"I'm actually really optimistic about the future of quants. The industry is more technical than ever, and there is as much need to understand the risks in the system as ever." – Robert C. Merton, quoted in "Risk," August 2012
In 1997, when Robert Merton won the Nobel Prize in Economics for his work on the Black-Scholes option pricing formula, the demand in the finance industry for people with mathematics, statistics and computer programming skills was exploding.
Investment banks were creating and marketing more exotic derivative securities, and “quants" were employed to price and hedge them. After the financial crisis and Great Recession of 2007-2009, when Merton uttered the words above, the Dodd-Frank Act had essentially shut down the exotic derivatives business. Was Merton just whistling in the wind when he insisted there was a role, even an increasing role, for quants in the financial service industry just after the near collapse of the banking sector? I think not. To understand Merton's optimism about the role of quants in finance, we must first understand what has changed in the past 20 years.
Twenty years ago, quantitative asset management was an oxymoron. Today, there are hundreds of fund managers who tout their use of quantitative methods.
In 1997, the bid-ask spread for trading on the New York Stock Exchange was one-eighth of a dollar, and specialists who would pony up $4 million for a seat on the NYSE could capture that spread on the millions of shares traded daily. By 2001, those seats had lost 75 percent of their value. Today, essentially all trading on the NYSE and elsewhere is done by computers quietly humming along. High frequency traders, who contribute the liquidity once offered by the specialists, must content themselves with a bid-ask spread of one or two cents per share.
In 1994, when Carnegie Mellon founded the first professional master's degree in quantitative finance, the recommendations of risk managers in banks were too often not taken seriously by traders or their managers. Banks now have a chief risk officer who sits in the executive suite and can overrule the most successful trader in the firm. Until the time of Lehman Brothers' 2008 bankruptcy, major banks assumed their loans to one another had negligible risk of default. Today, there are large groups of people in these banks who seek to estimate and price the risk of counterparty default. The mathematics required to price the exotic derivative securities of yore pales in comparison to the challenge of determining the price and hedging strategy for the risk of counterparty default associated with even the simplest trade today.
The world has changed and in this new environment quants are needed more than ever. Students with degrees in quantitative finance or financial engineering are pursuing careers in the following fields:
Quantitative Portfolio Management
Asset management firms need quants. A portfolio of assets is a collection of dynamically evolving random processes. If the portfolio contains derivative securities (e.g., futures contracts) as well as stocks and bonds, there may be complicated relationships among the assets. The goal of the fund manager might be to track an index or to outperform a benchmark. In both cases, the target is also moving, and the relationship between the managed portfolio and the target is complex. To understand these relationships, information is required. Modern computing equipment provides the capability of collecting massive amounts of data about the history of assets, news events, inventories of companies whose stock belongs to the portfolio and a host of other information. To store and manage these data, find patterns in the data, and make decisions based on the data, scientists use methodology from the rapidly developing field of machine learning. This methodology sits at the interface between statistics and computer science.
Other employers in this career path are hedge funds that take bets – but want to take precise bets. For example, to make a dollar bet that a country will not default on its bonds, one needs to take a bet on the dollar price of the bonds. However, the bonds are generally not denominated in dollars – thus, it is necessary to separate the exchange rate risk from the bond default risk. What instruments and in what quantity should one buy to accomplish this? This is a classical hedging problem – one that requires a mathematical model, data analysis and computer infrastructure. The name “hedge fund” derives from the need to solve this problem.
Unlike investing, where managers seek longer-term returns by understanding the movements of a portfolio of securities, currencies and/or commodities, a trader strives to take profitable short-term positions in single instruments.
In a New York Times article, Ryan Sheftel, the head of Credit Suisse's automated Treasury bond trading said, “Our best traders spend a lot of their time pounding away writing code. The code is making the trading decisions." This is a consequence of the exchanges going electronic. The intellectual content of the trading codes is secret, but we know that the skills needed to develop them are a combination of financial savvy, statistics, mathematical modeling and computer programming.
Banks create and sell derivative securities. These are demanded by firms engaged in production who need them to lock in prices of raw materials, reduce the exchange rate risk faced by their global enterprises, or otherwise reduce their business risks. For the banks, trading in derivative securities has become more difficult because of the recently recognized need to take counterparty default risk into account. An additional complication has arisen since the financial crisis – banks have multiple sources of funding, and because their inherent possibility of failure has now been recognized, some of these sources have become more expensive. Before the crisis, this extra cost (if indeed there was an extra cost) could be neglected. That is no longer the case. When a bank enters a contract, it either provides capital, which must be raised from some source, or it receives capital, which can be used to fund another deal. To correctly price a trade, the cost or benefit of funding the trade must be taken into account, and these costs can vary substantially depending on the source of the funding. Mathematical models to handle this new reality are still under development.
The science of computing and managing the credit and market risk borne by traders and portfolio managers has grown greatly in complexity and importance since the financial crisis. Finding people with the technical expertise to handle these computations has become a challenge for financial firms and regulators alike.
The Dodd-Frank Act mandates that banks compute a credit valuation adjustment (CVA) for each of their trades. This CVA takes into account the risk of counterparty default and the potential loss in the event of such default. This must be done for the whole portfolio, which can contain hundreds of thousands of long-lived deals. The CVA must be recomputed on a regular basis, and the bank is required to set aside risk capital that depends on the degree of CVA fluctuations observed. Consequently, banks enter trades that dampen these fluctuations, which is again a hedge, this time against counterparty default.
Another part of risk management is asset and liability matching. Insurance companies sell long-lived variable annuities, whose payoffs are partially guaranteed and partially market dependent. The market dependent part often allows the annuity owner to direct investments of the capital in the annuity. One of the challenges, in the face of uncertain markets and changes in longevity, is to determine the assets the insurance company should hold in order to offset the liabilities it has accumulated. The demand for quants in this area will continue to grow along with aging populations in the developed world.
A healthy financial system is a prerequisite for a healthy economy. We saw during the Great Recession following the financial crisis that when there is turmoil in the financial sector, people in many sectors of the economy lose their jobs. In the 21st century, a healthy financial system depends on the talent of well-educated professionals with a background in mathematics, statistics and computer science. In order for an economy to compete globally, it must have an efficient, competitive financial system.
It is easy to understand the need for people with an education in Science, Technology, Engineering and Mathematics (STEM) in medicine, manufacturing and Internet firms. The demand for STEM graduates in finance is just as great. Positions in finance are intellectually challenging, critical to the economy, and well compensated. It is possible to enter the finance industry with a STEM education at any level, bachelor's, master's or Ph.D. Many people choose to enter after a professional master's degree in quantitative finance, a degree that replaces the MBA for this career track.
The website, QuantNet, lists thirty professional master's programs in quantitative finance in North America.
The ticket of admission for these programs is an undergraduate education that includes multi-variable calculus, differential equations, linear algebra, calculus-based probability, statistical inference and data analysis, experience programming in an object-oriented language (e.g., C++), good written and oral communication skills, the ability to work as part of a team and an interest in or some knowledge of financial markets that is not necessarily learned in a classroom.
For people with such an education, Merton's optimism about careers in quantitative finance is entirely justified.