Constantine Yannelis
Assistant Professor of Finance
New York University
Leonard N. Stern School of Business
44 West 4th Street, KCM 9-98
New York, NY 10012
Research
Working Papers
Strategic Default on Student Loans
View Abstract
Student loans finance investments in human
capital. Incentive problems arising from lack of collateral in human
capital investments have been used to justify the differential
bankruptcy and other recovery treatment of student loans, despite a
lack of empirical evidence of strategic behavior on the part of student
borrowers. This paper uses policy induced variation in non-repayment
costs, that is unrelated to liquidity, to test for a strategic
component to the non-repayment decision. The removal of bankruptcy
protection and increases in wage garnishment reduce borrowers’
incentives to default, providing evidence for a strategic model of
non-repayment. The results suggest that rein- troducing bankruptcy
protection would increase loan default by 18%, and eliminating
administrative wage garnishment would increase default by 50%.
Consistent with strategic behavior on the part of borrowers, the
incentive effects of bankruptcy are larger for borrowers with large
balances, and smaller for very low and high income borrowers. The
results provide novel evidence that strategic behavior plays an
important role in student loan repayment.
Asymmetric Information in Student Loans
View Abstract
Student loans finance investments in human
capital, and adverse selection and hidden actions can lead to credit
limits and inefficient investment. This paper tests for and quantifies
information asymmetries in the student loan market. I first identify
selection parameters and the causal effect of increased borrowing on
loan non-repayment by using (i) non-linearities in borrowing rules for
federal student loans and (ii) changes in tuition for students already
enrolled. The paper finds that larger loan volumes induce default - a
40% increase in borrowing increases non-repayment by 10%. Adverse
selection effects are large in the 1980s when default costs were low,
and selection is ad- vantageous in the 2000s following increases in
non-repayment costs. Larger loan volumes inducing loan non-repayment is
consistent with both strategic and non-strategic models of default. The
pa- per uses policy induced variation in non-repayment costs to test
for a strategic component in the repayment decision. The removal of
bankruptcy protection and increases in wage garnishment re- duce
default without affecting pre-default liquidity, providing evidence for
strategic non-repayment. Reintroducing student loan bankruptcy
discharge would cost approximately $90 billion. Estimates from a
decomposition indicate that hidden actions resulting from increased
borrowing can explain approximately one quarter of the increase in
default in recent years, while most of the rest can be explained by
changes in the selection of borrowers and the institution type attended.
Debt and Human Capital: Evidence from Student Loans
View Abstract
This paper investigates the dynamic
relation between debt and investments in human capital. We document a
negative causal effect of the level of undergraduate student debt on
the probability of enrolling in a graduate degree for a random sample
of the universe of federal student loan borrowers in the US. We compare
students (i) within school and cohort, and (ii) across cohorts within
the same school at the time of a large tuition change. The latter
strategy exploits the fact that students who face a tuition increase in
earlier grades end up with significantly more debt than students who
face the same tuition increase in later grades. We find that $4,000 in
higher debt causes a two percentage point reduction in the probability
of enrolling in graduate school relative to a mean of 12%. Further
results suggest this effect is largely driven by credit constraints, is
monotonically weaker with family income, and is attenuated for students
who had compulsory personal finance training in high school. The
results highlight an important trade off associated with debt-financing
of human capital, and inform the debate on the effects of the large and
increasing stock of student debt in the US.
Does Climate Change Affect Real Estate Prices? Only If You Believe In It
View Abstract
Completion of registered report in progress.
A Day Late and a Dollar Short: Liquidity and Household Formation among Student Borrowers
View Abstract
The federal government encourages human
capital investment through lending and grant programs, but resources
from these programs may also finance non-education activities for
students whose liquidity is otherwise restricted. This paper explores
this possibility, using administrative data for the universe of federal
student loan borrowers linked to tax records. We examine the effects of
a sharp discontinuity in program limits—generated by the timing of a
student borrower’s 24th birthday—on household formation early in the
lifecycle. After demonstrating that this discontinuity induces a jump
in federal support, we estimate an immediate and persistent increase in
homeownership, with larger effects among those most financially
constrained. In the first year, borrowers with higher limits also earn
less but are more likely to save; however, there are no differences in
subsequent years. Finally, effects on marriage and fertility lag
homeownership. Altogether, the results appear to be driven by liquidity
rather than human capital or wealth effects.
