Before the global financial crisis of 2008, interest rates were low, and investors were hungry for higher yields. Financially engineered mortgage-backed securitizations offered high yields with seeming low credit risk. Though these arcane financial instruments often consisted of risky loans to less credit-worthy borrowers, they were rated highly by the credit agencies. The easy availability of credit created a bubble in the housing market that eventually burst, leading to defaults on mortgages and crashing the value of the securitizations, the stock market, and the economy.

Now fast forward to the present. In 2020, strong corporate profitability, low interest rates, substantial debt, and rising stock market prices set the stage for a stock market crash that could be triggered by an adverse event. Financially engineered collateralized loan obligations (CLOs) consisting of leveraged loans to highly indebted companies are today’s counterpart of the collateralized debt obligations (CDOs) of the last crisis, which contained subprime mortgage loans to highly indebted homeowners. This time, unlike 2008 and many other crises, the catalyst for the market collapse did not come internally, from within the financial system, but rather externally — the pandemic that brought the economy to a halt. However, internal factors like CLOs and high levels of leverage in general were potential time bombs in the financial system that could exacerbate and prolong economic disruption.

These insights into so-called “free lunch” strategies and their role in destroying wealth come from Bruce I. Jacobs, co-founder, co-chief investment officer and co-director of research of Jacobs Levy Equity Management in Florham Park, New Jersey, which manages equity portfolios for institutional clients worldwide. Jacobs is the author of Too Smart for Our Own Good: Ingenious Investment Strategies, Illusions of Safety, and Market Crashes published by McGraw-Hill Education in 2018. He also is the co-author with Kenneth (Ken) N. Levy of Equity Management: The Art and Science of Modern Quantitative Investing published by McGraw-Hill Education in 2017. He initiated the creation of the Jacobs Levy Equity Management Center for Quantitative Financial Research at Wharton and helped create a new quantitative finance major for MBAs by establishing the Dr. Bruce I. Jacobs Professorship in Quantitative Finance and the Dr. Bruce I. Jacobs Scholars in Quantitative Finance.

Knowledge at Wharton interviewed Jacobs about why markets regularly experience such crashes and the role that quantitative finance can play in predicting and potentially protecting investors against episodes of wealth destruction.

Knowledge at Wharton: In your book, Too Smart for Our Own Good, you wrote about some quantitative free-lunch investment strategies that led to the formation of market bubbles that eventually burst, destroying enormous wealth. Do you see any similarities between those crises and the present stock market collapse?

Bruce I. Jacobs: Quantitative finance offers tools and approaches that provide enormous benefit to investors, from security valuation to portfolio construction and performance measurement. Unfortunately, quant strategies are also subject to misuse and misinterpretation; this is especially the case with strategies that are complicated and lacking in transparency.

Free-lunch investment strategies are a good example of this. Free-lunch investment strategies are hard to resist—they offer high returns at low risk, an investor’s dream but an illusion of safety. This is contrary to Finance 101, where we learn there is a relationship between risk and return, and that higher expected return requires taking more risk, not less risk.

In the years leading up to the global financial crisis of 2008, interest rates were low, and investors searched for higher yields. Financially engineered mortgage-backed securitizations offered high yields with seeming low credit risk. Although the securitizations often consisted of subprime mortgage loans to less creditworthy borrowers, they were rated highly by the credit agencies. We all know what happened then — the easy availability of credit created a bubble in the housing market that eventually burst, leading to defaults on mortgages and crashing the value of the securitizations, the stock market, and the economy.

In 2020, unlike 2008 and many other crises, the catalyst for the crisis did not come internally, from within the financial system, but rather externally — the pandemic that brought the economy to a halt.

This time around, in 2020, strong corporate profitability, low interest rates, substantial debt, and rising stock market prices set the stage for a stock market crash that could be triggered by an adverse event. Of course, this time, in 2020, unlike 2008 and many other crises, the catalyst for the crisis did not come internally, from within the financial system, but rather externally — the pandemic that brought the economy to a halt.

Nevertheless, some quantitative approaches are contributing to the instability. For example, financially engineered collateralized loan obligations (CLOs) consisting of leveraged loans to highly indebted companies are today’s counterpart of the collateralized debt obligations (CDOs) of the last crisis, which contained subprime mortgage loans to highly indebted homeowners. If the risks were not disclosed, nor transparent, and hence not properly priced, trouble could lie ahead.

Leveraged loans grew dramatically in the low-interest-rate environment of the last few years. By the fall of 2019, the amount of outstanding leveraged loans denominated in U.S. dollars was estimated at $1.2 trillion, roughly equivalent to the outstanding value of subprime mortgages as we entered 2008. More than half of these loans ended up in CLOs.

Knowledge at Wharton: The three crises you analyzed — the October 1987 crash, the Long-Term Capital Management hedge fund collapse in 1998, and the subprime mortgage crisis of 2008 — all involved positive feedback loops that reinforced the fantasy of very high returns at low risk. Was there any evidence of this before the current crash?

