|
on Central Banking |
| By: | Leonardo Gambacorta; Byeungchun Kwon; Taejin Park; Pietro Patelli; Sonya Zhu |
| Abstract: | We introduce central bank language models (CB-LMs) - specialised encoder-only language models retrained on a comprehensive corpus of central bank speeches, policy documents and research papers. We show that CB-LMs outperform their foundational models in predicting masked words in central bank idioms. Some CB-LMs not only outperform their foundational models, but also surpass state-of-the-art generative Large Language Models (LLMs) in classifying monetary policy stance from Federal Open Market Committee (FOMC) statements. In more complex scenarios, requiring sentiment classification of extensive news related to the US monetary policy, we find that the largest LLMs outperform the domain-adapted encoder-only models. However, deploying such large LLMs presents substantial challenges for central banks in terms of confidentiality, transparency, replicability and cost-efficiency. |
| Keywords: | large language models, gen AI, central banks, monetary policy analysis |
| JEL: | E58 C55 C63 G17 |
| URL: | https://d.repec.org/n?u=RePEc:bis:biswps:1215 |
| By: | Ekaterina Pirozhkova (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Giovanni Ricco (CREST, Ecole Polytechnique, 5 Av. Le Chatelier, 91120 Palaiseau, France); Nicola Viegi (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa) |
| Abstract: | Four factors drive the high-frequency impact of monetary policy announcements in South Africa: affecting short-, mid-, and long-term yield curve, as well as country risk. Controlling for information effects, we build IVs to study the transmission of conventional monetary policy, forward guidance, term premia, country risk and information shocks. Our findings reveal textbook contractionary effects of conventional monetary policy. Policy communication, particularly forward guidance, has persistent effects on output and prices. Country risk is a novel and powerful channel of monetary policy communication in emerging markets. By defending its independence, re-stating its inflation target objective, and addressing external shocks, the central bank can mitigate country risk and generate strong expansionary effects. |
| Keywords: | Monetary policy, Small Open Economy, Inflation Targeting, Exchange Rates. |
| JEL: | E5 F3 F4 C3 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202442 |
| By: | Carlos Alba; Julio A. Carrillo; Raúl Ibarra |
| Abstract: | This paper analyzes, using a VAR model, the effects of US central bank monetary policy announcements, and information shocks from this authority regarding its economic outlook on Mexican financial and macroeconomic variables. Shocks are identified by combining a high-frequency strategy with sign restrictions, which exploits the co-movement between the policy rate and the stock market in the US around FOMC announcements. A restrictive monetary policy shock in the US is identified by an increase in the interest rate and a drop in stock prices, while a positive information shock is identified when both variables rise simultaneously. The results show that positive information shocks from the US central bank improve financial conditions in Mexico, appreciate the peso/dollar exchange rate, lower the sovereign risk premium and forex volatility, and increase stock prices, real activity and prices in Mexico. In contrast, restrictive US monetary policy shocks tighten financial conditions, and reduce real activity and prices in Mexico. |
| Keywords: | Monetary policy;international policy transmission;high-frequency identification;central bank information;VAR model |
| JEL: | E43 E52 E58 F42 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:bdm:wpaper:2024-14 |
| By: | Rajeswari Sengupta (Indira Gandhi Institute of Development Research); Harsh Vardhan; Akhilesh Verma (ESRI & Trinity College) |
| Abstract: | This paper explores the role of bank capital in monetary policy transmission within the Indian economy, where the banking sector is the primary channel for financial intermediation. Using a panel dataset of 18 commercial banks from 2002 to 2018, we assess how varying levels of bank capital influence monetary policy transmission. Our findings reveal that though monetary contractions reduce credit growth, yet banks holding higher capital show significantly lower sensitivity to monetary policy changes compared to those with lower capital. This effect is stronger in well-capitalized private sector banks. Our results suggest that bank capital helps mitigate the adverse impact of higher interest rates on credit supply, thereby weakening the overall effectiveness of monetary transmission. However, the buffering effect of bank capital on credit growth diminishes during periods of high nonperforming assets (NPAs). These results highlight the Reserve Bank of India's challenge in balancing financial stability with effective monetary policy implementation. |
