Investors are facing a strategic conundrum in a world of vast political, macroeconomic and market risks. In this environment, generating a better risk/return balance requires a deeper understanding of the sources of risk across asset classes and new data analysis techniques.
More than a decade has passed since the global financial crisis, yet the impact is still being felt. Years of accommodative monetary policy have left interest rates historically low and equity markets fairly valued around the world. As a result, expectations for future returns are more muted and investors need to make their assets work harder—in many cases by embracing more risk.
While risk-taking is a normal part of investing, the challenges are formidable. That’s because the sheer scale and variety of risks that investors face are daunting. These include the following:
- Populism continues to put strains on traditional political organizations. Brexit and electoral shifts across the European continent have rattled traditional mainstream politics. In the US, the upcoming election in 2020 features a highly unorthodox President Trump seeking reelection against a Democratic party entertaining redistributive economic policies of a type not seen in our lifetimes. The common denominator in all this is a sense of deep dissatisfaction within a large segment of the population in developed countries, and they are letting their voices be heard.
- Political friction, primarily but not exclusively between the US and China, is driving down global trade. This is beginning to have a noticeable impact on economic activity as reflected in PMIs and other sentiment surveys. It also reflects a broader retreat from decades of increased globalization, which has served to drive economic growth and limit inflation.
- Governments’ stimulative policies have increased their debt balances. In many countries, including the UK, US and Japan, debt-to-GDP ratios exceed 100%. Despite this, a third of sovereign bonds “pay” negative interest, contradicting the very first finance lesson we all learned: the time value of money.
For managers of capital, these seemingly unprecedented risks create a difficult bind in the need to stretch for more return. Risk management has evolved markedly over the decade since the twin market shocks of the 2000s. But, the scars earned during those downdrafts have left many asset owners wary of unforeseen risks and much more conscious of the need to protect capital.
So, asset owners, consultants and investment managers alike are increasingly approaching the question of risk, and how it interacts with returns, in new ways. What are some of the ways that risk management has evolved?
To answer that question, let’s examine four risk-management techniques:
- Asset Allocation
- Managing Factor Exposures
- Security Analysis
- Use of Big Data
Level 1: Asset Allocation and Beta Management
Asset allocation remains the most powerful lever most asset owners can pull to manage risk. Over time, we’ve moved from simplistic asset-allocation structures and measurement toward the use of more asset classes, in more varied ways, while also paying more attention to return patterns.
Choosing the most appropriate risk-management tools often depends on an investor’s risk priorities and time horizon. For investors with immediate defensive needs, options and tactical allocation strategies can help protect against short-term tail risk. But some long-term investors may be less concerned with short-term protection than with managing funding gaps many years into the future.
Alternative strategies, broadly defined—in liquid and illiquid forms—are effective as long-term diversifiers. That’s because they tend to have uncorrelated alpha with equities. Liquid assets offer more short-term flexibility and can be cheaper. However, many investors prefer illiquidity, as these assets offer a premium associated with less mark-to-market risk along the way, a prized feature in a volatile world.
Within traditional equities and bonds, the roles have also evolved. Both now serve as both return-seeking and risk-reducing asset classes. There is a greater appreciation, for example, that a high-yield credit portfolio has more in common in terms of economic sensitivity with an equity fund than with a core investment-grade bond fund. Likewise, a low-beta or long/short equity strategy may play more of a risk mitigation role than previously envisioned for traditional equity allocations.
Importantly, though, beta to an index is only one measure of a security or portfolio’s sensitivity. Beta can be tied up in complicated ways with other fundamental financial measures like leverage, volatility of earnings and asset intensity. Similarly, risks change over time: China and Brexit dominate investors’ concerns today but are likely to be surpassed by new and unforeseen risks in a year or two. As a result, we think it’s important to measure the correlations of investments to a wide variety of risks in the market, of which many may be temporary. This requires active monitoring and management.
By freeing ourselves from traditional categories, and instead relying on empirical relationships between asset classes, institutions can more creatively access pools of alpha and put them together in the best way to meet their plan’s return needs and risk tolerances.
Level 2: Factor Management and Measuring Alpha
Risk management is also evolving in the ways that investors access beta and alpha, with more attention being paid to the importance of factor exposures. Increasingly, investors are separating exposures to beta and alpha—and how they pay for them. Technology has empowered investors to easily move beyond cap-weighted exposures and incorporate other systematic exposures to sectors or factors. Although this can provide more efficient accessing to different return premia, it also comes with a different set of risks.
