He advocated planned large-scale investment programmes in industrialization for countries with a large surplus workforce in agriculture, in order to take advantage of network effects - economies of scale and scope - to escape the low level equilibrium "trap". As Rajiv Sethi wrote in a recent blog post, "the argument is based on the fact that the development of any particular industry may only be privately profitable if an entire set of interlocking industries were emerging simultaneously".
In a 2008 NBER working paper, Fred Bateman, Jaime Ros, and Jason E. Taylor found that the economic rebound of American South was the result of a 'Big Push' from large public capital investments during the Great Depression and World War II. They described the 'Big Push' development strategy thus,
"According to the 'big push' theory of economic development, publicly coordinated investment can break the underdevelopment trap by helping economies overcome deficiencies in private incentives that prevent firms from adopting modern production techniques and achieving scale economies. These scale economies, in turn, create demand spillovers, increase market size, and theoretically generate a self-sustaining growth path that allows the economy to move to a Pareto preferred Nash equilibrium where it is a mutual best response for economic actors to choose large-scale industrialization over agriculture and small-scale production."
About the success of the "Great rebound" in America's South, they wrote,
"Specifically, and consistent with big push theoretical literature, the infusion of public capital — roads, schools, waterworks, power plants, dams, airfields, and hospitals, among other infrastructural improvements — fundamentally reshaped the Southern economy, expanded markets, generated significant external economies, increased rates of return to large scale manufacturing, and encouraged a subsequent investment stream. These improvements helped create the conditions that allowed the region to break free from its low-income, low-productivity trap and embark on its rapid postwar industrialization."
David Beckworth though points to the other factors that sustained the South's postwar economic recovery - improved human capital formation, industrial policy, increased political competition, and the climate or Sunbelt effect.
In 1989, Andrei Shleifer, Kevin Murphy, and Robert Vishny examined the Rosenstein-Rodan idea of simultaneous industrialization of several sectors of the economy "in the context of an imperfectly competitive economy with aggregate demand spillovers and interpret the big push into industrialization as a move from a bad to a good equilibrium".
Under the 'Big Push', industrialization in one sector "raises the demand for other manufactures directly and so makes large-scale production in other sectors more attractive". The complementarities between the simultaneously industrializing sectors, which is the underlying rationale in favour of a large-scale planned industrialization, gets amplified through market size effects. Their findings have a strong echo in China's development story over the last quarter century,
"A program that encourages industrialization in many sectors simultaneously can substantially boost income and welfare even when investment in any one sector appears unprofitable. This is especially true for a country whose access to foreign markets is limited by high transportation costs or trade restrictions. The net payoff from a program of simultaneous industrialization can also be high when all markets are open, but a shared infrastructure - such as a railroad or a stock of managers - is necessary to operate profitably in any given sector. In the latter case, simultaneous development of many export sectors may be necessary to sustain any one of them...
countries such as South Korea that have implemented a coordinated investment program can achieve industrialization of each sector at a lower explicit cost in terms of temporary tariffs and subsidies than a country that industrializes piecemeal. The reason is that potentially large implicit subsidies flow across sectors under a program of simultaneous industrialization. Any cost-benefit analysis of subsidies or of temporary protection should reflect both the lower direct costs and the higher net benefit of a program that is coordinated across sectors."
In another influential 1991 paper, Paul Krugman brought in the role of expectations and argued that "the willingness of firms to invest depends on their expectation that other firms will invest, so that the task of development policy is to create convergent expectations around high investment". This argument is different from the standard competitive view of the world. He highlighted the important role played by self-fulfilling expectations (about growth path and prospects), in addition to history and traditions (past events set the preconditions that drive the economy to one or another steady state), in shaping the development path and determining the final outcomes.
He argues that in view of the presence of multiple equilibria (of outcomes), the choice among them will be determined based on the contributions from both history and expectations. Further, the "relative importance of history and expectations depends on the underlying structure of the economy - in particular, on the costs of adjustment".
Using a simple one-factor, two-sector (or two goods) model with increasing returns in one sector and constant returns in the other, two long run equilibria (each of which involves complete specialization in one of the two goods), and an initial state of incomplete specialization, he explored the possibility and direction of movements in resources across the two sectors. Rajiv Sethi nicely summarizes Krugman's findings
"If shifts in resources across sectors cannot be costlessly reversed, then such movements will depend not only on current inter-sectoral wage differences, but also on anticipated future differentials, which in turn depend on expectations about the future movements of resources across sectors. Krugman shows that there is a range of initial conditions, which he calls the overlap, from which either one of the two long run states can be approached if and only if it is expected to be realized.
If initial conditions lie outside this range, then history is decisive, otherwise expectations matter a great deal in affecting the eventual pattern of specialization.
The overlap may be viewed as a zone of uncertainty within which coordinated optimism can have major economic effects. Outside this zone, for better or worse, we are shackled by our history. Within it, expectations become crucially important."
The question, then, with the development prospects of any country is whether or not it is now in or around this zone of uncertainty (overlap) where expectations can be a critical determinant of its future development performance.
In a his blog post, Rajiv Sethi draws attention to the possibility that encouraging developments over the last few years in many parts of Africa may have laid the ground for these countries joining the group of high-growth emerging economies provided they are supported with a "big push" on investments in education and infrastructure. In order to finance this, Ngozi Okonjo-Iweala of the World Bank proposes the diversion by donor nations of a portion of anticipated future foreign aid in order to back a current bond issue, effectively securitizing these flows.
The "big push" model may also have a critical role to play in understanding India's economic growth prospects. This is especially so if the decade-long high GDP growth rates are to be sustained and the fruits of development are to reach the millions of poor living in the country's villages. Like in the American South in the inter-war years, rural India requires a co-ordinated push towards investments in roads, schools, water and sewerage, power plants, dams, telecommunications, hospitals, and other infrastructure. These investments and the externalities generated by them would create the conditions that unleashes the full potential for a long period of manufacturing and services-led economic growth.
Further, the economic reforms over the past decade-and-half and resultant spurt in economic growth may have pushed the Indian economy into Krugman's "overlap" zone, from where convergent expectations about large-scale investments can contribute towards achieving development in the country's rural backwaters.