Working Papers

Expectations for Gasoline Prices and Inflation: Evidence from Households With Karl Whelan, Constantin Burgi.

Conditionally Accepted JMCB

Abstract

Gasoline prices are highly salient to consumers and, for this reason, they may have an outsized influence on their thinking about inflation. We examine how people’s expectations about gasoline prices influence their expectations for overall inflation. We find little evidence from two US household surveys that people over-react to their beliefs about expected gasoline prices when formulating their expectations about overall inflation.

Firm Climate Investment: A Glass Half-Full With Ivan Yotzov, Nicholas Bloom, Philip Bunn, Paul Mizen, Gregory Thwaites.

NBER, VOXEU

Abstract

We analyze the importance of climate-related investment using a large economy-wide survey of UK firms. Over half of firms expect climate change to have a positive impact on their investment in the medium term, with around a quarter expecting a large impact of over 10%. Around two-thirds of these investments are expected to be in addition to normal capital expenditure, with some firms investing less elsewhere. These investments will be driven by larger firms as well as those in more energy-intensive sectors. Climate investments are expected mainly in switching to green energy sources and improving energy efficiency, and firms expect to finance these mainly using internal cash reserves. Overall, although firms are expecting to invest more resources in adapting to climate change, under reasonable assumptions, these investments are still not sufficient to meet the estimated targets implied by the UK Net Zero Pathway.

The Drivers of Household Inflation Uncertainty With Philip Schnattinger.

Draft

Abstract

Applying the round-number methods proposed in Binder (2017), we infer an individual's cognitive uncertainty about product groups. We use the responses of individuals point expectations about gasoline, food, medical, education, rent, and gold prices in the NYFed Survey of Consumers Expectation. Of these individual product groups, food prices are found to be the main driver of an individual's aggregate uncertainty about future inflation. We then show that a monetary policy is most effective at reducing food price uncertainty.

Work in Progress

Inflation Uncertainty and Household Wage Growth Expectations With Philip Schnattinger.

Draft coming soon!

Abstract

This paper investigates how subjective household inflation uncertainty—capturing second-moment beliefs—shapes employed individuals’ expectations of nominal wage growth. Utilizing detailed microdata from the Federal Reserve Bank of New York’s Survey of Consumer Expectations (SCE), we document two key empirical findings: (i) individual-level inflation uncertainty is positively associated with wage growth expectations, and (ii) this relationship is significantly stronger for low-income and low-wealth households. To address potential endogeneity arising from simultaneity in wage-price dynamics, we propose a novel instrumental variable strategy that exploits variations in forecast imprecision for highly salient consumer goods (gasoline and food). Our identification leverages the cognitive heuristic that individuals use ”round numbers to represent uncertain forecasts,” generating quasi-exogenous variation in inflation uncertainty. To interpret these empirical findings, we develop a search-and-matching model of the labor market with heterogeneous worker wealth, extending the framework of Krusell et al.(2010) to incorporate wage bargaining under uncertainty, combining the alternative offer bargaining wage bargain proposed Hall and Milgrom (2008) with the solution for bargaining under uncertainty developed in White (2008). In our model, nominal wages are negotiated before the realization of inflation. Risk-averse workers, facing uncertainty about their future real purchasing power, demand higher nominal wages as compensation for bearing inflation risk. This compensating risk premium mechanism plays a pivotal role, explaining why increased inflation uncertainty leads workers to form higher nominal wage growth expectations and negotiate higher wage increases. The mechanism is particularly strong for those workers with lower income and wealth who are less able to smooth consumption when inflation risk is high. Our findings highlight the importance of second-moment beliefs in wage determination and contribute to broader debates on inflation dynamics, labor market behavior, and the optimal design of monetary and fiscal policy under uncertainty

Salience in Inflation Expectations: Evidence from the UK.

Draft coming soon!

Abstract

This paper investigates which components of the UK’s Consumer Price Index (CPI) disproportionately shape both consumers’ and professional forecasters’ perceived and one-year-ahead inflation expectations. By leveraging a highly granular breakdown of over 200 CPI categories, we aim to understand how specific price changes—rather than the aggregate index—drive inflation beliefs across heterogeneous population groups. We merge detailed price data from the UK’s Living Costs and Food Survey with inflation expectation data from the Bank of England’s Inflation Attitudes Survey, creating a rich panel that allows us to trace how individual components of the consumption basket influence inflation perceptions. Our empirical strategy employs a two-stage machine learning framework. In the first stage, we use Random Forests and LASSO for feature selection to identify which CPI components are most predictive of expectations. In the second stage, we use these features to forecast inflation beliefs based on the selected features. Our findings show that a small subset of salient and frequently purchased items plays a disproportionately large role in shaping inflation expectations—particularly among lower-income households. These insights not only improve our understanding of expectation formation but also enhance the forecasting performance across different demographic groups.

Shedding light on economic growth: Nowcasting in Venezuela With Eurydice Fotopoulou, Maria Belen Sbrancia

Draft coming soon!

Abstract

Economic forecasting in Venezuela presents significant challenges due to the discontinuation of key macroeconomic data publication by the Banco Central de Venezuela since Q1 2019. The economy has experienced a sharp and prolonged contraction, particularly between 2013 and 2020, shrinking to a quarter of its 2012 size. The absence of reliable data on fundamental economic indicators—such as real GDP, trade flows, and manufacturing activity—combined with structural shifts in the economy, has severely constrained the ability to conduct accurate macroeconomic analysis and forecasting. This paper explores alternative approaches to addressing these data limitations by integrating traditional and non-traditional data sources with machine learning and econometric techniques to estimate real GDP. Specifically, it evaluates the applicability of Random Forest, Stacking methods, and the Dynamic Factor Model in the case of Venezuela. The analysis assesses the advantages and limitations of these methodologies, highlighting their potential to bridge critical data gaps and enhance economic forecasting in environments where direct data collection is limited. These approaches may offer broader applicability for forecasting in data-scarce economies, informing policy design and economic decision-making in similarly constrained contexts.

Media Sentiments, Politics and Expectations: Survey data from Fed Cleveland With Constantin Burgi, Edward S. Knotek II

Draft coming soon!

Abstract

Subjective Household Philips Curve: Survey data from ECB With Anushka Mitra, Einar Paz

Draft coming soon!

Abstract