B, Chase Chandler, FRM, CFP
Feb. 29, 2020
Equity markets sold off at quite a pace this week. It was the worst one week decline since 2008. We do not usually comment on weekly market movements. Given the rarity of this decline—and the fact that we have been positioned for such an event for a while now—here is a quick update on where we are and how we think about recent happenings.
Valuations had been stretched coming into this year and became more so in early February. A combination of events turned the tide. The market shrugged off Coronavirus (COVID-19) concerns for weeks before finally taking a dive. We believe the catalyst was Bernie Sanders' win in Nevada last weekend (Feb. 22) and the rising probability of his securing the Democratic nomination. If he keeps winning primaries it will be difficult for the party to deny him his spot in the November Presidential election. Health care stocks, which typically act somewhat as safe havens during corrections, have sold off at a more severe rate than the general market. The combination of a potential pandemic virus and Sanders as President of the United States has proven detrimental to such lofty market expectations. Markets will continue their slide to the extent the likelihood of each risk continues to grow. On the bright side (which we are extremely excited about), all of this could create some of the best buying opportunities we have seen in many years.
Performance Update
Through February, YTD performance for our Core Equity (“CTG CoreEq”) and Core Tail-Hedged Equity (“Core[TH]Eq”) are -2.5% and +2.0%, respectively. Over the same period the S&P 500 is down -8.5%, small-cap stocks have declined -10.4%, and Vanguard's Total World Stock ETF is down -8.7% (Figure 1). Figure 2 shows the value of $100 over the past 12 months compared to the S&P 500 and a hypothetical 70/30 portfolio split between the total equity market (ticker: VT) and the total bond market (ticker: BND).
Figure 1
Figure 2
Let's keep in mind that we have not yet fallen all that far—still less than -13% from the highs in mid-February. The decline in Q4-2018 was around -19%. The difference here is that the potential problems are systemic in nature whereas the 2018 drop was akin to a temper tantrum thrown by a toddler (the market) wanting more candy (low interest rates). This is not that. This is something that, even if the probabilities are low, will be catastrophic to the global economy. At best, we will see China’s economic growth decline dramatically with knock-on effects to the rest of the world. But the potential is for a global recession the likes of which we haven’t seen in a very long time. Something worse than last decade’s Great Financial Crisis.
Reiterating what we said in our 2019 year-end letter (sent out in January), the populous tends to overreact to trivial headlines and underreact to potential systemic risks. Some say the Coronavirus is not a big deal or that panic will cause more harm than the virus itself. The flu or people being hit by buses cause more deaths, they say. This is dangerous thinking, ignorant of power laws and exponential, higher-order effects. The flu and accidental deaths follow a normal distribution, which does not imply we should trivialize them, but they do not have the multiplicative potential to kill millions over a short span.
This is a risk-based issue. What happens if we panic and it turns out to be nothing? vs. What happens if we do not panic and it turns out to be something? In the first case, maybe we lose a few points of GDP growth for a quarter or two. The latter case will be much worse. There is significant downside asymmetry to events like pandemics, wars, nuclear explosions, and severe recessions. The cost of preparation and protection is extremely small relative to the risk, and little is lost if nothing happens. When something does happen, the cost of neglect is enormous. The misperception is that the probability of a catastrophic event (e.g. pandemic, attack on U.S. soil, market decline of -50% or worse) is small, hence can be disregarded. A specific event may have a low, single period probability, but the probability that something will eventually happen is for all practical purposes 1.
Figure 3
How we think about large drawdowns
Worthwhile investment managers do not get paid for accommodating the moment by moment desires of investors. Those who try to satisfy every investor’s wants do not make it through more than one or two cycles. Over the past few years, we have heard innumerable commentators, journalists, and everyday investors compare quarterly or annual returns for nearly every type of asset class/style to the S&P 500. After 2018 they said “[that] fund was down by 10% but the S&P was down by less.” In January they said, “the S&P was up by over 30 percent for 2019; [x] fund trailed by 20!” The explicit inference of this commentary—that the S&P 500 index has been and forever will be better than essentially all investment managers—has created major vulnerabilities and fragilities in the system.
Many who are first shall be last, and the last first[i]
When things are not going well, and even when they are at times just okay, it is important to take a step back and pinpoint what we are doing well. We need to remind ourselves that momentary appearances in markets are frequently deceiving to the untrained eye. Attractive payoff profiles where, for example, we risk one to make ten can have long stretches that give off the appearance of poor performance. But, when one really understands what is going on, they know what is coming. In these regimes (periods) we must be patient and prudently manage risk. We must resist the pressures and temptations to move to recent high-flying securities that have ostensibly made others rich. Those who do not maintain sobriety only set themselves up for problems later.
