ABSTRACT
This study investigates whether the transmission of monetary policy depends on the degree of competition in the banking sector. Applying an interacted panel VAR methodology consisting of a panel of 23 advanced economies, we find that GDP and credit respond less strongly to monetary policy shocks in economies where the bank market is more concentrated. We use the Lerner Index as a measure of banking competition and find evidence – albeit less clear – that monetary policy transmission is stronger when the banking sector is more competitive.
Acknowledgments
The views expressed herein are the authors’ own and do not necessarily reflect those of Bank of Korea or KDI.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 In an attempt to shed light on the role of banking market structure in monetary transmission to real magnitudes, we employ country-level data. We, nevertheless, acknowledge that combining bank-level data and real sector variables, such as firm-level investment and households’ consumption spending, could provide greater clarity about the effect of bank-specific characteristics in monetary transmission.
2 The index ranges from 0 to 1 and implies greater market competition as it approaches 0.
3 Among the endogenous variables, the investment, consumption, house prices, credit, share prices, and effective exchange rates are transformed into real terms. Components of the GDP and the CPI are seasonally adjusted by using X-13 ARIMA.
4 We also employ some alternative schemes for identification of monetary shocks, that is, zero and sign restriction identification, as well as recursive SVAR identification. We find that the results obtained by imposing the alternative identification schemes are in line with those from our baseline sign-restriction VAR.
For comparison, we impose an alternative restriction of short-term and longer-term interest rates rising and prices and GDP dropping in response to contractionary MP shock. We find that the result of our benchmark model is robust to various sign restrictions. We have also examined if there is any substantial change in the impulse responses when altering the restriction on the horizons and do not find any significant difference in the responses.