The downside of downside risk management.
Despite the postelection rally that delivered robust returns for U.S. equities, 2016 was a decidedly volatile year for markets. With heightened market volatility came increased discussions about risk management. This is a common response, and when volatility returns—which it always does—the popularity of risk-focused approaches and products will spike, too.
I take issue with most risk-management approaches and risk-based products (for example, controlled volatility or tail-risk hedging strategies) that stem from modern portfolio theory, or MPT, especially for long-term investors. My disagreement starts with the MPT definition of risk as short-term return variation, runs through the practice of portfolio diversification, and goes all the way to how risk-based strategies are managed and marketed.
A long-term investor can benefit from short-term volatility and can be harmed by overdiversification and costs from strategies that neither reduce the real risk of losing purchasing power nor improve the investor’s ability to achieve long-term financial goals. I’ll explore my thoughts on risk in greater detail below, but let’s start with a look at traditional risk management.
Risk Management, MPT-Style
Given investors’ rational desire to maximize the upside and minimize the downside, the definition and measurement of risk matter. If we’re going to manage risk, we first must know what it is and why it’s not always good for returns. And if we—or our clients—are long-term investors, the returns we should think about are long-term ones.
However, risk management in the MPT world (which most professional investors live in) is largely represented by the management of the movement and comovement of asset returns over short intervals using historical return data. Past relationships are extrapolated into the future, directly or indirectly. The cornerstone of MPT is measuring risk via the relative movement of market returns to produce metrics such as beta and alpha—terms related to the defunct capital asset pricing model, or CAPM. These frameworks rely on econometric analysis of monthly return data and use various forms of short-term return variation to measure and attribute risk.
In short, because of MPT’s focus on short-term variations, generating smooth returns has become the holy grail in our profession. For many investors, this is not only irrelevant but also can be expensive. The desire for smoothness reduces returns over time while not necessarily reducing the risks the investor should be worried about.
Early last century, a few investment theorists laid out those risks—namely, the risk of permanent loss of capital, the risk of being overly or needlessly diversified, and others. Somewhere along the way, the focus shifted from things that affect the long-term range of outcomes but are hard to measure to things that don’t matter as much to long-term investors but are easy to measure— that is, very short-term variations in asset returns.
Myopic loss aversion—or the combination of investors feeling more hurt by losses than helped by gains and constantly reviewing portfolio returns despite having a long-term focus—seems to have been codified into many investing frameworks and processes. This came not from inexperienced investors but from the professional investment management industry itself.
Practically, this has led to a revealed preference for investment strategies with embedded managed volatility, downside protection, or variable-annuity riders, despite their costs to long-term investors. As a long-term capital allocator who used these tools a lot early in my investing career, I’m very skeptical. Some of the best lessons come from your errors—as Oaktree Capital Management cofounder Howard Marks has brilliantly observed, “Experience is what you got when you didn’t get what you wanted.”
What I learned was that this type of short-term, volatility-based downside protection doesn’t generally work for long-term investors because of three main kinds of costs. First is opportunity cost. Every dollar invested in a protection strategy that accepts lower upside in return for downside protection is one not capturing the long-term “miracle of compounding” (as Vanguard founder John Bogle puts it) of reasonably priced equity markets or other assets that can be volatile in shorter periods. Short-term smoothness of returns often comes at the cost of lower long-term portfolio value. To paraphrase Berkshire Hathaway vice chairman Charlie Munger, for investors with a long horizon it can be better to take a lumpy 12% return over a smooth 9%.
The second cost is a matter of value. My sense is that few downside-focused investors consider the value they get for the fees paid or how they might accomplish the same goal differently. This is especially true for downside strategies whose costs vary over time, such as annuities. When the price you’re paying is high relative to the thing you’re trying to avoid, the implied probability of the bad thing happening rises—i.e., it costs more to hedge the risk. Often, investors collectively dislike these bad things at the same time, and the cost can become excessive. But for long-term investors, volatility can create opportunities, as prices tend to move around more than fundamentals. Paying a fee to remove an opportunity seems illogical.
