eThesis - School of Business electronic theses

Objectives

The objective of the thesis is to evaluate the equity durations implied by consensus cash flow estimates on American stocks. Previous research of equity duration is scarce, and as far as I know, this thesis is the first research using a wide analyst forecast database as the primary source of determining the implied equity duration.

Using generally approved methods and market standards, a consensus cash flow driven measure of implied equity duration is derived. Furthermore, using historical average returns, the implied equity yield curve is plotted. The performance of long-short portfolios is analysed and multivariate regressions executed to find whether the results are statistically significant.

Data

The consensus cash flow estimates are from Institutional Broker’s Estimate System (I/B/E/S), the total return and market capitalization data from The Center for Research in Security Prices (CRSP), and the other data such as share prices, book values, and Treasury Bill rates from Thomson Financials database.

The sample period of the research is January 1990 to December 2008, and the monthly regressions cover all AMEX, Nasdaq, and NYSE listed companies with relevant data (totaling 2,804 companies). The survivorship bias has been eliminated.

Results

The empirical results of the research indicate a downward sloping implied equity yield curve, which is in accordant with the previous (though scarce) research. Between 1990 and 2008, the duration is typically 5-12 years, and has a slight upward trend until the crisis 2008, which drops the durations.

The “lowest 10% (equity duration) minus highest 10%” portfolio has an annual average return of 19.75% from January 1990 to December 2008. The equity duration is a statistically significant variable to explain equity returns in both univariate and multivariate regression models including beta, book-to-market, size, and momentum factors as other explanatory variables.