In my August 30 column, I examined India?s inflation growth record, and compared it to four other large developing countries: Brazil, Indonesia, Mexico and Russia (BIMR). India has recently done better than all of them. Its good performance on the inflation front goes with an apparent lack of capital account openness. This would make India an exception to the general pattern of a negative correlation between inflation rates and openness, the explanation being that its governance institutions perform better than those of the BIMR quartet.
But there is another wrinkle to this tale. The comparison of capital account openness used the popular Chinn-Ito index, which combines various measures of legal restrictions on cross-border capital movements, using a purely statistical technique to determine how to achieve the ?best? combination. By this index, India has the lowest score (least capital openness), which is -1.1. A score of 2.6 is the best possible, achieved by several developed countries. The BIMR quartet?s scores are, respectively, 0.2, 1.2, 1.2, and -0.1. The comprehensiveness of the exercise undertaken by Chinn and Ito, and their use of the index in various analyses of the links between financial development and growth, have together contributed to its popularity. However, it tells only part of the story. Ultimately, de facto financial openness and financial integration are best indicated by relationships in asset prices. The idea is as follows.
The condition known as covered interest parity (CIP) states that, in the absence of market imperfections or transaction costs, the interest differential between financial assets of the same term denominated in different currencies will equal the cost of covering in the forward market the currency risk from arbitrage between the two assets (arising from possible movements of the exchange rate before the assets mature). If there are frictions, then there will be a no-arbitrage band (defined by two inequalities), rather than a single equality condition. This observation essentially goes all the way back to John Maynard Keynes, writing in the 1920s, but recently, better data and new empirical techniques have allowed the idea to be applied systematically. In the current context, capital controls contribute to market frictions, and so the nature of the arbitrage thresholds gives insight into de facto openness, consequent on legal restrictions as well as market imperfections.
A brand new paper by Gurnain Kaur Pasricha, at UC Santa Cruz, applies the CIP-threshold analysis to nine developed and nine emerging market economies, for post-1995 data. Aside from the novelty of her data, studying several emerging markets, she constructs a new index of financial openness or integration, which I shall call the Pasricha index. The Pasricha index uses several components in addition to the threshold bandwidth, measuring various characteristics of the sample data outside the thresholds (how often, how much, and how continuous). The index simply averages these components, a more intuitive approach than that used in Chinn-Ito. The numerical values are specific to the sample of countries used, so the Pasricha index is purely ordinal. However, comparisons with the Chinn-Ito index can be performed for any given set of countries and data. For Pasricha?s sample, her index aligns well with that of Chinn-Ito: with a correlation of 0.73. However, India emerges as the biggest outlier.
For Pasricha?s data, the Chinn-Ito index ranges from 2.62 (shared by five developed countries) to -1.09 (South Africa). India measures up at -0.95, second last in financial openness. However, according to the Pasricha index, India ranks 11th out of 18 nations, with a value of 0.01, where the range is from 1.02 (the UK) to -1.31 (Malaysia). Of the BIMR quarter, only Mexico is in Pasricha?s data set. Its Chinn-Ito index is 0.72 (again much higher than India), but its Pasricha index is only -0.03.
The upshot of all these numbers is that India?s de facto financial integration is way ahead of its rules and regulations. India?s record of good inflation outcomes, in this view, is perfectly consistent with the typical negative relationship between financial openness and inflation rates. This does not negate the importance of institutions. For example, it could be that the capital controls that are on the books are being enforced by sophisticated regulators in a manner that reduces actual frictions. It could also be the case, however, that sophisticated financial institutions and private sector participants are thwarting regulators? intent. Finally, there may be tacit collusion among regulators and financiers in the presence of political obstacles to overt capital account liberalisation. Analysts and policy makers alike can do their jobs better by understanding the specifics of the situation encapsulated in the two indices for India. For example, these specifics may lead policy makers to simplify controls, or change how they view the efficacy or benefits of the rules currently in place.
Whatever the details of its institutional underpinnings, India?s de facto financial integration is quantitatively established by the Pasricha index, which promises to be an important complement to the Chinn-Ito index in assessing capital account openness. Perhaps the most interesting extension of the analysis for emerging markets would be to the other three members of the BIMR quartet, and to countries such as China and Turkey, which also have low Chinn-Ito indexes. A final gloss on the issue of financial integration concerns financial crises. The Pasricha analysis and index calculations exclude crisis periods, but can be extended to such situations, to examine how market frictions, as measured by the index, change during periods of turbulence.
?Nirvikar Singh is professor of Economics at the University of California, Santa Cruz
