Clearbridge Value Trust commentary for the fourth quarter ended December 31, 2015.
“You can never do merely one thing.” – Garrett Hardin
Few of mankind’s inventions are as interconnected as markets. Yet, we often tend to simplify markets by studying things in isolation and treating certain variables as constants. Ultimately, these simplifications break down under the pressing weight of reality, as simple stories are extrapolated to extremes that cannot withstand the countering force of valuation and unachievable expectations. With this caveat in mind, this letter will focus on risk through the lens of crowding, and the current state of two well-known financial factors: price momentum and value. Through this two-variable simplification of markets, we will try to answer a question we get a lot: are we finally due for a turn in the market cycle that favors value?
To tackle this question we will start by defining some terms and reminding people of how we frame value in our investment process:
- From a factor perspective, we will use Fama-French data that frames value on book-to-price and price momentum as relative returns between winning and losing stocks. To be clear, we are not factor-based investors, but we are certainly aware of our active factor exposures as part of our regular portfolio construction reviews.
- For crowding, we will use Bernstein data that measures crowding based on institutional ownership (the number of active managers that are overweight), sentiment (price momentum and sell-side analyst ratings), and expectations (earnings forecast and valuation). We have found that Bernstein’s crowding framework has done a great job of capturing popular stocks, where forward risk potential from downside price volatility is much higher than would be indicated by traditional risk measures like beta.
- On the critical aspect of value in our investment process: We frame value, whether applied to traditional growth or value stocks, through the prism of business value. In this regard, we typically estimate business value by discounting cash flows under a range of scenarios, from the nightmare to the dream. This process focuses our attention on the true long-term drivers of value, which are growth and the ability to invest capital at a return above the cost of capital. In all cases, we are looking to exploit a gap between price and value, which converge to drive portfolio returns if realized fundamentals exceed the original market expectations embedded in the price.
With these broad concepts out of the way, we can provide some context for 2015 by comparing the performance of these two factors, and by the contrasting examples of the portfolio’s best- and worst-performing holdings in 2015.
Clearbridge Value Trust – Momentum Versus Value
In Exhibit 1 below, we show the calendar year performance of momentum versus value over the last 20 years. In 2015 through November, momentum outperformed value by over 25%, which is the fourth-largest gap in favor of momentum over the past 20 years, and the second-largest gap since the Tech Bubble. You can also observe that post the financial crisis in 2009, momentum has outperformed value in five out of six years, as this market cycle has flattered a growth and momentum investing style. We are also struck by the much greater volatility of the momentum factor relative to value, with the extreme collapse in momentum in 2009 creating an initial condition that favored momentum during this cycle. Not surprisingly, after several years in favor of momentum, including substantial absolute and relative performance for the factor in 2015, Bernstein’s crowding measures suggest that momentum is now quite crowded relative to value. To be sure, the current crowding into momentum and the historic volatility of the momentum factor suggest risks are building, but this exhibit alone does not argue for an imminent reversal. We will, however, take a stab at potential paths for momentum and value going forward when we look at cumulative returns of momentum and value cycles later in the letter.
One reason we believe crowding is a robust measure of risk is the increasingly extreme behavior of markets, which in many ways reflects much wider outcomes between winners and losers in the economy. We think more extreme outcomes are likely the result of digitization, and potentially the “commoditization” of equity trading that digitized markets enable through new, primarily passive, investment strategies. Either way, crowding seems to enable feedback loops that have elongated swings between price and underlying value. Wider dispersions between price and value are an important risk consideration for our process, but also present us with greater long-term opportunities. Our goal is to try and manage the challenges and risks of crowding through our continuous effort on portfolio construction, where we explicitly track the portfolio’s exposure to a myriad of risk and return factors, including crowding. We are comfortable owning crowded stocks but we carefully monitor the risk of being run over when the crowd heads for the exits. Meanwhile, we want to be a provider of liquidity in uncrowded names as long as we are being paid to take the risk of investing in these under-appreciated stocks.
The extreme nature of markets was on full display in 2015 with the extreme dispersion of two of our portfolio holdings: Amazon and CONSOL Energy, which happened to be the second-best- and second-worst-performing stocks, respectively, in the S&P 500 last year. We detail both positions below, but unfortunately the mathematics of declines is such that the mistake on CONSOL Energy helped contribute to a disappointing performance year for the portfolio.
Starting with the winner, we added materially to Amazon in 2014 when the stock was anything but crowded, and our analysis indicated the stock was priced potentially well below long-term business value. The narrative at the time was that Amazon’s CEO, Jeff Bezos, did not care about shareholders and would never show any expansion in profit margins. Not surprisingly, when we valued the stock in late 2014, consensus was calling for minimal margin expansion over the next few quarters. Our analysis also indicated that embedded expectations for margin expansion were pushed out into future years. These pessimistic expectations created an attractive long-term opportunity as we expected Amazon to continue to scale its legacy ecommerce operations, and to enjoy the rapid scaling of its dominant commercial cloud business, Amazon Web Services. This dual scaling would likely drive profit margins at some point, which would drive the stock much higher. In fairness, we had modest expectations that the stock would reflect single-digit operating margins in 2015, which was enough.
However, after an almost 120% move in the stock last year, Amazon stock fully phase transitioned to now reflect double-digit long-term operating margins. We actually don’t think this is unreasonable, but the stock going up has attracted a crowd, and placed it in the top 10% on Bernstein’s measures. The narrative now is that Amazon is perfectly positioned strategically to run the tables in disrupting legacy retail and legacy tech, which also seems reasonable. The issue is that many of the shareholders now own Amazon for one reason: the stock has gone up. If Mr. Bezos doesn’t give these new shareholders a linear ride on margins, which has never been his long-term motivation, the stock could suffer some short-term volatility as the crowd seeks to exit their investment on the first signs of margin degradation. Thus, we remain long-term believers in Amazon, but the new bullish narrative and crowding give us some pause.
