Sunday, March 30, 2014

Liquidity everywhere

A few weeks ago I claimed that the so-called value premium was really just a liquidity premium. The value premium, illustrated best by the HML, or high-minus-low strategy (shorting stocks that have high price-to-book ratios while buying stocks that have low ratios), is one of the more well-known market anomalies. By following this strategy, investors can supposedly do better than their counterparts on a risk-adjusted basis.

My point was simply that stocks with low price-to-book ratios get those low ratios in the first place because they are illiquid relative to stocks with high ratios. Anyone who buys the former while shorting the latter is acting as a liquidity creator for which the HML return is a reward. Fund managers who uses this strategy to drive fund returns aren't necessarily earning alpha, they're earning a fair return for acting like a liquidity-providing bank.

I got some push-back in the comments, on Twitter, and on Reddit from readers, some who were skeptical that the value premium could be a reward for bearing illiquidity, and others who were unhappy with my lack of data. I poked around a bit. Here are two empirical papers that try to describe the value premium as a liquidity phenomena: Time-Varying Liquidity Risk of Value and Growth Stocks (Akbas, Boehmer, Gene, Petkova; 2010), and A Liquidity-Augmented Capital Asset Pricing Model (Liu, 2006). Eat your heart out, folks.

But the main point of this post isn't to beat around the HML bush. My basic strategy in the last post was to take an abnormality and explain it by resorting to an unseen liquidity factor. I'm going to wash and repeat this strategy a few more times today. Liquidity is an invisible vector, or a missing plug variable, that can be used to explain all sorts of puzzles, anomalies, abnormalities, oddities, and weirdness. It's sort of like the Force, it's all around us.

The size premium is a market puzzle whereby small firms outperform large firms on a risk-adjusted basis. Again, we can introduce liquidity to explain at least part of this anomaly. Since the shares of small firms will typically be less liquid than the shares of larger firms, anyone who buys the former and shorts the latter is creating liquidity, an activity for which they should be duly compensated.

Torchio and Surrana (2013) reconcile the size premium with liquidity in this paper, noting that the size premium subsumes a liquidity premium. In the case of the smallest stocks in their study, the majority of the size premium is entirely explained by a lack of liquidity.

Then there's the mother of all pricing puzzles, the equity premium puzzle. Studies show that the return to holding stocks in the S&P 500 is far too high on a risk-adjusted basis relative to the return on U.S. treasury bills. This anomaly seems like a no-brainer to me. You can't compare the pecuniary return on stocks to t-bills because the two asset classes are like apples and oranges to each other. The U.S. Treasury bill is one of the most liquid assets in the world. By comparison, the average stock in the S&P 500 trades by appointment. If t-bills have seemed to underperform equities over the decades, it's only because t-bills provide a compensating stream of liquidity services that equities don't. When all is said and done, there's a good chance that once we account for liquidity returns, the total returns of equities and t-bills balance out.

Nor would I be the first to make this claim. Amihud (2002) [ungated version] tries to resolve the equity premium puzzle with a liquidity explanation, noting that the equity risk premium is
in part a premium for stock illiquidity. This contributes to the explanation of the puzzle that the equity premium is too high. The results mean that stock excess returns reflect not only the higher risk but also the lower liquidity of stock compared to Treasury securities.
Another interesting anomaly is the closed end fund puzzle. Closed end funds issue non-redeemable shares to the public and use the proceeds to invest in assets like stock or real estate. Oddly, shares in closed-end funds often trade at large premiums or discounts to underlying net asset value. Once again, liquidity seems like it could be a decent explanation. If the underlying assets that the fund invests in are highly illiquid, but the share units themselves are highly liquid, then those units provide an extra stream of liquidity services and should therefore trade at a premium to illiquid underlying assets. This premium could turn to a discount as the liquidity profile of underlying assets improves, or the liquidity return provided by share units deteriorates.

A neat paper that uses liquidity to explain the closed end fund puzzle is A Liquidity-Based Theory of Closed-End Funds (Cherkes, Sagi, Stanton; 2007).

Back in the early 1980s, Robert Shiller posited an excess volatility puzzle. The price of equities seem to fluctuate far more than one would expect based on the dividends that they are expected to pay. Let's introduce liquidity once again. Investors value shares not only for the their pecuniary yield (both dividends and price appreciation) but also for their moneyness, or their liquidity. In calculating the price at which a share should trade at, investors must estimate not only the discounted value of dividends thrown off by the firm, but also the discounted value of liquidity services it provides. If share prices seem to volatile relative to dividends, it may be estimates of liquidity services that are driving the results.

