30 In addition to the founder-CEO index that has been developed by EntrepreneurShares and published and dissemi- nated by Thompson Reuters, a company known as Solac- tive AG published a founder-run index in August 2015. The index includes an equal-weighted basket (annual rebalance) of stocks from Canada, Hong Kong, Japan, the Netherlands, Switzerland, the United Kingdom, and the United States. The founder CEO of Solactive (Steffen Scheuble) cites as his inspiration “an index being less focused on short-term profit but rather targeting long-term growth.” Moreover, Solactive cites several key insights (on its web page) from its licensing partners at BNP. Kemal Bagci from BNP Paribas (participant in licensing the product for launch of structure products) noted that “Founder-CEOs usually work towards a vision through which they can contribute their share to the world. They are willing to invest more into Research and Development, on average over 9%, than others. These sustainable investments result in lasting outperformance of over 8% per annum.” Another member from BNP (Florian Stasch) noted, “Many Founder-CEOs consider their business a lifetime achievement, willing to work hard for the long- term. This approach can have a positive impact on the growth of a company.” To qualify for the index, stocks must have a minimum market capitalization of $1 billion and an average trading value of $1 million over the past three months. It should be further noted that the only other known product that follows a similar approach would be EntrepreneurShares Global Fund, which has a similar investment strategy across comparable geographic base. EntrepreneurShares Global Mutual Fund was launched November 11, 2010 (five years prior to the Solactive founder-run index) and includes key variables other than founder CEO. 31 For purposes of analytical comparison, we used the S&P 500 Growth Index, the Russell 1000 Growth Index, and the Vanguard Growth ETF (ticker: VUG). The Vanguard Growth ETF seeks to track the performance of the CRSP U.S. Large Cap Growth Index and has net assets of approxi- mately $30 billion, a very low expense ratio (0.06%), and low turnover (10.7%). It has 310 holdings. We select this ETF because our factor analysis (as we discuss later) requires spe- cific holdings (not a summary of returns) to properly compute the factor relationships (e.g., value, momentum, size). More- over, given the extensive number of computations required for a detailed factor analysis over an extended period of time (including the 2008–2009 recession), we need an index with fewer than 500 securities to perform the computation from 2006 through 2015 (Bloomberg Analytics—one of the most detailed and powerful factor tools available in the marketplace—has data limitations on number of computa- tions it can provide). Rather than selecting a broader market index (such as S&P 500 or Russell 1000 Growth) for a shorter time period (e.g., multiple periods of three-year duration),
founder-CEO index because neglecting Type II errors could result in survivorship bias.
26 By way of example, on December 2, 2016, the Wall Street Journal listed top-performing mutual funds, year to date, in its “Category Kings” section and included a U.S. large-cap mutual fund by EntrepreneurShares that performed in the top five among 698 similar strategies. The identified fund employs a smart beta strategy with a founder-CEO overlay that is very similar to the one described in this article. 27 The author has encountered numerous anecdotal situ- ations in which other scholars and investment professionals who researched founder CEOs employed a f lawed method- ology based only on convenient word searches to uncover company founder CEOs. Such an approach will inevitably lead to a variety of Type I and Type II errors in developing the proper database and associated investment characteristics. 28 We start with the listing of publicly traded securities within the Russell 1000 Growth Index and initiate the search process at the beginning of our research period. We use the Russell 1000 Growth Index as a starting point because we are interested in a broad-based index with many constituents that have a growth orientation (consistent with our population). We recognize that many companies eventually may not meet the rules for our test (30 largest market cap entrepreneurial companies) but choose to begin with a large starting point of eligible securities. In searching through our databases, we note that the term “founder” does not appear in many databases or is applied inconsistently across years within the same database. Moreover, many founders are never identified as such within any database. We initially began our search within Capital IQ for founder CEOs but discovered that many founders disappear in searches for historical periods (titles only apply to current position). Moreover, titles associated with founders in Bloomberg often did not correspond with Capital IQ and also did not include historical elements. Finally, titles captured in ExecuComp did provide historical information but are limited to the S&P 500 list for the current year and may not be consistently applied over time. Consequently, we need to review each company using company disclosures, websites, and individual biographies on a one-by-one basis to capture precise title and founder information. We start our process on a historical basis utilizing the S&P 500 list and then expand to the Russell 1000 Growth Index with the development of the project. 29 As we discussed in a prior note, though the term “founder” may appear as a transparent variable with rela- tively minor deviation in interpretation, in practice, many databases, and even company websites, apply this term in inconsistent ways. We do not depend on search criteria or company websites to correctly assess the term but rather apply a consistent rule to each firm in the dataset.
W INTER 2017
T HE J OURNAL OF I NDEX I NVESTING
Made with FlippingBook - Online magazine maker