Commentary on Bekaert, Harvey, and Lundblad's "Equity Market Liberalization in Emerging Equity Markets"

Bekaert, Harvey and Lundblad (BHL) are to be congratulated for producing another paper on equity market liberalizations in emerging markets, and it is a pleasure to discuss their work. Yet, there are three reasons why I may not be an impartial discussant. (1) Having devoted most of my fledgling career to the study of capital account liberalization in emerging markets, I am favorably disposed to the research topic. (2) My published work contains extensive citations to the authors' papers, and (3) I am in broad agreement with the lion's share of the authors' conclusions about the effects of equity market liberalization on the cost of capital. The BHL paper has three central themes. First, liberalization reduces the cost of capital. Second, dating liberalizations is difficult and we should try to do a better job of pinning down precise liberalization dates. Third, and most importantly, the liberalization-induced fall in the cost of capital increases the growth rate of GDP per capita by one percentage point per annum. I believe the first message. There is broad consensus that liberalization reduces the cost of capital by up to 100 basis points, depending on how you date the liberalization (Bekaert and Harvey, 2000; Henry 2000a; Martell and Stulz, 2003; Stulz, 1999). All of the evidence we have supports this qualitative conclusion and suggests that the effects are economically significant, even if we can't precisely pin down the magnitude of the effects (Henry 2000b, 2003). I also believe the second message. Liberalizations are difficult to date. While there is broad agreement that liberalization reduces the cost of capital, there is some disagreement about the exact timing of liberalizations. This matters, in principle, because the size of the effect depends on what liberalization date one chooses. On the other hand, changing liberalization dates has virtually no effect on the qualitative conclusion that liberalization reduces the cost of capital. Because more precise dates are likely to strengthen our previous conclusions about the financial effects of liberalization, and because this is a conference on the real effects of finance, most of my comments will be directed towards the third message, which is summarized in Table 4 of BHL's paper--equity market liberalization increases the growth rate of GDP per capita by one percent per annum. I don't believe the third message. The claim that stock liberalizations increase the growth rate of GDP per capita by one percent per annum is inconsistent with the assumptions of the neoclassical growth model on which the analysis is based. The rest of my comments will be devoted to developing this thought in detail, but first, a small digression. The paper uses the terms "equity market liberalization" and "financial liberalization" interchangeably. Doing so is potentially confusing. Financial liberalization refers to the removal of domestic financial repression--government imposed interest rate ceilings, restricted use of savings for consumer credit purposes, and the like (McKinnon, 1973; Shaw, 1973). The McKinnon-Shaw literature studies the effects of financial liberalization on interest rates and growth, but financial liberalization per se has nothing to do with granting foreigners access to domestic capital markets. In contrast, the BHL paper summarizes the empirical effects of equity market liberalization, a decision by a country's government to allow foreigners to purchase shares in the domestic equity market. Strictly speaking, equity market liberalization is a specific type of capital account liberalization, which is a decision to allow capital in all forms to move freely in and out of the domestic market. In other words, the distinction between financial liberalization and capital account liberalization is worth making, because the two terms mean very different things in the literature and none of the BHL results have anything to do with financial liberalization in the traditional sense. For the sake of clarity, I would hold to the traditional nomenclature.