Research

Abstract: We conducted a survey of Turkish firms, using randomized treatments to provide various types of information about inflation in a high-inflation environment. By matching the survey data with detailed firm-level information on exports, employment, credit, and foreign exchange transactions, we explore the impact of exogenous variations in inflation expectations on firms’ behavior, borrowing decisions, and expectations. Our findings reveal several key insights. First, inflation expectations significantly impact firms’ expected prices, wages, and unit costs, with pass-through rates reaching as high as 90% in a high-inflation environment—demonstrating a much stronger connection compared to low-inflation settings. Additionally, firms with elevated inflation expectations are more likely to increase foreign currency purchases while reducing foreign currency sales. Second, firms with heightened inflation expectations substantially increase their demand for credit, shifting from short-term to long-term loans through the refinancing channel, with a significantly reduced cost. Specifically, a 10 percentage point increase in inflation expectations leads to a 3 percentage point rise in credit usage, accompanied by a 3 percentage point decline in borrowing costs. Third, firms facing higher inflation expectations tend to reduce employment and sales, reflecting a stagflationary outlook. Despite the negative impact on sales and employment, total credit usage increases, driven by long-term borrowing as firms seek to lock in lower long-term rates in anticipation of further inflation, highlighting the importance of the refinancing channel.

Presented: Bundesbank Conference on Expectations of Households and Firms, 2025; TEA, 2025; CBRT, 2024; Southern Economic Association, 2024; TGU, 2024

Abstract:  Using a large cross-country dataset covering over 150 countries and more than 10 macroeconomic variables, this study examines the consistency of IMF World Economic Outlook (WEO) forecasts with the full information rational expectations (FIRE) hypothesis. Similar to Consensus Economics forecasts, WEO forecasts exhibit an over-reaction to news. Our analysis reveals that this overreaction is asymmetric, with more measured response to bad news, bringing forecasts closer to the FIRE benchmark. Moreover, forecasts align more closely with FIRE hypothesis during economic downturns or when a country is part of an IMF program. Overreaction becomes more pronounced for macroeconomic variables with low persistence and for forecasts over longer horizons, consistent with recent theoretical models. We also develop a model to explain how state-dependent nature of attentiveness may drive this asymmetric overreaction.

Abstract: This paper examines how monetary policy committee announcements and inflation surprises affect Turkish firms' inflation expectations and FX management decisions during the 2015-2024 period, a time characterized by both relatively stable and highly volatile inflation. Using survey data, we define monetary policy (inflation) surprises as the unexpected component of interest rate (inflation) forecasts measured within a narrow window around policy announcements. First, we show that firms' aggregate inflation expectations change significantly in response to monetary policy shocks, but their response is highly state-dependent: In relatively stable inflation environments, firms react in line with standard theory, anticipating a decline in prices while becoming pessimistic about the economic outlook. By contrast, in volatile settings, firms adopt a supply-side interpretation, revising their expectations upward amid pessimism. Second, inflation surprises increase inflation expectations and heighten pessimism regardless of the state, providing additional evidence for the supply-side view. Third, a positive inflation surprise induces an increase in FX holdings, reflecting concerns about potential currency depreciation, whereas an unexpected policy rate hike prompts firms to reduce FX positions in anticipation of currency appreciation. Our findings demonstrate that in a high-inflation environment, firms are especially attentive to macroeconomic developments, revising their inflation expectations and adjusting their financial strategies accordingly. These results underscore the importance of monetary policy communication and inflation news in shaping firm-level decisions, particularly when inflation is elevated and volatile. 

Abstract:  We investigate the optimal exchange rate policy for a small open economy facing oil price shocks. Extending the framework of Itskhoki and Mukhin (2023), we demonstrate that the monetary authority can achieve the first-best outcome by closing both the output gap and uncovered interest parity (UIP) gap through a combination of interest rate adjustments and foreign exchange (FX) interventions. To achieve this, monetary policy focuses on closing the domestic output gap, while FX interventions play a dual role: stabilizing the nominal exchange rate and mitigating the impact of oil price shocks on UIP gaps. Simultaneously, they allow the nominal exchange rate to adjust in line with shifts in the natural equilibrium exchange rate driven by oil shocks. We further show that even in the absence of currency demand shocks, fluctuations in oil prices alone can justify FX interventions in a small open economy.

