Economists should rethink the role of government in promoting investment as a way of improving economic growth and social welfare, according to Professors Roger Farmer and Amartya Lahiri. Their research, published in the October 2006 issue of the Economic Journal, argues that the economics profession has erred in favouring the neoclassical model of growth over so-called ''endogenous growth'' theory.
Why do some countries grow faster than others? Is there anything that governments can or should do to promote growth? These are important questions, but they are ones that economists know much less about than they would like to. In the 1970s, economists didn't have much to say about growth; instead, they were more concerned about what causes business cycles. In the mid-1980s, a series of papers by Paul Romer and Robert Lucas changed the direction of the profession.
Romer and Lucas argued that growth is the issue if one is interested in economic welfare and they proposed a new way to think about growth that suggested that governments can and should do something about it. In the dominant paradigm that predated Romer and Lucas, due to Robert Solow, growth was left unexplained as an exogenous increase in productivity. In the world after Romer and Lucas, growth was explained ''endogenously''.
The new paradigm spurred a spate of research that tried to test the Romer-Lucas theory of endogenous growth. Empirical work was aided by the emergence of a new data set, due to Robert Summers and Alan Heston, which compiled comparable cross-section and time series data for hundreds of countries over a span of 30 years.
When researchers examined the data through the lens of their model, they found that the Romer-Lucas model did not seem to provide a good explanation of the facts. In particular, they found that countries within similar groups appeared to be converging in per capita income. This prediction follows from the Solow model that predated Romer-Lucas but it is not an implication of the Romer-Lucas theory.
The convergence finding caused opinion to swing back towards Solow''s explanation and most economists are now of the view that by promoting higher investment nationally, a country cannot do much to improve its long-run growth rate.
The paper by Farmer and Lahiri challenges the new orthodoxy. As in the Romer-Lucas theory, they construct a model of endogenous growth that allows for two kinds of capital.
One kind they call human capital to represent the stock of skilled workers in a country and the other they call physical capital to represent the stock of factories and machines. Their innovation is to permit international trade in physical capital and to follow its implications for convergence.
Like the Romer-Lucas model, the theory permits endogenous growth but by allowing for free international trade in physical, but not human, capital, the researchers are able to replicate the convergence facts that caused economists to favour the Solow model over the endogenous theory of Romer-Lucas.
Moreover, the model has the advantage that it can accommodate international trade in capital without implying implausibly high flows of capital across countries. In contrast, versions of the Solow model where countries are allowed to trade with each other predict implausibly high levels of international trade in capital that are typically not observed in the data. Most researchers subscribing to the orthodox view avoid this problem by constructing models where the world is thought of as a collection of economies that do not engage in international trade.
Farmer and Lahiri conclude:
''The economics profession erred in favouring the neoclassical model over endogenous growth theory. Economists should rethink the role of government in promoting investment as a way of improving social welfare.''
''Economic Growth in an Interdependent World Economy'' by Roger Farmer and Amartya Lahiri is published in the October 2006 issue of the Economic Journal.
