Making money off illiquidity: Two Strategies

At first sight, illiquidity is bad news for investors, since it gives rise to transactions costs, which, in turn, can lay waste to investment strategies. In a post from a few months ago, I examined how transactions costs can explain why so many strategies that look good on paper don't deliver their promised upside.

However, in this post, I want to take the "glass half full", optimistic view of the phenomenon. Illiquidity or potential illiquidity is not all bad news for investors. After all, to beat the market, you have to have an edge over other investors and here are two competitive advantages that can be created out of illiquidity.

a. Illiquidity arbitrage: Not all investors value liquidity equally. To the extent that you need or care about liquidity less than the typical investor in the market, you should be able to exploit this difference to make money. How? Wait for a period of illiquidity (either on the entire market or on an individual stock), where asset prices are marked down by typical investors, who observe the illiquidity and price it in. Then, step in and offer to buy assets. You will get these assets at a bargain price (from your perspective) but at a fair price (from the perspective of the median market participant). Wait for the illiquidity to ease and then sell the asset. This may very well be the biggest weapon that an old-time value investor brings to the market. In fact, Warren Buffet did exactly this type of bargain hunting during the banking crisis of 2008, taking large positions in Goldman Sachs and GE, during their most illiquid days. (I know... I know.. technically, this strategy is not arbitrage, since it is not riskless.. )

"This is easy. I too can be a liquidity arbitrageur", you may say, but it is easier said than done. There are two factors that are at least partially under your control. The first is the use of financial leverage in your investment strategy. Borrowing money to fund investments may increase your expected upside, if things go well, but it also increases your need for liquidity.  The second  is a combination of patience and a strong stomach. Buying during periods of illiquidity will expose you to down side risk, at least in the short term, and you have to be able to ride it out. But the desire for liquidity is also a function of  factors that are not  in your control. First, if your income stream is stable, predictable and in excess of your spending needs (Do you have tenure?) and you have have less need for liquidity. Second, it is subject to what I will loosely term "acts of God". A sudden illness, accident or unforeseen event (Did you invest with Bernie Madoff?) may quickly eliminate whatever buffer you thought you had. Third, if you manage other people's money, it is their need for liquidity that will drive your decisions, not your own. It is one significant advantage that you and I have on the most skilled portfolio manager.

b. Illiquidity timing: Both the level of illiquidity and the price demanded for it change over time in the market. An investor who can forecast changes in illiquidity well can profit off these changes. But how does forecasting liquidity translate into a payoff? You have to be able to shift into liquid assets, before the market becomes illiquid, and into illiquid assets, ahead of periods of liquidity. With the former action, you cut your losses and with the latter, you gain as the illiquid asset regain their value. This is particularly true, if you use financial leverage and invest in illiquid assets, as many hedge funds do. In fact, one study argues that liquidity timing may be one of the biggest competitive advantages in the hedge fund business.

How do you get to be a good liquidity timer? First, you have to track not just the standard investment measures - multiples and fundamentals - but also liquidity measures - trading volume, short selling and bid-ask spreads. In fact, those technical indicators that fundamentalists view with such contempt, such as trading volume and short sales, may be useful in detecting shifts in liquidity. Second, you have to be clear about how exposed an individual asset is to market shifts in liquidity - a liquidity beta, so to speak. In my extended paper on liquidity (linked below), I describe ways in which you may be able to estimate this beta.

For more on liquidity betas and the potential for making money off illiquidity, you may want to look at this paper that I just posted on liquidity and its effects on financial theory and practice:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1729408

Read More

Asset selection & Valuation in Illiquid Markets

In my last post, I looked at how the asset allocation decision can be altered by differences in liquidity across asset classes, with the unsurprising conclusion that investors who desire liquidity should tilt their portfolios towards more liquid asset classes. Assuming that you have made the right asset allocation judgment, how does illiquidity affect your choices of assets within each class? In other words, if you have decided to invest 40% of your portfolio in stocks, how does illiquidity affect which stocks you buy?