The third cost relates to how investor demand shifts with recent performance. After steep losses in late 2008, many investors rushed into downside protection strategies in 2009— akin to shutting the barn door after the horse has bolted. Investors who entered these strategies likely ended up with much lower returns than the returns of the strategies they left behind. This underperformance comes from timing—they sold something with a low valuation and bought something that was overvalued. Performance-chasing is a destructive behavior for investors.
Meanwhile, some solutions can be cost-free. Continuing Munger’s example, if you have trouble stomaching a 12% bumpy return, consider advice from behavioral economist Richard Thaler and look at your investment statement less frequently. Doing nothing can be a surprisingly effective investment tool.
It’s also possible to pay too much for diversification and to overdiversify. MPT considers diversification, as it does all things, through the lens of backward-looking data—namely the comovements of short-term asset prices. Recall the simple premise that diversification aims to offset losses with gains—yet, as with all investment problems, this one is set in the future.
Asset A may zig when Asset B zags, but if Asset A is overpriced, why would I hold it? Few assets are diversified by just one other asset. It seems more sensible to hold assets that not only add diversification but also are underpriced. What doesn’t seem sensible is overloading a portfolio with assets in the vague hope that they will add diversification. If fundamentals are the primary driver of returns, they must also play a major role in diversification. The stocks of two companies that are fundamentally similar should not be relied upon to provide diversification, irrespective of the historical correlation of short-term returns.
Fundamentals also drive diversification at the asset-class level. Corporate profits supply the returns from stocks, as interest rates do for bonds. Stocks provide cash to investors via dividends and buybacks, bonds via coupons and repayments of interest and principal. When the economy falters, profits tend to slip, leading to falls in stock prices. Bonds, on the other hand, tend to see policy rates fall along with inflation expectations, leading yields to fall and prices to rise.
It is this fundamental linkage that drives the diversification benefits of stocks and bonds. This extends to commercial property, with cash flows linked to contractual rental income often tied more closely to inflation. Diversification at the asset-class level links back to the cash flows that are generated by the underlying assets. Provided valuations and ownership levels are not extreme, the offsets come from this source.
Because fundamentals matter in the long run, the short-term comovement and beta are less important than these fundamental linkages. The most important inputs should be valuation and fundamental risks, at the asset and portfolio level. These drive the variation in long-term values and returns. Diversification helps manage uncertainty, reduces the risk of being wrong (primarily through the intrinsic value), provides offsetting gains and losses during extreme environments, and helps set the level of risk consistent with an investor’s tolerance.
A Potentially Better (But Harder) Way
An approach that relies heavily on historical price-based statistics isn’t investing. Benjamin Graham, the father of value investing, would call adherents to that approach speculators instead of investors. “Most [speculators] are guided by charts or other largely mechanical means of determining the right moments to buy and sell. The one principle that applies to nearly all these so-called ‘technical approaches’ is that one should buy because a stock or the market has gone up and one should sell because it has declined. This is the exact opposite of sound sense everywhere else,” Graham writes in the introduction to The Intelligent Investor (emphasis original).
We, like Graham, prefer to rely on things other than recent price movement when investing— specifically, how the intrinsic value of securities compares with their market prices. This is especially true when making decisions under uncertainty, and there’s always considerable uncertainty with investing.
Why do we prefer valuation? Valuation assesses a company’s prospects—the heart of a business— so it must be at the heart of investing. Buying stocks or markets that produce free cash flows and are available at attractive prices is straightforward, easy to understand, and delivers more value to the patient investor than most fads, theories, or formulas. The output will be subjective, qualitative, and judgment-driven. But that doesn’t make it any less rigorous.