First, factors are not all created equally. A rules-based index still needs someone to define those rules, and those definitions can have a significant impact on how an index might behave. This is particularly true in less developed areas like quality or low volatility, where there is little agreement on what these terms mean. This can result in very different holdings, and thus different risk exposures between portfolios that sound similar in what they aim to offer investors.
Second, factor performance is not random—it is often linked to different parts of the economic or interest-rate cycles. Value factors might do particularly well during an economic recovery but struggle during a contraction, a period when more defensive factors show their strength. By measuring the balance of factor exposures across a range of strategies, asset owners can gain a clearer picture regarding when an allocation could be expected to win or lose.
However, because return streams can be disaggregated, it is also possible to separate out style effects from true, idiosyncratic alpha driven by manager skill. New tools can help investors systematically determine whether active managers are indeed generating alpha from stock-picking skill.
Many investors have turned to passive exposures to redefine their allocations and reduce costs. However, passive strategies only limit relative risk, and while that might help allocators manage career risk (by limiting risk relative to a benchmark), it won’t necessarily cushion an allocation from absolute risk when markets hit a rough patch.
Level 3: Security Analysis—Revisiting the Sources of Risk
With the increasing popularity of passive investments, some investors are paying less attention to individual security risk. But fundamental research and analysis is not obsolete. In fact, portfolio managers today have better technological tools at their disposal and access to more data that can help improve risk reduction through a rigorous security selection process.
To truly manage the absolute and relative risk of portfolios requires in-depth security analysis, in our view. Macroeconomics can be helpful, especially in the arena of interest rates or currency, but most alpha stems from insights into industries and companies. And while the hunt for a high-returning investment is always alluring, paying close attention to risk is as important to driving superior performance.
In our view, four sources of volatility can help us better understand what causes one security to be riskier than another:
- Business models— Sensitivity to economic growth, shifting supply-demand balances, high fixed costs and operating leverage can all create more volatile cash flows, and thus more volatile security prices.
- Capital structure—Companies with more leverage tend to have more volatile securities volatility and are also more exposed to interest-rate risk.
- Sensitivity to exogenous factors—Businesses don’t operate in a vacuum, and their exposure to risks outside of their control, like regulation, can be an important driver of risk.
- Market sentiment or crowding—As much as we like to think that investing is purely analytical, we know that emotion, fashion and the power of crowds all influence security prices.
You can’t completely avoid these risks—and you shouldn’t. Indeed, taking risk is part of earning a return. Rather, understanding the sources of risk can allow a portfolio manager to better diversify them and establish the appropriate discount rate for any individual investment. Risk isn’t inherently bad if you ensure that you’re being adequately compensated for taking it.
Level 4: Big Data, New Frontiers
While risks have mounted around the world, the technological means of addressing market hazards have also evolved dramatically. New technology means that we have more data than ever to analyze our investments. And tools like machine learning and natural language processing continue to expand the information we have at hand. Big Data allows investors to tackle the same questions and controversies they have always faced, just more efficiently and with more available data points. But there is no substitute for human judgement in evaluating this data and reaching insightful investment conclusions.
We can also tap new analytical techniques. For example, here at AllianceBernstein, some portfolio managers are successfully applying cluster analysis, a technique borrowed from scientific research, to investing. Rooted in empirical observations of security prices, cluster analysis segments securities into groups that have been performing similarly in recent periods. Rather than using groupings based on sector, country, profitability or valuation, cluster analysis takes the facts as they are: these holdings are all behaving similarly, regardless of whether you think they should or not. Our analysts then seek to uncover explanations for the behavior patterns and identify potential hidden risks in a portfolio.
Be Careful What You Wish For
Risk management can seem technical at times. But it’s important to recognize that risk is usually rooted in some fundamental causes and that there’s no magic tool or statistic that can solve our problems.
Indeed, the rapid adoption of common risk models, and the increased focus on short-term returns, could be sowing the seeds of problems down the road, in our view. If everyone uses the same tools and decision rules to set allocations or trade portfolios, a shock to the system could cause everyone to pile into the same set of trades at once. This would only exacerbate volatility.
And while it’s important to learn lessons from prior crises and the mistakes we made, it would also be a mistake to learn "too much" from the past. Our challenge is to manage the risks that drive the market today—and tomorrow. This is especially important given the current high degree of political and economic uncertainty around the world.
For additional perspectives on refining risk management in today’s uncertain market conditions, read more from our investment teams in this edition of AB IQ. In Portfolio Risk Management: A Multidimensional Perspective, our multi-asset portfolio managers provide a detailed analysis of how different risk diversifiers perform and protect over time. And our senior quantitative equity analysts present an exciting new technological application to risk management in Cluster Analysis: Managing Risks You Didn’t Know You Had.
