Why Do Countries Use Capital Controls?


WP/98/181-EAWP/98/181


Why Do Countries Use Capital Controls?
R. Barry Johnston and Natalia T.Tamirisa


This study presents initial systematic evidence on the structure of capital
controls and their determinants in a cross-section of 45 developing and
transition economies. It first examines the structure of controls on individual
capital transactions and the extent to which it would be appropriate to
aggregate the individual transactions for the purposes of analysis. The paper
then reviews the various motivations for controls on capital movements that are
generally suggested in the literature. Finally, the paper examines empirically
the extent to which the different motivations can explain the use of controls
on different groups of transactions.


The research indicates that balance of payments and macroeconomic management,
market and institutional evolution, prudential, and other factors are important
in explaining countries. recourse to capital controls. The results point,
however, to significant differences in the importance of these factors in
explaining controls on inflows and outflows and on different types of capital
transactions. For example, macroeconomic variables appear primarily to motivate
controls on capital inflows, while institutional and market structure appear to
motivate financial regulations related to the operation of banks and
institutional investors. The relationship of capital controls to the balance of
payments is not robust to simultaneous equation analysis. These findings
generally support the view that capital controls have a limited role in balance
of payments management, that controls on inflows are used to support
macroeconomic objectives, and that recourse to capital controls reflects the
overall framework for economic regulation and the degree of financial market
development.


The paper concludes that the design of capital account liberalization program
requires proper understanding of the factors motivating the network of capital
controls, which is likely to require a more disaggregated analysis of capital
controls than has been the case to date.