Published Papers
"Corruption and Growth Under Weak Identification," with Marina-Selini Katsaiti and Marius Jurgilas, forthcoming in Economic Inquiry
Abstract: The goal of this paper is to revisit the influential work of Mauro (1995) focusing on the strength of his results under weak identification. He finds a negative impact of corruption on investment and economic growth that appears to be robust to endogeneity when using two-stage least squares (2SLS). Since the inception of Mauro (1995), much literature has focused on 2SLS methods revealing the dangers of estimation and thus inference under weak identification. We reproduce the original results of Mauro (1995) with a high level of confidence and show that the instrument used in the original work is in fact "weak" as defined by Staiger and Stock (1997). Thus we update the analysis using a test statistic robust to weak instruments. Our results suggest that under Mauro's original model there is a high probability that the parameters of interest are locally almost unidentified in multivariate specifications. To address this problem, we also investigate other instruments commonly used in the corruption literature and obtain similar results. Download here!
Working Papers
"Educational Corruption and
Growth"
Abstract: Educational corruption is a worldwide
phenomenon yet its impact on economic growth is unknown. In the paper we
formulate a macroeconomic model to explore the impact educational corruption
may have on growth, educational attainment, and the education wage premium.
We find that our model can produce a negative relationship between economic
growth and educational corruption as well as a positive relationship between
the education wage premium and educational corruption. We find strong
empirical support for these relationships in a cross-section of countries.
Our model also produces a negative relationship between the level of
educational attainment and educational corruption as found in a
cross-section of countries. To add to a recent line of literature on social
status and its implications for growth, we also show that attaching status
to education can be growth enhancing in countries that experience low to
medium levels of educational corruption but growth reducing in countries
that experience high levels of educational corruption. We also show that
borrowing constraints can exacerbate the impact educational corruption has
on economic growth, wage inequality, and educational attainment rates.
Download here!
"Endogenous
IVs, The Death of the t-stat, and More"
Abstract: In this paper we offer a naive empirically
corrected t-stat in an attempt obtain robust inference for a commonly used
statistic in the presence of weak instruments. When examining the empirical
t-stat’s performance we find it less than desirable when compared to three
test statistics robust to weak instruments including the AR, LM, CLR
statistics. We also examine some commonly used IV methods under weak
identification and carefully examine their properties with and without the
restriction that the IVs are exogenous in the structural equation. We show
that for point estimation the LIML estimator performs optimally in terms of
minimum median bias especially when the researcher has a large number of
exogenous IVs available to them. We also show that although the power of the
Sargan statistic is very low across small samples, it dominates the AR
statistic as a test for overidentification restrictions. After allowing for
endogenous IVs we show that this can have serious implications for
hypothesis testing even when using “robust” statistics such as the AR, LM,
and CLR test. Endogenous IVs create serious size distortions for a typically
used endogeneity test. This result is particulary problematic because of the
inability of the Sargan statistic to pick up these endogenous IVs even at
large deviations from the null.
Download here!
"The Determinants of Educational Corruption: The Case of Ukraine"
Abstract: Educational corruption is a relatively new area of study in economics due mainly to the lack of available data. This paper utilizes a unique data set to examine educational corruption of various types including bribing on exams and term papers as well bribing to obtain credits and bribing to enter institutions. The data was gathered from 1588 students attending educational institutions throughout Ukraine. The paper attempts to identify the determinants of bribery and corruption perceptions across various institutions and cities throughout Ukraine. The results reinforce the importance of corruption perceptions and the relationship they have with actual bribing behavior. The paper concludes that women tend to have a higher probability of bribing on exams and for entrance after controlling for job market perceptions. The results also suggest that students whose fathers are in agriculture have a higher probability of bribing when compared to students with fathers in the private or entrepreneurial sector. Bribing during secondary school is also a strong predictor of bribing at tertiary school across all forms of bribery.
"Nonparametric Instrumental Variable Estimation in Practice," with Mike Cohen and Tao Chen
Abstract: In this paper we examine the finite sample performance of two estimators one developed by Blundell, Chen, and Kristensen (2007) (BCK) and the other by Gagliardini and Scaillet (2007) (TIR). This paper focuses on the generalization and expansion of these estimators to a full nonparametric specification with multiple regressors. In relation to the classic weak instruments literature, we provide intuition on the examination of instruments relevance when the structural function is assumed to be unknown. Simulations indicate that both estimators perform quite well in higher dimensions. This research also provides insights on the performance of bootstrapped confidence intervals for both estimators. We document that the BCK estimator's coverage probabilities are near their nominal levels even in small samples as long as the sieve order of expansion is restricted. The coverage probability for the TIR estimator's bootstrapped confidence intervals are near their nominal levels even when the order of sieve approximation is large. These results suggest that in small samples the TIR estimator has a much smaller bias then the BCK estimator but its variance is much larger. We provide two empirical examples. One is the classic wage returns to education example and the other looks at the relationship of corruption and economic growth. Results here suggests that the impact of corruption on growth depends nonlinearly on a countries level of development. We also discuss how to estimate the partial effects using a consistent estimator that only relies upon a consistent estimator of the unknown function. Download here!
"A
Conditional Moment Test for Endogenous Regressors,"
with Mike Cohen "Endogenous
Search Frictions in the Market for Small Business Loans,"
with Kenneth Peterson
| 
