The Performance of the Banking Sector - Essay Example

Many studies have evaluated the performance of the banking sector. Empirical research is critical to systematically evaluate and draw conclusions from such performance data. This paper aims to critically compare two such research papers, and will show support for the research design of Cebenoyan and Strahan. The study of Sathye (2002) investigated the productive efficiency of banks India. Her results were that the mean efficiency score of Indian banks was comparable with the world mean efficiency score, as well as the efficiency of private sector commercial banks as a group were, ironically, lower than that of public sector banks and foreign banks in India.

Alternatively, Cebenoyan and Strahan (2003) investigated how active management of bank credit risk exposure through the loan sales market affected capital structure, lending, profits, and risk in the USA. Their findings were that banks which improve on their ability to manage credit risk may have greater leverage, and subsequently may lend more of their assets to risky borrowers. Sathye's hypothesis was that the mean efficiency scores of Indian banks (i.e., publicly, privately or foreign owned) would be different to each other.

Using three groups allowed Sathye to compare all three form s of banks that exist in India. And comparison data can determine if differences in efficiency are in fact real, or due to chance. Using financial ratios as the DV, Sathye also used archival data from the latest available publications for the year 1997-98 compiled by the Indian Banks' Association (IBA, 1999). She used the non-parametric test while the non-parametric approach is Data Envelopment Analysis (DEA). The test itself is suitable in that it is recognized to be robust for efficient frontier estimation.

And it is a popular tool for the evaluation of the efficiency of a number of producers. DEA can be used with multiple input and multiple output models also. However, a critical limitation is that measurement error can skew the results. And although it may result in how on e bank is doing in comparison with the others, it fails to compare to a theoretical maximum. Finally, Sathye does not report the limitations of her study, to inform the reader who may be considering replicating her research.

Turning now to the hypothesis for Cebenoyan and Strahan's study, which was that when banks with access to bigger internal capital markets (e.g. big banks and banks owned by multi-bank BHCs) hold less capital and lend more, then the same ought to be true for banks that use the external loan sales market to manage their credit risk (Cebenoyan & Strahan, 2003). They collected archival data from the Reports of Income and Condition (the ''Call Report'') for all domestic commercial banks in the United States.

Archival data is secondary data and its limitations are that the original reasons for collection are different to that of the present study, and so may not accurately reflect operationalized variables. The test of significance was to estimate a series of cross-sectional, reduced form regressions that would show any pre-existing relationships between the variables. A regression test was the ideal multivariate test to identify potential relationships between numerous DVs and the dichotic IV. It is therefore not surprising that their results were significant, and that their hypothesis was supported.

Also, unlike Sathye, Cebenoyan and Strahan used a parametric test which meant that their results could be generalized to a wider population of banks. Although the study of Cebenoyan and Strahan had a large sample initially (74,045 bank/year observations), the reader was not informed as to the number following attrition due to missing data. Sathye did not report any incidences of attrition. Finally, the large sample of data for Cebenoyan and Strahan would provide a better likelihood of a normal distribution of the data, and subsequently the results can be generalized to other large banks that meet the criteria for either of the IVs.

In contrast to Sathye's study, Cebenoyan and Strahan's had a larger sample, and so increased the confidence for the reader in the result that they found.