Summary

On June 6, 2019, Ivan Huljak, Senior Advisor, Croatian National Bank (HNB), and Reiner Martin, Lead Economist, Joint Vienna Institute (JVI), presented a paper on ‘The cost efficiency and productivity growth of euro area banks’. Thomas Richardson, Director, Joint Vienna Institute (JVI) chaired the event.

The efficiency of banks is highly relevant for financial stability, especially when interest rates are low, as they are now. In fact, the latest ECB Financial Stability Review identifies poor profitability and overcapacity in the banking sector as serious issues for financial stability in the euro area. In fact, in recent years the profitability of euro area banks has trailed that of banks in other parts of the world, such as the US and Scandinavia, whereas cost-to-income ratios of euro area banks are higher than in other developed economies.

Researchers have traditionally used accounting indicators, such as the average cost (AC) or the cost-to-income ratio (CIR), to assess banking sector efficiency. These are easy to compute but ill-equipped to capture efficiency properly. AC depends heavily on a bank’s business model, its size, and various country-specific factors outside bank control. The CIR captures several important aspects of bank performance, but it also depends partly on country-specific factors, such as the cost of labor. 

Recognizing these shortcomings, the presenters used an industrial organization approach to compute growth in total factor productivity (TFP) in the euro area banking sector. They estimated a trans-log cost function to compute the different components of TFP growth—overall technical efficiency, technological progress, equity, and scale effects. The analysis covered a sample of commercial, cooperative, and savings banks from 17 euro area countries from 2006 to 2017.

The presenters found that cost efficiency for the median euro area bank averaged 84% over the period. In other words, if the median bank would operate on the technical efficiency frontier, it could produce the same level of output at 84% of current costs. Moreover, the persistence of bank inefficiency over time suggests that structural long-term factors (location, client structure, macroeconomic environment, regulation, etc.) are more to blame than time-specific factors. 

Larger institutions tend to have lower efficiency scores, because they are more complex to manage. For 2007–16 the rate of technological progress for the median euro area bank averaged 1.7%. The shadow cost of equity had a substantial role after the global financial crisis began, suggesting that the reward of banks for being well capitalized increased in times of financial stress but declined thereafter. Finally, the presenters found economies of scale to be relatively larger for smaller institutions. Considering all components together, TFP growth in the euro area banking sector fell from above 2% in 2007 to below 1% in 2017—an undesirable development given the need to enhance the profitability of euro area banks.

The presenters suggested that banks should direct their productivity efforts to such areas as branch rationalization, digitalization of business processes, and possibly mergers and acquisitions. However, such cost-cutting activities take time to bear fruit and usually require substantial upfront investments.

During the subsequent discussion, audience members suggested expanding the policy recommendations stemming from the analysis. For instance, it was suggested that identifying best practices from certain banks or countries might improve the performance of euro area banks. It was also recommended that the presenters look in detail at the performance of other types of banks, e.g., foreign-owned banks or banks that benefited in recent years from state aid. How easily banks can exit the market was also mentioned as a possible determinant of cost efficiency and productivity. Finally, more detailed comparative analyses with other parts of the world (e.g., the CESEE region, the US, and Scandinavia) was also recommended.

Reiner Martin, Lead Economist, JVI