Ravikumar and Shao (2010) try to solve Shiller's volatility puzzle by explaining how the dual role of an asset as both a good yielding a flow of dividends, and a medium of exchange, might explain observed excess volatility.

Uncovered interest parity (UIP) is the idea that an investment in t-bills in two different countries should provide the same overall expected rate of return. Say that a Canadian t-bill yields 2% but U.S. t-bills yield 4%. If people are willing to hold low-yielding Canadian debt, UIP says that it must be because the exchange rate is expected to appreciate, providing Canadian debt holders with an extra 2% forex gain to bring their net return in line with the return on U.S. t-bills. UIP says, in short, that low yielding currencies should appreciate over time.

The uncovered interest parity puzzle, or forward premium puzzle, is that UIP is almost always violated; high-yielding currencies tend to appreciate over time rather than falling. A carry trade in which an investor borrows in the currency with the low interest rate and invests in the currency with a high rate is usually profitable.

Liquidity might be able to help explain violations of UIP. If low-yielding t-bills provide a superior stream of liquidity services than higher-yielding bills, then the exchange rate doesn't need to do as much "work" in resolving the lack of interest parity between low- and high-yielding t-bill rates across nations. UIP "violations" might be no more than shadows of an invisible liquidity premia. Carry traders that make their living shorting low-yielding t-bills of one nation and buying high yielding bills in another aren't earning excess returns, they are simply acting as liquidity creators—and getting fairly rewarded for it by liquidity buyers. (See my bit on the on-the-run off-the-run trade in my previous post.)

Linnemann and Schabert (2013) try to explain UIP violations in terms of the liquidity premia on treasuries.

Lastly, there's the credit spread puzzle. A credit spread is the difference between the yield on corporate bonds and risk-free treasuries. Data shows that the credit spread has historically been far too high to be explained by risks like expected default loss. Owners of corporate bonds earn too much, and owners of t-bills earn too little.

The answer to this puzzle is similar to that of the equity premium puzzle. Risk-free treasuries are some of the most liquid securities in the world, corporate bonds are not. Because t-bills provide an extra liquidity return, they don't need to provide as high a yield. If we factor this liquidity return into the equation, then what seems like an anomalously high spread between t-bills and corproate bonds probably isn't so anomalous after all.

There are a number of papers that try to explain the credit spread puzzle by resorting to liquidity. Parraudin and Taylor (2001), for instance, find that a large part of the AAA- to A-grade bond spreads are explained by liquidity.

So yes, I see liquidity premia everywhere, but as this survey of papers shows, so do a lot of people. If you haven't incorporated liquidity into your model of the world, whether that model be CAPM or something more specific to yourself, then both your investing and your way of doing economics will probably suffer.

Sunday, March 23, 2014

Dismantling a central bank

In a previous post, I made the point that banknotes aren't Samuelsonian bubble assets, say like a chain letter or a ponzi scheme or bitcoin. Upon the dismantling of a central bank, each note has a senior claim on a central bank's remaining assets. Rather than being mere "oblongs of paper", as Samuelson described them, banknotes occupy the very top of the capital structure hierarchy, above stock and bonds. This quality of being "well backed" might be sufficient for notes to trade at a positive value in the first place, and also help explain subsequent fluctuations in the purchasing power of those notes. In this post I revisit these ideas.

The process of dismantling a central bank would go like this. Imagine that the Reserve Bank of Fiji announces that it will wind up operations next week and recall all notes. Its assets consist entirely of Fijian dollar-denominated bonds. With the eminent end of Fijian dollars, the entire Fijian economy will have to quickly re-denominate existing debts and contracts into a new unit of account, say U.S. dollars. After this redenomination, the Reserve Bank of Fiji now finds itself holding a large quantity of U.S. dollar-denominated debt. It proceeds to sell these bonds, as well as its printing presses and premises, for dollars, and then uses dollars to cancel all Fijian notes. The Reserve Bank of Fiji is no more, and neither are Fijian dollars.

Because the central bank's hypothetical dissolution would result in noteholders ending up with some ultimate quantity of U.S. dollars in their pockets, Fijians can use this terminal value as a basis for computing the present value of Fijian banknotes. They would go about doing this computation in the same that they would with a stock and bond, both of which also have hypothetical terminal values, or liquidation values.