Abstract:  We run a survey of Turkish households in which randomized treatments provide different types of information about the first and second moments of inflation, in a high inflation environment. First, while factors commonly cited in the literature—such as education, income, gender, or age—do not significantly explain the cross-sectional variation in aggregate inflation expectations, the city of residence emerges as a significant determinant, reflecting spatial heterogeneity in the cost of living. Second, our information treatments, based on publicly available data, successfully generate significant exogenous variation in inflation expectations, even in a setting of very high inflation. Third, we analyze the causal impact of these changes in inflation expectations on households’ consumption and saving plans. We find that higher inflation expectations reduce expected spending, reflecting households’ stagflationary outlook: a 10 percentage point increase in inflation expectations corresponds to a roughly 3.5 percentage point decline in total and nondurable expected spending, along with a 3 percentage point decrease in the likelihood of durable goods purchases. This provides empirical evidence that the expected income channel dominates the intertemporal substitution channel. 

Presented: Koç Winter Workshop, 2024

Abstract: We document hand-to-mouth (HtM) ratios for European countries using the HFCS dataset and assess their role in monetary policy transmission. Using ECB monetary shocks and panel local projections, we find that countries with higher HtM ratios have a less pronounced response to monetary policy shocks compared to those with lower HtM ratios. This aligns with the predictions of heterogeneous agent New Keynesian models with both wage and price rigidity. When wages are stickier than prices, real wages for HtM agents may decline despite interest rate cuts, which hinders the demand boost typically expected from the New Keynesian Cross. Consequently, monetary policy is less effective in stimulating aggregate demand in countries with higher HtM ratios.
Presented: CBRT, 2024

Abstract: We study Euro Area inflation and unemployment dynamics by addressing the identification challenges inherent to estimating the Phillips curve and providing new frameworks and results for inflation forecasting. A well-known problem for Phillips curve slope estimation is that the simultaneity between monetary policy and economic conditions biases OLS estimates downward. To overcome this challenge, we adopt a panel local projection framework to estimate the cumulative impacts of monetary policy shocks on inflation and unemployment over a given horizon. This estimated trade-off is inversely related to the traditional sacrifice ratio, quantifying the unemployment cost of disinflation. Moreover, we use U.S. monetary policy shocks as IVs to ensure a causal interpretation of our results, and show that the trade-off identified through IV estimation is not just qualitatively, but also quantitatively, different from OLS estimation. We find no evidence that this trade-off has changed recently in the post-pandemic period. Finally, we provide novel tools for inflation forecasting based on the panel Phillips curve model. We document that panel data models yield accurate one-year ahead forecasts for 17 member states in the 2009-2023 period and clearly outperform the commonly used, naive single-country models.

Abstract:  This study investigates the elasticity of credit demand and examines how firm-level credit supply shocks influence borrowing costs and, in turn, affect key firm outcomes in Türkiye. Leveraging a comprehensive administrative dataset at the firm-bank level, we estimate the interest rate elasticity of credit demand to be approximately -2, indicating a high sensitivity of borrowing to cost changes. We then analyze how exogenous variation in borrowing costs affects firm-level outcomes such as sales, investment, employment, and default probability. Our results show that a 1 percentage point increase in interest rates reduces investment by 0.49 percent, lowers employment by 0.36 percent, and raises the probability of firm exit by 0.18 percentage points. These findings highlight the critical role of borrowing costs in shaping firm behavior, with important implications for performance and survival.

