If we are to obtain an overall picture of where the financial sector is, or is not, performing well, then the performance of financial intermediaries and markets—in terms of total assets, scope of activity, depth, efficiency, and penetration—can be compared to a carefully chosen set of comparator countries. National authorities are likely to be interested in countries in the same region, as well as those of a similar size and a similar level or higher levels of per capita income.3 The type of indicators that would be appropriate is discussed in chapter 2 and summarized in box 4.1.
Ideally, given data availability, it may be possible to use the results of research studies that have identified causal factors for cross-country differences in depth, efficiency, and other dimensions of financial development. For example, several studies have attempted to explain differences in average bank margins—key indicators of the price efficiency of banking in terms of policy, institutional, and macroeconomic variables. Those variables include the bank’s size, a measure of property rights protection, and other bank – and country-level characteristics, such as bank concentration, output gap, and interest rate level.4 If those policy and institutional variables are available for the country in question, the results of the studies can be used to throw light on potential improvements that could be achieved through better policies and better institutions. The residual between the expected value of average bank margins in the country predicted by the study and the actual margins, if positive, will point to the need for closer analysis of idiosyncratic features in the country—features that may be contributing to the gap. (For an illustration of this technique in practice in Kenya, see appendix E.) A similar approach can be used
for banking depth where macro-variables, such as inflation and the level of gross domestic product (GDP) per capita, are key determinants along with institutional variables, such as shareholder and creditor rights (e. g., see Beck, Demirgug-Kunt, and Levine 2003).
There are also some cross-country studies of other dimensions, including insurance penetration, stock market capitalization, and turnover, although those studies may not yet be sufficiently well established for heavy reliance to be placed on them for benchmarking purposes. Along with other dimensions, including access to financial services, cross-country research is not yet sufficiently developed to support this kind of benchmarking. In those cases, simple cross-country comparisons against peers can, nevertheless, be informative and can point to areas of deficiency.