Economists have long been aware of the difficulties of job search, of properly matching up buyers and sellers in the highly non-standardised widget called labour. If the matching problem were solved better, the natural rate of unemployment would go down, and output would be higher. The frictions of this market, equally, imply that there are opportunities for innovations which help people achieve better outcomes at lower cost. The more inefficient a market, the bigger the business opportunity for someone who's able to make a contribution to fixing it. In May 2007, there was an NBER conference on the subject of `labour market intermediation' which touches on many aspects of this problem.
Anand Giridhardas has a fascinating article Internet revolution reaches India's poor in the International Herald Tribune which describes attempts by a few computer engineers at building two new systems -- babajob and babalife - which utilise computer technology to improve on the age-old Indian mechanisms of job hunting, based on a `circle of trust'. The driver that you recruit in India is almost always recommended by someone you trust. The question is: Can these trust mechanisms be made more efficient using the Internet? Can they thus reduce the frictions of the labour market? I believe these entrepreneurs are onto a very good thing.
I am reminded of TeamLease, which is also an innovative startup animated by the inefficiencies of the Indian labour market. I'm not sure about the exact facts, but teamlease is either the biggest or the second-biggest employer in India today. I find it interesting that both these projects are relatively recent startups. On the subject of temp jobs, see Do temporary jobs improve workers long-term labour market performance? by Fredrik Andersson, Harry J. Holzer, and Julia Lane, and see the NBER conference mentioned above.
I recall having a discussion about the Internet and entrepreneurship a few months ago with Viral Shah and Vikram Aggarwal. It's relatively easy to take a known business model like amazon.com and try to transplant it into India, using relatively minor innovations like running a system of warehouses spread over India so as to avoid the irritating difficulties of buying from amazon.com and delivering to an address in India. Sometimes, this might make money - ebay might like to buy an Indian auction company that's built up a head start. But at a deeper level, this doesn't make too much sense, because the global firms are all very focused on India as an important market, and they have the economies of scale. What would be really interesting, in contrast, is to do business models in India which utilise the Internet as a tool but come up with innovations in meeting the needs of customers in the Indian institutional setting and reflecting the very different prices of alternative production technologies that are seen in India. The babajob and babalife examples involve understanding the informal sector and contributing to making it more efficient.
On the issue of cost function, the single big difference about India is low prices of labour. Manual labour in urban India runs at $0.5 to $1 an hour. This has far reaching implications for the products and processes that one can imagine; what's optimal when faced with Indian labour costs might often not be optimal anywhere else in the world. My favourite example of this issue is travel sites. Of course, it's nice and convenient and useful to be able to buy plane tickets on the web - cleartrip.com is an outstanding example of a graceful and well done implementation of that. But labour in India is unusually cheap. This means that a travel website should be able to affix the name of a support person and a telephone number onto the e-ticket. The customer should be able to call this person on the fly and make changes to the reservation. This kind of personalised customer-support is infeasible in the West, given high prices of labour. Hence, the big travel websites in the West don't think like this. But solving problems under Indian conditions needs to analyse such business processes which are figured out from first principles.