Measuring entrepreneurship over time

Abstract

This article is a review of three measurement issues that impact the study of entrepreneurship over time: (1) level of analysis difference between firms and individuals, (2) differences between rate and stock measures, and (3) the effects of choosing particular time frames on subsequent analytical results. Based on theory that views entrepreneurship as depending on ownership rights (Hawley 1907), this article develops a longitudinal measure of entrepreneurship in the United States—the number of organizations per capita. The problems and advantages of using a measure based on organizations per capita as an indicator of entrepreneurship is examined. Understanding how and what is being measured in studies of changes in entrepreneurship over time is an important issue for academic researchers and public policy makers. Measurement of changes in entrepreneurship over time can shed light on important research questions and public policy issues about entrepreneurship. For example, is entrepreneurship increasing in the United States? If entrepreneurship is increasing, is this trend comparable to any previous time periods, or is the current increase in entrepreneurship a new or unique social phenomenon? What influences changes in entrepreneurship over time? The choice of certain measures of entrepreneurship is likely to influence the answers to these questions. For example, the measures used in an entrepreneurship study are an implicit specification of one's views of entrepreneurship (e.g., entrepreneurship as self-employment or firm creation), and few measures of entrepreneurship reflect critical changes in important environmental influences (e.g., technological change) because of short measurement time frames. Therefore, determining the contribution of entrepreneurship to the well-being of an economy is dependent on understanding what measures of entrepreneurship are utilized for making these evaluations and the time frame used for these measures. We review three measurement issues that impact the study of entrepreneurship over time: (1) level of analysis differences between firms and individuals, (2) differences between rate and stock measures, and (3) the effects of choosing particular time frames on subsequent analytical results. Based on theory that views entrepreneurship as depending on ownership rights (Hawley 1907) this article develops a longitudinal measure of entrepreneurship in the United States—the number of organizations per capita. Studies of entrepreneurship focus on either individual level activity (e.g., self-employment) or on firm level activity (e.g., new incorporations). Self-employment research focuses on individuals who employ themselves; that is, individuals who report wages, but not wages paid to them by other individuals or organizations. In most studies of the self-employed, firm founders would, therefore, be classified as wage earners and not as entrepreneurs. New firm research focuses on the rate of new firm entrants, typically measured as new incorporations, so such businesses as proprietorships or partnerships will not be counted. A rate is a change from one state to another (e.g., the number of people who become self-employed for a specific year, or the number of new firms created for a specific year), whereas a stock specifies a particular level (e.g., much like a stock of inventory), such as the number of self-employed, or a particular number of firms, for a specific time period. It is important to recognize that entry into business does not necessarily guarantee remaining in business. For example, although the number of people who become self-employed may increase for a specific year, the number of people who remain self-employed may actually decrease if more of the self-employed fail to remain in business than those that enter. The stock of the self-employed can decrease if the outflow of self-employed is greater than the inflow. For those persons interested in whether entrepreneurial activity results in wealth creation (either individually or societally), the realization that entry into business may not lead to a sustainable business (and therefore no creation of wealth) should be a signal that measuring rates of business formation and rates of self-employment may not be appropriate for this type of research. The factors that drive changes in the rate of entrepreneurship are not likely to be manifest over short timeperiods. Changes in values, attitudes, technology, government regulations, and world economic and social changes have a significant influence on changes in entrepreneurship over time. Studies that have measured entrepreneurship over recent time periods are, therefore, likely to miss the influence of these variables. We introduce a measure of entrepreneurship (organizations per capita) based on a theory of entrepreneurship as ownership. This measure shows the stock of organizations in the U.S. economy over time (from 1857 to 1992). The problems and advantages of using a measure based on organizations per capita as an indicator of entrepreneurship is examined. We conclude with some suggestions for improving entrepreneurship research by recognizing the limitations of particular longitudinal entrepreneurship measures and by challenging the field to seek convergent validity among measures.