«Equity Style Investing RONG, WU How to cite: Equity Style Investing, Durham theses, Durham University. RONG, WU (2013) Available at Durham E-Theses ...»
When talking about a person’s occupation, one can be classified as ‘golden collar’, ‘white collar’ or ‘blue collar’, depending on the nature of his work involved. Similarly, a country can be classified as ‘developed’ or ‘developing’ based on the average overall wealth its people have and the current stage of its economy development. A capital market can also be classified as ‘developed’ or ‘emerging’ depending on whether the underlying economy needs growing liquidity, stability, infrastructure and other positive features. The mechanism of categorisation can help us to better understand the underlying objects because objects within the same category share common characteristics.
The idea of classification of objects into categories is also pervasive in the financial markets. The investable assets in the marketplace can be broadly classified into several groups as differentiated by the characteristics like return patterns or risk factors. Within each asset class there also exist some subgroups that share properties similar to their major asset class but are unique along specific dimension.
For example, investors can separate the assets from each other by classifying them as bonds, stocks, real estate and cash etc. Assets can also be further sub-categorised within each category (e.g. bonds can be subdivided into government bonds or corporate bonds; stocks can be sub-classified as value and growth, etc). In the investment community, ‘style’ refers to such classification of assets by market segments, and ‘equity style’ refers to systematic classification among stocks in the equity market. Style is by no means fixed, as time goes by due to market innovation or research discovery, new styles may evolve and old styles may die off1.
A number of descriptors can be used in empirical research to define equity styles. Firm characteristics like market values that lie in the size dimension and valuation multipliers in the value and growth dimension are most commonly used. While it is intuitive to subdivide the stocks according to their market capitalisations, categorising the stocks into the broad group of value and growth is perhaps more natural because value and growth stocks tend to follow different return patterns and therefore counterbalance each other. Moreover, the dispersion of value and growth returns is perhaps more likely driven by economic fundamentals. Hence value and growth stocks are often considered as two different asset classes. In addition to the common value-growth styles, stocks can also be classified according to their past performance and the winners and losers are identified.
Barberis and Shleifer (2003) suggest two possible reasons for the emergence of
new styles: financial innovations (e.g. inflation-related bonds) and the detection of outperformance of one asset group over another (e.g. momentum effect). On the other hand, some old styles are no longer available to investors due to change of the market condition. For example, inflation-linked bonds used to be attractive to investors in a high inflation economy, such products die off when the economy turns into deflation states.
Indeed, the concept of style is well recognised in today’s global equity markets. There are many index providers to offer equity style indexes as benchmarks to serve the investment community. Over the past decade leading financial markets have witnessed the availability and popularity of Exchange Traded Fund (ETF) and the introduction of style index futures that offer low trading cost and high liquidity for investors.
Figure 1-1 shows a typical equity style box that is widely used by market practitioners. This figure provides a visual representation of the major investment characteristics of stocks in the market. Such ‘equity style box’ was first created by Morningstar to define the riskreturn structures of stocks. The equity style box is comprised of nine categories with the underlying investment features defined by two dimensions. Horizontally, all stocks in the market can be divided into three categories: value, blend (i.e. a value/growth mix) and growth. Vertically, stocks are divided into three sizes based on their market capitalizations, representing small, medium and large, respectively. Since different category represents different risk-return profiles, investors with dedicated risk-return preference could generally confine their stocks to a specific category or combination of categories.
1.2 Equity style investing Equity style investing refers to the investment strategy based on the common stock classifications. Despite the introduction as a new investing concept in 1980s, the idea of style investing is by no means novel. The classic works of equity style analysis can be traced back to 1934 when Benjamin Graham and David Dodd published their groundbreaking book ‘Security Analysis’ and set out the concept of value investing. In this book, Graham and Dodd argue that some fundamental criteria like the intrinsic value, the future value and the market factors should be considered when evaluating a stock value.
Similarly, John Burr Williams develops the concept of fundamental analysis. His book ‘The Theory of Investment Value’ published in 1938 introduces the theory of dividend based valuation approach.
While Graham and Dodd (1934) advocate that investors should buy value stocks because the future growth of growth stocks tends to be exaggerated and hence uncertain, Thomas Rowe Price, Jr., on the other hand, publishes a paper entitled ‘Picking Growth Stocks’ in
1939. Price argues that buying growth stocks could offer hedge against the inflation because the earnings and dividends of growth stocks could be expected to grow faster than the overall economy.
Contrast with Graham who is regarded by many to be the ‘father of value investing’2, Price is best known for developing the growth stock style of investing and is regarded as ‘father of growth investing’.
Apart from value-growth style investing, the momentum investing is characterised as to buy the past winners and to sell the past losers (Jegadeesh and Titman (1993)), while contrarian investing does the opposite (DeBont and Thaler (1985)).
