Interesting piece from Murray Stahl of Horizon Kinetics September letter, see the full thing below.

H/T ValueInvestingWorld

The Russell 2000 Index has passed a moment of truth relative to the S&P 500 Index. From year-end 1994 through year-end 2013, the Russell 2000 outperformed the S&P 500 by only 23 basis points per annum. However, extend the time period by only eight months to August 29, 2014, and the Russell 2000 underperforms the S&P 500 by 20 basis points over an almost 20-year period.


These results would most likely severely confound  the many academics engaged in asset allocation  theory, because they beg the following question:  why would a presumably rational investor hold the  more volatile Russell 2000 for a long period of time  only to see it fail to outperform, or even  underperform, the less volatile S&P 500?


As of July 31, 2014, according to iShares, which offers the Core S&P 500 exchange-traded fund, or ETF, (IVV) and the Russell 2000 ETF (IWM), the ETF  tracking the S&P 500 traded at 4.42x book value,  while the ETF tracking the Russell 2000 traded at  3.79x book value. According to Russell Investments,  however, the Russell 2000 traded at 2.18x book  value on the same date. The most recent S&P 500  factsheet from S&P Dow Jones Indices provided the  price-to-book, or P/B ratio as of March 31, 2014;  therefore, one cannot make a precise comparison  between the numbers calculated by the index  provider and the ETF provider for the S&P 500.

In fact, there are many different methodologies for  calculating the price-to-book ratios. For example, one may calculate the price-to-book of every single  security in the index, and then determine the weighted-average price-to-book by calculating the
weighting factor of each security, and arrive at the weighted-average price-to-book. Or one could collapse  the shareholders’ equity of every company in the index and compare that to the aggregate market  capitalization of the index to get a different price-to-book. The latter methodology seems to be the more  accurate way to do it, but there are many other methods.


The salient point here is to consider the consequence of these differing ratios. If Russell is right—with the price-to-book value ratio of the S&P 500 approximately twice the Russell-calculated price-to-book-value ratio— then in order to maintain financial performance parity with the Russell 2000, the S&P 500 needs to earn at least 2x the returns on equity that the Russell 2000 earns, and probably more than that, because the Russell companies generally reinvest more of their earnings in their businesses. The S&P 500 companies pay out more of their earnings as dividends; accordingly, the S&P 500



The Russell 2000 ETF trades at a higher price-to-earnings, or P/E multiple (28.62x for IWM versus 22.05x for IVV as of July 31), but this is due to the fact that it comprises a variety of development-stage companies that have yet to generate meaningful earnings. For example, there are biotech companies that are modestly profitable but have yet to produce significant earnings, but that does not mean they will not in time. It is important to note that iShares employs some computational heuristics for its ETFs: P/Es above 60 are automatically reset to 60x, irrespective of how high these may be in reality, and negative earnings are excluded from the calculation. Russell calculates a P/E of 21.26x for the Russell 2000 as of July 31, which figure also excludes negative earnings. S&P Dow Jones Indices calculates a P/E of 18.57x for the S&P 500 as of the same date.


The P/E is a much more simplistic, maybe even naive, way to view the difference between the Russell 2000 and the S&P 500. Price-to-book value is actually a much more reasonable methodology. Generally speaking, book value is less prone to manipulation than are earnings, and book value growth is reflective of business performance. Furthermore, as P/E is driven by recent or current earnings, it may not be meaningful for development-stage companies (such as many of those that comprise the Russell 2000) for which current earnings are not necessarily indicative of potential earnings. This makes is difficult to use P/E when comparing established companies such as those that comprise the bulk of the S&P 500 to nascent companies such as those in the Russell 2000.

The more frequent use of P/E than P/B as a valuation metric is an example of availability error. The P/E ratio is much more readily available in the news media than is the price-to-book-value ratio and, therefore, people talk about it more. That does not mean, however, that it is the more important statistic, and the discrepancy between the different calculations highlights the importance of considering various financial metrics in context. Furthermore, the valuation gap may widen between the S&P 500 and the Russell 2000.


At current valuation levels, it would seem reasonable to move money from the S&P 500 to the Russell 2000. Yet, year-to-date through August 29, 2014, IWM experienced net outflows of $2.1 billion1, which, considered against its $24 billion in total net assets (as of August 29), is a considerable loss of assets under management in only eight months. In order to place these outflows in context, consider that between 2001 and 2013, there were only three years in which this ETF experienced outflows: 2007, 2009, and 2011, and in none of those years did the outflows reach $2 billion. Thus far in 2014, as IWM was experiencing outflows, IVV recorded $2.9 billion of net inflows. It appears, therefore, that asset allocators are moving money from the Russell 2000 to the S&P 500 when it might be time to do the opposite. If small cap stocks are indeed a separate asset class that should exhibit higher return for their higher risk, we are quite perplexed by recent flows given current valuation. As contrarian investors, we frequently invest in the very areas of the market that have fallen into disfavor by the broader markets—the long-term historical performance of the Russell 2000 has not justified the higher volatility experienced by those investing in it, but looking forward, smaller cap stocks as an asset class appear to be trading at more attractive levels than do larger cap stocks.