CER-ETH Research Seminar, Fall Term 2021

Everyone who is interested is cordially invited!

If you would like to receive our weekly invitation via e-mail, or if you have any other question, please contact AlenaMiftakhova.

Speakers

Peter Wakker

Title: Belief Hedges: Correcting for Subjective Beliefs when Measuring Ambiguity Attitudes even when Those Beliefs are Unknown

Abstract: Since Keynes (1921) and Knight (1921) we know that uncertainties usually do not come with probabilities (“ambiguity”). The first half of this lecture presents history, explaining the importance of ambiguity throughout but its popularity rising only since the 1990s. The second half solves a long-standing open problem:
To measure/apply ambiguity aversion, we must control for subjective beliefs. As yet, this could only be done for artificial events: Ellsberg-urns with secretized compositions or experimenter-specified probability intervals. It was unknown how to handle application-relevant events. We introduce belief hedges to solve this problem. That is, we combine uncertain beliefs so that they neutralize each other, whatever they are. In the same way as hedging in finance protects against uncertain returns. Now we can measure and apply ambiguity aversion to all events, greatly increasing the applicability of ambiguous beliefs.

Armon Rezai

Title: A Bayesian-inferred climate module for robust climate policy

Abstract: Large uncertainty remains about the effects of greenhouse gas emissions on Earth’s climate. We employ Bayesian methods to estimate a state-of-the-art climate module using historical data and simulations from complex Earth System Models. The resulting joint distribution of parameter values provide a co-variance matrix and constrains the choice of parameters in modelling. We sample one thousand realizations of the Earth system from this distribution and derive optimal policy by coupling the climate module with the economy of the DICE2016R model. Such a Monte Carlo approach is then contrasted with a robust decision making approach where climate policy (and all other decision variables) are forced to be uniform across states of the world. Even in the absence of risk aversion, carbon prices in this robust setting are significantly higher than the Monte Carlo ensemble’s mean and median. This suggests a strong insurance motive in choosing climate policy when the Earth system is uncertain. Our Bayesian approach also permits the computation of optimal policy for probabilistic temperature target, providing a natural conduit to concepts used in the atmospheric sciences.

Mar Reguant

Title: The distributional implications of real-time pricing

Abstract: While the benefits of Real-Time Pricing (RTP) of electricity are well known, less is known about their distributional impacts. We examine the distributional impacts of RTP by leveraging on a country-wide field experiment which started in 2015, when RTP became the default option for most Spanish households. Access to hourly consumption data during more than a year for over 4M households allows us to compute the bill impacts of the switch from flat rates to RTP. By examining the households' sociodemographic characteristics, we document who wins and who loses from RTP. We propose a two-step approach to infer consumers' unobserved income combining detailed household data with information of the distribution of income at the zip-code-level. Our results suggest that the distributional impacts of RTP were quite small and, once household income heterogeneity is accounted for, mostly progressive. We also find strong differences in the impacts across regions, even when controlling for income. Last, while households can significantly reduce their electricity bills by being more price responsive, the distributional impacts across income groups depend on the correlation between elasticity and income.

Kenneth Gillingham

Title: Equilibrium Trade in Automobiles

Abstract: We introduce a computationally tractable dynamic equilibrium model of automobile markets with heterogeneous consumers who choose to keep their car or trade for a different make, model and age. We focus on stationary flow equilibria, where outflows of cars due to accidents and endogenous scrappage equal inflows of new cars. We introduce a fast robust algorithm for computing equilibria and use it to estimate a model with eight household types and four car types using nearly 39 million observations on car ownership transitions from Denmark. The estimated model fits the data well and counterfactual simulations show that Denmark is over the top of the Laffer curve: it could raise total tax revenue by reducing the new car registration tax rate. We show that reducing this tax rate while raising the tax rate on fuel increases aggregate welfare, tax revenues, and car ownership, while reducing car ages, driving, and CO2 emissions.

Benjamin Moll

--CANCELLED--

Florian Brandl

Title: A Natural Adaptive Process for Collective Decision-Making

Abstract: Consider an urn filled with balls, each labelled with one of several possible collective decisions. Now, draw two balls from the urn, let a random voter pick her more preferred as the collective decision, relabel the losing ball with the collective decision, put both balls back into the urn, and repeat. In order to prevent the permanent disappearance of some types of balls, a randomly drawn ball is labelled with a random collective decision once in a while. We prove that the empirical distribution of collective decisions converges towards the outcome of a celebrated probabilistic voting rule proposed by Peter C. Fishburn (Rev. Econ. Stud., 51(4), 1984). The proposed procedure has analogues in nature recently studied in biology, physics, and chemistry. It is more flexible than traditional voting rules because it does not require a central authority, elicits very little information, and allows agents to arrive, leave, and change their preferences over time.

Kenneth Judd

Title:  The insurance perspective on climate change policy

Abstract: Much of the discussion of climate change policy focuses on consensus predictions of the future economy and climate, and admit only a modest amount of uncertainty in projections for the future climate, such as 2100 temperature. They also focus on the expected outcomes in, say, 2100. The damages from climate change are convex in the temperature, implying that we need to consider the range of plausible outcomes in 2100. Typical economic models ignore the very large uncertainty in economic growth. Adding this uncertainty to the DICE model, which we did in the DSICE model of Cai, Judd and Lontzek, shows that outcomes much warmer than typically considered have nontrivial probability. We buy fire insurance on our homes, not because we expect a fire in the near future but because we want to be prepared for an unlikely but very bad event. This is the approach that is done in DSICE and should be adopted much more widely.

Loic Berger

Title: Three layers of uncertainty and the role of model misspecification

Abstract: We explore decision-making under uncertainty using a framework that decom- poses uncertainty into three distinct layers: (1) risk, which entails inherent ran- domness within a given probability model; (2) model uncertainty, which entails subjective uncertainty about the probability model to be used; and (3) model mis- specification, which entails uncertainty about the presence of the correct probability model among the set of models considered. Using a new experimental design, we show that there exist distinct attitudes towards the three layers of uncertainty and examine the role of each of them in characterizing attitudes towards ambiguity. In addition to providing new insights into the underlying processes behind ambiguity aversion, we provide the first empirical evidence of the role of model misspecification in decision-making under uncertainty.

Harris Dellas

Title: Public Debt as Private Liquidity: Optimal Policy

Abstract: We study optimal policy in an economy in which public debt is used as collateral or liquidity buffer. Issuing more public debt raises welfare by easing the underlying financial friction; but this easing lowers the liquidity premium and increases the government’s cost of borrowing. These considerations, which are absent in the basic Ramsey paradigm, help pin down a unique, long-run level of public debt. They require a front-loaded tax response to government-spending shocks, instead of tax smoothing. And they explain why a financial recession, more than a traditional one, makes government borrowing cheaper, optimally supporting larger fiscal stimuli.

David Hemous

Title: Adverse selection as a policy instrument

Abstract: This paper applies principles of adverse selection to overcome obstacles that prevent the implementation of Pigouvian policies to internalize externalities. Focusing on negative externalities from production (such as pollution), we evaluate settings in which aggregate emissions are known, but individual contributions are unobserved by the government. We propose giving firms the option to pay a tax on their voluntarily and verifiably disclosed emissions, or pay an output tax based on the average rate of emissions among the undisclosed firms. The certification of relatively clean firms raises the output-based tax, setting off a process of unraveling in favor of disclosure. We derive sufficient statistics formulas to calculate the welfare of such a program relative to mandatory output or emissions taxes. We find that our mechanism would deliver significant gains over output-based taxation in two empirical applications: methane emissions from oil and gas fields, and carbon emissions from imported steel.