This text is a partial abstract of a paper by Alfaro L., Kalemli-Ozcan S. Harvard Bus. School; Volosovych V., Houston U published 2005.
Summary: Capital does not flow to poor countries in ways which theory would predict. Apologists for the theory evoke near-tautological arguments about national differences. The authors examine national differences in detail across a wide range of countries for the thirty years to 2000. As noted elsewhere on this site, it transpires that institutional quality is the key differentiator between those countries which attract capital and those which do not. If Peru's institutions approached the quality of those of Australia, the foreign investment which it is able to attract should quadruple. The authors note that while much has been written in the development literature about the importance of institutions, this finding offers a practical channel through which improvements in these are linked to long-term prospects for a country.
Neoclassical economic theory is not an intuitively accessible model for most non-economists. It is, however, well validated for the developed economies, linking three assumptions about individuals:
These assumptions allow the creation of highly mathematical, equilibrium-oriented models of factor pricing and allocation, of the firm and competition. The emphasis is microeconomic, with the general balances of the economy derived from this. Institutions are taken as a given: a crucial point in the context of the current discussion.
The body of neoclassical economics suggests that capital should flow from rich countries to poor ones, which it does not in fact do. This is called the "Lucas paradox". Theoretical predictions for the productivity of capital in India, given its low costs, are fifty eight times higher that that of the US, but this is not what is observed in practice. If theory were to be met with fact, all free capital should flow to countries such as India, where the returns to investment ought to be much higher than they are in the rich world. The failure of theory is, on the one hand, crucial to economic development and, on the other, equally crucial to the ability of the wealthy world to retain its standard of living.
Two kinds modification have been made to theory in order to bring it into line with observation. One approach looks at the difference in the so-called fundamentals of the productive side of the economy: of access to technology, state policies and poor supplies of factors of production, such as skills. The other approach focuses its attention international capital markets, noting their responses to volatility, political risk and information asymmetries. Lucas himself tended to downplay the second approach, insofar as capital failed to flow even during the colonial periods, when such risks were presumably equivalent to those in the rich world.
It may well be that these concepts are different facets of the same thing: that poor institutions lead to uncertainty and volatility, for example. The research ignores categories, rather digging into practical and measurable things that have been done across a range of 23 developed and 75 developing countries during the thirty years to 2000
These measures are then grouped into indices. For example, the the composite index which addresses institutional quality is the sum of the indices of investment profile, government stability, internal conflict, external conflict, absence of corruption, non-militarized politics, protection from religious tensions, law and order, protection from ethnic tensions, democratic accountability, and bureaucratic quality. Other factors, such as physical separation and transport, measures of public health, economic and stock market volatility, bank assets, labour supply and the like are also submitted to regression analysis.
The analysis shows that differences in institutional quality is a robust explanation for differences in capital inflows. In practical terms, it suggests that bringing the institutions of Turkey to the level of those in the UK would increase capital inflows by 60%, and bringing those of Peru to the level of Australia would quadruple its capital inflows. The figure shows the relationship between weighted average international capital flows and institutional quality. Corrected for this correlation, the differences in capital flows between nations are essentially randomly distributed.
The authors cite a number of examples where one can see the force of this analysis as it plays out in practice. Intel chose to site a plant in Costa Rica over Mexico. Both had similar incomes and similar levels of educated labour. Indeed, Mexico was favoured both by its geographical location and the absolute size of its economy and pool of engineers and other skilled workers. However, the decision in favour of Costa Rica was tipped by its political stability and lack of corruption. Mexico offered to change the law specifically to favour Intel - at the expense of other companies - and this was set in vivid contrast with Costa Rica's determination that any incentives offered would be available to all companies. (The author of this note was repulsed by a similar offer in Bolivia in favour of a blending plant: for if a Presidential decree can assist you today, what is to stop them from tearing it up or superseding it tomorrow?)
Turkey is undergoing profound reforms in its public service and banking sectors in preparation for its application to join the EU. The consequence has been a flood of foreign direct investment: a sum of under $1 bn in 1990 has grown to $5 bn in 2005. Investors such as France Telecom has justified their investment specifically in terms of the improvements in the institutional protection and reduced overheads that Turkey has created. Investors now have to go through three official procedures, as opposed to 15 in previous years.