How Deleveraging Makes for Market Craziness

Virtually every presentation, conference, or discussion these days opens with the statement “These are challenging times for markets,” or some equivalent. Part of the challenge, of course, is that it is simply difficult to make money from investments in down markets. This is especially true if there are long-only restrictions on one’s portfolio or mindset, but even when short positions are possible, a non-deriviative or margined short will at most double one’s investment, and this happens only when an investment goes all the way to zero.

It turns out that the fact that today’s crisis started as a credit crunch and evolved quickly into a leverage crisis adds an extra layer of confusion into markets that has many participants throwing up their hands and wondering if there is any logic left to markets. Bill Miller, the legendary CEO/CIO of Legg Mason Capital Mangement, noted in his second quarter letter to shareholders of the Legg Mason Value Trust :

“A point [John Rogers of Ariel Investments] made that I have likewise noted to our staff is that this is the only market I have seen where you could just read the headlines in the papers, react to them, and make an excess return. I have used the mantra to our analysts that if it’s in the papers, it’s in the price -- which used to be correct. Indeed, it borders on cliche in the business that by the time something makes the cover of the major news or business publications, you can make money by doing the opposite. There is solid academic research to back this up. But in the past two years, you didn’t need to know anything except to sell what the headlines were negative about (anything related to real estate, the consumer, or finance) and buy anything that was going up and that everybody liked (energy, materials, industrials).”

What Miller is complaining about is how many of the standard tenets of professional investment knowledge - that markets are relatively information-efficient, excess returns come from reacting differently than the crowd, etc. - have recently flown out the window. And Miller is not the only one. Quantitative funds, celebrated for their mechanical ability to de-emotionalize investment decisions have also been hammered as the rules-for-investing-success have essentially changed underneath them.

The craziness is because of deleveraging. As simple as that.

Superficially, this might seem self-evident. This is a deleveraging crisis, and the crisis is driving things down unpredictably, and so, naturally, deleveraging is the reason that things seem crazy. But what I’m trying to say here is that the fact that portfolio investors are trying to roll back their leverage ratios has specific effects that are different from a simple liquidity freeze or from earnings contractions in the real economy.

Why deleveraging confuses traditional market logic

To understand how deleveraging creates confusion, it helps to think about how investment decisions are made in normal market environments, and then think about how the availability and/or loss of leverage affects these decisions and the market as a whole.

Markets in “normal times”

The market portfolio (assume no leverage for now) represents the collective valuation of all assets at a moment in time by all potential investors in the market. It is essentially a weighted average of individual valuations, where the weight of any given investor’s valuation is proportional their total portfolio size. Given that the largest portfolios tend to be run by professional investors whose job is to be informed about markets as a whole and what they invest in, it is reasonable to assume that market prices and the weights of assets in the market portfolio usually incorporate a more-or-less sensible analysis of available information at any point in time. This is why the risk adjusted performance of passively replicating the market is generally difficult to beat: numerous very-large players with rapid access to the best information and analysis have already weighed in with their valuations. If one assumes that the average professional is fairly informed and has a sensible analysis of the majority of issues, then it is understandably difficult to outperform their average.

One can potentially outperform with an actively managed portfolio, but it requires the investor to have realistic knowledge of their comparative advantage. A manager could have access to better data, better analysis, a unique perspective or model, faster analysis and execution, or some other quality that offers some investment advantage. To maximize that advantage, it helps if at least some large-scale investors lack it, because a high-weight laggard will create a larger deviation between the current market valuation and one’s (presumably superior) valuation, and therefore a profit opportunity.

In normal times, active investors are placing their investments according to their valuation estimates, and passive investors are basically “piggybacking” and reinforcing those valuation decisions when they choose to use a passive index. Although there is some diversity in specific active strategies, a large body of financial and economic knowledge informs these investment allocations, much of it follows the received wisdom of what are “sound” investment strategies and the logic of how markets operate. Whether they think consistent outperformance is possible or not, most managers agree that outperformance is difficult, because the most common insights into valuation are likely already embedded in prices.

Cheap credit, low volatility, and creeping leverage

Interest rates and credit spreads have been quite low for the last decade, in part because of Fed policy to keep base rates low, and in part because of loan securitization that (rightly or wrongly) reduced premia for credit risk. Low credit spreads have two major effects on investment portfolios. First, low returns to risk-free investments decrease the attractiveness of saving and shift theoretically optimal portfolio allocations to hold a higher proportion of risky investments (defined as any asset that has the potential to lose value and therefore pays a premium for taking on risk). Second, easy credit actually enables portfolios to leverage themselves up inexpensively, increasing exposure to risky events. For many fund managers and allocators, the low-volatility environment of the last 5 years also made this look like a sensible strategy. On the whole, risky assets were returning substantially more than low-risk and risk-free assets, volatility was low, and it looked sensible to increase exposure to risky assets in general. Moreover, when nearly everyone had access to easy credit, cheap borrowing increased the demand for virtually all risky assets, sending the prices (and thus total returns) even higher year after year.

What this meant, on average, was that relative sector and security weightings were being determined based on defensible financial and economic principles, but taken as a whole, society was buying too much of them because easy credit and low volatility understated the risks. The fact that so many investors overallocated their portfolios to risky assets drove asset prices up and made these investment decisions look brilliant, when in fact these were simply unsustainable gains that had yet to correct. The more money made available through borrowing, the more ended up invested in risky assets. Within risky assets, these investments may have been placed based on rational economic insights about relative growth, competitive advantage, and so forth, but the total amount allocated to these assets was unsound.

