خروجی های شرکت به عنوان عامل تعیین کننده ورودی جدید : آیا از تخریب خلاقانه محلی، شواهدی وجود دارد ؟

This study posits that a local process of creative destruction provides an impetus to regional industrial renewal. We argue that exits of older firms release resources that stimulate local entry. New entrants add value to these resources by redeploying them in more productive uses. We test our hypotheses with a unique longitudinal database encompassing the entry and exit of Canadian manufacturing enterprises. We find that exits of old firms increase entry and that on average new entrants are more productive. Persistent high local rates of exit, however, deter entry.

The market must clean itself out by taking resources away from the losers, so it creatively destroys the losing companies and reallocates resources to new companies”. Former US House of Representatives Majority Leader Dick Armey (March 2002) The 21st century ushered in a shift from an economy where innovative activities by large established firms were the prime cause of economic development (Schumpeter 1954) to an economy where entrepreneurs play a key role in economic growth and renewal (Schumpeter, 1942 and Audretsch and Thurik, 2001). This new economy, ‘the entrepreneurial economy’, is driven by innovation processes through which entrepreneurs entering the market constantly seek to fill unmet demands or meet demands in more effective ways. Schumpeter (1942) suggested that through their innovations entrepreneurs that enter the market render their rivals obsolete, destroy their profits and lead them to exit. This process of creative destruction, facilitated by episodic switches in technological regimes and markets throughout the economy, was posited by Schumpeter to be the key endogenous driver of economic growth in capitalistic systems. Recently, several studies have linked measures of regional entrepreneurial activity to regional economic growth (e.g. Audretsch and Fritsch, 2002 and Acs and Armington, 2004), suggesting the possibility that in addition to the global Schumpeterian process of creative destruction, there are processes of industrial renewal at the local level driven by entrepreneurial entry. The overarching research questions of this paper are: does a process of creative destruction, distinct from the Schumpeterian process, exist at a local level? What are the dynamics of this process? Is it bounded geographically? We integrate perspectives from evolutionary economics, organizational ecology, and entrepreneurship to articulate the characteristics of a process of local creative destruction, its spatial boundaries and its temporal underpinning (i.e., short-run versus long-run drivers). We test our key assumptions through three interrelated hypotheses. To test the hypotheses we estimate Zero-inflated Poisson (ZIP) models of entry events belonging to five manufacturing sectors in Canada between 1984 and 1998 at the census subdivision level. We use a unique database made available by Statistics Canada that accurately identifies firm entries and exits. In addition, we compare estimates of productivity of firms that enter and exit. Our study informs several important areas of investigation in the entrepreneurship literature. First, we contribute to the debate over whether failure is good (Knott and Posen, 2005). Davidsson (2003: 42) argues that “…one of the first things entrepreneurship scholars should try to get rid of is the bias against failure… both theory and empirical evidence suggest that experimentation that may end in failure as well as the demise of less effective actors are necessary parts of a well-functioning market economy.” Knott and Posen (2005: 617) explore whether “… the same creative destruction process fueled by successful ventures may also be fueled by unsuccessful ventures.” In this paper we provide evidence that failure of older firms promotes the birth of new ones and that new entrants contribute to improvements in productivity.2 We show that this process operates in small geographical areas. Second, we provide some insight into the question of why locations differ in their propensity for entrepreneurial activity (Low and MacMillan, 1988 and Shane and Venkataraman, 2000). We estimate the magnitude of the impact of some key location characteristics (e.g., agglomeration, competition) and the characteristics of neighboring locations on entry. We show that localization economies (resulting from sectoral co-location) dominate urbanization economies (derived from industrial diversity) in attracting entrepreneurial entry to a location, and that in most sectors local competition tends to reduce entry. Third, we contribute to key issues in the emergent fields of entrepreneurial geography and cluster dynamics (e.g., Baum and Sorenson, 2003 and Audretsch and Feldman, 1996). The existence of high correlation between entry and exit rates in many countries is well documented (e.g., Dunne et al., 1988, Eaton and Lipsey, 1980 and Cable and Schwalbach, 1991). These correlations are attributed, in the organizational ecology and economic literature, to causal links flowing from entry to exit (Geroski, 1995). These links include the ‘liability of newness’, the relationship between entry and exit barriers, and adverse selection. We show that in small geographical areas other links between exit and entry exist and that these links may operate differently depending on their temporal dimension. Understanding both why firms choose specific locations and the dynamics of exit and entry, particularly its spatial dimensions, may help policymakers understand better the process of regional growth and renewal. Our results highlight the positive aspects of firms' exits in improving regional productivity, effects that could be suppressed when government support is provided to failing firms.

