By most accounts, the last few years have been a very tough time for deep-value investors. Persistent slow world economic growth and prevailing uncertainties about fundamental macroeconomic issues like the strength of the world financial system, the ability of central banks to ward off deflation, the growth of the Chinese economy and the survival of the Eurozone have led to enormous risk aversion among equity investors. This has meant that the traditional arbitrage of very inexpensive stocks, particularly in the cyclical area of the market, has not been taking place.
The flip side of that risk aversion has been a very high preference for quality among investors. Perennially profitable companies that predictably earn their costs of capital and generate free cash flow have now been outperforming for several years. Another factor has been new entrants to the quality investing scene. Since quality investing without valuation governance is not exactly rocket science, all the rocket scientists in the business have been saturating the market with low volatility and quality factor funds and ETFs. Such participants are usually the heralds of a crowded trade. To use a tug-of-war metaphor, quality has been exerting a strong pull the last few years, and valuation has been disappearing among quality stocks.
This valuation conundrum underlines a difficulty with the quality-value style that we employ at Burgundy. High-quality business models can appear in any industry group, but tend to be concentrated in a few industries. We play in a small sandbox. As a result, when a compelling narrative of buy and hold for quality stocks establishes itself, consensus investors tend to enter and drive up prices in this narrow universe, sometimes to absurd levels. We saw this with the Nifty Fifty market in 1972, and again with the Buffett mini-bubble in 1998, when quality became seriously overvalued.
In the last couple of years it has been a constant struggle to balance value and quality in the Burgundy portfolios. A portfolio constructed solely on the basis of quality today risks becoming just another growth strategy if it is not also rigorously governed by valuation – which brings me to a highly simplified thought experiment that I presented in a 2014 issue of The View from Burgundy titled “Confessions of a Buffetteer.”
Consider that you are running two portfolios. In one portfolio, you propose to keep low statistical valuations constant throughout the market cycle, adhering rigidly to a program of low price/earnings ratios, low price/book ratios, etc. In the other, you wish to keep quality constant, as measured by strong balance sheets, high returns on invested capital and low volatility streams of free cash flow.
Assume that you start the process in a bear market trough, when there are plentiful undervalued stocks in the capital markets. There may initially be some overlap in the portfolios. But as the bull market unfolds, the portfolios will diverge in several respects.
In the statistical-value portfolio, as price targets are reached and multiples expand, the manager must scour ever deeper for discounts of all sorts. Activity can be quite high in this portfolio. As the risk preference of the market rises, by the late cycle it is only risky securities that remain cheap and it is likely that there is a decline in the quality of the statistical-value portfolio over time. Remember, I am making this a purely statistical exercise so this portfolio will never see a discount it does not like, be it due to cyclicality, complexity, secular decline, managerial incompetence or geopolitical tensions.
In the quality portfolio, some positions will be falling by the wayside as the relentless forces of capitalism lay siege to businesses through technological change, shortened product life cycles or globalization. In the case of American companies, managements will be pillaging the business and diluting shareholder value through their compensation arrangements. Turnover will be lower than in the statistical-value portfolio, but valuations will tend to rise significantly from the trough of the market. Given the rather homogeneous nature of the quality investment universe, there are very few pockets of opportunity to improve valuation in the portfolio without sacrificing quality.
Consequently, in one portfolio, if statistical valuations are held constant, quality declines. In the other, where quality is held constant, valuation suffers.
Now, of course, this is simplified to the point of naivete – markets do not operate in straight lines and the idea that quality and value could be completely mutually exclusive is unlikely. But there is no doubt that value and quality often coexist uneasily, sometimes for prolonged periods. And unfortunately the last couple of years has been one of those times.
Both the quality portfolio ungoverned by value and the value portfolio ungoverned by quality will ultimately be risky for clients. As value investors, we are trained to view valuation as the primary risk control in our portfolios, based on the commonplace observation that something that is already very cheap is unlikely to get too much cheaper, even in a market downturn. Value is supposed to be a defensive, conservative strategy. Ben Graham exhorted us not to lose sight of quality as a risk control in our portfolios, but it was left to Warren Buffett to come up with the idea of quality with value as an investable theme in the markets. But since value is much more quantifiable than quality, the value investment community has never been as comfortable with this approach to investing.
We certainly see why this is in today’s market, with many converts to quality investing and a plethora of new products purporting to follow a quality investing approach, which crucially usually will not incorporate a valuation discipline. We can be certain that many of these investors will embarrass themselves when quality goes out of style and when expensive quality, of which there is no shortage today, gets pummelled. They will then bleat that buy and hold investing just doesn’t work, and go away and leave us alone for another decade or two.
To say that quality and value are involved in a kind of tug of war is very simplistic, but I think it captures an essential truth – the tension between those two things is useful and perhaps essential to producing good returns at an acceptable level of risk. Value and quality need each other. You may let value pull you in one direction or let quality pull you in the other, but you should always feel that tension. And whatever you do, whether you’re pulling from the value side or the quality side, don’t let go of the rope!