Jacobs: Low interest rates, or the availability of cheap money, give rise to more borrowing by individuals for housing and other purchases, including securities, and more borrowing by companies for business investment, share buybacks, and dividend payments. This in turn leads to more consumption, more economic growth, and higher stock prices. As the good times roll on, individuals and companies take on more and more leverage because they perceive the risks are low. As I mentioned earlier, securities based on loans to highly indebted companies have been a troublesome feature of corporate borrowing.

This positive feedback or self-reinforcing behavior, where leverage increases profitability, which lowers the apparent risk, and increases the willingness to take on more leverage, resembles what happened with the interplay between mortgage securitizations and the housing market before the 2008 global financial crisis.

When positive feedback causes prices to reach a level unsupported by underlying fundamentals, a shock to the system — homeowners defaulting on their mortgages en masse or a global pandemic — can topple the house of cards. Then the leverage that was a tail wind can transform into a head wind so severe that it levels financial markets and the economy. Investors can then be hit with margin calls that force liquidations at adverse prices, deepening price declines.

Knowledge at Wharton: You speak about excessive leverage and say that modern portfolio theory doesn’t take into account leverage risk. Yet when a portfolio is leveraged, there is additional volatility. Please explain in what way leverage is not considered, and how you have addressed this with a quantitative model.

Jacobs: Modern portfolio theory is about portfolio diversification. It is the math behind the saying, “Don’t hold all your eggs in one basket.” But it is totally inadequate, especially now. Modern portfolio theory works fine for investors who buy securities with cash. But nowadays investors use margin loans to finance stock purchases, and they also borrow shares to sell short—a strategy for profiting from falling stock prices. Both are forms of leverage. Using leverage is like stacking baskets of eggs on top of one another. You know the higher the stack of baskets, the more likely it will topple.

Using leverage is like stacking baskets of eggs on top of one another. You know the higher the stack of baskets, the more likely it will topple.

Modern portfolio theory takes into account the portfolio’s expected return and volatility of return, but there is no consideration of leverage risk. Yes, as a portfolio’s leverage increases, so does its volatility, but there are additional risks that are unique to leverage. These unique risks include the risk of a margin call, which can force investors to liquidate securities at adverse prices, and the possibility of losing more than the amount of capital invested.

My partner Ken Levy and I reinvented portfolio theory by extending it beyond the standard two dimensions of expected return and volatility to consider a third dimension representing leverage risk. This means that investors can now determine the most appropriate portfolio by considering their preference for more expected return along with their aversion to volatility risk and to leverage risk. If individual investors take into account their own aversion to leverage risk, the total level of leverage in the financial system may be lower than it would otherwise be, thereby reducing systemic risk.

Knowledge at Wharton: You have warned in the past about the “illusion of liquidity.” Describe what you mean by liquidity, why is there an illusion, and what ill effects it can have. Is it related to the “illusion of safety,” which you have also warned about?

Jacobs: There is both trading liquidity—the ease with which you can get into and out of positions—and funding liquidity—the ease with which you can raise funds. When interest rates and volatility are low, most investors assume that both trading and funding liquidity are nearly infinite, that they can trade with little market impact, and can borrow whatever amounts are needed. But that’s an illusion; as we’ve seen in every crisis, both trading and funding liquidity evaporate rapidly.

In the 1980s, for example, those using portfolio insurance thought they could reduce their risk by selling equity on a timely basis. But this turned out to be a false assumption when the market declined sharply on October 19, 1987. In the 1990s, managers of the highly leveraged hedge fund Long-Term Capital Management thought that they could liquidate positions when they needed to, but this was not possible during the summer of 1998, when a Russian default led investors to flee from risky assets. LTCM suffered margin calls, was unable to trade out of positions or raise new capital, and had to be bailed out by a consortium of Wall Street firms.

When interest rates and volatility are low, most investors assume that both trading and funding liquidity are nearly infinite, that they can trade with little market impact, and can borrow whatever amounts are needed. But that’s an illusion; as we’ve seen in every crisis, both trading and funding liquidity evaporate rapidly.

Both portfolio insurance and the supposedly low-risk arbitrage strategies of LTCM were free-lunch strategies, designed to deliver high returns at low risk. But the delivery of high returns at low risk depended on liquidity, and when liquidity disappeared, the safety of these strategies proved to be an illusion, too.

Knowledge at Wharton: You once wrote that “seemingly rigorous mathematics, plausible theory and early performance success has proved time and again to be a dangerously enticing cocktail.” Did quant models and strategies exacerbate the effects of the current crash?

Jacobs: Following the financial crisis of 2008, volatility-based strategies became very popular. Risk-parity strategies, for example, use leverage to try to keep risk balanced across different asset types. Other strategies involve using options to hedge or speculate on the level of price volatility in the stock market, as measured by the VIX index. These types of strategies seek to control portfolio risk by trading based on volatility levels. They typically use leverage, and when volatility shoots up, they can suffer margin calls, which require an abrupt unwinding of positions at unfavorable prices. These forced liquidations further increase volatility. We saw this in February 2018, when several mutual funds with leveraged VIX trades lost most of their assets and closed. We saw this type of forced selling again in the current crisis, and it added to the panic.

