I think this is one of many bits of data that I add to support my personal hypothesis that public equity markets are increasingly failing at their job of being an efficient way for our economy to allocate capital.
They almost certainly can't be too efficient. There's a proof that if markets are efficient then P = NP.
This isn't really about whether markets are or can be efficient. This is about corporate governance. Boards of directors should not permit CEOs to play these games, and compensation of CEOs should be structured to not create such destructive incentives.
Beware of "perfect" as a requirement. For instance, if you can do perfectly accurate calculations with real numbers in constant time, that also leads to P=NP (because you can encode entire problems in very long strings of digits and solve them with a few operations.)
You can have efficient-enough markets, like ±0.001% efficient, without requiring P=NP.
Perfect efficiency does not exist. Some efficiency does. There are three variants of the hypothesis: "weak", "semi-strong", and "strong" form. [1] Insider information isn't some conspiracy -- people have gone to jail over it year after year. With information disparity, efficiency is always going to be a fraction of what it could be. Just think about two parties -- one thinks the stock should be high because of insider information. And lets say they are also mega-rich big bucks. The other party is people without the insider info, they think the stock should be low. So Mr. Mega-Rich buys a lot of shares, the stock goes up. Then it falls back down because no one else wants to buy it at that high price. These oscillations will continue as Mr. Mega-Rich takes advantage of his insider information - and the more trading volume on the buying end, that he provides, the more choppy the market for that equity will be. If Mr. Mega-Rich is continuously buying, that choppyness won't go away - lest other parties take his actions to be information itself, and adjust their appraisals.
Wall Street I and II aren't the most accurate pictures, but they certainly aren't fiction. If you aren't the shark, you're getting sharked. But because of 21st century prosperity, you still might be happy with your returns, even while paying the tax of giving money to those with better information.
People say this a lot, but I think it's misleading to most who hear it. This is certainly true in a trading context (i.e. where the buy/sell time horizon is short enough where the underlying asset's economic value doesn't change, though the market price might), but not in an investment context (where it's expected that the inflow of capital will create economic value and the stock price will reflect that).
The takeaway is that you should leave daytrading to the pros, and instead bet on humanity improving the means of production over long time horizons. Unless you're actually a pro and have good reason to believe you can beat the market.
I think it's not efficiency in the sense that you're talking about, it's that the picking and pricing of stocks as a signal of their effectiveness back to the company that is broken.
Does the apparently dubious proof assert that there are exactly that many functions? Because if the assertion is rather that there at least that many, that assertion is true but really understating the case. E.g. for n=3 there are 2^(2^n)=256 functions, and after that it really blows up.
It gives that as an upper bound on the size of the search space of boolean functions with n-bit input, representing technical strategies that consider the last n market movements. It goes on to claim that if there is a historically winning technical strategy, a nondeterministic Turing machine can write out a full description of that strategy in polynomial time.
The remaining problem is how to uniquely describe in polynomial space an element of a set with doubly exponentially many elements, but the authors don't bring that up because they're working under the assumption that there are 2^n functions rather than 2^(2^n).
The assertion that the efficient market hypothesis is equivalent to P=NP also comes with no proof that future performance of a trading strategy will match its past performance or that anyone will actually deploy a perfect strategy if one exists.
The first step in correcting a problem isn't proposing an alternative. It's identifying the problem.
Given that the powers that be clearly have no consensus on the existence of the problem (nevermind that they may have a vested interest in not identifying the problem), and that identifying a problem, while non-trivial, is typically orders of magnitude easier than identifying a solution, it's a little disingenuous to dismiss the point by asking for a solution. Just because the solution is unknown doesn't mean we can't discuss the existence of the problem.
The promise of public markets isn't perfection though, it's just "better than the alternatives." Austrian economics doesn't claim that short-term inefficiencies do not exist in markets, but rather that they are less harmful than attempts to fix them.
[edit]
The sibling comment that asks if the markets are failing more than in the past is more productive because if they are that means that something is different; I personally am disturbed by the degree to which "a rising tide lifts all boats" is no longer true.
I wonder if it's because the wealthy and powerful have gotten better at cornering the market on gains (which would beg for some check on their power), but I also worry that it's actually because material growth has actually stalled, and the market gains are mostly illusion (which would require some other correction).
