Horizon Kinetics commentary for the third quarter ended September 30, 2016.
Also see
Horizon Kinetics – Preface to What is Defensive or Moderate Risk Investing?
Depending on the particular strategy, our equity portfolios generally have cash balances in the 25% to 35% range. It’s the result of a process we began over a year ago. It is worth noting that this cash has not impeded the perfor-mance in any obvious way. Some of our equity strategies are ahead of the market so far this year, some behind. Some that are behind were ahead a month or so ago, and some that are ahead now were behind. In fact, on about 20% of the trading days this year, our portfolios returns were in the opposite direction than the S&P 500, and typically by over a half percentage point, often by a full percentage point. Let’s just say, for the moment, they’re neither better nor worse overall.
[drizzle]
But our portfolios contain two features of great value that the market does not: 1) a lot of cash; and 2) idiosyncratic securities – stocks (and maybe a couple of interesting bonds) that we believe will behave more in accord with their own particular business and valuation characteristics than with the systemic factors that drive the market indexes in lock-step. Nowadays, you don’t want to be too close to that.
Using Core Value, only 5 of the holdings are in the S&P 500; for the Small Cap strategy, 1 holding is in the top 500 companies in the Russell 2000; for the Research strategy, only 2. That top portion of the Russell 2000 accounts for 61% of its value. If one wished to find an ETF with that little exposure to those two representative indexes of large- and small-cap companies, it would be very challenging – maybe it can’t even be done. So there is perhaps a third, related feature of value in these portfolios: scarcity.
Attention to these two issues, liquidity and securities with idiosyncratic characteristics will become increasingly important. Particularly for people who believe they are appropriately diversified and in defensive equity sectors in the manner now generally practiced.
This was the topic we presented a few weeks ago at the Grant’s (Interest Rate Observer) Fall Conference. That the channeling of the $2+ trillion flood of assets into index funds since the 2007 Financial Crisis forced ETF organizers to focus on the largest companies with the most share trading liquidity. The sheer weight of that much money into the rather limited number of companies large enough to absorb that much demand for their shares has distorted prices to an extreme degree. It’s simply a matter of supply and demand – no one who has ever visited a Middle-East souk or a farmer’s market or art auction is confused by what moves the prices. Yet in the stock market, the influence of the same forces is still generally unnoticed (although not by the Grant’s audience).
So we’ll just focus on how the modernized practice of indexation has distorted “defensive” equity investing. This should be prefaced by an understanding that indexation as originally conceived by the academic founding fathers of indexation involved simply owning the entire market, simply being a participant in the results of all stocks: large, small, fast growing, mature. After making a long-term asset allocation decision, one would decide how much to place in stocks, how much in bonds, how much in cash as a risk balancer. Very little cause for adjustment. The idea of industry sector funds, style funds, country funds, runs counter to the very mechanism they were proposing to avoid the perennial challenge: how does one even begin to understand the risks entailed by the 1,594 different ETFs traded in the U.S. as of year-end 2015 (specifically, 360 broad-based, 266 sector funds, 592 international, 81 commodity, 21 hybrid and 274 bond ETFs). Which do you buy, when do you sell, and why?
Let’s Test an MBA
To start, let’s take an MBA from a top business school and challenge this recent graduate – perhaps an applicant for a position on the Horizon Kinetics research staff – to construct a defensive equity portfolio. Such a graduate will have some fluency with the statistical measures of risk, such as standard deviation of the share prices of a company or sector; ?, or beta, which takes price volatility and compares it the market’s volatility. There will be knowledge of differing business models, such as the inflation-beneficiary aspects of a portfolio of commercial real estate with long-term leases. Or of utilities, with their regulated rates of return and near-term protection from changes in interest rates and the business cycle. These sectors would certainly find favor over industries subject to frequent product cycles and obsolescence risk like technology or biotech.
This person would be conversant with Efficient Market Theory and the uncertain results of active management versus the reliability of an index. There are other important analytical tools such as the dividend discount model. In the current low interest rate environment, this model will demonstrate the attractiveness of equities that have high dividend yields relative to Treasuries. For instance, a “dividend aristocrats” ETF will incorporate both a yield that is almost twice the 10-year Treasury yield, and an expectation of dividend increases over time, because these companies can reinvest a portion of their profits for future growth – a win/win formula. (REITs must pay out almost all their income, so they lack the same ability to self-fund their growth, and utilities have very little growth opportunity, so likewise pass out most of their earnings.)
Here is what an MBA might suggest.
- The iShares Select Dividend ETF (DVY), current yield 3.17%. Of more than 1,350 different ETFs in the U.S., with $16.1 billion of AUM this is the 21st largest – it’s a top 1 percenter – a testament to its univer-sal popularity.
- The iShares Cohen & Steers REIT ETF (ICF), $3.8 billion of AUM, 2.89% yield
- The iShares U.S. Utility ETF (IDU), $0.8 billion of AUM, 2.98% yield
- And, for a touch of growth, but with a particular eye toward safety, the PowerShares S&P Small Cap Low Volatility ETF (XSLV), $0.6 billion of AUM and a 1.99% yield.