Abstract:  This paper leverages a rare quasi-experiment—the unexpected September 2021 interest rate cut by the Central Bank of the Republic of Türkiye—to examine how inflation expectations shape firm behavior in a high-inflation environment. Drawing on a rich dataset that combines monthly survey responses with administrative records, we exploit the heterogeneous revisions in firms’ inflation expectations triggered by the policy shock. Firms that significantly increased their inflation forecasts (treated) subsequently became more pessimistic about economic conditions, reduced employment, and curbed domestic sales. At the same time, they strategically raised procurement, acquired more foreign currency assets, and boosted borrowing in local currency—even at higher costs—in anticipation of debt erosion. These patterns suggest that firms’ heightened inflation expectations drive both defensive and opportunistic behaviors, ranging from hedging against currency depreciation to locking in lower financing costs. Overall, the findings highlight the critical role of inflation expectations in guiding firm-level decisions and underscore the importance of policy credibility in volatile macroeconomic settings.

Abstract:  Expectations of households, firms, and market participants differ significantly, with particularly wide dispersion during volatile periods. We examine how macroeconomic news impacts the inflation expectations of these groups. Our findings reveal that households’ and firms’ inflation expectations increase in response to a positive monetary policy (tightening) shock, whereas professionals’ expectations decline. This divergence arises from fundamentally different interpretations of economic dynamics: households and firms view monetary tightening as a signal of worsening economic conditions—such as rising costs, weaker supply capacity, and potential production disruptions—leading them to expect higher inflation. As a result, they associate tightening with higher inflation. In contrast, professionals adhere to a demand-driven perspective, interpreting monetary tightening as a mechanism to reduce aggregate demand and thus lower inflation, consistent with standard economic theory. This pattern is further confirmed by how the three groups respond to unexpected inflation news (inflation surprises). These findings help explain the persistent disagreement in inflation expectations across households, firms, and professional forecasters, particularly during periods of economic uncertainty.

Abstract:  We examine how real-time fluctuations in U.S. presidential election probabilities affect stock valuations of firms with distinct partisan affiliations. Using high-frequency data from blockchain-based prediction markets (PolyMarket and PredictIt) and corporate political contribution records from OpenSecrets during the 2024 campaign, we implement a local projection framework to identify the impact of electoral probability shifts. Our results demonstrate that increases in Donald Trump's election probability generate statistically significant positive differential returns for Republican-affiliated firms relative to Democratic-leaning counterparts. These effects exhibit a dose-response relationship with the degree of partisan alignment, emerge within hours of probability shifts, and persist for several weeks. The documented pattern remains robust across multiple temporal aggregations (hourly and daily), alternative prediction market data sources, and various econometric specifications. By identifying how corporate-political connections mediate market responses to electoral information, this research provides novel empirical evidence on the transmission mechanism through which political uncertainty affects asset pricing dynamics in an era of heightened polarization.

Abstract:  This paper analyzes how unexpected oil supply shocks shape firms' inflation expectations and real activity using a dataset of Turkish manufacturing firms. At the aggregate level, oil supply shocks significantly increase both actual CPI inflation and firms’ average inflation expectations, with effects persisting for up to 15 months. Our firm-level analysis reveals substantial heterogeneity: smaller and highly leveraged firms respond more strongly to oil shocks, significantly raising their expectations for inflation, own prices, and unit costs compared to larger, financially robust firms. Furthermore, these shocks worsen firms’ business outlook and lead to tangible reductions in capacity utilization. Leveraging administrative firm-to-firm transaction data, we show that oil shocks also reduce firms’ sales, purchases, and the number of trading partners—particularly among financially constrained firms—highlighting real dislocations that propagate through production networks. In contrast, carbon price shocks and global temperature changes have no significant impact, consistent with the absence of a binding carbon pricing mechanism in Türkiye during the study period. Our findings highlight oil supply shocks as a crucial driver of firm-level expectations and real activity, emphasizing the importance of incorporating energy-cost dynamics into inflation-targeting frameworks.

