Five Growth Mysteries in Search of a Broader Innovation Policy

June 2, 2014


Left to right: William Maloney,  Asli Demirguc-Kunt, Gerardo Corrochano

  • Increasing countries’ ability to adopt and adapt existing technologies is critical; strong innovative capacity is far more important than what particular combination of goods countries produce.
  • To spur innovation, investments in R&D are insufficient—they must be matched by a private sector that has the capacity and demand for technological upgrading
  • Hence, our conception of the National Innovation System and of innovation promotion policies must go beyond a narrow focus on universities or R&D and cover access to finance, managerial capacity, and a business environment that encourages risk taking.

In this May’s Policy Research Talk, William Maloney, lead economist in the World Bank’s research department, considered five mysteries relating to innovation policy: 

  • Why did countries experience such different development outcomes despite similar structures of production?
  • Why did immigrants often dominate the industrialization process in Latin America, despite facing the same weak business climates and institutions as the locals? 
  • Why is it that the quality of developing country exports does not seem to catch up to that of advanced countries? 
  • Why is it that—given the enormous potential gains in terms of growth, poverty reduction, and the generation of widespread, high-quality employment—poor countries seem to invest so little in R&D? 
  • Finally, why does China do so much?

He opened his discussion with a striking example of one good resulting in two very different development outcomes. At the end of the 19th century, Chile’s copper production fell dramatically, its share in world production plummeting from 35 percent to 5 percent. Local observers pronounced the industry and its contribution to Chile’s development over.  However, in a matter of decades, copper production rebounded dramatically by multiples and today the industry remains central to the country’s economic growth. 

The explanation for the reversal was the introduction of European advances in metallurgy and chemistry that had changed the mining sector worldwide. In the United States, copper production was leveraged to build a critical mass of advanced human capital and establish a knowledge network of universities specializing in these new technologies that became the foundation for the subsequent industrialization process. Clearly, what Chile was producing was not limiting its growth. The ability to manage and introduce new technologies made the difference between whether production of a very homogenous product led to economic stagnation, or gave impetus to a long period of dynamic growth. Latin America lacked this ability as reflected, for instance, by the region’s low numbers of engineers.

But the second mystery suggests that simply educating engineers is not enough. Despite weak institutions and an adverse business climate, foreigners had some human capital, experience, or managerial expertise which allowed them to dominate the industrialization process in the region when the locals did not. Prior to their arrival, and without this managerial ingredient, there would have been no demand for knowledge, engineers, or technical education and no one to convert knowledge into productivity growth. Data presented from the World Management Survey suggests that Latin America still lags greatly in this area. Similarly, the fact that we do not see catch up of the quality of exports to advanced country levels seems to be due to another missing ingredient: credit markets and other institutions that would mitigate the risk from introducing new techniques and products.

" It’s not enough to focus on narrow measures of technological progress: number of engineers, patents and R&D. We have to think of these other complements, like management and financial markets. "

William Maloney

Lead Economist, World Bank’s Research Department

“We know that innovation is a very important driver of productivity and growth, and it’s essential to producing more challenging jobs—and for poverty reduction,” said World Bank Research Director Asli Demirguc-Kunt, who hosted the event. “Estimates for returns to R&D even for developed countries tend to be over 50 percent. The farther the countries are behind the frontier, the higher the return. But we also know that poor countries, with the exception of China and perhaps India, invest far less as a share of their GDP than rich countries in R&D. The puzzle is: why don’t they invest more, if the returns are supposed to be so high?” 

Maloney addressed this particular paradox in his talk, drawing on research in his recent working paper on “Why Don’t Poor Countries Do R&D?”.  His research demonstrated that the relationship between a country’s level of poverty and the returns it experiences from investing in R&D is not a simple, linear one. Rather, it follows an inverted U shape: middle-income countries see a strong return on their investment, while the poorest countries experience low or even negative returns. Why? Poor countries often lack the necessary components—human capital, research infrastructure, managerial quality and financial markets— needed to convert spending on R&D into growth and poverty reduction.   As he summarized in his talk, “countries may invest in R&D, but there’s no private sector to absorb these advances in innovation, or there may be barriers to the private sector doing so.” 

The final mystery, then, was why China did so much R&D. Is it that the country was able to assemble all these ingredients, and hence was able to profitably conduct so much R&D? Its investment climate is not friendly and the management data suggest that while Chinese firms were among the best for rapid production of non-innovative products, they were among the worst in long-run planning, vision of the future, and human resource strategy. The answer may lie in the fact that most R&D and patenting is done in foreign multinationals which, effectively, draw on their own established National Innovation System to make doing R&D profitable. It is not clear that the local Chinese innovation system alone could justify the rates of R&D investment documented. 

Maloney went on to argue that our conception of innovation policy and the National Innovation System needs to include a broader range of complementary ingredients. “It’s not enough to focus on narrow measures of technological progress: number of engineers, patents, and R&D. We have to think of these other complements, like management and financial markets.” He also suggested that more developed countries have a higher appetite for or ability to manage risk, which positively influences their ability to innovate and increases their rewards. 

Gerardo Corrochano, Director of Innovation, Technology and Entrepreneurship at the World Bank’s Financial and Private Sector Development network, echoed the need for practitioners to consider the formulation of innovation policy in terms of supply and demand, and devote their development efforts accordingly, “Roughly there’s this one-third/two-third breakdown in innovation spending between public and private sectors. If you look at most successful countries, it’s about one-third public and two-thirds private. But in many of our clients in the Bank, it’s the other way around.” He emphasized the need to establish effective linkages between public sector spending on innovation or more specifically in S&T, and industry, academia, and the outside world, in order to increase the quality of public spending and build the capacity of developing economies to absorb innovation.