Knowledge at Wharton: What are the most significant risks in markets today and are they quantitative products? What should be done to lessen the risks in the future?

Jacobs: Financially engineered collateralized loan obligations (CLOs) are today’s counterpart of the collateralized debt obligations (CDOs) that played a major role in the global financial crisis. CLOs, like CDOs, consist of below-investment-grade loans, in this case, corporate loans rather than mortgages. Like CDOs, they offer tranches for various levels of risk and return, with the AAA-rated tranches supposedly offering safe, high yields — seemingly a free lunch. The underlying loans are often very shaky. CLOs were the biggest buyers of leveraged loans, which became a favored financing vehicle for private equity firms in the low-rate environment of recent years. Many of these loans have been downgraded and may default.

Knowledge at Wharton: Now that markets have dramatically fallen, what should investors do? How can investors preserve assets and still benefit from appreciation?

The golden rule is to stick with your investment discipline. Don’t sell when investors are panicky or buy when investors are exuberant.

Jacobs: Aside from shopping for bargains, the golden rule is to stick with your investment discipline. Don’t sell when investors are panicky or buy when investors are exuberant. Your allocation across different types of assets should be balanced in accordance with your risk tolerance, so that you can bear the pain in a distressed market, or even buy into a panicked market when prices are cheap, and sell in an exuberant market when prices are dear. This type of “rebalancing” typically benefits portfolios as markets tend to revert to prior levels over time, that is, they mean-revert.

Knowledge at Wharton: You are an equity investment manager. Do your clients only benefit from stocks that appreciate, or can they also benefit from stocks that decline?

Jacobs: Some time ago, we introduced to the investment community 130-30 long-short portfolios that sell stocks short in addition to buying stocks as long positions. Our idea was to be able to benefit from stocks that we believe are likely to decline in addition to benefiting from stocks that are likely to rise.

The design of 130-30 long-short portfolios is to invest, for every $100 of capital, $130 in long positions and $30 in short positions, so that the net market exposure is $100. With this approach, the portfolio can benefit from both winning and losing stocks, while retaining a full market exposure. A portfolio like this is for investors who want a full market exposure with no market timing. Our clients are long-term institutional investors who want for this portion of their pension or endowment funds a full market exposure with the opportunity for value-added relative to the market index.

We used our model of portfolio risk that I discussed earlier, which takes into account the risk of leverage, to determine that a 130-30 long-short portfolio was the optimal amount of long and short position leverage for an investor having typical aversions to volatility risk and leverage risk.

Knowledge at Wharton: You have said that “misconceptions and gullibility cause some investors to make the same financial mistakes over and over again…. They fall for the same flawed narratives about risk management, liquidity and leverage.” Why does this keep happening? What is the antidote?

Jacobs: Human nature has evolved very slowly over the centuries. Human beings are prone to making the same mistakes and they are prone to forgetting the lessons, especially from generation to generation. People are always looking for the elixir of higher returns at lower risk.

People are always looking for the elixir of higher returns at lower risk.

Also, the causes of crises can change. In the global financial crisis, the initial quakes came from the financial sector, from the leverage of subprime mortgage securitizations. Today’s crisis emerged from the coronavirus pandemic. Risky debt is now concentrated in the collateralized loan obligations based on leveraged loans to highly indebted companies.

The most effective antidote is education, disclosure, and unfettered debate. It’s critical for investors, executives, and regulators to understand the types of instruments and strategies created by quantitative finance, because some of these can have systemic implications.

The global financial crisis of 2008 revealed how little executives and regulators understood about the mortgage instruments underlying that crisis. It’s also important for the creators of financial products to disclose the risks and to foster or at least not impede open discussion and debate. This can be more difficult than it might seem. For instance, I warned of the systemic dangers of the seemingly benign portfolio insurance strategy in the 1980s in open forums, but I was closed out of investment journals by the vested interests of the creators who were academicians occupying board positions. And consider the number of risk managers who were ignored or fired when they warned that mortgages underlying securitizations in the 2000s were unsound.

Knowledge at Wharton: Can quantitative finance modeling help predict future crises? How?

Jacobs: One approach that seems to hold promise is agent-based modeling. Agent-based modeling uses simulation to estimate the effects of different environments on market prices and liquidity. Ken Levy, Harry Markowitz, and I developed one such model—the Jacobs Levy Markowitz Simulator, or JLMSim. Simulators such as JLMSim are capable of modeling the agents and market mechanisms behind prices.

JLMSim considers the behavior of investors, security analysts, and traders, and how they affect markets. Security prices are determined by the interactions of these different market participants in the simulated market. One of our findings was that a simulated market with too many momentum investors led to bubbles and crashes. This is what happened during the internet bubble in 1999, when prices bubbled up and subsequently crashed, and this can explain other market bubbles as well.