The Consequences of Student Loan Credit Expansions: Evidence from Three Decades of Default Cycles
View Abstract
This paper studies the link between credit
expansion and student loan repayment using administrative federal
student loan data. We demonstrate that changes in federal student loan
credit availability to high-default institutions explain most of the
historical time series variation in defaults. Between 1981 and 1988,
eligibility for federal loans was expanded, leading to the entry of
institutions with borrowers more likely to default. From 1988 to 1992,
credit access was tightened, leading to the exit of many institutions
with high default rates. We exploit policy rules to show that changes
in credit availability led to high-default school exiting.
When Investor Incentives and Consumer Interests Diverge: Private Equity in Higher Education
View Abstract
This paper studies the effect of private
equity buyouts in the for-profit postsecondary ed- ucation sector.
Employing novel data on 88 private equity deals and 994 schools with
private equity ownership, we find that private equity buyouts lead to
higher enrollment and profits, but also to lower education inputs,
lower graduation rates, higher tuition, higher per-student debt, lower
student loan repayment rates, and lower earnings among graduates.
Neither selection ability of the private equity firms nor changes to
the student body composition seem to explain our results. An important
mechanism for the effects we observe is that private-equity owned
schools are better able to capture government aid.
Managerial Ownership and Firm Performance: Evidence From the 2003 Dividend Tax Cut
View Abstract
We
identify the causal relationship between managerial ownership and firm
performance using the 2003 Dividend Tax Cut as a quasi-natural
experiment, which increased the net-of- tax executive managerial
ownership. Consistent with predictions from agency theory, we and a
significant and hump-shaped improvement in firm performance measured by
Tobin's Q due to the tax cut. Moreover, the relation between managerial
ownership and firm performance relies on the strength of other
co-existing channels of corporate governance. In particular, only firms
with weak internal corporate governance proxied by G-Index or E-Index
are significantly affected in a hump-shaped manner by the tax cut. At
the same time, firms with weak institutional governance increase their
performance significantly more. These findings strengthen our
interpretation that the improvement in firm performance is due to
increased managerial incentive. Our results are robust to examination
of pre-trend and placebo tests, accounting measures of firm performance
as well as other considerations.
Greener on the Other Side: Estimating Consumer Sensitivity to Local Sales Tax Changes
View Abstract
We estimate the sensitivity of retail sales
to sales taxes using policy discontinuities in sales tax rates at state
borders using novel national retail sales data. We estimate the border
substitution effect of sales taxation and the price elasticity of
demand for traditional and internet retailers. Our approach generalizes
the case study approach by exploiting a new nationally representative
data set and all sales tax changes between 2008 and 2010. We estimate a
price elasticity of 5.9 in ZIP codes on state borders, and a price
elasticity of 1.7 for internet retail. We find that the effect of sales
taxation on retail is largest in geographic areas close to a state
border, and sales taxes in neighboring states have no effect on retail
activity in geographic areas more than 75 miles away from a state
border. The effect is largest for electronics, appliances, and
clothing. We find no effect for goods that are not taxed such as
services and food.
The Minimum Wage and Employment Dynamics: Evidence from an Age Based Reform in Greece
View Abstract
This paper analyzes minimum wage employment
effects and dynamics using an age-based reform in Greece. In 2012 the
minimum wage was reduced differentially for workers above and below age
25. Administrative labor force data, graphical evidence and a
difference-in-difference approach are used to identify a negative
effect on employment. The bulk of this effect is substitution from
older workers to younger workers, however there is also a significant
disemployment effect. Identifying elasticities of substitution and
comparing younger to older workers yields an employment elasticity of
between -.28 and -.46 for young workers. The result is present in
several outcome variables including unemployment, full time employment,
and hours worked, and is robust to a number of specification and
falsification tests. The employment effects of the reform operate
primarily through a reduced rate of new hires for workers with a higher
minimum wage, with no effect on job destructions. Consistent with the
predictions of the jobs ladder model, the group subject to the higher
wage is less likely to transition to another job, or seek employment
while on the job.
Published or Forthcoming Papers
The Rise in Student Loan Defaults in the Great Recession
View Abstract
We examine the rise in student loan defa
ults in the Great Recession by linking administrative student loan data
at the individual borrower level to student loan borrowers’ individual
tax records. A Blinder-Oaxaca style decomposition shows that shifts in
the composition of student loan borrowers and the massive collapse in
home prices during the Great Recession can each account for
approximately 30 percent of the rise in student loan defaults. Falling
home prices affect student loan defaults by impairing individuals’
labor earnings, especially for low income jobs. By contrast, when
comparing the default sensitivities of home owners and renters, we find
no evidence that falling home prices aff ect student loan defaults
through a home equity based liquidity channel. The Income Based
Repayment (IBR) program introduced by the federal government in the
wake of the Great Recession reduced both student loan defaults and
their sensitivity to home price fluctuations, thus providing student
loan borrowers with valuable insurance against negative shocks.