The inestimable Mike Freimuth, who blogs here, pointed out that there is a problem with this. How can Fijian noteholders actually quantify the amount of dollars to which they are entitled upon dissolution? This is easy if the Reserve Bank of Fiji is already on a dollar standard (say it redeems notes for US$1). Once all of the central bank's assets have been sold for U.S. dollars, each noteholder would get US$1 until all Fijian noteholders had been satisfied, upon which less senior stakeholders like bondholders and shareholders would get their portion of the central bank's remaining stash of U.S. dollars.

However, if Fiji is on a floating standard, the task of valuing noteholders' claims gets quite thorny. While fiat notes may have a senior claim on assets upon dissolution, it isn't evident how many actual dollars this entitles noteholders too. And if their quota of remaining assets, or terminal value, is not known at the outset, how can Fijians arrive at a value for notes in the first place?

One answer to Mike's criticism is that Fijian note owners come up with their best estimate of how much dollars the central bank (or a judge) would decide to award them upon a hypothetical dissolution. If the market's collective guess is that no assets would be forthcoming, then Fijian dollars would be worth nothing upon windup. If Fijians on average assume that the sale of bank assets would net $100m, and they think a judge would award them half of that, then each Fijian banknote will earn a prorated share of $50m. If they think that the all assets will be awarded to them, then they'll get a prorated share of $100m.

While these expectations about the terminal value of notes would probably be sufficient to jumpstart the positive value of Fijian dollars in the first place, this isn't a very satisfying explanation for subsequent variations in the price level. After all, the Fijian price level on any given day would be a function of Fijians' many and divergent expectations concerning their as-yet-unstated share of some future pie. As noteholders take potshots at guessing what this unknown size of this slice would be, the Fijian dollar would fluctuate wildly, sort of like a penny stock. Penny stocks owners have wildly fluctuating estimates concerning their ultimate, and unfixed, share of the corporate pie, which is only determined after debtors have had their pickings.

However, the behavior of currencies is not like penny stocks—the former tend to vary only a little in price from day-to-day. To explain price level fluctuations, it would seem that something other than the terminal value of the Reserve Bank of Fiji's assets must be at work.

Here I would like to reiterate my point from an earlier post that once a banknote has been jumpstarted into having a positive value, the dissolution-value of a central bank assets will only have a distant influence on those note's subsequent value. By far the more immediate effect that central bank assets have on the value of notes emerges via their ability to be mobilized in the maintenance of price stability. By selling assets for notes and retiring them (or threatening to do so), a central bank can prevent the value of their notes from flagging. Alternatively, the income earned by those assets provides the central bank with the resources to pay more interest (on reserves at least), a feature that will also ensure price stability.

So it seems to me that any impairment to a central bank's assets will affect the value of notes not so much because it lowers their value come future dissolution, but because it limits the central bank's ability to repurchase notes and pay interest in the present so as to maintain their price target. If a central bank didn't provide a target supported by a repurchase facility, then the value of notes would be dictated by something like their terminal value—and prices would be much more volatile then they are now.

Sunday, March 16, 2014

Credit cards as media of account

What is this gas station using as a medium of account? Visa/Mastercard dollars or Federal Reserve dollars?

In this post I'll argue that in many cases, a nation's medium-of-account doesn't consist of base money issued by its central bank, but credit card money created by Visa and Mastercard. This may have some interesting implications for monetary policy. Whoever issues, creates, or manages a nation's medium-of- account determines the general level of prices, and this makes it a monetary superpower.

But before I get to that, let's revisit the meaning of the word medium-of-account.

I've written a number of posts on the idea of medium-of-account because it has always seemed to me like an important concept, although admittedly it's taken me a while to zero-in on a satisfactory understanding of the term. What I like about medium-of-account is that along with the ideas of unit-of-account and moneyness, it allows us to pretty much remove "money" from the list of terminology we use when talking about monetary phenomena. No single word is so widely-used yet so imprecise as money. And because of this, no word has bred as many bitter econblog battles. By splitting apart the various ideas associated with "money" and passing these meanings on to alternative words like medium-of-account, some of this morass can hopefully be unclogged.

Without further ado, here are the definitions. By the way, these aren't mine. I've picked them up from folks like Jurg Niehans, who coined the term medium-of-account; Scott Sumner; and Bill Woolsey—hopefully nothing has been lost in translation.

The unit-of-account is a word or symbol like $, ¥, £. Inherent in the idea of UOA is the subdivision of the unit, so that $1 is comprised of 100 cents. (1)

The thing (or things) that defines that unit is (are) the medium-of-account. When a merchant chooses to sell a painting for $100, for instance, he is selecting the unit in which he prices, say the $, as well as the specific medium that defines the $ unit. This last choice is important because dollars might appear in any number of different mediums, or forms, including Federal Reserve paper money, Federal Reserve deposits, branded private bank deposits, cheques, credit cards, and more.

