Lessons From Silicon Valley

Creating industrial clusters like Silicon Valley could be the best way forward for Start Up India. But we must also tailor our model to our own unique circumstances.

The visit by Prime Minister Narendra Modi to Silicon Valley is an implicit signal that his administration views technology innovation as an important part of his development agenda for India. On the one hand, this is a laudable goal, as Indians have become a big part of both innovation and entrepreneurship in the Valley. On the other hand, there have been many attempts to replicate the magic ingredients that have made the Valley such a fount of innovative ideas; most of them have been unsuccessful.

A few months ago, I spent time in Palo Alto, the one-time epicentre of the Valley, and I was told that the innovation ecosystem had moved further north, with San Francisco, traditionally a bohemian city more known for the arts, now becoming a magnet for entrepreneurs (with the unfortunate side effect of gentrification and soaring rents driving a lot of the colourful older residents out).

Recently, in Hoboken, New Jersey, looking out over the water at the new tower that stands where the 9/11 World Trade Centre once stood, a professor at Stevens Institute of Technology told me about the new technology entrepreneurs in New York City. Thus, it looks like there are a few centres that are attracting the kind of Schumpeterian creative destruction that made the Valley what it is.

Similarly, Berlin, Singapore, and even London have succeeded in creating incubators and accelerators that are generating a new set of innovative companies. But it seems that at one time or another, almost every government has attempted to duplicate the secret sauce of Silicon Valley. In some cases, these resulted in no more than “plans that either come to naught or half a page of scribbled lines”.

I think the reason is that there are many factors, both hard and soft, that Silicon Valley embodies. First, there are the great research universities. Second, the ecosystem of venture capitalists. Third, the very appealing quality of life, including in San Francisco and Berkeley. Fourth, the culture of the area reflects a certain footloose-ness: many of the residents are migrants either from other parts of the US or the rest of the world, which means they are inherently more risk-seeking and less rooted.

The diversity of the Valley is an intriguing cultural artifact that may well contribute to the explosion of innovation there. The effects of a diverse population are hard to pin down, but I have two data points. One, Bell Labs in its glory days; and two, Silicon Valley itself. If one reads books like The Idea Factory, a history of Bell Labs by tech journalist Jon Gertner, one is struck by a remarkable fact about Bell Labs—a preponderance of shy, hard-working farmboys from the US Midwest who did the invention there. A sort of Lutheran work ethic? Maybe it was their frugal Midwestern values, but Bell Labs seldom threw anything away, and neither does 3M.

Silicon Valley has a different ethic altogether: one that includes a certain amount of hype and razzmatazz, a bit of fast-talking hype that may be more New York City and Los Angeles. It is able to reinvent itself periodically, and quickly: for instance, it went from the chip-dominated early days smoothly into the engineering/graphics workstation era, and then onto the e-commerce and social media eras without missing a beat; its motto is to fail fast and often, and “pivot” as needed.

The great diversity of the Valley’s population seems to help. When I first arrived there from a mostly-white area of New Jersey, I felt comforted that there were plenty of brown (Hispanic) and yellow (East Asian) people; today entire townships such as Fremont are dominated by Indians and Chinese. Doubtless, the Confucian work ethic and the Hindu work ethic are somehow involved in the surprisingly large number of startups in the area with Indian and Chinese co-founders.

If diversity were the secret to success, in India I suppose you’d find Bangalore, Mumbai and Delhi to be likely winners in future; and particularly Bangalore, where I lived for years without ever having to learn more than a smattering of the local Kannada, as broken Hindi and broken Tamil suffices. I also found that—at least before today’s terrible traffic jams—anybody from any part of India would be willing to move to Bangalore to live, partly because of the balmy weather.

And sure enough, Bangalore has the broadest set of clusters of technological strengths in India: from electronics to avionics to software to e-commerce. And it is mostly youngsters from small-town India who crowd into the place, ambitious and upwardly mobile. Bangalore surely also has many of the other ingredients for innovation: a vibrant investor community, fairly good colleges (although, sadly, with the exception of IISc and IIM, nothing remarkable), and a fairly good quality of life.

The idea that industrial clusters are self-perpetuating multipliers of competencies has been around for a while. There are obvious clusters around the world: Detroit’s automobile companies, the Ruhr Valley’s steel, or Singapore’s financial services. India does have many clusters of the Old Economy where strong supply chains and human resources have made for local competencies, such as small-scale manufacturing in Punjab. But new-ish clusters with some of the characteristics of Silicon Valley may well be the answer. For instance, the automotive cluster in Chennai, the IT services cluster in Hyderabad; and in future, we will see more.

I am helping set up an electronics incubator in Kerala, and I hope that this will lead to a cluster of hardware startups in the Internet of Things and Robotics areas. One of my pet ideas is an aerospace cluster in the industrial/freight corridor that should come up between the ports of Tuticorin on the east coast of the peninsula and the proposed twin Vizhinjam/ Colachel ports on the west coast. Given the number of ISRO installations in the vicinity, and availability of (arid) land, this may make sense.

Especially as the government rolls out its Smart Cities initiatives, the nurturing of clusters in new areas should be a priority. Interestingly, the Department of Industrial Policy and Promotion (DIPP), which owns the Make In India initiative and the industrial corridors (for instance, Delhi-Mumbai and Bangalore-Chennai), is now also responsible for the development of Start Up India.

While it is true that governments have a big role in encouraging startups, it should be careful to learn from the mistakes made in previous efforts. First, the incubators have become a simple real estate play, and end up in transferring public funds to private entities. I was amazed to hear from a bureaucrat that there was a 90 per cent success rate in an incubator, which is essentially impossible, as the global average for startup mortality must be something like 10 per cent.

The issue is that success has been redefined as mere survival. The criterion has to be a lot stricter. Otherwise, public money is wasted to give free office space to private parties who essentially have no prospects of ever succeeding. Besides, creating large numbers of—especially student—“entrepreneurs” also has a major downside: they drop out of college; their startup fails; and they have no degree, and therefore few prospects. This has already blighted the lives of many.

In addition, as a former advisor to a State-run and a bank-run venture capital fund, I found that few viable business plans came our way; and what was called “risk capital” was really almost always private equity, more comfortable with the prospect of funding expansion in a going entity rather than truly taking risks.

In the US, it is true that generous Defense Advanced Research Projects Agency (DARPA) funding, and National Science Foundation (NSF) grants have acted as industrial policy in practice, and directed funds in areas of interest to great commercial benefit.

A friend who got an NSF grant pointed to a study that suggests that the funds invested by the government are in effect recovered through payroll taxes paid by new companies.

In India, I believe that has not been true: a lot of good money is being thrown after the bad. If DIPP wishes to channel funds as a reflection of its policy direction, it may be a good idea if it requires matching funds being raised by entrepreneurs from other sources, especially from industry. The lack of an industry connection is often a fatal flaw for startups. Thus, a set of industrial clusters created based on local competitive advantage, and chartered also with encouraging startups, may well be a good model.

Finally, we must take existing western models of industrial policy with a grain of salt. It is increasingly evident that there is a certain “Indian exceptionalism”. What works here may well be a variant of conventional wisdom, tailored to our circumstances.

Rajeev Srinivasan has worked for innovative companies such as Bell Labs, Siemens and Sun Microsystems in strategy and product management. He has taught innovation at several IIMs, and is a member of the Think Tank working on India’s national IP policy. He has been a conservative columnist for almost 20 years, and has degrees from IIT Madras and Stanford Business School.