There is general recognition that business creation is an important element of economic growth and regional innovation. In this paper we contribute to the understanding of the role that enterprise exit plays in the industrial regional renewal process. Our study suggests that exits of firms in a location generate more opportunities in that location and therefore attract new entry. The study shows that this phenomenon is essentially local, with weaker spillovers to neighboring locations and little if any impact in more distant locations, highlighting the role of local knowledge and social networks in finding and acting on opportunities. The major research objectives of the study were to examine the existence of a local creative destruction process, its prime drivers and characteristics. The process of local creative destruction we uncovered in our study is complementary to the process of technological change and industrial renewal articulated by Schumpeter. “[R]search supports the idea that technologies evolve over time through cycles of long periods of incremental change, which enhance and institutionalize an existing technology, punctuated by technological discontinuities in which new, radically superior technologies displace old, inferior ones” (Baum, 1996: 97). While the Schumpeterian process of creative destruction explains the consequences of radical technological shifts, where new entrants introduce superior new technologies or products, making technologies and products of incumbents obsolete and forcing them to exit (or innovate), the local creative destruction process explains the process through which technologies change incrementally. In this process new technologies are built on know-how and inputs embodied in existing technologies. Inertial forces constrain change processes and adaptive capabilities of some older incumbents, leading to erosion of their fit with their environment. Unable to maintain their competitive position, they exit. We show that their exit creates a stimulus for entry of new firms. We also demonstrate that exit of new entrants (i.e., exit that can be attributed to the liabilities of newness and smallness) does not create as significant a stimulus for entry. This suggests that while churn may correct for mistaken entry decisions and thereby enhance regional productivity, the stimulus for new business can be attributed mainly to the exits of more mature firms that release resources (including knowledge resources) that new entrants can use. The succession of exiting old firms, which find it difficult to change, by young ones, unconstrained by inertial forces, that can recombine released resources in new ways, is an effective process where value of released resources is not only conserved but also enhanced. Our results show that productivity is increased through the process sufficiently to compensate for those new entrants who fail soon after entry. A process where failure stimulates entry may lead, however, to a suboptimal location pattern within the region as some locations with high failure rates may not be suitable for some types of enterprises. We found that a persistent high failure rate in a location reduces entry to that location. Indeed lower long-term failure rates in neighboring locations encourage a shift of entry to them from the focal location, thus minimizing risks. The existence of local competitive markets implies pressure on prices and elimination of abnormal entrepreneurial profits, and thus a reduction in the attractiveness of a place. Our study found that local competition in micro-clusters deters entrepreneurs from entry into the location. Prior research results showed that agglomeration economies matter to location selection between economic regions; our results suggest that they matter also to within-region choices. We also show that localization economies (i.e., economies generated by collocation of firms from the same sector) are more important than urbanization economies (i.e., economies generated by the scale and diversity of the overall economic activity in the region). Our results contribute to a better understanding of the spatial dimensions of economic activities, in particular the dynamics of entry to and exit from micro-locations and the interrelationships between different micro-locations within a region. Consistent with the sociological literature highlighting the importance of local social and professional networks to the formation of new enterprises, our study suggests that a variety of key economic processes which influence entry and exit are centered in micro-locations, and that regional spillovers, while important, do not have as strong an impact as processes operating within the location. We show that spatial sectoral specialization has some distinct advantages (strong localization versus urbanization economies and strong impact of same sector exits on entry). These results suggest that regional policymakers must craft their policies with sensitive attention to the characteristics and processes of small geographical areas (i.e., “one size does not fit all”). Furthermore, our results suggest that diversity in the metropolitan region and specialization within a sub-region create an attractive environment for new enterprises. Given the higher impact of localization economies on attracting new entrants, policies should emphasize the concentration of resources around one sector within small regions rather than the dispersion of resources among a variety of sectors, while promoting diversity across the economic region. Our results concerning the local nature of creative destruction suggest that regional policies should be careful not to suppress evolutionary processes of local creative destruction that increase local productivity. In particular, policies aimed at supporting failing companies through subsidies and tax concessions should be carefully evaluated. Such policies should distinguish between assistance offered to new entrants that suffer from the liability of newness and smallness and mature companies that due to inertial forces fail to adjust to new economic environments. Young entrants may have a potential for growth but may not withstand random economic shocks that occur immediately after entry. In such circumstances, temporary aid may help these companies to survive and have a chance of realizing their full potential. Our study shows that exit of young entrants may be wasteful and does not trigger a process of creation and renewal where released resources are recycled. In contrast, the exit of mature firms triggers a process of creation where resources are recycled, experience is gained and recaptured, and innovation is fostered through the entry of new firms. Governments can facilitate the release and recycling of the valuable resources and knowledge locked up in mature underperforming firms through flexible regulatory and tax policies, and educational and training programs. Such programs and policies should attempt to avoid the promotion of excess entry and the wasteful shake-up process that often accompanies it. This study is not without limitations. As with all empirical studies, claims of direction of causality need to be interpreted with caution, since we cannot rule out all plausible competing explanations. Our study, however, provides a comprehensive array of controls, thus reducing the problem. High exit levels in a CSD may result from repeated entry attempts by less competent entrepreneurs (serial exits) who prefer to re-establish new ventures in the same location. Thus exit levels may be a proxy for other phenomena not accounted for in our model such as failing entrepreneurs' lack of networks, capabilities, talent, and industry experience.20 Given that our population consists of all CSDs in Canada, we believe that the magnitude and significance of the results is not driven by some unobserved heterogeneity. More detailed economic environmental measures would allow more specific tests of the local process through which exits trigger employment transition, search for opportunities and entrepreneurial entry to a location. The dynamics of manufacturing spin-offs represents another intriguing alternative explanation for the differential in entrepreneurial entry levels across locations. Klepper, 2002 and Klepper, 2003 finds that an important determinant of geographic concentration in the U.S. automobile industry involves the role of spin-offs of automobile manufacturers. We cannot differentiate between entrepreneurial entry through spin-offs and unaffiliated entry. This remains for future research. Our study examined only entry within five manufacturing industries. Immobility of some tangible assets and partial immobility of employment may better characterize the manufacturing sector than the service sector. Moreover, in sectors with rapid technological changes obsolescence may reduce the importance of the existence of a market for used assets. Consequently, we cannot generalize the effects of exit on entry to all service or manufacturing industries. Future research should expand the population studied to include different sectors, in particular high-technology sectors and services which may exhibit different patterns of information diffusion and resource recycling associated with turnover. Finally, examination of the dynamic relationships between the global Schumpeterian and the local creative destruction processes and the innovation diffusion processes associated with them could be an important area for new research.