| Keywords: | Bank capital, Monetary transmission, Balance sheet channel, Non-performing assets, Public-sector banks |
| JEL: | E4 E5 G2 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:ind:igiwpp:2024-019 |
| By: | Ekaterina Pirozhkova (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Nicola Viegi (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa) |
| Abstract: | This paper studies the bank lending channel of monetary policy transmission in South Africa in a context where the bank-loan level data, typically used for this type of analysis, are unavailable. Supply-side changes in credit provision are measured with data on composition of homeloan supply by banks versus nonbanks. High-frequency surprises in forward rate agreements are used to instrument for exogenous shifts in monetary policy in a proxy-SVAR model. The bank lending channel is found to be operative, as banks reduce the supply of homeloans following monetary tightening with a negative effect on housing market. The effectiveness of the deposits channel is shown: banks widen the deposit spread after monetary tightening, and the volume of deposits shrinks. As retail deposits provide a unique stable source of funding for banks, the deposits channel underlies the operativeness of the bank lending channel in South Africa consistent with theory. |
| Keywords: | monetary policy transmission, bank lending channel, credit channel, nonbank financial institutions, housing finance |
| JEL: | E52 G21 G23 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202443 |
| By: | Ángelo Gutiérrez-Daza |
| Abstract: | Empirical evidence suggests consumers rely on their shopping experiences to form beliefs about inflation. In other words, they "learn by shopping". I introduce this empirical observation as an informational friction in the New Keynesian model and use it to study its consequences for the transmission of aggregate shocks and the design of monetary policy. Learning by shopping anchors households' beliefs about inflation to its past, causing disagreement with firms over the value of the real wage. The discrepancy allows nominal shocks to have real effects and makes the slope of the Phillips curve a function of the monetary policy stance. As a result, a more hawkish monetary policy reduces the volatility and persistence of inflation, increases the degree of anchoring of households' inflation expectations, and flattens the slope of the Phillips curve of the economy. |
| Keywords: | Inflation;Inflation Expectations;Monetary Policy;Business Cycle;Informational Frictions |
| JEL: | D84 E31 E32 E52 E58 E70 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:bdm:wpaper:2024-12 |
| By: | Ester Faia; Karen K. Lewis; Haonan Zhou |
| Abstract: | We re-examine monetary policy spillovers to Emerging Market Economies (EME) in the form of capital flow reversals, using sectoral-level securities holdings data for Euro Area investors. In response to a surprise monetary tightening, active investors such as investment funds re-balance their portfolios away from EME, while more passive, long term investors such as insurance funds and banks exhibit no significant reaction on average. For active investors, the reallocation out of EME appears stronger under synchronized monetary tightening between the Fed and the ECB. However, these investors may even inject more capital to EME securities when the monetary tightening surprises contain positive news about the Euro Area economy. Issuers' monetary-fiscal stability may explain the heterogeneous impact of these spillovers. |
| JEL: | E44 F32 F33 G15 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:32986 |
| By: | Bhattacharjee, Arnab; Holly, Sean; Wasseja, Mustapha |
| Abstract: | Transcripts from the US Federal Open Markets Committee provide, albeit with a lag, valuable information on the monetary policymaking process at the Federal Reserve Bank. We use the data compiled by Chappell et al. (2005b) on preferred interest rates (not votes) of individual FOMC members. Together with information on which monetary policy decisions are based, we use these preferred rates to understand decision making in the FOMC, focussing both on cross-member heterogeneity and interaction among the members of the committee. Our contribution is to provide a method of unearthing otherwise unobservable interactions between the members of the FOMC. We find substantial heterogeneity in the policy reaction function across members. Further, we identify significant interactions between individuals on the committee. The nature of these interdependencies tell us something about information sharing and strategic interactions within the FOMC and provide interesting comparisons with the Bank of England's Monetary Policy Committee. |
