Abstract
Can a place be hardwired to support entrepreneurial activity more than somewhere else? Could the way the place is physically designed bring people together to share their ideas and work proactively to develop new businesses and work towards social change? Would this make a place inherently more prosperous? If this was the case, then urban design could have an as yet unrecognised economic value. This paper presents research into how urban design can affect a community's ability to act entrepreneurially, investigated through an empirical analysis of the 155 district centres of Greater London. Using self-employment as a proxy for entrepreneurial activity, the analysis tests a potential relationship between self-employment and the spatial configuration of each centre. The research finds that while income shows the strongest relationship with self-employment, in certain low-income neighbourhoods spatial configuration becomes important. The paper concludes that places that provide the most variety, density and proximity can enable diverse face-to-face interaction in the public realm and overcome the barriers to entrepreneurship that are inherent in neighbourhoods of multiple deprivations.
Notes
†This paper is partially based on research carried out for an MA dissertation in Urban Design at Oxfod Brookes University.
1 Proxy 2 showed a good relationship with Proxy 3 (r = 0.859**), and there were reasonable relationships between Proxies 2 and 3 and Proxy 1 (r = 0.312**, 0.517** respectively.) *Significant result (prob > |t| less than 0.05). **Significant result (prob > |t| less than 0.01).
2 The representation is achieved by multiplying the proxy for the first place (Proxy 1), with the average of the proxies for the second place (Proxies 2 and 3).
†Colour versions of all the figures in this article appear in the online publication of this article. Available at: http://www.informaworld.com/rlce.
3 800 m represents approximately a 10 minute walk and therefore a reasonable sized district centre.
4 It is acknowledged that such approximations may leave the analysis vulnerable to the phenomena of ecology fallacy; therefore, a more robust sampling technique may be needed if this initial research is developed further.
5 Proxy 1 was available at SOA level representing approximately 1000 people while Proxies 2 and 3 were only available at MSOA level, representing 4000–5000 people.