In light of the experience of the last few years, the object of this paper is to re-examine the role of supply shocks, labour market tightness and expectation formation in explaining bouts of inflation. We begin by showing that a quasi-flat Phillips curve, which was popular prior to the pandemic, still fits the post-2020 US data well with the implication that (1) labour market tightness likely played a limited role in generating recent inflation and (2) changes in short term inflation expectations induced by supply shocks likely played a major role. We then explore how best to capture the join dynamics of inflation and inflation expectations in response to supply shocks. Given the difficulty of capturing these dynamics under rational expectations, we propose and evaluate a model with bounded rationality. In our model, supply shocks that affect many goods affect inflation expectations and this drives persistent inflation dynamics, while supply shocks that are concentrated in a sector lead to much more temporary changes in both inflation and inflation expectations. Although our departure from full rationality is minor, it allows perceived common shocks to dis-aggregated inflation series to be amplified and propagated over time in a manner consistent with observation.

Paper

In this paper, I propose a flexible non-parametric method using Recurrent Neural Networks(RNN) to estimate the dynamic structure of most expectation formation models in macroeconomics. This approach does not rely on restrictive assumptions of functional forms and parametric methods but nests the standard approaches of empirical studies on expectation formation. Applying this approach to data on macroeconomic expectations from the Michigan Survey of Consumers(MSC) and a rich set of signals available to U.S. households, I find qualitatively new results: (1) agents' expectations about the future economic condition have asymmetric and non-linear responses to signals; (2) agents' attentions shift from signals about the current state to signals about future: they behave as if they were adaptive learners in ordinary periods and become forward-looking as the state of economy gets worse; (3) the content of signals on economic condition, rather than the volume of these signals, plays the most important role in creating the attention-shift. My method also allows me to apply the Double Machine Learning method to assess the statistical significance of these empirical findings. Finally, I show these stylized facts can be generated by a model with rational inattention, in which information endogenously becomes more valuable when economic status gets worse.
Paper Online Appendix

Households perceive macroeconomic variables to move together in a way different from the patterns seen in their realizations, professionals' forecasts, and the standard macroeconomic models. In particular, they expect unemployment rate to go up when expecting higher inflation and hears about inflation news. We perform a structural test based on a noisy-information framework and provide empirical evidence using survey expectations and newspapers' economic reporting to show that such perceived correlations reflect subjective models instead of simply imperfect information.
Paper

India witnessed a sharp wage catch-up by the historically disadvantaged scheduled castes and tribes (SC/STs) towards non-SC/ST levels during the period 1983-2012. We provide a structural explanation for the catch-up, using a multi-sector, heterogenous agent model where individuals differ in ability as well as their caste identity. Castes differ in the costs of schooling and accessing sectoral labor markets which results in caste-based talent misallocations. We show that exogenous productivity growth can explain 72 percent of the observed wage convergence. As in the data, education convergence is the primary driver of the wage convergence in the model. We provide independent evidence in support of this mechanism.
Paper

Over the past three decades India and China have experienced rapid growth and structural transformation. Underneath this similarity however was one significant difference: rural-urban wage gaps declined in India, but widened in China. In both countries, the majority of these wage dynamics are unexplained by worker attributes. We formalize a two-sector-two-location model in which structural transformation and urbanization respond endogenously to productivity shocks. While the structural transformation effect widens the urban-rural wage gap, the urbanization effect reduces it. We attribute the contrasting wage gap dynamics in the two countries to the higher costs of urban relocation for workers in China.
Paper