"Under the Lender's Looking Glass" Journal of Real Estate Finance and Economics (2016) doi:10.1007/s11146-016-9561-4
Abstract: This paper studies the impact of bank monitoring on the risk of US equity REITs. Using a unique, hand-collected data sample of mortgage balances, I show that bank screening and monitoring of REIT assets via utilizing secured mortgage financing (vs unsecured, public debt) lowers the overall company risk of a REIT. At the asset level, screening results in retail and office assets with higher acquisition values and located in primary markets, i.e., more transparent assets, being pledged as collateral. Further, I find evidence consistent with the role of lender monitoring for secured mortgage loans and show that properties located in closer proximity to a REIT’s headquarters are more likely to be pledged as collateral for a mortgage.
"Calendar Anomalies: Do REITs Behave Like Stocks?" (with Su Han Chan and Mehmet Akbulut) International Real Estate Review, 2015 Vol. 18 No. 1: pp. 177 – 215
Abstract: This study addressees the unsettled question of whether REITs behave similarly to stocks with respect to calendar anomalies. We determine the magnitudes of several types of calendar anomalies for both the REIT and stock market index returns in 22 countries between 1990 and 2012. In general, our evidence shows that the calendar effects are not universal across countries and that REITs behave differently from their stock counterparts in a number of countries. The difference in behavior is especially evident around the turn of the month where REITs exhibit significantly higher returns than general stocks in the global sample as well as in 7 of the 22 countries examined. This result may possibly be linked to the higher level of institutional ownership in REITs than in stocks in their corresponding market during the period examined.
“Decomposition of Debt and the Road to REIT Returns” (with Linda Allen) Revise and Resubmit, Journal of Real Estate Research
Abstract: Analyzing a hand-collected loan level database of heterogeneous REIT borrowings and controlling for REIT risk and loan collateral as well as endogeneity of access to public debt markets, we fi.nd that mortgage loans include a rate premium to compensate banks for monitoring. Access to public debt markets raises financing costs, inconsistent with a bank hold-up problem for REITs. However, non-monitored debt financing costs are reduced when bank lenders and equity analysts monitor REIT management. Equity alpha reflects positive abnormal returns from equity analyst monitoring, but not from costly bank monitoring. Equity gains from analyst monitoring are not found during recessions.
"Housing Location Choice in a Non-monocentric City" (with Hyoung Suk Shim) Revise and Resubmit, Journal of Regional Science
Abstract: Our study explores housing location choice and distance to work in both a theoretical and empirical frame work. We infer two theoretical predictions: i) income and distance to work have a positive relationship (consistent with Alonso(1964)) and ii)leisure time and distance to work have a negative relationship. We use the 2009 national household travel survey (NHTS) to test our predictions empirically and the findings support our predictions. In addition to supporting our predictions, we are able to conclude that female heads of household prefer housing closer to work as do higher educated, older workers, people that own their home and those that use public transportation. We utilize an instrumental variable approach to address the endogeneity of unobserved amenities and find that the results are robust. Finally we consider a sub-sample of high amenity cities and find that the magnitude of income coefficients is still positive and greater in magnitude, which could be explained by households gravitating to amenities and thus locating further from work. In the high amenity city, spousal education is found to be a positive and highly significant factor in locating a greater distance from work, whereas it is insignificant in the remainder of the sample.
"Agree to Disagree: NAV Dispersion in REITs" (with Stace Sirmans and Stacy Sirmans) Submitted, Journal of Real Estate Finance and Economics
This is the first study to analyze REIT Net Asset Value analyst coverage and dispersion. We find that NAV analyst coverage has a positive relationship with REIT value and a negative relationship with REIT volatility. Subsequently we analyze NAV analyst estimate dispersion and find that it has a positive relationship with REIT leverage and volatility. We break down our sample by property type and find that retail REITs have the greatest NAV coverage and hospitality REITs have the greatest NAV analyst dispersion. Finally, we compare the significance of NAV forecast dispersion to earnings (FFO) forecast dispersion. We find that NAV dispersion has a significant negative relationship with REIT value whereas FFO dispersion is not found to have a significant relationship.
"Better the devil you know: reluctance to refinance expensive loans" (with Meagan McCollum)
Abstract: Prior research has demonstrated that individual borrowers hold on to sub-optimal mortgages. That is, when interest rates decline so that refinancing the mortgage would generate substantial savings in interest, borrowers choose to keep their existing, expensive mortgages. Our study utilizes the closed form solution put forth by Agarwal et al.  to identify the resistant borrowers and puts forth the potential drivers of their behavior, using empirical data on over half of all residential mortgages in United States.
Presented at American Real Estate Society 2016 Meetings in Denver, CO