Isn't it the case that a merchant chooses "all of the above media of account" when choosing to price in dollars? After all, one dollar is just as good as another.

Not necessarily. For instance, we know that in the early to mid-19th century a plethora of dollar-denominated exchange media circulated, much like now. There were dollar coins, which the U.S. Mint coined out of a certain number of grains of silver and/or gold. There were also privately-issued dollar banknotes, these being the most prevalent exchange media since coins rarely circulated. However, when merchants set sticker prices, the medium they had in mind when defining the $ unit was the less-common coin, not the more-prevalent notes. Notes were only accepted by merchants at varying discounts to their face value, despite the fact that most banknotes were branded as "dollars".

For example, if our merchant listed a painting for $100, then it could be purchased with one-hundred one dollar coins, or, alternatively, $102 in banknotes from a certain bank, or $103 from another.

If the value of all banknotes simultaneously inflated, what would happen to prices? Given that the value of the coin had remained constant, the merchant's price as well as the general price level would not have changed during this inflation. All that would adjust would be the varying discounts applied to the whole range of private banknotes. The painting would still be listed at $100, and it could still be purchased with the same quantity of coins, but it might take $110 or $115 worth of notes to purchase it.

However, if the U.S. Mint had chosen to reduce the quantity of gold or silver in a coin, then the merchant would increase his sticker price for the painting to $110 or so. The general price level would inflate.

So a unique feature of the medium of account is that the general price level pivots around the MOA's value. If the owner or issuer of the medium of account, in our example the U.S. Mint, has the wherewithal, it can control these economy-wide price changes by modifying the nature of the media it emits, say by reducing the metal content of coins. Few institutions have this sort of monetary superpower because only a few institutions create media that also happen to be media-of-account. Because 19th century private banks didn't issue media of account, they were not monetary superpowers.

Let's bring this back to the present. What is the modern medium of account? Who controls it and thereby earns the mantle of the U.S.'s reigning monetary superpower? Scott Sumner argues that central bank base money serves as the medium of account. I don't doubt that he's right. But in a large subset of transactions, I'd argue that Visa and Mastercard dollars are the medium of account. And this means that Visa and Mastercard rival (in theory at least) the Fed as monetary superpowers.

To understand why Visa and Mastercard dollars serve as media of account, you need to know a bit about how credit cards work. Merchants who accept credit cards as payment must pay a small percentage of each transaction's value to the credit card networks (comprised of the Visa and Mastercard associations, plus the banks that issue cards and process payments). So if someone buys $1.00 worth of stuff, the merchant might get $0.995, the remaining half cent going to the card network.

The fee that the merchant must pay varies by the quality of card. Basic cards might result in the merchant giving up 0.5% to 1% to the card network while premium cards, those offering better rewards, might bring a fee of 2-4%. Merchant fees have been rising over time, especially as card rewards become more exotic.

Merchants hate seeing credit cards, especially premium cards. They hate them because they are required to pay the card fees but cannot pass these costs off to the customer. Why? Well the best way to pass these costs off would be for the merchant to put a surcharge on each credit card transaction equal to the fee the card network charges the merchant. A surcharge policy would mean that it would cost any customer wishing to buy a $100 painting with Visa or Mastercard $102 or $103.

However, as a condition of using the card networks, merchants are prohibited from discriminating against card users. Surcharges are 'illegal'. Visa and Mastercard can extract these sorts of promises from merchants because they are oligopolies. If you are exiled from their networks for breaking their rules, you're as good dead.

So the upshot is that if a customer buys a $100 painting with cash (or debit), the merchant gets $100; if they buy it with a basic Visa card, the merchant might get $98; but if the customer buys it with a premium card, the merchant will only get $96. I'd hate premium cards too if I only got 96 cents on the dollar.

To get around these rules, merchants who accept cards have come up with an ingenious strategy: change the medium of account. Basically, the unit of account that merchants use, the $, stays the same, but whereas the merchant's original medium of account was Federal Reserve dollars, they now switch over to defining the $ in terms of Visa/Mastercard dollars. In the eyes of a merchant, a credit card dollar is only worth around 97 or 98 cents. Having adopted Visa/Mastercard as his MOA, our merchant will proceed increase his sticker prices by a percent or two across the board. The painting which retailed for $100 is now priced at ~$102. When someone buys the painting with a credit card, two dollars of this amount goes to the card network, leaving the merchant with $100. He earns the same real income as before.

