices. Of course, the corollary is that the best way to outperform the market is to have inside information that is not public. So you buy then, and sell when that inside info becomes public.

Now, of course when you get into 2nd year economics when assumptions start being relaxed, you find that the rules for small cap stocks, illiquid stocks, and so on have additional or different informational qualities.

It is no longer 1975. We all know about asymmetric information , transactions costs, and so on. Even Harvard Business School has taught its MBA graduates the EMH is crap for over a decade.

Also, I don’t know anybody – trader, investor, manufacturer – who makes their investment and production systems assuming that ” financial markets are the best possible guide to the value of economic assets”. Sure, calculating your cost of capital might involve theories that assume some form of EMH, but most of your decision-making would come from company reports and market intelligence, cash-flow analysis, etc.

I don’t know how we could measure whether everyone would be driven down to their own cost price in a real market because I don’t know how we would measure “own cost price”, so I’m unable to answer your other question. Of course the fact that I don’t know how to do something doesn’t mean that it’s impossible, I could just be showing off my ignorance yet again.

John Quiggin – that’s good to know.
D-squared – thanks for the link.

In order to refute this, is it not necessary to propose a superior guide to the value of economic assets? I don’t believe the EMH claims that market prices are accurate, only that they are relatively better than other methods.

The existence of bubbles do not disprove the EMH. If assets are obviously overpriced, why would wise investors not make long bets that the price of assets will fall? I suspect it is because they don’t know the when or by how much prices will change any better than those running up the prices.

Warren Buffet lost billions betting that the US dollar was overvalued. Almost all commentary I read supported his position based on the fundamentals. Finally after his contracts ended the dollar went down and everyone said he was right about the fundamentals but simply had the timing off. Then the dollar shot up even higher and now everyone “knows” it is overvalued. Is this a bubble? Is there a better guide to the future value of the dollar than the current value? If so, put your money where your mouth is and make some cash!

In this case, the standard “hard-nosed” analysis is one that completely ignores the economic character of the activity (a part of the entertainment industry) and acts as if the true professional is one who wants to win the game/series/retain Ashes at all costs. I was just pointing that out, in an overstated fashion as you note.

Obviously, there are plenty of reasons why the outcome in (say) tomato markets doesn’t reflect consumer preferences, but it would be silly to discuss tomato producers as if their primary concern was winning a Tomato of the Year award or similar.

Obviously there are known, and probably unknown, unknowns about what is going into the finished product, but just based on looking at this and the intro from the point of view of one whose economics is pretty rudimentary, I think you might want to step back a bit and deal with some of the concepts involved, like (path-dependent Pareto-) efficiency, information (perfect and otherwise), price-taking (everyone a price-taker! There’s a nice bit of fetishism.), perfect competition (with little bits of tinkering to provide a modicum of realism), revealed preference, etc etc.

Otherwise, to my eye anyway, the broad sweep of neoclassical macroeconomic doctrine – and its counterparts – tends to come across as just so much free-floating jargon (i.e. more so than it actually, is I suspect).

Obviously you don’t want to (or rather aren’t here trying to) write an economics textbook, though judging from what I was given for A-level and during my brief undergrad flirtation with the idea of reforming the subject from within, a good one is sorely needed – ‘Economics as if people/reality really mattered’. Actually that might provide an alternative genre of titles – cos if the original allusion isn’t accessible enough (and I certainly didn’t get it), the main appeal of that whole range of possibilities seems to me to be reduced significantly.

But anyway, in the finished collection anyway, you might want to/are planning to address some or all of those dogmas-dressed-as-definitions-or-methodologies – and the order in which that is done could help a lot with exposition.

The EMH seems pretty wierd to me – like a lot of economic theorising, rather convoluted and either a strong vacuity/tautology suspect, or a pretty onviously silly assertion that no one can beat the market because everyone already knows everything. Isn;t it basically a distributed crowdsourcing sort of idea, combined with the assumption that people will only reliably aim at accuracy if it’s in their own self-interest? It could certainly do with some very detailed unpicking (not suggesting that no-one has thought of that, nor more relevantly that you haven’t.)

Shorter: non-economists are being chucked in the deep end.

FWIW, some sort of highly perspicuous running analogy might be useful (and a magic pony).

And titles – The Dead Hand of Adam Smith? Dismal Dogmas? Holding Economists to Account? Just brainstorming, not suggesting these are much good.

& BTW (not entirely germane), there is a prevalent tendency on the part of economists to ignore the sheaf of caveats that they keep in the bottom drawer, apparently for the sole purpose of producing them occasionally in misleading rebuttal of allegations that they haven’t taken them into account. I’m probably wrong, and don’t mean to be impudent, but something rather similar seems to be exhibited here on the Ashes thread: As regards unsporting behavior and so on, it all depends on what the fans want. If we cheer on our side when they engage in sledging and time-wasting, we’ll get more of it. If we cheer the player who walks without waiting for the umpire, that’s what we’ll get.

I think the simplifications involved in the no barriers, perfect info, non-tuism, zero transaction cost, instantaneous, no externalities etc. model are just too easy to accept, so that the (acknowledged) complications which lead far, far away from the idealised version have to shout very loudly indeed to get incorporated into any given analysis. Which is why, even though it might be possible to produce a realistic model along ‘ideal efficiency/general equilibrium + concessions to reality’ lines, the odds are always stacked against it actually being done properly – and a different approach is IMO needed. Unfortunately I haven’t got round to developing that just yet. Maybe for your next book?

I stand in awe of how much efficiency, in the engineering sense of that word, such activities add to the ability of the human species to feed and clothe itself, preserve its planet, and prepare for its future. I mean, there are just ALL KINDS of decisions involved in farming, manufacturing clothes, pumping oil, housing the elderly, etc that require changes in strategy, tactics, management, and capital investment every microsecond or so and which can go horribly wrong in the 75ms a ping packet takes to travel from Farmville Iowa to New York City.

Not Really

Not, in case that isn’t clear.

That looks like two different meanings of the word ‘perfect’, one descriptive and one normative. Put it another way: is it possible for public information to be so good, and competitive pressures so high, that everyone trying to sell gets driven down to their own cost price? And even if it’s not possible, isn’t that one definition of a market that clears perfectly?

I would definitely advise further research before entering into the highly competitive and complicated world of algorithmic trading. The Themis Group White Paper is a good place to start in getting a handle on how technical these things can get. Basically, the problem is a) limited capital to place positions and b) neither markets nor computers are black boxes – everything depends on the very exact market microstructure of how an order gets filled, and on the way in which the computer is set up to minimise latency (Counterstrike players have absolutely nothing on algo hedge funds when it comes to whining about their ping times).