Geography Reference
In-Depth Information
have a concentration of organizations of a certain kind (in a specii c industry) tend to
generate a relatively large number of new organizations of that same kind. This pattern
has been shown for industries in general (Armington and Acs, 2002), and for specii c
industries like footwear (Sorenson and Audia, 2000), accounting (Cattani et al., 2003),
biotechnology (Stuart and Sorenson, 2003), computer workstations (Sorenson, 2005),
motorcycles (Wezel, 2005).
There are several reasons for why industrial clusters foster entrepreneurship (Audia
and Rider, 2005; Rocha, 2004). Clusters provide established relationships and better
information about opportunities. They open up niches of specialization because of
the low degree of vertical integration. Clusters foster a competitive climate and strong
rivalry among i rms that impose pressure to innovate because of the presence of close
competitors. They provide role models with the presence of other i rms that have 'made
it' (see Fornahl, 2003; Vaillant and Lafuente, 2007), and a cultural environment where
establishing one's own business is normal and failure is not a social stigma. Clusters
provide access to physical, i nancial, and commercial infrastructure; easing the spin-of s
of new companies from existing ones. Especially because potential entrants will know
how the local industry functions and have the technical skills to operate in it.
Industry localization may have negative ef ects on new i rm formation: increased
concentration and vertical integration raise entry barriers (Beesley and Hamilton, 1994).
The shift in the direction of the ef ect of localization on new i rm formation might be
explained by the life-cycle stage of the industry: in the early stages geographical concen-
tration has positive ef ects (or is even driven by new i rm formation; see Feldman, 2001;
Feldman and Francis, 2003; Feldman et al., 2005), while in late stages (stagnant or even
declining markets, and increased relevance of scale economies) the negative ef ects domi-
nate. Spatial concentrations of activities in mature industries might still have high new
i rm formation rates (still cognitive ef ects and low barriers to entry), but high levels of
competition (and decreasing demand) lower the performance of these entrants (Sorenson
and Audia, 2000). This means that there is still industry localization, but there are no
localization economies any more.
Urbanization
Urban areas have important advantages for entrepreneurship. Population density has
been found to positively af ect entrepreneurship (Reynolds et al., 1994; Wagner and
Sternberg, 2004). Some authors have argued that this positive ef ect of population density
(most emphasized in big cities) might be a temporary phenomenon: the resurgence of big
cities in the 1990s is connected both to a reduction in the negative social interactions
(e.g. crime) and to an increase in positive social interaction (Glaeser and Gottlieb, 2006).
However, urbanization and its most straightforward indicators, population size and
density, cover many mechanisms, which might have dif erent weights and values in dif-
ferent contexts. This is for example rel ected in the large heterogeneity in entrepreneur-
ship rates in world cities (Acs et al., 2008). Advantages stemming from high population
density include the relative ease of access to customers as well as the inputs required
(capital, labour, suppliers) to produce the goods or services. The classical 'incubation
hypothesis' in urban economics states that persons aspiring to go into production on a
small scale have found themselves less obviously barred by a high cost structure at the
centre of the urban area than at the periphery (Chinitz, 1961; Dumais et al., 1997; Hoover