| Keywords: | Monetary policy, Interest rates, FOMC decision making, Spatial Weights Matrix, Spatial Lag Model |
| JEL: | E42 E43 E50 E58 C31 C34 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:hwuaef:303042 |
| By: | Gonzalo Cisternas; Aaron Kolb |
| Abstract: | Expectations can play a significant role in driving economic outcomes, with central banks factoring market sentiment into policy decisions and market participants forming their own assumptions about monetary policy. But how well do central banks understand the expectations of market participants—and vice versa? Our model, developed in a recent paper, features a dynamic game between (i) a monetary authority that cannot commit to an inflation target and (ii) a set of market participants that understand the incentives created by that credibility problem. In this post, we describe the game, a type of Keynesian beauty contest: its main novelty is that each side attempts, with varying degrees of accuracy, to forecast the other’s beliefs, resulting in new findings regarding the levels and trajectories of inflation. |
| Keywords: | central banks; credibility; forecasts |
| JEL: | D82 E5 |
| Date: | 2024–09–30 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fednls:98892 |
| By: | Dvoskin, Ariel (CONICET – University of San Martín, EIDAES); Libman, Emiliano (CONICET – Center for the Study of State and Society) |
| Abstract: | Following Aspromourgos’ (2007) steps, in this paper we examine the role of the interest rate as a “conventional” variable under different assumptions regarding price and quanti-ty determination, that aim to characterize the reaction functions of Central Banks, repre-sented in standard New Consensus models. More specifically, we lay out a minimal model and suggest a taxonomy that helps examining under which conditions prices and quantities can be determined independently of each other, and whether there is or there is not a unique natural rate of interest consistent with the equilibrium level of output. We argue that the natural rate of interest need not exist even if, as the New Consensus ar-gues, we allow prices and quantities to be somehow connected. |
| Keywords: | Inflation Targeting; Interest Rates; Monetary Theory of Distribution; New Consensus Model |
| JEL: | E31 E52 E58 |
| Date: | 2024–09–26 |
| URL: | https://d.repec.org/n?u=RePEc:ris:sraffa:0068 |
| By: | Tomás E. Caravello; Alisdair McKay; Christian K. Wolf |
| Abstract: | In a rich family of linearized structural macroeconomic models, the counterfactual evolution of the macro-economy under alternative policy rules is pinned down by just two objects: first, reduced-form projections with respect to a large information set; and second, the dynamic causal effects of policy shocks. In particular, no assumptions about the structural shocks affecting the economy are needed. We propose to recover these two sufficient statistics using a ``VAR-Plus'' approach, and apply it to evaluate several monetary policy counterfactuals. |
| JEL: | E32 E58 E61 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:32988 |
| By: | Adrien Auclert; Matthew Rognlie; Ludwig Straub |
| Abstract: | In the past decade, a new paradigm for fiscal and monetary policy analysis has emerged, combining the canonical macro model of income and wealth inequality with the New Keynesian model. These Heterogeneous-Agent New Keynesian (“HANK”) models feature new transmission channels and allow for the joint study of aggregate and distributional effects. We review key developments in this literature through the lens of a unified “canonical HANK model”. Monetary and balanced-budget fiscal policy have similar aggregate effects as in the standard new Keynesian model, while deficit-financed fiscal policy is much more expansionary. We discuss the split between direct and indirect effects of policy, and also the implications of cyclical income risk, maturity structure, nominal assets, behavioral frictions, and many other extensions to the model. Throughout, we highlight the benefits of using sequence-space methods to solve and analyze this class of models. |
| JEL: | D1 E21 E31 E32 E43 E52 E62 |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:32991 |
| By: | Ratcliff, Ryan D. |
