NEW APPROACHES FOR ASSESSING NEW VENTURE PERFORMANCE
Kenneth C. Robinson
Charles W. Hofer
Robinson & Hofer
University of Georgia
Terry College of Business
Department of Management
Brooks Hall 419
Athens, GA 30602-6256
Scott W. Kunkel
University of San Diego
School of Business Administration
San Diego, CA 921 10
The relationships between multiple measures of economic performance for -new ventures which had undertaken an initial public offering (IPO) between 1980 and 1983 were examined. The economic measures of performance examined include: (1)Sales (level) (2) change in sales; (3) net income; (4) return on Sales; (5) return on assets; (6) return on invested capital; (7) return on equity; (8) earnings per share; and (9) shareholder value created. The impact of new venture strategy on each of these measures of economic performance for new ventures was also examined.
A total of 289 IPO prospectuses were examined for firms which had undertaken an IPO between 1980 and 1983. The new ventures selected for inclusion in this study were chosen based on the following criteria: (1) independence from corporate sponsorship; (2) less than six year old at the time of the IPO; and (3) operating under the founding management team A total of 82 firms met the above criteria. However, the sample consisted of 75 firms as complete information for all of the variables examined in the study were not available for 7 firms. The nine measures of economic performance were examined over a three year period of time to determine the extent of association between the measures. The underlying distributions of each variable were examined and tests of association were performed with both nonparametric and parametric statistics. The data for clusi** now venture strategy was gathered from the IPO prospectus for each venture. Analysis of variance and multiple t-tests were utilized to examine the impact of new venture strategy on each of the nine measures of economic performance for new ventures.
The associations between the economic measures of new venture performance differed substantially between the parametric and nonparametric tests of correlations. An examination of the underlying distributions found that serious departures from normality existed for all but one variable of economic performance, the parametric correlation test revealed fewer correlations between the nine measures of economic performance when compared to the comparable nonparametric correlation test. The Spearman nonparametric correlation analysis revealed that most of the nine measures of economic performance exhibited significant correlations between each other although the extent of correlation between the measures differed an individual basis. There were also significant differences in economic performance among new venture strategies. The impact of now venture strategy on economic performance exhibited the clearest differences when shareholder value created was chosen as the dependent variable.
Different measures of economic performance convey different information and may or may not exhibit strong associations with increases in shareholder value. The use of multiple measures of economic performance would enhance comparisons of results across studies as these measures are often not reasonable substitutes for one another. These measures should also be examined for conformance to the assumptions of the statistical test(s) being utilized as departures from those assumptions may lead to serious errors in the interpretations of the results. It also appears that the impact of new venture strategy on new venture economic performance is most clearly determined by shareholder value created as the dependent measure of economic performance when studying publicly held companies.
|Return to Babson College
Main Home Page
©1996 Babson College. All rights reserved.
Last updated November 22, 1996 by Cheryl Ann Lopez