Clearbridge Value Trust – Energy sector performance
The energy sector was by far the worst performing sector in 2015, and CONSOL Energy was further burdened by its dual positioning as a coal and natural gas producer. Fortunately, CONSOL is one of the lowestcost producers of coal, and they wisely diversified into natural gas years ago when they recognized the longterm challenges facing coal. However, in 2015 CONSOL’s earnings collapsed as the demand for coal plummeted, and North American natural gas remained acutely oversupplied. As a result, CONSOL lost money on every hydrocarbon molecule it produced. The extreme pain of current pricing is not sustainable for the industry over the long run, as supply will inevitably be curtailed through a Darwinian culling of high-cost and overly leveraged producers. At this point, the main questions are who will survive and what commodity price is discounted. We think CONSOL has the liquidity and asset strength to survive the severe downturn, and our numbers indicate the stock discounts roughly $2.50 gas, which is a fairly attractive long-term level. However, we certainly wish we had fully appreciated the volatility that CONSOL’s coal exposure would add to the stock, and we have let the position shrink to one of our smallest positions. Like all cycles, low prices will inevitably lead to higher prices and a new up cycle, but we will remain disciplined in positioning for this turn with a primary focus on the companies with the balance sheet strength to survive the turn.
With the extreme examples of Amazon’s momentum and CONSOL’s deep value pain in mind, let’s take a look at the cumulative returns for value and momentum cycles since 1970.
The next two exhibits show the magnitude and duration of momentum and value cycles since 1970, where each up cycle is represented by 10% or higher trough-topeak performance and each down cycle is represented by -10% or larger peak-to-trough correction. For instance, there were 17 up and 16 down cycles for the momentum factor between 1970 and 2015, whereas the value factor experienced nine up and nine down cycles over the same time period.
Starting with momentum, as shown in Exhibit 2, the current cycle has been the longest up cycle for momentum since the late 1990s equity bubble. This is not necessarily damning, as this cycle did start from a position of extreme underperformance for momentum. We also have not matched the return magnitude of many of the previous up cycles. However, to boldly bet on strong forward returns, momentum investors have to start making a repeat-bubble argument. This may not be too much of a stretch given the competitive dominance of some of the current winners, like Amazon, coupled with an environment of low interest rates that is starved of cyclical growth. The key is to be open-minded to this possibility given the current context, while fully realizing that the cycle contains the seeds of its own demise as it ages and that realized returns give way to rising downside risks.
The current down cycle for value, in Exhibit 3, looks somewhat average by comparison. Yes, the value factor has not done well on an absolute basis, but with most value indices down low single digits in 2015, the real recent pain for value has been relative to the recent strength in momentum as evident in Exhibit 1.
This leads back to our original question on whether we are due for a value up cycle. The answer is not immediately, in our opinion, as the crowd will stay with the momentum devil they know until they experience the ultimate catalyst for behavioral change: psychological pain from either losing money from a big reversal in momentum or a lost opportunity from a big move in value. This behavioral reality is why value alone is not a good timing tool, except at dramatic extremes that we will detail below. On the momentum side, a reversal in momentum will likely require either an economic recession, odds of which we peg at less than 25% in 2016, or a big increase in global growth—which is extremely hard to forecast, but does not appear imminent.
What we can gauge more directly is the potential for a big move in value, and we do this continuously by looking at valuation spreads. Our preferred chart of valuation spreads is in Exhibit D below, where Empirical Research Partners calculates the spread by taking the cheapest 20% of the market and gauging how cheap this value bucket is relative to the market average over time. The big thing to notice is that big valuation opportunities come along roughly once every ten years or so. These big value opportunities can ruin or make a career, and the key for value managers is to survive when the line is rising violently so that you have the capital to make the most of the extreme value opportunity when it peaks and turns the other way. Like all good exponential market moves, these events create violent spikes, so the art of timing is critical. At the peaks, the valuation math becomes incredibly easy, but the behavioral hurdle is incredibly difficult as these opportunities are typically born of severe crises like the Great Financial Bubble, or generational diversity breakdowns like the Tech Bubble.
Clearbridge Value Trust – Widening valuation spreads
Currently, valuation spreads are widening due to the developing pain in commodities. Quite simply, balance sheets and cost structures that reflect the halcyon days of the great 2000s commodity boom cannot withstand current distressed commodity prices. If current prices are sustained or move lower, stress will continue to build as legacy equity capital is wiped out, and the remaining debt is restructured. This violent process will ultimately set up the next up cycle as the collapse in capital spending will ultimately lead to an undersupplied market.
Our strategy is one of constant preparation and patience as the opportunity develops. We are still underweight energy and basic commodities, but we have some capital committed to companies we deem long-term survivors. As valuation spreads most likely continue to expand and we get paid more for the risk, we will gradually commit more capital. If the pain is severe enough to cause a two to three standard deviation blowout in spreads, commodity-related equities will likely become a major part of the portfolio. In this scenario, the extreme pain of past mal-investment in commodities will certainly give rise to a value cycle, as the expected returns from betting on the surviving equities will be extraordinary.
Outside of this emerging drama and opportunity, the price-to-value return potential embedded in the portfolio is at the most attractive levels in roughly three years. Critically, our excellent investment team continues to find long-term valuation opportunities in several different areas, which allows us to match the valuation-driven upside potential with the portfolio construction discipline that we think is so critical to surviving the current headwinds for value managers. Even for value managers, it is important to be good at more than just one thing, and appreciating the cyclical dance between momentum and value is critical.