| Abstract: | Badinger and Schiman (2023) use a narrative high-frequency analysis of news and financial markets to develop a small set of restrictions on the structural shocks of a VAR of the Euro area. Their approach does not uniquely identify a structural representation, so their results are based on the distribution of a randomly generated set of parameters that satisfies the restrictions. Their method generates impulse responses that are consistent with macroeconomic theory, but that differ from previous studies that use alternative highfrequency identification strategies. They use this difference to argue that, unlike previous studies, their method is able to separate monetary policy surprises from confounding central bank information shocks - an important new contribution to the literature. I conducted two replication studies of their work on behalf of the Institute for Replication (I4R). First, I used the code provided in their replication package to replicate all of their main results, aside from the small variations expected in replicating a Monte Carlo study. Second, I attempted to use their original data to recreate their results using a different statistical software (Eviews 13). I was unable to replicate their results for two reasons. First, my program is unable to exactly replicate the custom prior they used to generate their reduced-form results. Second, my models routinely generated nonstationary VARs that nevertheless satisfied the identification restrictions. This differs from the author's results, but is not surprising given the ambiguous stationarity of the underlying macro variables. |
| Keywords: | Structural VAR, Residual Sign Restrictions, Replication |
| JEL: | C32 E43 E44 E52 E58 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:i4rdps:160 |
| By: | Arpan Chakraborty |
| Abstract: | Modern macroeconomic models, particularly those grounded in Rational Expectation Dynamic Stochastic General Equilibrium (DSGE), operate under the assumption of fully rational decision-making. This paper examines the impact of behavioral factors, particularly 'animal spirits' (emotional and psychological influences on economic decisions) and 'inflation extrapolators', on the communication index/sentiment index of the US Federal Reserve. Utilizing simulations from a behavioral New Keynesian model alongside real-world data derived from Federal Reserve speeches, the study employs an Auto-Regressive Distributed Lag (ARDL) technique to analyze the interplay between these factors. The findings indicate that while the fraction of inflation extrapolators do not significantly affect the Fed's sentiment index, various aspects of animal spirits exert a notable impact. This suggests that not only is the US output gap influenced by animal spirits, but the Federal Reserve's communication is also substantially shaped by these behavioral factors. This highlights the limitations of rational expectation DSGE models and underscores the importance of incorporating behavioral insights to achieve a more nuanced understanding of economic dynamics and central bank communication. |
| Date: | 2024–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2409.10938 |
| By: | Viral V. Acharya; Nicola Cetorelli; Bruce Tuckman |
| Abstract: | In recent years, assets of nonbank financial intermediaries (NBFIs) have grown significantly relative to those of banks. These two sectors are commonly viewed either as operating in parallel, performing different activities, or as substitutes, performing substantially similar activities, with banks inside and NBFIs outside the perimeter of banking regulation. We argue instead that NBFI and bank businesses and risks are so interwoven that they are better described as having transformed over time, rather than as having migrated from banks to NBFIs. These transformations are at least in part a response to regulation and are such that banks remain special as both routine and emergency liquidity providers to NBFIs. We support this perspective as follows: (i) the new and enhanced financial accounts data for the United States (“From Whom to Whom”) show that banks and NBFIs finance each other, with NBFIs especially dependent on banks; (ii) case studies and regulatory data show that banks remain exposed to credit and funding risks, which at first glance seem to have moved to NBFIs, and also to contingent liquidity risk from the provision of credit lines to NBFIs; and (iii) empirical work confirms bank-NBFI linkages through the correlation of their abnormal equity returns and market-based measures of systemic risk. We discuss some potential regulatory responses, including treating the two sectors holistically, recognizing the implications for risk propagation and amplification, and exploring new ways to internalize the costs of systemic risk. |
| Keywords: | nonbank financial intermediaries; nonbanks; shadow banking; bank regulation; regulatory arbitrage; systemic risk; credit lines; derivatives margin |
| JEL: | G01 G21 G23 G28 |
| Date: | 2024–09–01 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fednsr:98820 |