Do public banks reduce monetary policy power? Evidence from Brazil based on state dependent local projections (2000–2018)

ABSTRACT

We test the hypothesis that public banks reduce monetary policy power. Previous studies have shown that companies with access to government-driven credit present smaller fall in investment and production after a contractionary monetary policy shock. Nevertheless, these studies are based on microeconomic data and ignore macroeconomic effects, especially the cost-push effects, of monetary policy. We employ state-dependent local projections to compare monetary policy power – the sensibility of inflation to changes in policy interest rate – between periods of high credit of public banks and periods of high credit of private banks. We do not find evidence that monetary policy is less powerful in periods of high credit of public banks. Even though periods of high credit of public banks present a lower effect over output, those periods present less persistent price puzzles than periods of high private credit. Robustness of results is enhanced by performing several tests. We attribute our results to lower flexibility in interest rates of credit from public banks, what leads to lower transmission in financial costs, lower reduction in capital stock and lower variation in the exchange rate.

KEYWORDS:

• Public banks
• monetary policy
• local projections
• credit-channel

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed here.

Notes

1. Formerly, the main benchmark for BNDES loans was the Long Term Interest Rate, a rate defined by the National Monetary Council and usually fixed in lower levels than the policy rate.

2. In addition to lower cost shocks, the existence of public banks could lead to less pronounced reductions in capital stock following increases in interest rates (Feijó and Sousa 2012; Modenesi and Modenesi 2012; Castro 2018). Through this channel, the short-term impacts of targeted credit on the power of monetary policy could be offset by allowing potential output to expand, reducing the impact of demand pressures on inflation.

3. Analogously, Jordà, Schularick, and Taylor (2019) compares the credit cycles defined by different debtors, whether households or companies. Authors conclude that cycles led by household debt are more influent for monetary policy interventions.

4. Estimations are performed with the lpirfs package, implemented in R by Adämmer (2018).

5. Function implemented by Adämmer (2018), and originally suggested by Granger and Terasvirta (1993).

6. These effects are lower than during the period of low credit of public banks, what seems to be at odds with the impulse response functions. Nevertheless, this is the case because the increases in interest rates during the period of high credit of public banks is higher. Thus, considering the higher interest rates increase in the high public credit state, GDP is more resilient to the monetary policy in the high public banks credit regime than in the low public credit regime.

Additional information

Funding

This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq); Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ).