Research

My research interests lie in the intersection of macroeconomics and financial economics.  I am particularly interestead in the use of experimental methods to answer questions related to the macroeconomic aspects of finacial markets. 

I am also interested in research on economic education, and improving student outcomes in principles of economics.

Publications

In-Person versus Online Instruction: Evidence from Principles of Economics, with Kenneth Elzinga, Southern Economic Journal, 2023, 90, 3-30.

COVID-19 required many professors to switch from in-person teaching to online instruction, allowing exploration of a pivotal question in education: are learning outcomes better when instruction takes place in-person or online?  We compare student performance across two semesters of the same large introductory economics course – one taught in-person in 2019, the other taught online in 2020.  We analyze test scores from over 2,000 students for exam questions common to both instructional formats.  At the aggregate level, we find no difference in student performance between online and in-person instruction. When dividing questions by required reasoning skills, we find that online instruction improves student performance on questions requiring knowledge of a definition or formula.  Additionally, student course evaluations rated the online course over in-person pedagogy.  

Capital Constraints and Asset Bubbles: An Experimental Study, with Lee Coppock and Charles Holt, Journal of Economic Behavior and Organization, 2021, 183, 75-88.

This paper evaluates the effects of credit constraints in an asset market experiment with present value considerations induced by interest payments on cash.  All markets exhibit price bubbles, with peak prices exceeding the present value of dividends and redemptions by 30-130%.  Starting with a baseline condition (low income, tight credit), a relaxation of credit constraints generates significantly higher price bubbles.  A price increase of similar magnitude results from an increase in exogenous income, holding credit tightness constant.

Working Papers

Tobin’s Q, Liquidity, and Speculation in Laboratory Markets, with Charles Holt

This paper uses a laboratory experiment to study channels through which excess liquidity and Tobin’s q affect capital investment.  Subjects produce and trade capital goods in a multiperiod market in which the capital stock depreciates and is subject to convex production costs.  Treatments vary the marginal cost of investment, the aggregate cash level, and the relative cash endowments of individual subjects, holding aggregate cash constant. Increasing aggregate cash increases price bubbles for capital goods.  In contrast to q-theory predictions of no effect, there is excess production of capital goods relative to a first-best optimum in markets with high cash endowments.  The degree of overproduction is positively correlated with the size of price bubbles. (Online Appendix)

Persistent Private Information in Experimental Asset Markets, with Charles Holt and Margaret Isaacson

This study evaluates the extent to which laboratory markets disseminate private information about the long-run payout of durable assets, and how persistent private information affects traders’ beliefs about future asset prices. Subjects trade dividend-paying assets, which are redeemed for a randomly determined value that is revealed in advance to “insiders.” The efficient market hypothesis predicts prices will incorporate insider information, which precludes bubbles. Markets exhibit price bubbles in all treatments, with magnitudes that are uncorrelated with the proportion of insiders. But these price surges do not permit insiders to earn more than outsiders on average. Price predictions are elicited, and insiders tend to make lower predictions. Using adaptive expectations models, insiders are less responsive to previous prediction errors in forming their price predictions. (Online Appendix)

Macroprudential Taxes, Ratios, and Rents: Equivalences and Not-Quite Equivalences, with Anton Korinek

This paper shows that most modern macroprudential instruments are designed so that the surplus from regulation accrues to the financial sector. We develop a model to compare the instruments that macroprudential policymakers have at their disposal to delineate which types of measures are equivalent, which ones are not, and how they differ. One crucial distinction is that the distribution of the surplus created by macroprudential instruments across borrowers and lenders depends on who is subject to the regulation, opening an important political dimension to the choice of instrument. This may lead to inefficient regulations such as second-best regulatory ratios that create distortions but keep the surplus with the regulated entities. By contrast, we establish equivalence conditions for taxes, ratios and quantity instruments under both borrower and lender heterogeneity based on a transformation-of-measure approach. Moreover, we show under what conditions restrictions from financial constraints and market power are equivalent to macroprudential measures. Finally, we show that our results extend to a triple-decker setting with an intermediation chain from savers to banks to borrowers.

Works in Progress

Financial Amplification: An Experiment

I develop an experimental methodology to test the theory of financial amplification in the lab. I develop a theoretical model designed for implementation in a laboratory environment.  When borrowing constraints depend on asset prices financial amplification occurs.  The model also allows for borrowing constraints that allow an exogenous amount of borrowing per asset held.  When these borrowing constraints are used financial amplification does not occur in the model.  I show that exogenous constraints can be calibrated to result in an equilibrium equivalent to the price-linked constraint equilibrium.  When liquidity shocks increase debt levels from this equilibrium, prices decrease more under price-linked constraints than exogenous constraints.  This model is implemented in an experimental design with two treatments, one for each borrowing constraint.  In each session subjects participate in the model multiple times.  In some rounds participants experience liquidity shocks.  Financial amplification is identified in the lab by comparing asset prices and allocations across the borrowing treatments for high and low liquidity shock rounds.