Capital controls and international interest rate differentials

Abstract

Since the Asian crises it is often taken as granted that capital markets have significant functional deficits. Often these deficits are believed to be so very strong that the ability of free capital markets to guarantee a more or less correct international allocation of capital is denied. It is argued that speculation dominates capital markets so much that capital allocation is purely random. This is one of the major arguments backing the present trend to re-establish capital controls, which emerged after the capital market distortions observed during the Asian flu. In the present article it is shown that capital markets, while certainly prone to many distortions, are well capable of roughly guiding capital to the proper place. Though allocation is not model-like perfect, this steals the thunder from the idea, that closed or government-guided capital markets were able to perform better.

Notes

¹Especially the question, whether there is a general tendency to use or not to use financial market regulation is debated in this context (Alesina et al ., 1993), where the authors were able to show the likeliness of capital controls being imposed under different regimes, but also found an annoying lack for any impact on growth that should intuitively exist). The presumably most quoted panel analysis is Rodrik's (1998) article ‘Who needs capital account convertibility?’, where he believes to prove the lack of growth impact of financial openness using data, which is completely inadequate for panel analysis with such sophisticated questions.

²Nevertheless this has been done for single countries, as in Chung and Ni (2002) for Korea.

³This statement refers to panel studies only. Studies which are only dealing with one country usually employ more detailed capital control measures as, for example, done in Goh (2005).

⁴Since 1996 the Gwartny–Lawson index is calculated as the share of uncontrolled sectors multiplied by 10. The more detailed information on the direction of capital flows which is not included in the table appendices is not included in the new index.

⁵The kind of capital transaction real outflow controls (in the sense that they are applied to the outflow of domestic capital) is not a matter of class building here. As a matter of fact even portfolio capital that was subject to an outflow is more likely to be re-organized within the global capital market than being re-transferred to the original country in the short run. Therefore, it is no major concern what kind of investment leaves the country but if investments leaves the country at all.

⁶Notice that this is not a classical de jure versus de facto comparison, where the question is, in how far certain rules are really enforced. The question is, whether rules can be circumvented. Because the search of alternatives usually carries its own costs, even circumvenable controls are hindering, meaning that the impact of this problem is only minor compared to a de jure and de facto problem.

⁷The correlation coefficient after Pearson between them is slightly above 75%.

⁸Full results are found in Table 1 and 2.