This switch in MOAs allows our merchant to inflate their prices and thereby pass off card fees to their customers without illegally imposing surcharges. Fed cash has ceased to be the MOA, but will still be a popular exchange medium. But now customers who prefer paying in cash must request a cash discount at the merchant's till. Given the $102 sticker price on the painting, they should be able to buy the painting for around ~$100 Fed dollars.

Since Visa and Mastercard now manage the medium of account for a large proportion of American merchants, they have become monetary superpowers and can exercise their own brand of monetary policy. If Visa and Mastercard increase the rewards on their cards, merchants will be docked larger fees. Merchants will react by increasing sticker prices across the board. Thus we get inflation. If rewards are lowered so that the merchant is penalized less, then merchants will lower their sticker prices. This is deflation. These price changes are independent of any action taken by the Federal Reserve.

That's not to say that the Fed would have lost its monetary superpowers. It can still cause inflation or deflation by engaging in open market operations are adjusting the interest rate on reserves. However, in an extreme scenario, we could imagine the Fed's effort to increase prices being offset by Visa and Mastercard's efforts to decrease prices. A monetary battle of sorts could erupt.

I think that a good analogy to help understand this is to return to the 19th century example of dollar coins issued by the U.S. Mint. If gold prices rose, the price level would fall. But if the U.S. Mint were to simultaneously reduce the gold content of dollar coins, the MOA, it could entirely offset this fall and create stable prices. Just like the U.S. Mint can offset any change in the value of gold by increasing or decreasing gold content of coins, Mastercard and Visa as issuers of MOA can (in theory at least) offset any change to the value of Fed dollars by increasing or decreasing the reward content of Visa/Mastercard dollars.

An interesting bit of news worth pointing out is that in 2013, Visa and Mastercard finally allowed U.S. merchants to introduce surcharges on credit card transactions. I'd expect that merchants will slowly start to transition back to using Federal Reserve dollars as their medium of account. We should see the various types of credit cards dollars being priced at varying discounts to Federal Reserve dollars, similar to how banknotes in the 19th century were priced, with each note earning a discount relative to the dollar coin. Premium cards will face large surcharges, and regular cards small surcharges.

This means that in the U.S., Visa and Mastercard have effectively lost their monetary superpowers. They can no longer effect the general price level. In other places like Canada, however, courts have allowed the no-surcharge policy to continue, which means that Visa and Mastercard dollars will continue to be MOA. The card networks will remain as Canadian monetary superpowers.

There's a lot more material that I'd like to add to this already-dense post, but I'll hold off for now. In sum, in this post I'm "kicking the tires" of the basic definitions that folks like Scott Sumner and Bill Woolsey provide us. In applying them to the world around us, it sure seems to me like credit card media-of-account currently coexist with the standard Fed dollar medium-of-account. But I'm curious to see if others agree with my interpretation.

P.S. Here are two interesting tangents I plan on writing about next month:

1) A bimetallic monetary system has two media of account; gold and silver. When the market rate between gold and silver shifts, the system suffers from Gresham's Law. If we have a monetary system that uses Federal Reserve dollars and Visa/Mastercard dollars as the two media of account, what does a modern version of Gresham's Law look like?

2) The Fed gathers price data so it can better target a 2% decline rate in the CPI value of the "dollar". But if some merchants are pricing goods in terms of a different dollar medium of account, isn't the Fed gathering inappropriate data? If credit card networks are pushing up prices via fee increases, the Fed might misinterpret these changes as being Fed-inspired and adopt the wrong monetary policy. How might the Fed adjust its methodology to account for the use of credit card MOA?

(1) As Tom Brown points out, some economists describe the unit-of-account not just as a sign, but also as a fixed quantity of the medium-of-account. So if the unit of account is the $, and the medium-of-account is gold, than the number of grains of gold that defines the dollar is rolled into the concept of unit-of-account. Alternatively, we can leave the unit-of-account as a mere sign, and refer to the medium-of-account not just gold but a given quantity of gold grains. Thirdly, we could give the quantity of the medium of account that defines the $ an entirely different term, say the "Tom Brown multiple". As long as we remember that there's a sign, the thing that represents that sign, and the quantity of that thing then we can avoid unnecessary semantic debates

Friday, March 7, 2014

Is the value premium a liquidity premium?

 The "High minus low" strategy: Source 

If you haven't read Clifford Asness and John Liew's recent article on market efficiency, you should. There's plenty of meat in the article, but the one sinew I want to chew on is this above chart.  It shows the cumulative returns to a strategy called HML, or "high-minus-low."