Credit flows dry up; hello deleveraging

Eventually, the credit faucet had to turn off, as lenders found it harder and harder to find borrowers that could pay back their loans. This happened first in the mortgage sector, but also involved auto loans, credit cards, and credit more generally. The securitization that in part enabled this cheap credit is not inherently a bad idea, but as in earlier posts, it did make it difficult to do appropriate due diligence on one’s borrowers, and the loan originators had few if any incentives to select good borrowers.

When sub-prime and no-documentation loans entered the bundling stream, lenders were taking on borrowers for whom there was little or no historical data on which to create rational pricing and risk models. As a result, lenders took on far more risk than they realized. Many may simply have had no clue about the risks they were taking; others may have hoped to exit their positions before the house came falling down.

As defaults began to accelerate, lenders started panicking, sending credit premiums to new highs. The result was that the days of cheap credit were over. Portfolio managers who had leveraged up to buy more risky assets suddenly needed to sell those assets to reduce their debt and interest expense.

Here is where the seeming craziness begins.

Deleveraging makes markets look crazy

Deleveraging makes for crazy markets because, when an investor is forced to delever, they reverse (“unwind,” in industry jargon) previous positions in order to turn assets into cash and reduce debt/leverage levels. If the investor and/or the quantity of leverage is large, then these changes are likely to have an impact on market prices. Moreover, the need to deleverage does not typically come from a re-evaluation of the market’s prospects, but usually from changes in the borrower’s credit position or liquidity needs.

If we assume that investors’ portfolio compositions are created by a mostly rational and realistic evaluation of company and market prospects, what this means is that investments are placed in logical allocations when there is no need for leverage or leverage is freely available. When enough investors deleverage, the market impact of large-scale unwinding simply reverses these sensible allocations sending prices in exactly the opposite direction that you would expect, given a logical analysis of the macro-economy.

For example, suppose large investors conclude that defensive companies with stable earnings and consistent dividends are logical investments for the current environment and start tilting their portfolio allocations to take advantage of that. When one or more of these large investors are forced to delever, these “favored” companies are sold, rather than bought, sending prices in the opposite direction, even if the macro-, sector, or company data suggests that one should be buying.

Therefore, as investors deleverage, financial markets will lurch forward and back, causing market observers to scratch their heads - perhaps worriedly - wondering “why did that price collapse, while this other price rose?” During a large deleverage, a number of seemingly attractive securities will move in counterintuitive ways precisely because they seem attractive.

The price fluctuations have a second-order effect as well: it makes even experienced and competent investors less sure of their analysis and more likely to second guess themselves. This actually exacerbates the effect, since the appropriate response to increased model uncertainty is to reduce the quantity of funds that are managed with that model.

Implications

To think about the future, it will help to have some idea of how the deleveraging process will play out. Deleveraging is a painful process for everyone involved, particularly because it typically entails recognizing net losses as one exits positions. Existing leverage starts to more and more excessive as asset prices fall, which forces selling at a loss.

As a result, companies and investors will generally try to deleverage as slowly as possible, but also in sudden jumps, when they realize there is no longer any choice. At these moments, there may be pressure to take a “big bath” loss to avoid needing to deleverage again in the future. This big bath method may or may not work in any one case, but the result is that markets will have sudden drawdowns as different investors realize they can no longer hold on. Like a scared child trying to slide down a long pole, companies and investors will deleverage in rapid unpredictable jumps until they reach the ground, and every time we encounter a deleveraging slip downwards, markets will jump in the opposite direction that the fundamentals seem to warrant.

From a portfolio manager’s perspective, one possible solution would be to deleverage by choosing portfolio assets to liquidate at random. This is hard to ask a portfolio manager to do, because so much of a portfolio manager’s job is about applying knowledge and information and reducing the effects of randomness, but by randomizing the choice of assets to liquidate, one might reduce the connection between deleveraging needs and the mal-performance of otherwise economically sensible assets.

It is hard to know exactly when the mass-deleveraging dynamic will end, but one clue will be that volatility has declined as a result of large-scale moves in either direction becoming less frequent. Whether volatility will return to the low levels of the pre-crisis years is difficult to say - I think it is unlikely to return to those levels any time in the foreseeable future - but the immediate implication is that we will simply need to get used to an environment with higher levels of volatility than we have in the recent past. In fact, if one simply accepts that volatility will be high for a while and creep downwards as deleveraging needs are flushed from the system, this may be enough to guide portfolio strategy for a while.

A major part of this argument is that the need for deleveraging has made market action seem to detach entirely from company and economic fundamentals. Part of this can be attributed to a lasting increase in volatility over the levels of the past few years, and part of this is because large financial and economic changes will themselves impact company fundamentals in non-linear ways. One implication is that technical analysis methods may turn out to be more useful than normal in this market environment. Certainly, Bill Miller’s above reflects that markets are exceptionally information-inefficient at the moment, which is one of the pre-requisites for effective technical analysis. The other reason is that as fundamentals themselves become more difficult to measure and predict, the psychological element becomes a more important aspect of investment decision making, and capturing the psychological element is where well-designed technical analysis really can add value.