The following maxim is apt:1 “Not everything that can be counted counts, and not everything that counts can be counted.” When measurability takes precedence over efficacy, there is room for error. For an investor with a long time horizon operating with no borrowed money and limited near-term need for funds, does her sensitivity to market beta matter? Does monthly tracking error matter? If she has tracking error relative to a benchmark of 3% versus 1%, is that relevant? These questions are not asked enough.
Many investors struggle with uncertainty and turn to these MPT-inspired metrics. These tools seem to produce tangibility and precision—some certainty out of the uncertainty—but they may be just adding risk for the long-term investor concerned with the permanent loss of purchasing power over a multiyear window. The messiness of reality is unavoidable, and, irrespective of how many decimal points we put on the end of a number, uncertainty is not reduced. One thing that can change is the confidence in the decision-maker through artificial precision. Buyer beware.
Absurdities in Practice
Now back to the application of this MPT form of risk management. Many strategies that seek to manage risk by using monthly, weekly, or daily price movements for investors often lead to expensive smooth return lines but not necessarily to lower risk of the permanent loss of capital. Think of managed volatility, options, or long-volatility strategies. A popular substitute for options or long volatility are trading strategies—I reserve use of the word investment for strategies that focus on the underlying ownership of assets and cash flows they generate for owners. Trading or speculative strategies are more interested in price appreciation over a relatively short period.
Without going into too much detail, and at the risk of oversimplification, these strategies are designed to sell when markets fall and buy when prices rise. Through attempting to limit downside risk or match a volatility target, the strategies exhibit strong procyclicality: The more markets fall (or volatility rises), the more they sell, and vice versa. Central to sound investing is buying low and selling high, as determined by price relative to intrinsic value. Not only do these trading strategies generally ignore valuation, they buy when prices are higher, not lower.
Pocket the Insurance Premium
What should matter to investors is whether they’ll make money over the long term and that the time horizon matches their liquidity needs. Risk, to paraphrase Warren Buffett, is a loss of future buying power.
An unleveraged investor who doesn’t need to sell his assets for many years typically has little need for a smooth line. Even retirees in the “danger zone,” or the years too close to retirement to recover from a market crash, usually do not need as smooth a line as is sometimes suggested (the exception being someone about to purchase an annuity for longevity reasons).
That is not to say bad things can’t happen in markets—they can and do. However, the price one pays for something matters. And I fear, as mentioned earlier, that many investors are overpaying for downside protection. The volatility market, or any investment product that “pays” for lower volatility through upside reduction, is prone to overpricing, and this tends to find its way into various forms of hedging.
Of course, overpaying isn’t the only cost; there’s also the opportunity cost of giving up better long-term returns and the risk of not meeting a financial goal. Insurance or the reduction of risk costs money, and investors must accept this trade-off, despite what those promoting their silver bullets might suggest.
Focus on Fundamentals
It’s appealing to think we could generate attractive returns without the pesky downside risk of the balanced portfolio. But this ignores the fundamental nature of markets. Selling when something is falling and buying when it is rising doesn’t seem like a rational strategy for a long-term investor. Above everything else, price relative to fair value is the most important principle.
For some of you, the parallel to the portfolio insurance of the 1980s will be clear. It illustrates that there really are no new ideas, and that even the bad ideas tend to get recycled. Economist John Kenneth Galbraith outlined this brilliantly in A Short History of Financial Euphoria when he stated that “the world of finance hails the invention of the wheel over and over again, often in a slightly more unstable version.” Let’s hope these prophetic words prove unnecessary for the latest version of portfolio insurance. Nevertheless, we should always stand guard.
There will certainly be people who disagree with my skepticism on MPT and “risk-managed” strategies. That’s OK, because diversity is healthy in every context. And my major concerns lie not so much with the tools but with their application and interpretation. At times, though, we lose sight of the bigger picture and slip into handling the question we know the answer to, even if it is the wrong question.
Instead, I urge investors to remain focused on the long term and the fundamentals. We need to remember that the underlying fundamentals drive most long-term investment returns, that price matters, and that investors’ goals and time horizons should inform strategy. That’s the way we approach investing.