Income Changes and Consumption: Evidence from the 2013 Federal Government Shutdown
View Abstract
We use the 2013 federal government shutdown
and a rich data set from an online personal finance website to study
the effects of changes in income on changes in consumption. The 2013
shutdown was a significant and unanticipated income shock for federal
government workers, with no direct effect on permanent income. We
exploit both the differences between unaffected state employees and
affected federal employees as well as between federal employees
required to remain at work and those required to stay at home to
generate variation in income and leisure time. We find strong evidence
for excess sensitivity of consumption patterns, violating the permanent
income hypothesis. The decline in spending can be largely explained by
increased home production, changes in spending allocations, and credit
constraints. We are able to discern detailed categories of household
spending with widely varying elasticities. The results demonstrate the
importance of household liquidity, leisure and home production when
constructing stimulus or social insurance policy.
Quantifying the Premium Externality of the Uninsured
View Abstract
In insurance
markets, the uninsured can generate a negative externality on the
insured, leading insurance companies to charge higher premia. Using a
novel panel data set and a staggered policy change that introduces
exogenous variation in the rate of uninsured drivers at the county
level in California, we find that uninsured drivers lead to higher
insurance premia: a 1 percentage point increase in the rate of
uninsured drivers raises premia by roughly 1%. We calculate the
monetary fine on the uninsured that would fully internalize the
externality and conclude that actual fines in most US states are
inefficiently low.
Credit Constraints and Demand for Higher Education: Evidence from Financial Deregulation
View Abstract
This paper uses staggered bank branching
deregulation across states in the United States to examine the impact
of the resulting increase in the supply of credit on college enrollment
from the 70s to early 90s. A significant advantage of our research
design is that it produces estimates that are not confounded by wealth
effects. We find that lifting branching restrictions raises college
enrollment by about 2 percentage points (4%). Our results rule out
alternative interpretations to the credit constraints channel. First,
the effects are largest for low and middle income families, while
insignificant for upper income families as well as bankrupt families
who would have been unaffected by the increased access to private
credit. Second, the effect of lifting branching restrictions subsided
immediately following periods of increased loan limit through
government student loan programs. We also show that household
educational borrowing increased as a result of lifting branching
restrictions. Our results provide novel evidence that credit
constraints play an important role in determining household college
enrollment decisions in the United States.
A Crisis in Student Loans? How Changes in the Characteristics of Borrowers and in the Institutions they Attended Contributed to Rising Loan Defaults
View Abstract
This paper examines the rise in student
loan delinquency and default drawing on a unique set of administrative
data on federal student borrowing, matched to earnings records from
de-identified tax records. Most of the increase in default is
associated with the rise in the number of borrowers at for-profit
schools and, to a lesser extent, 2-year institutions and certain other
non-selective institutions, whose students historically composed only a
small share of borrowers. These non-traditional borrowers were drawn
from lower income families, attended institutions with relatively weak
educational outcomes, and experienced poor labor market outcomes after
leaving school. In contrast, default rates among borrowers attending
most 4-year public and non-profit private institutions and graduate
borrowers—borrowers who represent the vast majority of the federal loan
portfolio—have remained low, despite the severe recession and their
relatively high loan balances. Their higher earnings, low rates of
unemployment, and greater family resources appear to have enabled them
to avoid adverse loan outcomes even during times of hardship.
Decomposition analysis indicates that changes in characteristics of
borrowers and the institutions they attended are associated with much
of the doubling in default rates between 2000 and 2011. Changes in the
type of schools attended, debt burdens, and labor market outcomes of
non-traditional borrowers at for-profit and 2-year colleges explain the
largest share.
Indiscriminate Discrimination: A Correspondence Test for Ethnic Homophily in the Chicago Labor Market
View Abstract
Numerous field experiments have
demonstrated the existence of discrimination in labor markets against
specific minority groups. This paper uses a correspondence test to
determine whether this discrimination is due to prejudice against
specific groups, or a general preference for the majority group. Three
groups of identical fabricated resumes are sent to help-wanted
advertisements in Chicago newspapers: one with Anglo-Saxon names, one
with African-American names, and one with fictitious foreign names
whose ethnic origin is unidentifiable to most Americans. Resumes with
Anglo-Saxon names generate nearly one third more call-backs than
identical resumes with non Anglo-Saxon ones, either African-American or
Foreign. We take this as evidence that discriminatory behavior is part
of a larger pattern of unequal treatment of any member of non-majority
groups, ethnic homophily.
.