by Leslie Papke, Professor and Associate Chair
Our professors are active researchers in both economic theory -- modeling the workings of our economic system -- and in empirical economics -- estimating the economic effects of world events like COVID-19 and climate change, individual choices, and government programs. Some of this work is done with each other! Here are brief summaries of a few of these articles published in the past year or so.
"Relationship Finance, Informed Liquidity, and Monetary Policy" in Journal of Economic Theory (Luis De Araujo and Raoul Minetti)
"It’s not you -- it’s me." Economic entities form relationships and have breakups too! When corporations want to expand their operations or invest in research and development, they may borrow from the financial sector often forming credit relationships. In recent decades these relationships have been weakened by mounting competition among financial institutions and by increasing geographical distance between banks and firms. In addition, the growing ability of financial institutions to re-sell these loans to others may have diluted their incentives to form relationships with firms. And on the firm side, corporations have accumulated large cash reserves so they are also less likely to seek out a credit relationship.
Luis Araujo and Raoul Minetti, along with their coauthor Pierluigi Murro, explore what these relationships mean for the overall level of economic activity. Does the relationship structure of the credit sector influence the efficiency of economic outcomes? Does it magnify or reduce the consequences of changes in economic conditions and effects of monetary policy? In the aftermath of the Great Recession, scholars and policy makers have debated whether this observed trend should be allowed to continue or if efforts should be made to preserve a more relationship-oriented financial system.
Their theoretical approach postulates a complementarity between banks’ desire to make loans and their effort in understanding firms’ projects -- the larger a bank’s investments the stronger its incentive to acquire project expertise and provide support in case of financial distress. The three economists find opposing economic effects of these credit relationships – they can lead to overinvestment in these firms. On the other hand, they can improve resilience to economic conditions as long as the state of the economy is not too poor. They conclude that in certain situations, policy makers should consider providing funds to banks, and in other situations, it is welfare-improving to provide the liquidity to the firms directly.
"School Segregation and Racial Gaps in Special Education Identification," in Journal of Labor Economics (Todd Elder and Scott Imberman)
Racial differences in economic and educational outcomes persist in the United States. Black economic mobility lags behind that of their White peers, and large Black-White wage gaps have long been documented. Some of these differences in adulthood can be traced to early-life educational attainment and achievement gaps. Todd Elder and Scott Imberman, along with their co-authors David Figlio and Claudia Persico, focus on a potential driver of racial gaps in adulthood -- special education identification. About 6.4 million public school students in the U.S. receive services at an annual cost of almost $40 billion. Special education provides a way for students to receive accommodations and possibly treatment for learning disabilities. So, early and accurate identification is important for adult economic well-being.
The authors examine the extent to which minority students are disproportionately placed in special education, and, in particular, how racial selection gaps vary by the racial composition of schools. How do peers and the larger social environment affect who is identified? A novel feature of their work is their data -- they use a rich data source that links education records to birth certificate records for every child born in Florida between 1992 and 2002. Importantly, birth records allow them to condition on initial health endowments and baseline demographics -- previous work has been unable to do this. They are also able to disaggregate identification into specific disability categories. Apart from school differences in resources, if special education identification is a function of one’s health relative to one’s peers, and if minorities have relatively poor health endowments, then school racial composition can lead to under-identification of minorities in heavily minority schools, and over-representation in heavily white schools.
Albert Einstein found that everything is relative, and the authors confirm that this is also the case for special education identification. Focusing on kindergarten and fourth grade, their estimates suggest that Black and Hispanic students in schools with relatively few minorities tend to be over-represented in special education relative to their predicted rates. Further, minority students in heavily minority schools are under-represented in special education relative to their underlying incidence of disability. If so, their long-term educational and economic outcomes could suffer.
"The Impact of the Affordable Care Act: Evidence from California’s Hospital Sector," in American Economic Journal: Economic Policy (Emilie Jackson)
The main provisions of the 2010 Patient Protection and Affordable Care Act (ACA) took effect in January of 2014. The ACA led to the largest expansion of publicly-funded health insurance since Medicaid and Medicare were introduced more than 50 years ago. For states who chose to participate, means-tested Medicaid insurance was expanded to uncovered individuals with family incomes at or below 138 percent of the federal poverty level (FPL), and those with more income but still less than 400 percent of the FPL could sign up for subsidized insurance.
Emilie Jackson, along with coauthors Mark Duggan and Atul Gupta, examine the initial effects of this insurance expansion using California hospital data. They compare data on the universe of hospital stays and emergency room visits combined with data on hospital finances from 2008 to 2016 -- the period before and after the expansion. More than 5 million Californians gained access to Medicaid over this period and an additional 1.4 million received subsidized insurance through the state’s ACA insurance exchange.
The authors note there is a pre-existing bump up in health insurance coverage at age 65 because nearly all individuals become eligible for Medicare at that age. The Medicaid expansion and publicly-subsidized private insurance through the ACA exchange sharply reduced this discontinuity in health insurance at age 65. They quantify the effects on hospital finances following the expansion by exploiting pre-ACA variation in the share of uninsured patients across hospitals. This allows them to estimate causal effects that are most meaningful for the near-elderly population.
They estimate that many of the newly insured patients replaced previously uninsured patients -- some of whom would have been subsidized by county-run safety net programs. Since the expansion was entirely financed by the federal government during this period, this implies a transfer to California taxpayers who previously financed the safety net programs. They find a net increase of 4-5 percent on average in hospital stays and arrivals at the emergency room, and a substantial decrease in out-of-pocket costs paid by the previously uninsured patients. And since Medicaid reimbursement rates are higher than payment from uninsured patients, California hospitals accumulated reserves in the short run. They also find a meaningful decline in in-hospital mortality but this finding is tentative since it is imprecisely measured.
"Do People Maximize Quantiles?" in Games and Economic Behavior (Antonio Galvao)
In making choices under uncertainty, economists have typically assumed that individuals maximize expected utility (EU) -- that is, the mathematical model used to formalize preferences assumes agents maximize their average utility across uncertain outcomes. But economists have noted that this sort of behavior is not commonly observed in small-stakes laboratory experiments. And, as Antonio Galvao and his co-authors Luciano de Castro, Charles Noussair, and Liang Qiao argue, it is not obvious why a decision maker would maximize an average of a function of payoffs. Instead, they may choose to maximize a point in the distribution that they most care about -- to maximize a quantile of their payoff distribution. For example, an individual may attempt to maximize the median payoff -- they maximize the payoff that they have at most a 50 percent probability of failing to attain or at least a 50 percent probability of attaining. This concept is common in oil and gas extraction, for example, where firms attempt to maximize a low quantile -- the P90, P50, and P10 indicators of a well correspond to the levels of production that are exceeded with 90%, 50% (median), and 10% probability.
Quantile preferences (QP) have several attractive features -- it is independent of one’s utility function so an optimal choice is relatively easy to compute. Yet, QP has received little attention in decision science. This paper is the first to estimate how common this method of decision-making is in any population! The paper describes a two-part experiment that allows the authors to uncover individuals’ intended choice rules. They find that a sizable percentage of individuals -- between 32% and 55% (depending on the test) are QP maximizers. The average estimated quantile that is being maximized is 0.42 -- about 80% of individuals have quantile estimates between 0.3 and 0.6. Most subjects are found to be risk averse and women are found to be more risk averse than men – that is, women tend to maximize significantly lower quantiles than men on average. These findings illustrate the real-life importance of QP and the authors suggest continued research into the properties of QP and their implications for economic and financial decision-making.