To select assets within each asset class, you can either value each one on its fundamentals (intrinsic valuation), compare its pricing to how similar assets are priced (relative valuation) or price it as an option (contingent claim valuation). In each case, illiquidity can affect value.

 a. Intrinsic Valuation: There are many different intrinsic valuation approaches but they all share a common theme. The value of an asset is a function of its expected cash flows, growth and risk. In discounted cash flow valuation, for instance, the expected cash flows discounted back at a risk adjusted discount rate yields a risk-adjusted value. In conventional valuation, the expected cash flows are unbiased estimates of what the asset will generate each period and the risk adjustment is for non-diversifiable market risk (with equity) and for default risk (with debt). Nowhere in this process is illiquidity considered explicitly. Not surprisingly, we tend to over value illiquid assets.

So, how do you bring illiquidity into intrinsic valuation? There are two choices. The first is to estimate the risk adjusted value, using the conventional approach, and to then reduce this value by an illiquidity discount. That discount can be estimated by looking at  on how the market prices illiquid assets. For instance, studies have looked at restricted stock (stock issued by publicly traded companies that cannot be traded by investors for one year after the issue), pre-IPO transactions (where co-owners sell their stake in the months prior to an announced IPO) and companies with multiple classes of shares traded on different venues (with different liquidity characteristics). These studies generally yield large discounts (25-50%) for illiquid assets and  private company appraisers have generally used these studies to back up the use of similar discounts when valuing non-traded businesses. Perhaps, this approach can be extended to publicly traded companies.

The second is to adjust the discount rate for illiquidity, pushing it up for illiquid companies. The illiquidity premium added to the discount rate is usually estimated by looking at the past. In its crudest form, you can assume that the premium that small cap companies or venture capitalists have earned over the market (about 3-4% on an annual basis over the last few decades) is due to illiquidity and add that number on to the cost of equity of any "illiquid" company. In its more sophisticated version, the adjustment to the discount rate can be linked to a measure of illiquidity on the company - its turnover ratio, trading volume or the bid-ask spread. One study concludes that every 1% increase in the bid-ask spread pushes up the discount rate by 0.25%. Thus, the cost of equity for a stock with a beta of 1.20 and a bid-ask spread of $0.50 (on a stock price of $ 10), with a riskfree rate of 4% and an equity risk premium of 6% is:
 Cost of equity = 4% + 1.20 (6%) + 0.25% (.5/10) = 12.45%
With both approaches, the value will decrease with illiquidity.

 b. Relative Valuation: In its most common form, relative valuation involves screening the market for cheap companies, with one screen for pricing (low PE, low price to book, , low EV/EBITDA) and one or more for desirable fundamentals (high growth, low risk, high ROE). If you ignore illiquidity, your cheap stock portfolio will end up with a lot of illiquid stocks. The simplest way to incorporate illiquidity is to add it as a screen. Thus, in addition to screening for high growth and low risk, you could also screen for high liquidity (high float, high turnover ratios, low bid-ask spreads, high trading volume etc.). The tightness of the liquidity screen can then be varied to fit your liquidity needs as an investor.

 c. Contingent Claim Valuation: All option pricing models are built on two principles: replication (where a portfolio of the underlying asset and a riskfree investment is created to have the same cash flows as the option) and preventing arbitrage (the replicating portfolio and the option have to trade at the same price) . Both principles require liquidity: you be able to trade the option, the underlying asset and the riskfree asset in any quantity and at no cost. Illiquidity in any one of these markets will throw a wrench into the process and cause the option pricing models to yield incorrect values, with the imprecision increasing with illiquidity. So, what are your choices for bringing illiquidity into the process? You can try to modify the models to incorporate illiquidity explicitly but option pricing models are complicated enough already and this adds an additional layer of complication. Alternatively, you can adjust the inputs into the option pricing model. My choice would be the underlying asset value (S): using a lower value for illiquid underlying assets will reduce the value of call options on those assets.

In summary, no matter which approach you use, illiquidity is not a neutral factor. The investments you make within each asset class will reflect both the illiquidity of the investment and your own liquidity needs (and preferences) as an investor.

I have a paper on the effects of illiquidity on financial theory, where I examine the effects of liquidity on valuation in more detail:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1729408

Read More

Asset allocation for illiquid markets

In my last post, I argued that illiquidity is not a minor problem restricted to a few stocks. In fact, it can affect all stocks, at least during some time periods, with its effect varying across stocks. I also noted that much of financial theory is built around the presumption that markets are liquid.