This strategy involves going long cheap U.S. stocks (as measured by their price-to-book ratio) and simultaneously going short expensive stocks. Over the last eighty-five years or so, cheap stocks have roundly beat out expensive ones. This is called the value premium. A large part of Asness and Liew's investing effort revolves around exploiting this premium for their clients at AQR Capital Management.

Now as the authors point out, this outperformance could be due to a combination of two things. The behavioural explanation is that people are irrational, subject to various psychological tics that drive aberrations in prices. Perhaps investors are fickle and would rather plunge into sexy and expensive concept stocks than purchase boring but cheap stocks. Asness and Liew propose the idea that investment committees don't have sufficiently long windows for evaluating the performance of their investments. Whatever the explanation, by trading against the various behavioural tics, investors can earn superior profits.

The other explanation is that the value premium has a very rational underpinning. Those who buy cheap stocks and sell expensive ones need to earn a higher return because they must be compensated for bearing some sort of inconvenience, or because they are providing the market with some extra service.

What is that something? My guess is that if you were to adjust the HML strategy's results to account for the superior liquidity return provided by stocks that seem expensive, you'd probably see the returns on cheap and expensive stocks converge. What appears to be an anomaly on the chart is just the shadow of a liquidity premium. Asness and Liew aren't exploiting an irregularity, they're producing liquidity—and getting paid a fair rate for doing so.

Take two companies that are identical except that the shares of the first are more liquid than the second (yep, I've been down this road before). Given a choice of buying either of the two, investors will always prefer the more liquid share. Owning a stock with low bid-ask spreads and plenty of depth provides investors with the comfort of knowing that should some unpredictable event arise, they can easily sell their shares in order to mobilize the necessary resources to cope with the event.

We can think of people "consuming" the comfort provided by liquid shares, much like they consume the peace of mind provided by a fire extinguisher stored away in a closet. If we want more peace of mind, we need to buy a better quality fire extinguisher and/or shares with a higher degree of liquidity. If we can do with less, then we can buy a smaller fire extinguisher and/or switch into illiquid shares.

We can decompose the price of a share into two parts—the price people pay for the share's earnings, and the premium they are willing to cough up to consume the peace of mind that its liquidity provides. Since the earnings on our two shares are the same, the portion of each share's overall price that is explained by earnings will be equal. However, people will put a small premium on the consumption value of the services provided by the illiquid shares and a large premium on the superior consumption yield provided by liquid shares. The difference in premiums means that the market price of the liquid share will be higher than the illiquid one.

Because of this price discrepancy, people will typically say that the liquid share is "expensive" and the illiquid one "cheap", but these are misnomers. The liquid shares provide a stream of valuable services that the illiquid shares fail to provide, and therefore logic dictates that they must trade at a higher price. They might seem expensive relative to underlying earnings, but only if we intentionally ignore the very real flows of consumption that they provide.

By selling "expensive" stocks and buying "cheap" ones, Asness and Liew are really just selling liquidity while taking on illiquidity. In short, in exploiting the HML line they are acting as liquidity providers. Just like Kidde (a major manufacturer of fire extinguishers) provides the world with a worthy service—peace of mind—Asness and Liew are producing and selling that very same good. They are willingly holding the illiquid long positions that others would prefer not to hold, thus forgoing the peace of mind enjoyed by others. And in borrowing and selling liquid shares, they are feeding liquidity into the market that would otherwise be stranded in someone's account at a depository.

Just like Kidde should be well-compensated for its product, Asness and Liew should be appropriately rewarded for their sacrifices. As compensation, their illiquid long position will typically appreciate at a faster rate than an equivalent liquid long position. And their liquid short position will appreciate at a slower than an equivalent illiquid short position.

This difference in expected price appreciation emerges because in equilibrium, an illiquid share needs to provide the same total expected return as a liquid share. Liquid shares already provide an outsized non-pecuniary return (their ability to act as fire extinguishers). In order to counterbalance this, the illiquid share must provide an outsized pecuniary return, or a faster rate of price appreciation. I'm ignoring dividends here. (Again, see this post).

So the HML line charted above illustrates the financial compensation that flows to folks like Asness and Liew who go out of their way to fabricate financial fire extinguishers. (I don't doubt there are also behavioural reasons for it.) Or, conversely, it represents the financial return that people are willing to forgo in order to consume the services provided by liquid shares.

