Abstract: This study addresses the limitations of traditional earnings risk measures, which often rely on statistical moments like variance, skewness, and kurtosis. In cases of heavy-tailed distributions, these moments may not exist in the population, challenging moment-based analyses. To overcome this, we introduce robust conditional Pareto exponents as novel and reliable measures of earnings risk, alongside methods for their estimation and inference. Analyzing data from the UK New Earnings Survey Panel Dataset (NESPD) and the US Panel Study of Income Dynamics (PSID), we observe that: (1) Moments frequently fail to exist; (2) Tail earnings risk increases over the life cycle; (3) Job stayers experience higher tail risk; and (4) These patterns are consistent both during the 2007{2008 recession and the 2015-2016 growth period.
Abstract: This paper presents empirical evidence on the nature of idiosyncratic shocks to firms and discusses its role for firm behavior and aggregate fluctuations. We document that firm-level sales and productivity are hit by heavy-tailed shocks, mostly unexplained by observable factors, and follow a nonlinear stochastic process, thus departing from the canonical linear AR(1). We estimate a state-of-the-art model to flexibly capture the rich dynamics uncovered in the data and characterize the drivers of nonlinear persistence and non-Gaussian shocks. We show the role these features play in achieving empirically plausible volatility and persistence of micro-originated (granular) aggregate fluctuations.
Abstract: This paper develops a novel approach to random effects estimation and individual-level forecasting in micropanels, targeting individual accuracy rather than aggregate performance. The conventional shrinkage methods used in the literature, such as the James-Stein estimator and Empirical Bayes, target aggregate performance and can lead to inaccurate decisions at the individual level. We propose a class of shrinkage estimators with individual weights (IW) that leverage an individual's own past history, instead of the cross-sectional dimension. This approach overcomes the "tyranny of the majority" inherent in existing methods, while relying on weaker assumptions. A key contribution is addressing the challenge of obtaining feasible weights from short time-series data and under parameter heterogeneity. We discuss the theoretical optimality of IW and recommend using feasible weights determined through a Minimax Regret analysis in practice.
Abstract: Food price changes have a strong and persistent impact on UK consumers’ inflation expectations. Over 60% of households report that their inflation perceptions are heavily influenced by food prices and display a stronger association between their inflation expectations and perceptions. We complement this finding with a Structural Vector Autoregression (SVAR) analysis, illustrating that food price shocks have a larger and more persistent effect on expectations compared to a “representative” inflation shock. Finally, we augment the canonical New-Keynesian model with behavioural expectations that capture our empirical findings and show that monetary policy should respond more aggressively to food price shocks.
Abstract: I propose the use of state-space methods as a unified econometric framework for studying heterogeneity and dynamics in micropanels (large N, medium T), which are typical of administrative data. I formally study identification and inference in models with pervasive unobservable heterogeneity. I show how to consistently estimate the cross-sectional distributions of unobservables in the system and uncover how such heterogeneity has changed over time. A mild parametric assumption on the standardized error term offers key advantages for identification and estimation, and delivers a flexible and general approach. Armed with this framework, I study the relationship between job polarization and earnings inequality, using a novel dataset on UK earnings, the New Earnings Survey Panel Data (NESPD). I analyze how the distributions of unobservables in the earnings process differ across occupations and over time, and separate the role played on inequality by workers’ skills, labor market instability, and other types of earnings shocks.