So, how would financial theory and practice change, if illiquidity is explicitly incorporated into the process? Let's start with the first step in investments, asset allocation, where you decide how much of your overall wealth you will invest in different asset classes - treasuries, corporate bonds, stocks, real estate, collectibles. In fact, defining asset classes loosely, private equity, hedge funds and mortgage backed securities can be considered new entrants in the game, vying for portfolio dollars.

In the classic mean variance framework, the optimum asset allocation mix is the one that maximizes expected returns, given a risk constraint. Thus, you feed in the expected returns and standard deviations of different asset classes, in conjunction with their covariances with each other, and let optimization work its magic. Here is the catch. The average returns, standard deviations and correlations all come from historical data. With illiquid asset classes, standard deviations tend to be under estimated (for a completely illiquid asset, there will be no trading and the standard deviation will be zero) and the covariances consequently will be misestimated. In fact, the least liquid asset classes often look like they offer the best risk/return tradeoffs, if you don't control for illiquidity. Plugging these values into the optimization framework will generate weights that are too high for the illiquid asset classes, for the typical investor. In the last decade, especially, this has led many endowment funds to over invest in real estate, private equity and hedge funds, categories notoriously over exposed to the vagaries of illiquidity.

So, how would you bring illiquidity into the mix and what are the consequences? There are two routes you can follow. In the first, you adjust the expected returns of illiquid asset classes downwards to reflect the expected cost of illiquidity.  That would make these asset classes less desirable and counter act the underestimation of standard deviations. The other is to restate the optimization problem thus: Maximize expected return subject to the constraints that risk be below a "stated" level and that liquidity be greater than a specified constraint.


With both approaches, the "right" asset allocation mix will vary across investors. Investors who desire or need more liquidity will tilt their portfolios towards more liquid asset classes (large market cap stocks, highly rated corporate bonds) . Investors who value liquidity less may actually gain by tilting their portfolios away from more liquid asset classes towards less liquid ones (real assets, small cap and low priced stocks, low rated corporate bonds).

I have a paper on the effects of illiquidity on financial theory, where I look at asset allocation in more detail:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1729408

Read More

All assets are illiquid

Much of financial theory is built on the premise that markets are liquid for the most part and that illiquidity, if it exists, occurs in pockets: it shows up only with very small, lightly traded companies, emerging markets and privately owned businesses. In fact, almost the prescriptions we provide to both investors and corporate finance reflect this trust that both security and asset markets are liquid.

To see how unrealistic the assumption of liquidity is, consider what a liquid market would require: you should be instantaneously be able to sell any quantity of an asset at the prevailing market price with no transactions costs. Using that definition, no asset is liquid and the only question then becomes one of degree, with some assets being more liquid than others.

Given the premise that all assets are illiquid, here is a follow up question: how do you measure illiquidity? One obvious measure, especially in securities markets, is the total transactions costs, including not only the brokerage costs, but the bid-ask spread and the price impact from trading. The other is waiting time. In real estate, for instance, illiquidity manifests itself in properties staying on the market for longer periods. Days on market (DOM) is a widely reported statistic in real estate and is used to measure the health (and liquidity) of different markets.

There is significant empirical evidence that illiquidity varies across asset classes, within assets in a given asset class and across time.

• Across asset classes, illiquidity is more of a problem in real asset markets (real estate, collectibles) than in financial asset markets. Within financial asset markets, the US treasury market is the most liquid, followed by highly rated bonds and developed market stocks, with low-rated (junk or high yield) bonds and emerging market stocks bringing up the rear.
• Within each asset class, there are wide variations in liquidity. In the treasury market, the just-issued, standard maturity treasuries (3 month, 6 month, 10 year) are more liquid than the seasoned, non-standard maturity treasuries. Within the stock market, larger market cap and higher priced stocks are more liquid than smaller market cap, lower priced stocks.
• Liquidity also varies widely over time. While the long term trend in liquidity in equity markets has been towards more liquidity, liquidity moves in cycles, increasing in bull markets and decreasing in bear markets. Punctuating the long term trend are crises, like the 1987 sell-off in the US and the 2008 banking crisis, where liquidity dries up even for the largest market cap companies. During these crises, illiquidity manifests itself in many ways: trading halts, higher bid-ask spreads and bigger price impact when trading.

