货币政策对贷款利率的传导：来自巴西的证据（英）

Monetary PolicyTransmission to LendingRates: Evidence fromBrazil Daniel Leigh and Rui Xu WP/25/152 IMF Working Papersdescribe research inprogress by the author(s) and are published toelicit comments and to encourage debate.The views expressed in IMF Working Papers arethose of the author(s) and do not necessarilyrepresent the views of the IMF, its Executive Board,or IMF management. 2025JUL IMF Working Paper Western Hemisphere Department Monetary Policy Transmission to Lending Rates: Evidence from BrazilPrepared byDaniel Leigh and RuiXu* Authorized for distribution byAna CorbachoJuly2025 IMF Working Papersdescribe research in progress by the author(s) and are published to elicitcomments and to encourage debate.The views expressed in IMF Working Papers are those of theauthor(s) and do not necessarily represent the views of the IMF, its Executive Board, or IMF management. ABSTRACT:Thispaper estimatesthestrength ofmonetary policy transmission to bank lending rates in Brazil.We identifymonetary policy shocksusingforecast errors fromBrazi’s dailyFocussurveyofprofessionalforecasters.We thenestimatethe pass-through to lending ratesbasedon aninstrumental variable applicationoflocal projectionsandfindan aggregate pass-through of 70percentafter four months,reflectingfull pass-through to market-based lending ratesand 20percenttogovernment-directed creditinterestrates. Analysisusing bank-level data reveals varyingdegrees ofpass-through across credit types, from 40 percent for payroll-backed loans to 80 percent for working capital loans, and stronger pass-through for larger banks.Estimatedpass-throughhasincreasedsince 2020due to more responsive corporate loans. RECOMMENDED CITATION:Leigh, Danieland Rui Xu.2025. “Monetary Policy Transmision to LendingRates: Evidence from Brazil.”IMF Working Paper. I.Introduction Despitedouble-digitpolicyinterest ratessince 2022, growthof credit and economic activityin Brazilhasbeenstrong. TheCentral Bank of Brazil’sbroad credit gapmeasure—the deviation oftotalprivatecredit fromitsestimated historical trend—expanded in 2024 toits highest levelsince 2016at5.4 percent of GDP(Figure 1).This hasprompteddiscussions about the effectiveness of monetary policyin Brazil.As in other majoreconomies, thereis a lackof consensus on how muchpolicy rate changespass through tolending rates forBrazilianhouseholds and firms, acentral channelthrough whichmonetary policy affectsthe economy. Somestudies suggest that the pass-through to lending rates in Brazil is constrained by idiosyncratic factors,includinga sizable proportionofgovernment-directed(earmarked)credit, a history of macroeconomicinstability, and high interestratemargins. Research by Elias and Guimaraes (2024) and Takedaand others(2005) indicates an insignificant pass-through from policy rates to household credit lending rates due to highinterestratespreads.Otherstudies,including those by theCentral Bank of Brazil (BCB,2022) and Divino andHaraguchi (2020), demonstrate a pass-through well above one for personal creditlending rates. De Mello andde Castro (2012) and Divino and Haraguchi (2020) find evidence of asymmetric pass-through, with lendingratesmore responsiveto tightening cycles than to easing cycles. Furthermore, the influence of bankconcentration and financial regulation on pass-through dynamics is emphasized in research by de Mello andPisu (2010) and Alencarand others(2020). In addition, it remains unclear to what extent the COVID-19 pandemic and recent changes in Brazil's financialmarket have influenced pass-through.Ontheone hand,pandemic-eraliquidity injections by theBCBandgovernment loan guaranteesmayhavemoderated pass-through bydistortingcredit pricing.On the other hand,structural changes, includingcapital deepeningandreformofthe Brazilian Development Bank (BNDES)in2018mayhaveenhancedlending rate flexibility andpass-throughfrom policy rates.Unlike for other majoreconomies,wherestudieshaveshown weakenedmonetary policy effectivenesssincethe pandemic(see, forexample,Barrett and Platzer 2024; Prabheesh and others 2021; Andaloussi and others 2024; and Beyer andothers 2024), evidence onBrazil’sinterest rate pass-through is limited tothe pre-pandemic period. To shed light on these issues, this paperaddressesthree questions related to the strength of monetary policytransmission. First,how strong is the pass-through from policy rates toaggregatelending rates in Brazil?Second,has the pass-through changed significantlysince 2020in the context ofthe pandemic and variousstructural changes in the credit market? And, finally,how does the pass-throughvary across loan types andbank characteristics?In addressing these questions,ourresearch introduces three main innovations. Thefirst innovation is toconstruct a daily dataset of monetary policy shocks in Brazilusingforecast errors ofpolicyrates from theFocussurveyof professional forecastersmaintained by the BCBforthe 2012-2025 period.We measure the forecast error by comparingtheactualpolicyr