Bubbles and the bias against shorts

Keynes’ words have become one of the most tired clichés in finance. “The market can remain irrational for longer than you can remain solvent.” I can’t find date attached to the quotation, but since Keynes died in 1946, it’s a fair guess that he uttered those words at least 60 years ago. Yet, his words are just as true today as they were then. That, I claim, is a blot and a shame on the profession of finance.

Markets are human institutions, evolved and invented to serve an incredible social purpose. They collectively are tasked with deciding how we make use of all that is precious, so that we productively employ rather than squander our providence. The financial press often treats markets like sporting events, like a kind of casino, or a thermometer for the “economy”, like all kinds of things. But markets are not weather, nor idle wagers. They guide what people do and do not do. They determine what is built and created, and what is left merely imagined. Markets guide us in deciding whether to learn computer skills or carpentry skills, whether to drive or to walk to work, where we will live and whether or not to have children. Markets are mind, we are all neurons. We are also the fingers and toes that do their bidding, individually rebellious but statistically obedient.

Update History:
  • 31-Mar-2006, 5:08 a.m. EET: Miscellaneous small clean-ups, added sentence about how no means of profiting from identification of bubbles implies that bubbles can grow unrestrained.
  • 21-Apr-2006, 10:53 a.m. EET: Some small clean-ups, removed phrase “with very diverse portfolios shorting low-cash-flow instruments” in description of investors who can take advantage of positive asymmetries in shorting, because the sentence was too long and wordy.

6 Responses to “Bubbles and the bias against shorts”

  1. HZ writes:

    I think Keynes said it during the inter-war years. He was a currency trader on the side (never mind insider info, conflict-of-interest and such modern notions). He shorted GBP at times. Made a bundle eventually but went (almost) broke once or twice in the process. Part of the reason why Keynes was a lot more practical than more academics.

    The upwards bias is rooted in human need to see nominal growth. For a safer short you need an instrument that let you bet the real value of sth going down. Otherwise shorts eventually lose unless the target go kaput.

  2. HZ — There’s definitely something to be said for academics who have faced the discipline of the market…

    The bias against shorts presented by inflation is material, although I think less offputting than the path-dependence of short-position sustainability. All forms of investment are time-dependent. A buy-and-hold long investor who identifies a stock that doubles in 2 years has earned a living. A buy-and-hold long investor who identifies a stock that doubles in 8 years when ten year bonds are paying 9% has purchased extra risk for no additional return. Inflation means that time is even more critical for a short: A short investor who correctly identifies 20% overvalued stock finds her purchase price only 10% overvalued after 3 years due to inflation, if the stock hasn’t reverted yet to its “value”. Stocks are after-all pass-through securities that automatically increase in nominal value as the nominal value of a company’s revenue stream increases. This indeed makes life much harder for shorts than for longs; shorts pay inflation twice, first in diminished capital gains (or increased loss) due to inflation, then in diminished purchasing power of any eventual gains. But longs and shorts both are willing to risk less-than-stellar or even slightly negative returns based on the uncertainty over when a security will find the investor’s estimate of its correct price. Only shorts have to face the prospect of being forced to liquidate their positions at a very large losses, when they remain confident in the correctness of their positions.

    Regardless, I do agree that resolving the inflation bias is important. It’s worth noting that large, creditworthy investors who can put the cash proceeds of a short position to use without penalty can hedge the inflation loss by going long TIPS. But under the (really disadvantageous) terms available to an individual short investor, at least in my experience, this kind of hedging is impractical. (One way or another, the brokers I’ve worked with skim away a significant fraction of the cash flow that should be available from the proceeds of the short-sale. In fact, I often pay net interest on accounts full of cash from short-sale proceeds, as the broker pays me below-market rates on short proceeds, but considers the liability represented by my short positions an expensive margin loan. Perhaps there are short-friendly brokerages out there that diminish these costs. If so, I’d like to know!)

    There are lots of institutional biases and extra costs that discourage shorts. An inability to effectively hedge the inflation risk is certainly one of them, as are high implicit brokerage costs and the risk of being forced to cover or liquidate during a transient spike. Our markets do effectively discourage shorts, and thus habitually blow bubbles.

    BTW, thanks again for continuing to drop by and comment!

  3. HZ writes:

    Steve, it is nice to have a quiet corner for a chat!

    You could buy puts but that is even more time-event dependent. Or you could short and buy out-of-money calls for insurance, which adds more cost. It is not easy to find out exactly how my brokerages compute margin interest in the case of short positions. With all the things we are talking about that are stacked against shorts, I never bothered to try. I believe even a smart guy like Chanos only eked out < 3% over the long term vs > 10% for S&P.

    I think even large players pay a hefty interest to borrow shares. Been wondering why that is the case. Maybe they have a side agreement that the one loaning the shares won’t call back the shares within certain amount of time and so gets paid for that promise.

  4. “I think even large players pay a hefty interest to borrow shares. Been wondering why that is the case. Maybe they have a side agreement that the one loaning the shares won’t call back the shares within certain amount of time and so gets paid for that promise.”

    You may well be right — I really don’t know. If it’s the case that shorting is expensive for everyone, though, then lots of the theoretical basis of derivatives pricing, which depend upon the availability of low cost shorting to create arbitrage constraints, would be expected to work very poorly in the real world. Futures prices on liquid securities, for example, do hew very close to zero-arbitage prices that presume an arbitrageur can short with no costs beyond covering the cash flow due the party who lent the security. This implies arbitrageurs somewhere can short at this theoretical cost, or else that there’s a tinkerbell effect, where market participants are so convinced of the theory never test the arbitrage bounds. I’d bet the latter, that somebody can and does short cheap, constraining the price.

    BTW, I do think, given real markets, shorting usually is a bad strategy (except as a means of “underweighting” in a generally long portfolio). Markets are biased to usually appreciate in real terms, and as you point out, inflation in the currencies that underlie them further augment the upward drift. But, “usually” is not always. Rational or no, we’ve created a financial economies thatt appreciate beyond sustainability and then pop, leading to massive destruction of financial value. (This may not be all bad — Maybe the creation of real capacity during the inflations is larger than the real-economy writedown of malinvestment during the pops. A case can be made for a world where the irrationally exuberant create madly and then go bankrupt in the bust; the world keeps their creations, and they create again to dig themselves out of the hole. Maybe that’s “good”. It’s not a very comfortable story, though, and very ineffecient.) With markets biased to inflate, shorting is like gambling at a table rigged against you 90% of the time. It’s the only arrogant presupposition that you can time the crash that ever makes it worth trying.

  5. HZ writes:

    I think derivitives are different. Options can be naked, there is no need to own underlying securities. Buyers of equities, OTOH, get both cash flow (as you indicated) and VOTING RIGHTS whether they buy from short sellers or plain vanilla sellers. While most investors may not care for the voting rights, they are finite and can’t be created out of thin air. It means that the ones loaning shares must give up those rights. (Maybe for a very short term short away from any proxy dates this is not important, creating opportunities for naked shorts — that is you borrow without the knowledge of the owner but then you have to cover and return the shares by next proxy date. I suspect for retail shorts the brokerages have to finagle sth behind the scenes to get all the accounting right.) I don’t know if this explains things as I don’t work within the securities industry. I did read story about how Buffet charges a significant interest in loaning out his USG shares to shorts a couple of years ago.

  6. HZ — No argument. It’s just that derivative pricing depends upon the availability of ordinary short-selling as an arbitrage strategy, in which shorts pay no more than the cosh-flow due the security holder. I agree that derivatives are a whole different ballgame than shorts. It’s just if it’s really true that shorting is expensive for everyone, than derivatives pricing makes no sense.

    (Naked shorting is usually illegal, which is makes sense for voting shares, but highlights an inconsistency with respect to other instruments. By what rationale is it illegal to naked-short a debt instrument, but okay to conjure credit-based derivatives whose notional principal far exceeds the size of the primary debt issuance? If there is a problem with “third party bets” creating misalignments of incentives and credit risks that harm the markets for base securities, than the scale of derivative markets should be regulated. If there aren’t such problems, naked shorting of pure cash-flow based instruments, and non-voting securities parallel to stock created by naked shorts, should be permissable. I think there are problems with having the scale of some financial markets be wholly out of sync with the actual real-economy financing they are based on, but I don’t think the question of how derivative markets ought to be regulated is well understood. We may have to see something break before we understand how we might have fixed it.)