Reforms and long-run inputs use efficiency of Indian and Pakistani commercial banks

Yaseen Ghulam, Ahmad Nawaz Hakro

Research output: Contribution to journalArticlepeer-review

18 Downloads (Pure)

Abstract

The banking sectors in Pakistan and India were reformed in the 1990s to promote competition and boost output through more efficient use of resources. This study assesses the input usage efficiency of both banking industries during pre- and post-reform periods via data spanning nearly four decades and addresses methodological concerns after applying the order-m frontier. According to the data, rather than operating on the predicted or notional frontier to deliver the specified output level, the average commercial banks in both nations appear to operate beyond the efficiency frontier. Indian banks, both domestic and foreign-owned, seem to use inputs around 30% more efficiently than Pakistani institutions of the same kind. Evidence of resource-use efficiency increases and their maintenance over a longer post-reform period is seen in industries from both nations (15% and 3% for Indian and Pakistani banks, respectively). Indian banks’ order efficiency did not significantly alter in the early post-reform period, but it later began to improve and kept improving over a longer period (averaging 5% improvement between 2005 and 2020). Evidence of improvements in Pakistani banks’ input use efficiency points to a notable improvement in the first post-reform period (about 12%) and then a longer-term trend (an additional 5% from 2005 to 2020).
Original languageEnglish
Pages (from-to)85-122
Number of pages38
JournalAsian Academy of Management Journal of Accounting and Finance
Volume21
Issue number1
DOIs
Publication statusPublished - 18 Jun 2025

Keywords

  • Efficiency
  • Productivity
  • Indian banking
  • Pakistani banking
  • Reforms
  • Banking industry
  • Order-m estimator

Fingerprint

Dive into the research topics of 'Reforms and long-run inputs use efficiency of Indian and Pakistani commercial banks'. Together they form a unique fingerprint.

Cite this