Problemi Ekonomiki (Dec 2018)

Evaluating Scale Efficiency of Ukrainian Banks Using DEA

  • Kyshakevych Bohdan Yu.,
  • Mazharov Dmytro V.

DOI
https://doi.org/10.32983/2222-0712-2018-4-233-240
Journal volume & issue
Vol. 4, no. 38
pp. 233 – 240

Abstract

Read online

The article there built DEA models with variable return to scale (VRS) and constant return to scale (CRS), which made it possible to determine the change in efficiency with a change in the scale of operations and to assess the scale efficiency of Ukrainian banks for 2017. It is assumed that a bank has an efficient scale if the volume of its active operations is optimal or, in other words, if any changes in their volume would have made it less efficient. The value of the scale efficiency of a bank is obtained by dividing the CRS by the VRS. The scale efficiency measures the closeness of the size of a bank’s operations to the scale that is most efficient for it. It is found that large state-owned banks had a higher average VRS of technical efficiency than other groups of banks. However, large state-owned banks demonstrated a low level of SE with the diminishing return to scale (DRS). This indicates the inefficient use by state-owned banks of their resources and the need to reduce the scale of their operations in order to achieve their most efficient scale of activity, which is obtained in the article on the basis of the CRS model. Moreover, there considered the negative impact on the banking business in Ukraine of significant preferences provided by the NBU to state-owned banks, namely: full guarantee of deposits in them, access to virtually unlimited refinancing, etc. The analysis demonstrates that the majority of small Ukrainian banks functioned with increasing return to scale (IRS), which indicates the potential for economies of scale and the need to scale up their operations to maximize returns on existing inputs. In the article there obtained the average values of VRS, CRS, and SE for the entire banking system of Ukraine for 2017 with the income and intermediary approaches, and calculated correlations between them.

Keywords