Horizon Kinetics market commentary for the first quarter ended Mach 31, 2015 – Readers can find original article by visiting the website horizonkinetics.com
So, we’ll talk about the portfolios. And we might talk about performance, but it won’t be about short-term performance, because—forgive a perhaps incendiary statement—it’s meaningless. We’ll review some portfolio holdings, using some objective valuation facts and predictive attributes as they relate to security selection, but only in a long-term context. We won’t reference much in the way of the macroeconomic factors that are considered a de rigueur element of the portfolio management process and security valuation models, such as the outlook for interest rates, GDP growth, future oil prices, and so forth. In actuality, they are more a source of bad decision making and return erosion than they are of assistance. Sound extreme? A couple of examples, then, to set the stage. These examples will, for contrast, include substantial data of the sort that are considered important or even critical in investment decision making by non-long-term investors, yet in reality are not able to reliably inform decisions that assure or improve returns.
Horizon Kinetics: What is Long Term, Anyway, and Would You Fire this Manager?
First, here are some actual 5-year performance figures for a certain investment partnership, to be identified later. These are net of management fees and expenses. Your task is to determine whether an investment consultant would have retained or dismissed this manager on behalf of clients in the fund. Without exaggeration, the underperformance is massive, a difference of roughly 28 points.
Is 5 years sufficient time to permit a reasoned judgment? Bear in mind that the manager underperformed during 4 of these 5 years. Worse yet, a standard statistical analysis of the fund’s return pat-terns, such as its beta or Sharpe ratio or alpha, would have revealed that it was far more volatile relative to its returns and the returns of Treasuries than the market, such that the manager would appear to have been taking excessive risk. It probably does not require much discussion to agree that late in year 5 or early in year 6, this manager—had he managed to last that long—would have received a letter of dismissal.
But what if we add a few more data points? The preceding 8 years look very different than the 5 above. This same manager outperformed the market in each year but one, and by a yet more massive 94% points. And in the original year 6, which was the final year of this 14-year record and which would have been the year of this manager’s dis-missal, the fund returned a rather startling 73.2%.
Over the full 14 years, the fund returned 13.6% annually versus the S&P’s 5.2%, or 6.0x a client’s original investment, versus 2.0x for the S&P 500. The true time period, as shown in the second table, was 1962 through 1975. The fund was Munger Partners, and the fund manager was Charles Munger, perhaps best known as the Vice Chairman of Berkshire Hathaway and longtime business partner and confidant of Warren Buffett.
No doubt, modern portfolio analytical techniques can reveal some interesting aspects of a fund manager’s style, but such analysis would also have dictated the firing of Charlie Munger. Such data is no substitute for understanding the thought process behind the construction of a portfolio. They don’t describe the degree of undervaluation or return expectations for the securities within it—would the decision to dismiss have been any different if the consultant understood the nature and valuations of the holdings? Perhaps not—one might, in order to know this, have to evaluate the consultant’s decision making and risk control processes in the conduct of that business.
One might pause, for just a moment, to again consider just how extraordinary the Munger Partners returns were for his investors, and just how extraordinary the opportunity cost was for the hypothetical dismissal of his services. Even so, the Munger Partner returns pale beside his later investment record.
Just as Warren Buffett ultimately closed Buffett Partnership and continued investing through the corporate vehicle of Berkshire Hathaway, so too did Charlie Munger eventually operate through his control of publicly traded Wesco Financial. During the 10-year period 1989 to 1999, the book value of Wesco Financial compounded at a 21.0% annual rate, from $39.54 per share to $266.21.
Horizon Kinetics: Volatility in Core Value
The failure of Munger Partners to control the downside volatility—or perhaps worse, the absence of any process to control such price risk—is more anathema in the world of modern portfolio management than is mere underperformance. How, exactly, does one control against such interim price risk? There is a bedrock presumption that one can anticipate events that will cause individual stocks or industry sectors to decline, that this can be done in an actionable way, that if it can be done it will not detract meaningfully from returns and, most importantly, there is a bedrock presumption that to not do so is a failing, if not an act of irresponsibility. Whether it is possible to prevent short-term downside volatility without damaging long-term returns (it isn’t, really) is a longer discussion, for another letter; so, we’ll stay here with the uncontroversial notion that investors appear very much to wish to avoid downside volatility.
That requirement is observable in the flow of funds: from the 2008 Financial Crisis year through the end of last year, if one adds together both the amount of capital that was withdrawn from equity mutual funds (which is a proxy for active security selection) and that was added to ETFs (which are, simply, indexed baskets of securities), the total is over $1 trillion – and that is a conservative estimate of the exodus from active to passive management. In February, over $30 billion of net new ETF units were issued, or about $1 ½ billion per trading day.
Active management, by the way, is not synonymous with long-term investing—the historical average turnover rate, or trading activity, in equity mutual funds is about 60% per year, and it is not unusual to see turnover of 100% or more. This level of activity presumably reflects an effort to anticipate and avoid a decline in the price of a holding. Why hold an energy stock that will do well over five years if, this year, it might decline due to lower oil prices? Nevertheless, investors are leaving this form of investment and risk control for index-based investing. One would think that indexing is an inherently low-transaction-volume form of investing—one simply buys and holds an index that, representing the market or a portion of the market, should be less volatile and have less security specific risk.
Rather than own a single energy stock in an uncertain oil price environment, one can simply hold a basket of energy stocks, such as the Energy Select Sector SPDR ETF. It is, with $12 billion in assets, clearly popular, and the largest holdings are ExxonMobil, Chevron and Schlumberger, all blue chips. Why, then is the average daily turnover 14.5%? That’s about 3,600% per year, incredibly more than the most aggressive mutual fund managers. In fact, the annual turnover of ExxonMobil itself is only 0.3% per day. By that measure, the trading or liquidity demand for the Energy ETF exceeds by 40x the daily liquidity of its largest constituent, ExxonMobil. Why buy a passive basket of stocks to avoid one sort of perceived risk in order to incur additional risk by trading, unless the trading is supposed to reduce risk? But on what basis are investors determining exactly when to sell the Energy ETF, and just where to put that money almost once every week (14.5% per day X 5 trading days equals 72.5% turnover), and then to buy it again almost once every week? Even a broad index, like the iShares Russell 2000 ETF, with $30 billion in assets, has average daily turnover of 11.6%, which is about 2,900% per year.
See full PDF below.