The exploration of style investing has gained growing popularity over the past decades. Since mid-80’s U.S. institutional investors have been found to follow some pre-defined investment strategies with specific market segments (c.f. Ahmed et al. (2002)). While value and growth investing are regarded as two most important investment styles, the most popular style investing is perhaps to combine value and growth with firm size to capture the interactions of basic style dimensions. For instance, strategies based on the combination of large value stocks, large growth stocks, small value stocks and small growth stocks. Such strategies could capture the interactions of different styles effects and could arguably yield better returns3.
One reason for style investing being well established and gained its popularity is perhaps due to its simplicity in the investment process.
Money managers face the complex and ever changing investment Graham’s work has remained influential in nearly half century in the investment industry. The merit of value investing is perhaps best demonstrated by Graham's most famous student Warren Buffett.
Asness (1997) documents a strong relation between value and momentum effects.
It is found that value premiums are strongest among loser stocks (low momentum) but are weakest among winner stocks (high momentum). Likewise, momentum is particular strong among growth stocks.
environment, they often experience the maze of investment choices given an overwhelming amount of assets available in the investment opportunity set. The classification of the investible assets into some categories simplifies the manager’s decision-making problem when dealing with asset allocation and therefore making the investment process less intimidating (Mullainathan (2002)). This is because instead of having to screen thousands of individual stocks for the investing portfolio, managers could simply make the dynamic asset allocation decision among the style level (Barberis and Shleifer (2003)). Indeed, formal market segmentation has become an integral part of today’s asset management industry. Recent studies find that professional money managers follow specific investment styles (c.f.
Brown and Goetzmann (1997), Fung and Hsieh (1997b), Chan et al.
(2002)), and the control of investment style is regarded as a critical aspect of investment monitoring and decision-making process.
1.3 Motivations and objective for the research
The concept of equity style and style investing offers a good example of the exchange of brilliant ideas between academic research and the investing practice. Despite the apparent simplicity of asset allocation process, manager’s incentive for engaging in equity style investing also stems from capitalising on the time-varying return differentials across equity styles. Institutional investors such as pension and endowment funds act as fiduciaries and therefore accept substantial responsibilities and assume significant liabilities. These investors often follow specific styles that determine the construction of their portfolios and generate unique return patterns compared to the benchmark. Such investment return patterns are caused by diverse behaviours of different asset classes. Financial markets have long observed the style return differentials as well as the tremendous swings of equity style dynamics. For example, over the past 70 years, while the US small-cap stocks outperform in the long-run, the largecap stocks are able to beat their small counterpart during 1950s and 1980s. The US value and growth stocks also perform differently over the past three decades. Value investing outperform during 1970s and 1980s, followed by the dominance of growth stocks during the 1990s. More recently, the market has again seen that value stocks outperform again since year 2000. Evidence of the divergence of style returns is also reported in other equity markets outside the US.
Overall, empirical evidence generally suggests that over the long term small-cap and value investing have been more advantageous in most equity markets around the world, but there can be periods where the size and value-growth returns reverses dramatically.
Style analysis adds to arsenal of portfolio management tools, and the dynamics of equity style returns have introduced the new risk-return structure for active portfolio management. But to have capitalised on its time-varying nature, money managers would need to not only identify the underlying driving forces that determine the relative style performance, but also to capture the mechanisms through which those underlying forces work. Most importantly, successful active managers must be able to capture the dynamic properties of those driving forces to forecast the future style trends. Over the past years, although the benefits of style investing have been well recognised globally, the academic view of the underlying cause for such benefits is open to debate. There is still no general consensus as why some asset groups are able to earn better returns than others do under the same economic regime. Since style investing is based on asset classification, arguably a sensible categorisation of assets should be based on the characteristics that relate to the asset's cross-sectional expected returns. In an efficient market where the price of stocks reflects all relevant information, style investing should not be more profitable than any other portfolios containing the arbitrary subset of stocks. Furthermore, if investors do not diversify across styles then any portfolios based on single styles would not be mean-variance efficient. Hence equity style investing maybe fundamentally risky, and the evidence of style premium would suggest that either the markets are inefficient or the traditional asset pricing models are misspecified. Previous studies suggest equity characteristics such as the market capitalisation and book-to-market ratios (BM) are closely associated with the cross-sectional expected stock returns (Fama and French (1992, 1996)). However, the mechanism of how such characteristics work remains controversial. Rationalist such as Fama and French (1993, 1996, 1998) argue that size and BM are risk factors 4, Behaviourist, on the other hand, argue that mispricing resulting from investor’s sentiment unrelated to the fundamentals plays the key role. Meanwhile, a variety of business cycle variables have also found to contain useful information in interpreting the expected stock returns. Hence the observed differentials of style return should be time-varying and related with shocks from the macro economy.
Chapter 3 is motivated by the empirical findings regarding the relationship between stock returns, equity characteristics and the business cycle fluctuations. The use of the business cycle framework is motived by the strong a priori relationship between stock returns and the business cycle conditions. Traditional financial theories link
Daniel and Titman (1997) contend that these characteristics are irrelevant to the
covariance structure of stock returns.