To some degree, Asness and Liew's HML strategy reminds me of the strategy used by Long Term Capital Management. LTCM did a lot of convergence trades in treasury markets. It shorted "on-the-run" bonds, the most recent vintage of debt issued by the government, while purchasing "off-the-run" bonds. On-the-run bonds attract more liquidity than an equivalent off-the-run bonds, and therefore trade at a premium. Thus we see different prices for what are otherwise two identical securities. After a few weeks go by, the current on-the-run bond is replaced by the next issue and suddenly loses its liquidity premium. As long as the short position in on-the-run bonds is held until the next treasury auction, it yields a guaranteed gain when offset against the long position.


I don't think that the profits that LTCM enjoyed by exploiting the on-the-run convergence trade should be thought of as arbitrage gains accruing as a result of other people's irrationality. Much like Asness & Liew and Kidde, LTCM was fabricating peace of mind for others; it held the illiquid securities others didn't want while borrowing and supplying the market with the liquid securities that others preferred. The "excess" return that LTCM enjoyed was the market's fair reward for providing this service.

The service that Asness and Liew provides also reminds me of what a bank does. A bank purchases illiquid personal IOUs issued by families and businesses while selling highly liquid deposits. The bank needn't offer much of an interest rate on deposits because deposits already provide a high liquidity yield. It requires a high interest rate on the IOUs it purchases because it must be compensated for absorbing their illiquidity. The spread the bank earns by holding illiquid assets and providing liquid assets is similar in nature to the HML spread earned by Asness and Liew, and the on-the-run/off-the-run spread earned by LTCM.

Of course LTCM eventually bit the dust. But that didn't have anything to do with the demise of the convergence trade, it had to do with leverage. When its debtors proved unwilling to roll over LTCM's funding, the fund was forced to liquidate at a loss before its positions had converged. If LTCM had been less aggressive, it could have easily held its positions until payoff. The firm might still be in the business of producing fire extinguishers for the financial community, just like Asness and Liew are doing by participating in the HML trade.

Saturday, March 1, 2014

Beware the financial Jeremiahs

Jeremiah, the prophet of impending disaster. By Rembrandt, 1690. See full version.

The 1929 analog model has resurfaced.

The 1929 analog is a recurring visual meme, usually a chart, that periodically plagues financial markets. All versions of this meme invariably map the bobbing and weaving of the 1929 Dow Jones Industrial Average onto movements in the present Dow, with the inevitable conclusion being that we are, by analogy, on the verge of a repeat of the 1929 crash.

The most recent reincarnation originates from noted market timer Tom DeMark. His claim has been amplified by newsletter writer Tom McClellan and irresponsibly blared all over the internet by Marketwatch (see here, here, here). I produce the chart below:

Source: Marketwatch

I've been following various flareups of the 1929 analog for over a decade. They usually crop up in September, just before the anniversary date of the October 29 crash. Extended bull markets are particularly fertile ground for 1929 analog behaviour as the long run-up to the 1929 crash will typically map quite well to the current bull market. Financial Jeremiahs, those whose bread and butter is to perpetually predict hard times, are a major source of these graphics. The meme typically dies a quick death as market movements subsequently fail to conform to the analogy. DeMark's version has received far more press attention than any of the other flareups I've followed, thus this post.

The 1929 analog chart always has been and always will be silly. Worse, there is always a small chance that the chart will have large repercussions (more on that later).

The chart is silly because there's no logic behind it. It simply doesn't follow that the alignment of prices today with prices from eighty years ago means that subsequent prices must adhere to the old path. There's little else to be said.

What makes the chart so effective isn't the logic that underlies it (there is none), it's because it harnesses our brain's automatic ability to rapidly complete patterns. Our brains are always trying to pick out visual regularities in the chaos, or to generalize. This is an incredible power, allowing us to recognize a face at night using only a few cues, or pick out a dalmatian against a camouflaged background (see picture below).

When we look at the 1929-2014 analog, the chart is virtually begging us to complete the pattern. Note, for instance, how the red line has been placed a constant distance below the blue line rather than having the lines cross over each other. This isn't an accident—it's a feature designed to crystallize the comparison in the mind of the viewer. Any crossing over of lines would only impede the viewer's ability to rapidly make the analogy.

In the same way that we get an aha! moment the moment that we finally tease out the dalmatian from its surroundings, the transferral of the 1929 crash onto the as-yet incomplete 2014 plot provides us with a burst of satisfaction. So we stop thinking, the puzzle seemingly complete. The long and sober thought processes that should go into forecasting a major turn like a crash is short-circuited by the superficial sense of completion that the overlay of prices gives us. And that's what the chart maker wants, to short circuit are deeper thought processes by appealing to our innate propensity to rapidly fill in the visual blanks.