None of this evidence would matter if investors did not care about illiquidity but there is clear evidence that they do: liquid treasuries have lower yields (and higher prices) than illiquid treasuries and investors demand higher returns on stocks with lower trading volume and higher bid-ask spreads. One study find that every 1% increase in bid-ask spreads increased expected returns by 0.25%, and these higher required returns push down asset prices. Adding to the problem, the price that investors charge for illiquidity also varies over time, spiking during periods of crises: illiquid assets get discounted even more during these periods.

If illiquidity varies across asset classes, across assets and across time, and investors price in this illiquidity, it seems prudent that both portfolio  and corporate financial theory be modified to reflect the potential for illiquidity. Alas, given the length of this post, that has to wait for the next one.

Read More

Are complex models the answer?

There seems to be consensus that conventional economic models did a poor job predicting the magnitude of the last crisis and that we need to do better. In today's Wall Street Journal, we see the beginnings of one response:
http://online.wsj.com/article/SB10001424052702303891804575576523458637864.html?mod=WSJ_economy_LeftTopHighlights
In short, a physicist, a psychoanalyst and an economist believe that they can build a bigger model that captures the complexities of the real world and does a better job of forecasting the future. Good luck with that! While I wish them well, my response is that this will go nowhere or worse, go somewhere bad.

To those who believe that complex models with more variables are the answer to uncertainty, my response is a paper by Ed Lorenz in 1972, entitled Predictability: Does the flap of a butterfly's wing in Brazil set off a tornado in Texas?, credited with creating an entire discipline: chaos theory. In the paper, Lorenz noted that very small changes in the initial conditions of a complex models created very large effects on the final forecasted values. Lorenz, a meteorologist, came to this recognition by accident. One day in 1961, Lorenz inputted a number into a weather prediction model; he entered 0.506 as the input instead of 0.506127, expecting little or no change in the output from the model. What he found instead was a dramatic shift in the output, giving rise to a Eureka moment and the butterfly effect. (One of my favorite books on the topic of Chaos is by James Gleick. It is an easy read and well worth the time.. for investors and economists)

Complex models work best with inputs that behave in thoroughly predictable ways: software and engineering models come to mind.  They break down when the inputs are noisy and the relationships are unstable: macro economic models are perfect lab experiments for chaos. The subjects (human beings) belong in strange and unpredictable ways, the variables that matter keep shifting and the relationships between them change over time. In fact, I will wager that the models that worked worst during the last crisis were the most complex models with dozens of inputs and cross relationships.

So, what is the solution? My experience in valuation suggests that you should go in the other direction. When faced with more uncertainty, strip the model down to only the basic inputs, minimize the complexity and build the simplest model you can. Take out all but the key variables and reduce detail. I use this principle when valuing companies. The more uncertainty I face,  the less detail I have in my valuation, recognizing that my capacity to forecast diminishes with uncertainty and that errors I make on these inputs will magnify as they percolate through the valuation. More good news: if I am going to screw up, at least I will do so with a lot less work!!

Read More

The insider trading scandal: Thoughts about the hedge fund business

As many of you are aware, the last week has been filled with news stories about imminent arrests in an insider trading scandal that supposedly entangles multiple hedge funds and bankers at a couple of large investment banks. I am being leery about naming names, even though they are now in the public domain, since these selective leaks  can be devastating not only to the individuals named but also to the entities that they represent. While some may feel that "they" deserve this, I think we still live in a country where you are innocent until proven guilty.

So, I am going to use this story to talk about a bigger question. Not insider trading, because I have put my views on the topic down in a previous blog post, but about the hedge fund business. Note that many of the targets in this investigation are hedge fund managers. Since I have no reason to believe that hedge fund managers are any more immoral or unethical than any other random group of money managers, the question then becomes: What is it about the hedge fund business that seems to drive this constant search for an information edge? Or why do we not see more traditional mutual fund managers involved in these scandals?