When things are going well, it is often a time to seize opportunity; to put our foot on the pedal, so to speak. In these periods, we must also look for areas where we can get improve our processes and execution in searching for antifragile payoff profiles. We believe we have entered this regime.
Markets do not provide predictable, normally distributed returns which can easily be modeled by anyone with a computer. One cannot assess the validity of an investment strategy or philosophy based solely (or even mostly) on recent returns. A counter example of something that is normally distributed is the time it takes to run a mile. Say we have two runners, each of whom ran one mile nearly every Saturday in 2019. For our purposes we will assume 50 runs. The first runner does not beat a 9 min/mile in any of the 50 attempts. We will not find him all the sudden beating runner number two, who ran a sub 5 min pace in each of 2019's 50 runs. You will also not find in a group of 100,000 runners, an outlier or two who run a two-minute mile. It is not possible. If something is not possible---if we can say that about a system---it is normally distributed.
People treat investing as something akin to the racing example above. They believe they can assess what is better based on a large enough sample size of recent results in the same way they can assess who the faster one-mile pacesetter would be given 50 runs. Even seemingly smart people (particularly Ph.D.'s, authors, journalists, financial advisors, and "data scientists") fall into this trap. They think being "data driven" is good across all domains, not realizing how useless that data will be in fat-tailed domains. They mistake backward-looking outcomes for probabilities, assuming the "runner" who has won most of the time will continue to do so. While this is usually true in normally distributed systems (running races), it is terribly misguided and dangerous in complex, nonlinear systems (markets).
Investors around the world are investing based on the idea that markets are normally distributed, even if they espouse differently. They bet their life savings on the 5 min/mile runner and are stunned when he suddenly can't finish in under an hour. Meanwhile the nine-minute runner, in a seemingly unbelievable turn of events, completes a mile in under 120 seconds. Now of course this would not happen in an actual race, assuming the runners are healthy. If it did occasionally occur, we would be hard pressed to find a right-minded person who would consistently bet on the "faster runner." Yet this is exactly what does happen in markets. Just substitute most investment strategies for the "5 min/mile runner", and the few principled investment philosophies/managers for the "9 min/mile runner."
In one system we can rely on what has happened recently as a guidepost for what will happen in the future. In the other system, the exact same actions that worked yesterday may not work today. And what didn't work yesterday may well work today. Great investing means managing risk when most are imprudently chasing returns. During this period, which is where we've been for the past few years, the apparent winners are most exposed. Those who appear to be losing will eventually be the winners precisely because they understand what is happening and are willing to trail other investment approaches in order to protect capital. One cannot allocate capital at attractive prices unless he/she still has the capital to invest.
An ironic fact of investing: many who understand the principle of the saying “many who are first shall be last...” apply it almost exclusively in one (eternal) context. They fail to transfer that knowledge to practical domains and miss its implications in this world. It is a risk management heuristic. It is Mark Spitznagel’s “seen vs. the unseen” (The Dao of Capital, 2013). Those who rely heavily on the “seen”—what is clear and obvious—can succeed only in linear fields. They have no chance in market-like arenas. Seeing the “unseen”—that which is inevitable but has not yet occurred—is not forecasting or predicting. It is a pragmatic understanding of markets and risk; which encompasses the real-world aspects of mathematics, statistics, and probability, in addition to, among other things, history, financial statements, valuation, market structure, and human behavior. Seeing the unseen means knowing (a) most of what could have happened in the past did not, (b) markets will not act in the future as they have in the past, and (c) how to structure investments accordingly. It is the certainty that something unpredictable will surprise and roil market participants who are so reliant on forecasting and fancy models.
We cannot predict what will happen, but we know something will happen. Risk is always in the system. It does not matter what the whatis. Our investment return profiles are structured around that something. And trying to predict what that something will be with specificity demonstrates someone naive to how markets, statistics, and probabilities work in the real world.
Those who take on poor risk-reward profiles (e.g. risking ten to make one) with so-called “high probabilities” will go bust. Historical results are irrelevant if one took asymmetric downside risk to achieve them. These investment strategies can look good for extended periods because they “win 99% of the time.” But what happens the other 1%? Nothing good. These people do not get cumulative probabilities vs. single period probabilities. And decent people will pay the price (or already have) for their ignorance. Even if you have a small chance of blowing up over a short period, your chance of going bust rises dramatically the longer you play. Be wary of this type of thinking. When it comes to investment returns, pushing to be continuously first today will cost an investor everything tomorrow.
[i] Mark 10:31
コメント