By definition, these ETFs are statistically among the safest of stocks. This is expressed by their Beta, the primary risk figure used in asset allocation programs. As a test, simply go to Yahoo Finance, type in a symbol like IBM, and right there, alongside market cap and P/E ratio will be the beta; there is no other risk statistic provided up front like that. So the beta of the Dividend ETF, as of September 30th, is 0.61, meaning that it is considered only 61% as volatile as the S&P 500. The REIT ETF has a beta of 0.86, 14% lower volatility than the S&P 500, and the utility ETF beta is an amazingly low 0.13, 87% lower volatility than the S&P 500. The small-cap low-volatility ETF has a beta of 0.79.
If this portfolio were simply apportioned in equal quarters to each ETF, the average yield would be 2.8%, versus a 10-Year Treasury’s 1.7%, and would be said to have much lower volatility than the market, as any back-test of the historical results would show.
Moreover, the 5-year annualized return of the Utility ETF is 12.2% through September 2016. For the REIT, the 5-year rate of return is 15.0%. The 5-year number for the Dividend ETF is 16.0%. And the low-volatility small-cap ETF only has a 3-year record, but it returned 11.8% annualized through October 12th, way higher than the 6.6% of the Small-Cap Index. Because of those high returns alongside the low volatility, there statistics like the Sharpe ratio that give these ETFs particularly high risk-adjusted return ratings and which make them highly valued in asset allocation programs.
Let’s Grade the Test
So that’s security selection the modern way: by external observation and categorization. A security or industry has a category description (such as large cap, blue chip, foreign, growth), it has a few valuation measures (P/E, dividend yield, recent earnings growth rate), it exhibits certain statistically reducible price behavior. These descriptive features, mixed and matched in any number of ways, are considered sufficient to select securities and assemble portfolios. And the resultant portfolio can be projected forward to estimate values or income at retirement.
It is not, however, company or security analysis. It does not suggest in any informed way how these businesses and share prices will actually behave in the future such as in the manner a business person contemplating purchasing a rental building or auto dealership might require.
Let’s demonstrate just a bit of traditional investment analysis as both a contrast to this mechanical approach and also to see if a different conclusion is reached about the essential problem that was posed: to construct a defensive equity portfolio. We’ll start with a couple of examples at the individual company level, then look at risk at the industry level as well.
Starting with the iShares REIT ETF, the largest 10 holdings account for 61% of the value of the ETF. This is a level of concentration antithetical to the thrust of indexation as a haven from security selection risk. Narrowly defined ETFs in particular, like sector funds, tend to get concentrated or top heavy relatively quickly as the companies sort themselves out into the more and less successful. Therefore, an accident with just one or two top holdings, even if there is no problem with the rest of the ETF, could sabotage the results in a way that is very unindex-like.
Company Example #1
The second largest holding in the iShares REIT ETF, at 7.0%, is Public Storage REIT. It is the largest such, directly owning almost 2,300 structures, and has a $46 billion market value. Like other storage companies, it raises its prices by about 6% a year, but expenses have been rising by only about 2% to 3%, which is a large reason for its 9% earnings growth rate. It earns a 22% return on its equity, so it is very profitable. It has also been an aggressive acquirer of smaller storage companies, which provides a path for growth. Given these unalloyed positives, the shares yield only 2.5%, less than the REIT ETF itself, and trade at 28x cash flow from operations. That’s the good news.
However, the barriers to entry in self storage are not high. Anyone who’s been to a modern facility can see it: the low capital investment (no foundation, per se, just a concrete apron, the corrugated metal buildings and rooms), the low operating expenses (the electronic key-card gate, movement-sensitive lighting, the security camera con-sole and single employee). Therefore, there is much competition, and other large companies are engaged in the same consolidation strategy.
In recent periods there has been an increasing scar-city of good quality properties to acquire. This is not yet visible in the income statement, nor will it appear in a database, because there is no searchable data-base field on a Bloomberg terminal called Storage Property Rental Rates. But you can see it in the annual and quarterly reports – the number of facilities acquired over the past 4 years has been trending lower. More important, the average rent of the acquired properties has declined steadily, from $15.15 per square foot in 2013 to $11.10 during the first half of this year. Essentially, only more marginal properties can be had, which will be less accretive to earnings. They might require greater expense to raise them to the company’s operating stand-ards, they might be located in less appealing demographic areas, the company might have to rely more on new construction. But other storage companies are facing the same reality, so competition is increasing. A firm that specializes in providing loans and development capital to self-storage companies now keeps a “Danger List” of cities in which the new supply under development exceeds 10% of the current supply of storage facilities.3
Which is all a normal part of a dynamic business sector. Except, Storage Properties is priced for growth, not for business risk. If growth slows or stagnates, if the dividend stops increasing, it is difficult to imagine the shares continuing to trade at 28x cash earnings. What if they were to trade at a 15x multiple, which is more normal for a flattish earnings profile – which would still equate to a pretty strong 6.7% earnings yield? In that event, its 7% position in the ETF will decline by near 50%, which would reduce the value of the entire ETF by 3.5%, which exceeds the annual dividend yield. Just that one stock, just for a valuation correction, not a business disaster, can erase a year’s worth of income for the ETF shareholder.
See the full PDF below.
[/drizzle]