At first sight, hedge funds and mutual funds share much in common. They both solicit money from investors, promising to deliver above-market returns to them and they get compensated for managing this money. But there are three significant differences:
1. Constraints on investing: Mutual funds are far more constrained in where and how they invest than most hedge funds are. Some of these constraints are imposed by regulators, some are self imposed and some are the created by clients.
a. Long versus long/short strategies: Most mutual funds can only buy stocks (The Investment Company Act of 1940 that governs mutual funds puts significant restrictions on short sales by funds), whereas hedge funds often can both sell short on some stocks and go long on others.
b. Investment choices:  There are several mutual funds that are judged based on "tracking error", measured by how far their returns deviate from a specified index's return. This constraint, usually imposed by clients, is designed to prevent fund managers from straying too far from the companies in the index. Hedge funds generally can invest not only in whatever company they want but many invest across asset classes.


2. Clientele mix: Hedge funds attract investments from either the very wealthy or institutions (pension funds, for instance). In fact, most of them actively discourage small, individual investors by requiring a large, minimum investment. Mutual funds tend to attract individual investors. At the risk of a gross generalization, institutional and wealthy investors are more demanding than individual investors; they move their money out of loser funds and into hot funds far more quickly than other individuals do. Put another way, if the herding effect is a phenomenon that affects all investors, it seems to affect wealthier and institutional investors more. (Some hedge funds put withdrawal restrictions on investors to counter.)

3. Financial leverage: Hedge funds are far more likely to use borrowed money to lever their bets. Most mutual funds either do not borrow money or do so on a very restrictive basis.

4. Compensation systems: Mutual funds generally make their money in two ways. All mutual funds cover their expenses from the money under management; the management; these management expenses are public information and can be accessed at services like Morningstar. A subset of funds also assess a one-time up-front sales charge (load), when you invest, where a certain percent of what you invest is taken by the fund at the time of the investment. Hedge funds assess an annual load (1% or 2% of the invested amount each year) and a percentage of the profit on the investment (10 or 20% of the profits).

You may be wondering at this stage what all of this has to do with insider trading. Consider why people seek out insider information. If they succeed, they can buy or sell a stock prior to that information becoming public; when it goes public, the stock will pop up or down, depending on whether the information is good or bad news.

All four differences highlighted play into why hedge fund managers see more gain from seeking out and exploiting private information:

1. Hedge funds can exploit both good news and bad news, by buying stocks in advance of the former and selling short ahead of the latter. Mutual funds can only buy on good news (though they can sell any existing holdings of companies on which bad news lies ahead).
2. While money management at any level is a rat race, where funds try to keep their own clients and coax clients away from their competitors, the race becomes more frenetic with hedge funds. A hedge fund that lags its peer group can enter a death spiral, where losses spur withdrawals which feed into more losses.
3. If the "inside information" is precise, you can use even more debt in your investment strategy, creating huge payoffs on your investment, when the information becomes public. Financial leverage acts as a multiplier on profits from insider trading.
4. The compensation system at hedge funds essentially gives every hedge fund manager a call option: they make 10 or 20% of all profits, no matter how large, and do not share in losses. Trading on information ties in well with this system, since it delivers skewed returns: most of the time, information turns out to be worthless, but when it is relevant information, the payoff is very large. Thus, even if insider information is noisy and provides little benefits or even creates costs for investors over the long term, hedge fund managers may still benefit personally from its use because of the upside call built into their compensation systems...

In summary, hedge fund managers are far more likely to be "information traders" than mutual fund managers and "insider information trading" happens to be one of the more lucrative (albeit illegal) manifestations of the same. So I am not surprised to see them ensnared in insider trading scandals more frequently. (I think the clustering of hedge funds around Stamford, Greenwich and other old-wealth Connecticut towns also adds into the mix. I would assume that there are hedge fund managers constantly rubbing shoulders with each other in every expensive restaurant in those towns...  trading stories and comparing returns...

In closing, though, I am not sure that all of this seeking out of information is generating a payoff for hedge funds. In the aggregate, they continue to either match or under perform actively managed mutual funds (which, in turn, under perform index funds), when fees and transactions costs are factored in. It is entirely possible that some of the "super" performers among hedge funds got there because they had access to private information that no one else had. I just don't think it is likely! As Shakespeare would put it, this seems like much ado about nothing!

