· The Innovation Growth Lab has now launched an updated database of over 150 RCTs in the areas of innovation, growth and entrepreneurship. It includes both developed and developing country examples, and has scrapped the registries to also include many ongoing trials. It is also nicely organized by topic (e.g. technology adoption, access to finance, business training, promoting innovation, etc.).
· Pay me later: Jason Kerwin summarizes a new paper of his which conducts an experiment with tea workers in Malawi – where over half of the workers sign up for a service that allows some of their pay to be withheld and paid out in a lump-sum at the end of the season – which he finds has lasting impacts on wealth.
· On the IDinsight blog, Dan Stein shares his summaries of four papers that stood out for him at NEUDC.
· Timothy Bartik has a new (free) Upjohn Institute short book on economic development incentives – by which he means business tax breaks or other government assistance, which goes to targeted businesses to create jobs in a local area. A local example is the competition to attract Amazon’s HQ2. The focus is on the U.S., but the evidence and discussion will be relevant to the debate about whether you should try to move jobs to people or people to jobs elsewhere. He summarizes the evidence as “At least 75 percent of the time, incentives are all costs, with no job creation benefits”. His discussion of wasteful incentives also reminds me of a classic Planet Money episode “Why did the jobs cross the road?” about the Kansas City-Kansas/Kansas City-Missouri border. Another interesting factoid for me was that almost half a billion dollars per year in the U.S. is spent on manufacturing extension services – offering advice and training to SMEs – and he views money spent on these services as far more cost effective than incentives to large corporations. He has a chapter devoted to evaluation, and his views on how programs should be designed to enable a comparison group, with useful insights about when different types of designs may be more or less applicable (e.g. on whether a scoring threshold should be used to determine eligibility and enable an RDD “Say you’re a staff person for a manufacturing extension service. How are you supposed to assign a quantitative score for “how much this firm needs advice” to each local manufacturer? The idea is absurd”).
· Some more nice tributes to the economics Nobel prize this year from Pam Jakiela on the CGD blog and Tavneet Suri and Nidhi Parekh on VoxDev, as well as an interview with the Economist
Join the Conversation