Clifton Green

Now, more than ever, investors are wondering just what the market
will do. As they desperately look to recover portfolio losses after the
market lows of 2008, investment strategy is certainly foremost on
everyone’s minds. According to T. Clifton Green, an associate professor of finance and finance area coordinator at Emory University's Goizueta Business School,
the key to investing, even in such a difficult environment, is in
understanding market trends, as well as in taking a long-term approach.
Nonetheless, his recent research suggests that investor behavior relies
too heavily on sentiment and the categorizing of similarly priced
stocks. This categorizing behavior, he notes, is not based on economic
fundamentals or firm results.

In a research paper titled “Price-Based Return Comovement,” Green
and co-author Byoung-Hyoun Hwang, a doctoral student in finance at
Goizueta, use stock split data from the Center for Research in Security
Prices to prove the point. The research paper is forthcoming in the Journal of Financial Economics.
The paper notes, “Stock prices are unique in that they can be changed
arbitrarily by altering the number of shares outstanding. This practice
makes cross-sectional comparisons of price per share relatively
meaningless.” Consequently, the pair looked at all ordinary common
shares listed on the NYSE, AMEX and NASDAQ between 1926 and 2004,
focusing on 2-for-1 splits and excluding stocks with post-split prices
under $5.00 a share. The final split sample included 5,424 events, with
the authors requiring stocks to have return data in the sample over a
12-month period ending one month before the split and over a 12-month
period beginning one month after the split.

Green and Hwang found that the price of similarly priced stocks do
appear to be relatively stable over a long period of time due to firm
stock split behavior. The research paper notes that managers will split
their stock when the per share price starts to deviate too much from
their peers. Green surmises that institutional tradition might be
driving this need to level out share price, keeping it in a set and
expected range. The duo notes that prior researchers in the field
determined that “nominal prices of common stocks have remained constant
at around $30 per share since the Great Depression as a result of firms
proactively splitting their stocks, which the researchers find
difficult to fully rationalize.”

Interestingly, Green states that the drive to split a stock is
driven by price-based stock categorization by investors. The paper
adds, “If investors group stocks based on price, a firm with a stock
price significantly different from its peers has the incentive to split
rather than risk facing a smaller pool of investors.” The results
indicate that price categorization, though often an arbitrary tool,
remains a popular investing strategy. The authors add, “In the full
cross-section, price-based portfolios explain variation in stock-level
returns after controlling for movements in the market and industry
portfolios as well as portfolios based on size, book-to-market,
transaction costs, and return momentum.”

Green adds that investors also place some sort of value or meaning
in the share price of the stock. Lower priced stocks tend to be of more
interest to individual investors. He admits that it can be a
complicated process to choose stocks for a portfolio, and investors can
begin to rely on a variety of different methods, such as share price
categorization, to come up with some sort of strategy. “But the price
doesn’t really convey any specific information about the stock.
Generally, splits are not based on capital structure issues.”

Nonetheless, the pair notes, “Taken together, our findings suggest
nominal prices are relevant to investors when constructing and
rebalancing their portfolios.” This irrational behavior does have its
consequences. They add, “Price-based categorization of stocks has a
material effect on return dynamics, and provides additional support to
sentiment-based explanations for return comovement. Our results also
offer a straightforward justification for 'trading range' motivations
for splits and provide a potential explanation for the observed
increase in volatility following splits.”

Overall, after a split, even without changes in a company, the now
lower priced stock will travel in a new realm, more closely shadowing
price changes (comoving) with lower priced stocks than with the ones it
was previously priced near. The duo’s analysis “involves looking for
shifts in split stocks’ comovement with price-indexed portfolios before
and after the split.” The pair notes that stock splits cause
significant changes in nominal prices with no resulting change in
firms’ fundamentals. The paper states, “As such, they provide a
relatively clean test of category-based investing with few confounding

Lower priced stocks are generally less liquid and receive much less
attention from more sophisticated investors and traders. Institutional
investors tend to favor higher priced stocks “beyond fundamental
considerations.” The authors conclude that stock splits can often be a
confounding influence on the market, deterring investment from more
knowledgeable traders and creating significant clientele effects.