Read More

How do you evaluate risk taking?

The GM IPO is the news of the week. The fact that GM has been able to go public and that the government may not only get its money back on its investment but may even make a profit has led to some celebration in the White House:
http://www.whitehouse.gov/photos-and-video/video/2010/11/18/president-obama-gm-ipo
I don't begrudge the White House its victory dance that the GM bet looks like it has paid off, but it is an auspicious moment to examine how we judge risk taking, in general. As I see it, risk taking can be judged on four dimensions.

1. Outcome: The nature of risk taking is that you win some of the time and lost some....  It is human nature to judge the quality of risk taking by looking at the outcome of the risk taking. Success is thus vindication and failure is calamity. If you follow this to its logical ends, if you succeed, you are a good risk taker and if you fail, you are not. This is the theme that is being tapped into by both Warren Buffet when he wrote his thank you note to the government for TARP and by the White House for its GM investment. "Things worked out well in the end... So, it must have been  a sensible decision up front"...


2. Process: A more complicated way of judging risk taking is to look at whether the risk taking made sense at the time that the risk was taken, with the information available at the time,  rather than with the benefit of hindsight. At least in theory, it is possible that even the best-deliberated risky decisions can have bad outcomes, if fate does not cooperate, and that terrible choices when faced with risk can have "good" outcomes. Playing devil's advocate, however, it is much more difficult and more work to evaluate process than outcome.

3. Side costs and benefits:When risk taking creates costs and benefits for others, judging risk taking by its outcome for just the risk taker may not be fair, since it is conceivable for risk takers to make money while creating costs for society. It is also possible for risk taking activity to create losses for the risk takers while generating benefits for society. A fair assessment of risk taking will require us to consider these side costs and benefits into account.

4. Future risk taking behavior:It should not be surprising that how we take and reward/punish risk taking in the present can affect how people take risks in the future. If risk taking of a certain type consistently is rewarded, you will see more of it in the future. If in contrast, risk taking of a different type leads to punishment/losses, you will see less of it in the the future. If markets are viewed as "too easy" on risk takers, there will be more risk taking in the future in the future.

When does it make sense to judge risk taking on outcome alone? If a risk taker takes many risks over time and is right on average on a consistent basis and there are no significant side costs and benefits, it is reasonable to argue that success is the result of good risk taking.  A portfolio manager who beats the market each year for 10 years must be doing something right in terms of risk taking. However, judging a large risk taking venture with significant side costs and benefits and consequences for future risk taking on whether it makes money or not may not be sensibel.

Returning to the GM investment, the judgment on whether it was successful becomes more nuanced when we consider the other factors, even if the government's original investment makes money.

a. Did the government investment in GM make sense at the time of the investment, given what was known then? Tough to tell, since we do not know what the government knew at the time of the intervention. Given what the rest of us knew at the time, it would have been difficult to justify the billions invested in the company. However, it is possible that the authorities had information about GM's assets and liabilities that we did not. So, I will give the benefit of the doubt to the government on this point.


b. What were the side costs and benefits of the government investment? On the plus side , GM's salvation prevented thousands of layoffs at the company and budgetary chaos at at least one state (Michigan). On the minus side, the government's intervention on GM's behalf has cost other carmakers a chance to make inroads in this market. To the extent that the other car makers are foreign (Toyota or Honda), this may seem like a good thing, but the belief in free markets cannot stop at the borders. It is also conceivable that GM's salvation may have cost Ford a chance to make inroads into the market and secure its position for the long term.

c. What are the long term lessons for risk taking behavior?
It is undeniable that GM made some really bad strategic and management decisions in the the last three decades. In the face of default, the government stepped in, upended bankruptcy and tax laws (which conveniently were written and interpreted by government officials) and saved the company. Other firms that took more prudent decisions in the face of risk were put at a disadvantage. It would seem to me that the lessons learned on risk taking from this experience are the wrong ones: if you take risks, make sure that you are a big firm with the right connections.

Bottom line. As a taxpayer, I am happy that the GM investment looks like it will break even or better.. As an investor, I am less happy about the long term consequences of this success. I am afraid that it sends the wrong signals on risk taking to the market and investors.

Read More