New firm formation in British counties: comparing the 1990s with the 1980s

Abstract

New firm formation has continued to play a key role in regional development policies across the globe for more than twenty years. Successful policy requires a clear understanding of the underlying determinants of the business birth rate and there is good reason to believe that the fundamentals may be subject to change. Armington and Acs (2002) argue for a fresh examination of the subject in the United States, in part because of developments in spatial theory including the new economic geography (Krugman 1991a, 1991b, 1994) and endogenous growth theory (Romer, 1990; Nijkamp and Poot, 1989) and because of the evolution of the regional economy. They contend that the motivation for much of the literature on new firm formation in the 1980s was high levels of unemployment in traditional US industrial regions. But the focus of the literature today is different, emphasising high technology start-ups as a critical driver of the new economy. In this paper, we pose a similar question for the UK regions. Specifically, we seek to apply and develop the model tested by one of the present authors and colleagues on 1980s data, to data for the 1990s (Ashcroft, Love and Malloy, 1991). The paper is in five parts. First, we discuss the measurement of new firm formation in the UK regional context. Secondly, we identify some stylised facts on the variation of new firm formation in British counties in the 1980s and 1990s. The third part of the paper specifies the model used in our work on the 1980s and discusses the earlier findings. Part 4 tests the model on data for the 1990s, retests for key variables rejected on 1980s data and introduces new variables to reflect developments in theory and the evolution of UK regional economies. The final part of the paper compares the explanation of new firm formation in the 1980s with the estimates for the later period. In this section implications for policy and future directions for research are considered.