This joint LSE- University of Birmingham project sets out to identify the causal effects of two leading UK accelerator programmes using quasi-experimental methods.
Return to the theme page
Innovation is a well-understood driver of economic development: ideas help move the technology frontier forward, influencing productivity growth. Similarly, entrepreneurs are held as important for economic prosperity. However, both innovation and entrepreneurship involve substantial amounts of experimentation and carry high risks of failure. Governments and the private sector have thus developed tools to de-risk these activities. Science parks and business incubators are two well-established approaches. ‘Accelerator’ variants are now growing rapidly, using intensive networking and mentoring opportunities for competitively selected firms.
Accelerators are highly selective business support infrastructures. Statistical analysis that does not deal with this selectivity cannot disentangle whether accelerators have a real effect on firm performance or simply attract firms that are likely to succeed – irrespective of their enrolment in the programme. The core objective of this project is to evaluate the ‘true’ causal effect of accelerator on firms’ economic outcomes, using regression discontinuity (RD) designs that compare treated firms with ‘losing’ applicants or firms that drop out of a programme early.
The team is currently evaluating two leading UK programmes. One programme focuses on startups in the financial services / fintech space. We are using RD methods and secondary data from OpenCorporates, Crunchbase and other sources to track outcomes. The second programme focuses on digital creative / tech firms. Here we are combining the first design with qualitative work with programme participants and managers.
Funder/client: Leverhulme Trust
Timescale: 30 months from 1 September 2017
Project lead: Max Nathan
Tel: 0121 414 4713