Abstract: I investigate the welfare maximizing steady-state inflation rate in a heterogeneous-agent New Keynesian model with Downward Nominal Wage Rigidity (DNWR). After matching the annual wage change distribution in the U.S., I demonstrate that DNWR has a significant impact on the economy, particularly when the inflation target is set low. The optimal inflation rate is estimated to be as high as 8.8%, and increasing the inflation target to the optimal level yields a welfare gain of nearly 3.50%. While the results exhibit sensitivity to parameterization, a broad range of calibrations indicates that the optimal inflation rate is consistently above 3%.
Presented: CBRT, 2024; 27th International Computing in Economics and Finance Conference, 2021

Abstract: It is well known that introducing sectoral heterogeneity in price stickiness amplifies monetary nonneutrality in standard New Keynesian models (Carvalho, 2006). Yet, less attention has been paid to the role of heterogeneity in wage stickiness. Using industry-level data, I document a statistically significant negative correlation of –0.27 between wage and price rigidity at the 3-digit NAICS level. I then develop a multi-sector New Keynesian model that incorporates heterogeneity in both wage and price rigidities. The model shows that, depending on the correlation between sectoral wage and price rigidities, the cumulative real response to a monetary shock can either be amplified or dampened relative to a benchmark economy with only heterogeneous price rigidity. Specifically, a perfectly positive correlation between sectoral wage and price rigidities amplifies the cumulative real response by up to 14 percent, whereas a perfect negative correlation reduces it by approximately 9 percent. However, in the model empirically calibrated to the 53-sector U.S. economy, the aggregate impact of wage rigidity heterogeneity is limited, as the weak observed correlation and the influence of large, moderately rigid sectors mute the underlying transmission channel.

Abstract: We explore the spillover impact of zombie firms in Türkiye by exploiting a rich administrative dataset that contains firm-level information on balance sheets, inter-firm sales, employment, and firm-bank level credit records. We document three key facts regarding zombie dynamics: (i) Leveraging matched firm-bank level credit registry data, we highlight the presence of an evergreening motive, leading to a misallocation of credit away from productive firms. (ii) Zombie firms, which are on average less productive than the nonzombie firms, impede investment and employment opportunities at healthier firms. Nonzombie firms operating in sectors with a high prevalence of zombie firms experience lower sales, assets, and productivity. (iii) Incorporating B2B sales data structured similarly to firm-level input-output data, our study reveals that firms with stronger upstream or downstream zombie connections tend to exhibit reduced sales, investment, and employment compared to firms without any zombie connections. Moreover, banks supply less credit to healthy firms operating in a zombie network than to firms with fewer zombie connections.
Presented: Özyeğin, 2024; Koç, 2024

Abstract: We document the demand- and supply-driven components of inflation in Türkiye by following the decomposition method of Shapiro (2024). The results suggest that the recent surge in inflation, which began with the COVID-19 pandemic and deviated significantly from global inflation rates, reaching as high as 80%, was initially driven by supply factors. As monetary policy loosened, demand-driven inflation also increased; however, throughout the post-COVID period, supply-driven inflation consistently exceeded the demand-driven component in this high-inflation environment. Consistent with theory, oil supply and exchange-rate shocks increased the supply-driven inflation, while monetary policy tightening reduced the demand-driven inflation. This decomposition can potentially serve as a useful real-time tracker for policymakers.

Abstract: In this paper, we investigate labor income profiles in Turkey. In doing so, we study the role of educational attainment, gender and the public versus private sector employment on labor income profiles by using the Turkish Statistical Institute’s Household Labour Force Survey micro-data from 2004 to 2018. We first report that while the average labor income profile in Turkey exhibits a moderate hump-shape over age, there exists an immense degree of heterogeneity in labor income trajectories over education, gender and the sector of employment. Second, while the public sector employment is more advantageous for low-educated Turkish employees, university graduates in Turkey’s labor market face a risk versus return trade-off in their choice of the sector of employment: the private sector labor income profiles display a similar level of average income but a higher degree of cross-sectional variation compared to their public sector counterparts. Third, we report a significant gender pay gap, especially among low-educated workers, which aligns well with historically low female participation rates in Turkey. Our findings via distributional clustering analysis, ordinary least squares and pseudo-panel estimations all indicate that in attempts to infer economy-wide average labor income profiles, abstracting away from any of these listed factors could lead to misleading inferences.