Working Papers
Bank–Firm Relationships under Aggregate Shocks: Monitoring Costs and Separation Friction [Job Market Paper] (Slides)
Abstract: This paper quantifies the aggregate cost of inefficient separations in bank-firm lending relationships and evaluates the policies used to preserve these relationships. Over the course of a relationship, the bank builds up information about the firm through costly monitoring, and this information is lost when the relationship ends. A separation is inefficient when the bank or the firm unilaterally ends the relationship, even though the joint surplus remains positive. I build a directed search model of the credit market, calibrated to the Portuguese credit register, in which monitoring costs decline with relationship tenure and loan terms are fixed when the relationship forms, thereby generating inefficient separations. Consumption is 8.3 percent higher under the planner's continuation rule than in the market economy, mainly because replacement relationships pay the high monitoring costs of a new match again. An aggregate downturn does not raise the number of inefficient separations; it shifts them toward banks ending young relationships, from 6 to 22 percent of all separations, and the least productive firms lose access to new credit. Of the downturn's 2.7 percent consumption loss, a full moratorium on foreclosure recovers a quarter, and a subsidy paid to firms recovers 12 percent. The same subsidy paid to banks increases the loss by a tenth: it encourages bank entry, which improves matched firms' outside options and raises the number of firm-initiated separations.
Potential Output in the Caucasus and Central Asia: Beyond the Growth Surge with Nasir Rao, Tongli Zhang, and Shuyu Wang (under IMF review: Draft available upon request)
Abstract: Recent shocks—including the pandemic and Russia’s war in Ukraine—were followed by stronger growth across much of the Caucasus and Central Asia (CCA), raising a key question: is the acceleration temporary, or does it reflect higher potential output? Using six complementary methodologies, this paper finds that the region's average potential growth increased from 4.2 percent in 2018–21 to 4.8 percent in 2022–25, with method-specific estimates clustered within a relatively narrow range. The increase was driven mainly by gains in total factor productivity, reflecting improvements in the quality and allocation of capital toward more productive uses, as well as more efficient resource deployment across the economy. Gains were larger in oil-importing economies, while oil exporters recorded more modest improvements amid hydrocarbon dependence and structural rigidities. Given data limitations and the inherent uncertainty surrounding potential-output estimation, the findings should be interpreted as indicative ranges rather than precise point estimates.
Abstract: This paper studies how monetary policy transmission changed at the zero lower bound. Using a time-varying parameter VAR with stochastic volatility and state-dependent local projections with high-frequency identified shocks, I find that credit spread transmission is amplified at the zero lower bound on average but turns negative at the peak of quantitative easing, consistent with tightening surprises being read as positive signals about the recovery. Real activity and inflation show no significant state dependence. The stochastic volatility results document a sharp, concentrated spike in credit spread volatility at the onset of the financial crisis, followed by rapid normalization, suggesting that unconventional policy stabilized financial conditions while altering the nature of monetary transmission through the financial channel.
Policy Publication