Supply Chain Disruptions and Global Inflation: In Need of New Approaches for Public Policies?
Paper Session
Friday, Jan. 6, 2023 10:15 AM - 12:15 PM (CST)
- Chair: Thierry Warin, HEC Montreal and CIRANO
The Demand and Supply of Information About Inflation
Abstract
In this article we study how the demand and supply of information about inflation affect inflation developments. As a proxy for the demand of information about inflation, we ex- tract Google Trends (GT) for keywords such as "inflation", "inflation rate", or "price increase". The rationale is that when agents are more interested about inflation, they should search for information about it, and Google is by now a natural source. As a proxy for the supply of information about inflation, we instead use an indicator based on a (standardized) count of the Wall Street Journal (WSJ) articles containing the word "inflat*" in their title.Predicting Individual Treatment Effects: Increasing Social Efficiency of Public Policy
Abstract
Experimental methods are tasked with recreating a fictitious world, which is known as the counterfactual. This paper brings forward a very simple and general concept for predicting the counterfactual using a Causal Time Machine. One strength of this method is that it is not plagued with challenges of selection bias that are commonly known to exist in the cross-section. Another strength is that the method generates true Individual Treatment Effects (ITEs). So far as I am aware, all current methods that claim to generate ITEs are Conditional Treatment Effect Models CATEs rather than ITE models. Another advantage of the time dimension is that it allows predictive models to be individually built for every individual within the cross-section, and predictive accuracy can be measured on data that the model has never been trained on. Model uncertainty is thus directly observed and known. This paper shows a simple blueprint for a Causal Time Machine that accommodates many different classes of Statistical and Machine Learning models. I apply a Causal Time Machine to the US-China trade war as a demonstration of how grossly inefficient homogeneous policy is within the context of treatment effect heterogeneity. This is a call for us to recognize the profound social inefficiency that we create because we rely on Average Treatment Effects for Policy Analysis rather than Individual Treatment Effects. As a result, policy is built for the average representative agent, who is an agent that is similar to very few individuals within a population. Policy ends up being over-applied or under-applied most of the time.The Political Economy of Export Bans and Commodity Price Volatility: Theory and Evidence from Agricultural Markets
Abstract
We show the importance of accounting for political risk to understand forward-looking price volatility in agricultural markets. We propose a theoretical model that shows uncertainty about the future world price of staple foods is positively related to the likelihood (and, counterintuitively, is further boosted by the actual imposition) of export bans in top producer countries. To test our model's predictions, we use option-implied volatilities (IVols) as a proxy for commodity market uncertainty. We construct a novel, daily dataset of major restrictions on grain and oilseed exports that were announced, adopted, or repealed in 2002-2019. We show that wheat and corn IVols are significantly higher on the day and the week when a ban is first imposed and also during the whole period when the ban is in effect. The effects of export bans are statistically and economically significant. The results hold even when we control for global macro-economic uncertainty and risk aversion (jointly proxied by the equity VIX) and for cash market tightness (including the state of grain inventories) prior to the ban.Intra-National Home Bias, Trade and Welfare: New Evidence from the Commodity Flows Survey
Abstract
We estimate the evolution of the US intra-national home bias by means of the Structural Gravity Model. We use data from the US Commodity Flows Survey from 1993 to 2017 focusing on the heterogeneity across states to determine whether home bias is driven by some states. Our estimates show that US intra-national home bias has increased since 2002. We use a General Equilibrium Poison Pseudo-Maximum Likelihood (GEPPML) estimator to obtain welfare effects due to the change in home bias. Our results show that the changes in GDP associated with the increase in the home bias vary significantly across States. Specifically, they suggest that smaller States would suffer significantly more as compared to larger States. This effect on GDP might be decomposed into effects on consumers (via inward multilateral resistances) and on producers (via factory-gate prices) for each of the States in the sample. Our findings points to the fact that the general equilibrium effects from the increase in the home bias on producers appear to be larger than the general equilibrium effects on consumers.Discussant(s)
Matthew Kidder
,
Newberry College
Florian Mayneris
,
University of Quebec-Montreal and CIRANO
Dalibor Stevanovic
,
University of Quebec-Montreal and CIRANO
Matthew Kidder
,
Newberry College
Matthew Kidder
,
Newberry College
JEL Classifications
- F4 - Macroeconomic Aspects of International Trade and Finance
- H0 - General