I wonder if rent-seeking and other forms of non-productive passive income aren't at the root of it all. We're generating wealth and value and "growing" the economy, but how real is that growth when a) nothing is actually being produced (i.e., there's no product with inherent value), and b) wages don't increase anyways.
All this wealth is being generated, but it's going disproportionately (historically speaking) to the wealthy. Well, the problem with that is that the wealthy have an extremely low marginal propensity to consume, so, all they do is invest. That means that both more wealth is available to build businesses and less wealth is available to purchase the resulting products. So, it's easier than ever to create a business because investment dollars are cheaper than ever, but it's harder than ever to make it profitable because nobody's buying. That should depress costs, but the race against inflation has got to catch up eventually. You can keep making it cheaper and cheaper to run a business, but if wages don't increase and cost of living continues to increase just due to inflation, then it doesn't matter how cheap your goods are. There's no middle class left with the money to purchase luxury goods. Then what?
Edit:
> Austrian economics doesn't claim that short-term inefficiencies do not exist in markets, but rather that they are less harmful than attempts to fix them.
This seems -- at least partially -- bogus on it's face. Sure, you can definitely regulate a market into collapse (see the Soviet Union) but deregulation can just as easily create supermonopolies which are equally unjust and ineffective. The point of the market isn't to generate wealth, it's to distribute resources efficiently so that the nation as a whole (or state, or world, or whatever collective noun you want to use for human civilization) gets their required products. Generating wealth is a complete side effect of the market as far as the benefits to society are concerned.
I assume you are at least passingly familiar with Piketty's Capital? He argues that as long as the rate of return on investments is higher than economic growth, wealth will concentrate, and the 2nd half of the 20th century was so equalizing because of rapid post-war growth.
This makes a lot of sense, but it doesn't automatically imply unilaterally implementing a wealth tax will solve the problem. It's not like there aren't a lot of people who want to tax wealth rather than income, but it is much harder to hide income than wealth (and yet we still have people who do so to a large degree).
> Austrian economics doesn't claim that short-term inefficiencies do not exist in markets, but rather that they are less harmful than attempts to fix them.
We've had these discussions for hundreds of years. You really think the powers that be will reach a consensus on this issue if we talk about it some more? The only path forward is really to suggest reforms or an alternative imo and convince people it's superior. Basic Income / Negative Income Tax / changes to corporate governance / whatever.
> You really think the powers that be will reach a consensus on this issue if we talk about it some more?
Do you really think ignoring the issue or remaining silent will fix the issue? Do you know what happened historically when wealth distribution and societal demands got wildly out of balance? Crime, famine, revolution, war, and death.
That only works if the equity price is a sufficiently correlated representation of the fundamental efficiencies of companies. That's the failure.
If CEOs are managing to manipulate the stock price for personal gain having nothing to do with how efficiently that company manages actually delivering products and services then that's a failure of the market as an economic allocation tool. Choose winners and loser stocks all you want - you can engineer a profit, but all that work might not do anything as a signal for companies to manage their fundamentals better. Now - you can argue about if this is just a short term influence and it doesn't matter because maybe it all still balances out in longer terms. But I lean towards thinking it's a systemic long term disconnect problem.
It's not. It is, at best, a short term incentive problem. The publishing of articles like this one is how this incentive problem corrects itself.
Consider: you are an investment manager considering allocating capital. You know that company A is engaged in short term manipulation tactics to the detriment of their long term business prospects, and company B doesn't care at all about short term market movements and focuses entirely on growing and sustaining their business. You also know that there was this McKinsey study recently that showed that companies like company B earn higher market returns long term. Where are you going to allocate your capital?
Now, I grant you, some investment managers are under pressure from their clients for short term performance. And that can skew the results here. But in aggregate, over time, the machine is efficient, and it will eventually price these things correctly. And in so far as i'm aware, there is no better known mechanism for allocating capital.
Now that this problem is published, to the extent that it's true, you, I, and anyone else who wants to can make it go away by allocating our own capital accordingly. And the nice thing is that we get compensated for doing so in the form of superior returns.
Because of buy backs the net of amount of investment financed in the US stock market is less than zero. The purpose of equity markets is not to raise capital, it's for equity owners to cash out.
Yes, if one looks at productivity numbers in the last decade, as well as more indirectly via new business creation, perhaps at lifetime of businesses, and at an even wider scale if one looks at inequality (failing at distributing of wealth). Just to name a few factors I'm thinking about.
