The Gavekal Knowledge Leaders Indexes: Capturing The Excess Returns Of Highly Innovative Companies by Gavekal Capital
By Steven Vannelli, CFA, Eric Bush, CFA, and Bryce Coward, CFA
In our first white paper, “The Knowledge Effect: Excess Returns of Highly Innovative Companies,” we identified a market anomaly that leads to persistent excess returns among highly innovative companies. We offered two explanations why companies that share a common risk factor—the Knowledge Factor—historically generate excess returns. First, the introduction of the semiconductor has enabled humankind to multiply its intellectual strength in a similar way that the steam engine and electric motor enabled humankind to multiply its physical strength. Corporate knowledge production takes the form of investment in research and development (R&D), advertising and employee training. Corporations spend more on knowledge than they do on property, plant and equipment. The second important root for the Knowledge Effect is the dearth of information about corporate knowledge activities that has been amplified by the poorly timed implementation of conservative accounting practices at the start of the greatest period of knowledge production in human history. This information deficiency has led investors to make a systematic error in the way they assess the prospects of companies that invest significantly in knowledge. Ultimately, this systematic error is reflected in a persistent risk premium, or excess return, for companies that invest significantly in knowledge.
In this second paper, we describe our process of creating the Gavekal Knowledge Leaders Indexes. These indexes are designed to capture companies that share a common risk factor: knowledge intensity. We begin with a history and discussion of index construction schemes. Next we review how and why we created our own Gavekal Capital International (GKCI) Indexes to serve as the selection uni-verse for the Gavekal Knowledge Leaders Indexes, comparing and contrasting our methodology with Morgan Stanley Capital International (MSCI) Index model. From there, we discuss how we adjust company financial statements for knowledge investments and outline the rules we use to identify the companies in our flagship Gavekal Knowledge Leader Indexes. We follow with a detailed review of the performance and risk history of each index, comparing and contrasting with the MSCI Indexes. We conclude with a factor based decomposition of the Gavekal Knowledge Leaders Indexes which quantifies the alpha specifically attributable to the Knowledge Factor.
At the end of October, the value investor Mohnish Pabrai gave a presentation and took part in a Q&A session at Boston College and Harvard Business School on the Uber Cannibal Investor Framework, which he has developed over the past decade. Uber Cannibals are the businesses “eating themselves by buying back their stock,” the value Read More
Index Weighting Schemes
Any indexing discussion starts with an acknowledgement that the index weighting scheme is crucially important to the results of the index. Different commonly used indexes use different methodologies, and it is important for investors to appreciate the differences.
In the United States, the longest running stock index, the Dow Jones Industrial Average, still uses a price-weighted methodology to calculate its index. This means that a stock that trades at $100 will comprise 10x more of the total index than a stock that trades at $10. It is well documented that the disadvantages of this weighting scheme, such as the arbitrary overweighting of a higher priced stock to lower priced stocks, creates a poor representation of the stock market as a whole.
Eventually in a move to make stock market indexes more representative of the broader market, the vast majority of stock indexes moved to a pure market-capitalization value weighting scheme. Under this regime, a stock index’s weights are calculated by taking the market capitalization of each individual security, adding them all together, and calculating the proportion that the market cap of each individual security is to the total market cap of all the securities in the index. This leads to a stock index where larger companies account for a greater proportion of the index than smaller ones. The S&P 500 used such a weighting methodology until 2004.
As technology made foreign investing easier and more accessible, a movement started in the early 2000s by the largest index providers to move to float-adjusted market capitalization weighting. The float-adjustment attempts to include only the shares available to purchase on the open market rather than simply the total number of shares outstanding. MSCI shifted all of its indexes to a float-adjusted methodology in 2002 and most large index providers followed suit soon thereafter. According to the index providers, float-adjusted indexes provide a more accurate set of investment opportunities for investors. They also reduce the cost of running index funds and ETFs because funds with less float, and consequently less liquidity, are a smaller proportion of the total index.
Academic work in recent years, however, is pushing back against the idea that float-adjusted indexes are more advantageous than pure-value weighted indexes. In “Pure Versus Float-Adjusted Value Weighting” Seifried and Zunft found that pure-value weighted indexes exhibit “favorable index properties” and that “float-adjusted indices fail to improve index practices and enhance distortions.” The main disadvantage of float-adjusted indexes is that “due to regulatory differences and different definitions of free float” float-adjusted indexes are “subject to a time lag, resulting in incomparability be-tween different countries and providers and best guesses when analyzing data.” This leads to a weighting scheme that is more subjective and less objective than a pure-value weighting scheme.
Float-adjusted indexes do offer a more investable universe than basic value weighted indexes. But, pure-value weighted indices with simple liquidity thresholds are better still as they offer an objective methodology that is not only more transparent and uniform, but more investable. We employ this value/liquidity hybrid model in our GKCI Indexes. While the intentions of the entrenched index providers may seem sound, regulatory and information discrepancies reduce the benefits of float-adjusted indexes and create significant disadvantages for investors.
Construction of the Gavekal Capital International Indexes
We begin the construction of the Gavekal Knowledge Leaders Indexes by first creating their selection universe indexes, the GKCI Indexes. These indexes are designed to be broad representations of the investable universe in each country.
We begin by taking the top 85% of stocks that are publicly listed in each country. In order to end up with a truly investable universe, we create a few rules to exclude hard-to-invest in securities or non-equity securities before we can take the largest 85% of stocks. The types of securities that we exclude are: Treasury Shares, American Depositary Receipt (ADR), Master Limited Partnership (MLP), Preferred Shares, Non-Equity Securities, Shares that are issued for founders, executives, and family for a dual share class stock (i.e., we include Berkshire Hathaway Class B shares but not Class A shares).
See full white paper below.