Unfortunately, this silly chart has a very small chance of having large repercussions.

Assiduous readers may remember that I wrote about the 1929 analog last October. In that post I hypothesized that the best explanation for the 1987 stock market crash was an emergence of the 1929 analog meme. The mechanism would have worked something like this...

At some point in 1987 stock prices began to randomly overlap with a plot of 1929 prices. Traders found meaning in this fluke and began to trade using the 1929 trajectory as a guide. Paradoxically, their trading helped push prices in the same direction as the 1929 plot, reinforcing the similarity between the two charts. This would have only increased the degree of belief they placed in the analog, causing them to increase their 1929-inspired trading, this activity creating ever more conformity between 1929 and 1987 prices. A feedback loop had been created, a loop that would only have expanded as traders told their friends about the pattern, thus expanding the size of the population who was driving the process. The feedback loop finally culminated in a self-realization of the 1929 crash on Monday, October 19, 1987. (This is just a short summary, go read the full article.)

This is why DeMark's 1929 analog, amplified by the likes of McClellan and Marketwatch, has the potential to be dangerous. Despite being no more than a silly picture, if enough people believe in it, that silly picture could actually inspire a stock market crash. Markets, after all, are reflexive. Fundamentals usually drive the ideas that people use to inform their trading behaviour. But at other times, ideas get a life of their own, and when enough people adopt them, these ideas create the very underlying reality that they only claimed to predict. In markets, silly beliefs can become true.

The way I see it, it's our duty to provide a counterbalance to destabilizing reflexive forces like these by either ignoring financial Jeremiahs or roundly vilifying their ideas. After all, sharp downturns are healthy insofar as they are justified by actual events and changes in the fundamentals, but if they're created by mass faulty thinking, everyone is made worse off.

I couldn't help but notice that DeMark was employed by Tudor Investments from 1988 to 1990. Interestingly, Paul Tudor Jones, founder of Tudor Investments, made a pile of money during the 1987 crash by basing his trades on a 1929 analog model (again, read my old post). It would seem that DeMark isn't doing anything new, he's simply repeating a time tested strategy once used by his former employer. A cynic would say that folks like Tudor Jones and DeMark spread the 1929 analogy not because they actually believe in it, but because they want to harness people's tendency to overgeneralize for their own gain. If enough proles take the hook, then markets could plunge, thus benefiting Tudor Jones's and DeMark's pre-existing trading positions. I'm sure that's not the case and that all parties are being genuine. But advertising a trade after one is already in it, i.e. talking one's book, is a time honoured strategy among finance professionals.

With the Dow having such a good performance in February, the simplistic analogy between 1929 and 2014 is slowly being stretched to the point that it no longer aligns. It looks like the DeMark's analog model could die a natural death. However, there's a simple strategy often used by those calling for the end of times. When Warren Jeffs, president of the Fundamentalist Church of Jesus Christ of Latter-Day Saints, predicted the end of the world on December 23, 2012, and it failed to happen, he changed the date to December 31. The 1929 analog can simply be redrawn, shifting the entire 1929 plot over to give more time for the our current market to ripen towards an imminent crash. Even if DeMark isn't the one to do it, someone else will draw the analogy. The longer the current bull market continues, the more fertile the ground will be for these sorts of destabilizing memes.

P.S.: Most commentators have been vilifying the chart, which is good. (See Matthew Boesler, The Reformed Broker, Matthew O'Brien, and the Wall Street Journal). Their criticisms are mostly along the line of... "the analog is less apparent if we rescale the axis." They use something like the chart below as their rebuttal in order to decouple the performance of 1929 and today.

Source: Business Insider

This rebuttal is a weak one since it gives too much ground to the supposed logic that underpins the 1929 analogy. Say that the two plots were to be correctly scaled and say that the prices in one era closely aligned with the other. There would still be no good reason to assume that the current period must follow the prior one into a nosedive. In attacking the scale of the chart, critics are missing the larger error that underpins the 1929 analog.

In short, don't give into your brain's rapid ability to complete these facile patterns. A truly well-reasoned crash prediction would require such a massive allocation of mental power to arrive at that no one would ever actually get there. Admittedly I'm being a nitpicker here. Though the various rebuttals all appealed to the same bad logic that the original chart did, at least they helped counter the reflexive properties of the most recent appearance of the 1929 analog. An enemy of my enemy is my friend, I suppose.