How Do the CEO Political Leanings Affect REIT Business Decisions?

Abstract

Business decisions made by the real estate industry have a profound effect on the well-being of people who live, work, or shop in these buildings. While these decisions may be informed by evidence, the available evidence is often incomplete or imperfect. Therefore, the personal opinions or judgments of senior executives can have an effect. In this paper, we study these effects in two parts: risk-taking and environmental, social, and governance (ESG) activities. Since a person’s political learning is a relatively stable measure, and is associated with preferences for risk and ESG activities, we examine how the political leanings of the CEOs are related to these effects. Using the data from 2003 to 2016, we find that real estate investment trusts with Democratic-leaning CEOs tend to take more risks, as evidenced by higher levels of leverage and more risk in stock prices. We further find that Democratic-leaning CEOs are more broadly engaged in environmentally oriented ESG activities.

Keywords:

• Politics
• environmental, social, and governance activities
• corporate social responsibility
• risk taking
• decision making

JEL CODES:

• D21
• G32
• M14
• Q56

Acknowledgments

We appreciate valuable comments from our referees, Shawn Bond, Seow Eng Ong, Jun Qian, Xiaolei Liu, Liang Peng, Bill Hardin, Masaki Mori, and conference participants at the 2017 ARES Annual Meeting in San Diego; the 2017 CEA Annual Conference in Nova Scotia; the 2017 AsRES Annual Meeting in Taichung; the 2017 AREUEA International Conference in Amsterdam; the 9th ReCapNet Conference in Mannheim; the 2018 AREUEA Annual Meeting in Philadelphia, as well as seminar participants at Fudan University and Peking University. Each of the authors on this paper contributed equally.

Notes

1 https://www.opensecrets.org/overview/industries.php. This information excludes donations at the state or city level and does not include expenses associated with lobbyists.

2 We thank an anonymous referee for noting that these concerns are becoming so well recognized that the notion of responsibility or good intentions associated with the term CSR is being replaced by corporate reports emphasizing specific ESG metrics and activities.

3 In the same book, Sah and Miller (2018) offer a complementary review of US REITs while Parker (2018) reviews the experience of REITs globally.

4 Some readers may think that a natural experiment to test both of these hypotheses would be to consider a subset of REITs for which the CEO changed from one who leaned toward the Democratic Party to one who leaned toward the Republican Party. While this thought is reasonable, it is complicated by endogeneity in the process which selects the new leader.

5 Eichholtz and Yönder (2020) add to this idea by studying the prices paid by REITs when buying property. Using a measure derived from the exercise of stock options, they find that CEO optimism affects the price paid when buying properties and that the magnitude of this effect varies with the optimism of the CFO.

6 www.fec.gov

7 Some readers may wonder whether the large contributions to “Other Parties” revealed by Appendix 1 indicate that firms are trying to purchase political capital. Our data on donations comes from the Federal Election Commission, which does not collect data on donations to local politicians such as mayors or city councilors. The fact that nearly 60 percent of the REIT’s CEOs donate nothing to any political party, and that those who do donate tend to focus on one party only, also suggest that the relevance of this issue to our study is likely limited.

8 In a further robustness check, we also define DEM as 1 if over 80% of CEO contribution amount goes to Democrats, otherwise, DEM equals 0, which we look at the extreme. For both measures, the results are qualitatively similar, and are available upon request.

9 In a robustness check, we also adopt yearly and quarterly standard deviation of the market premium to measure the systematic risk.

10 The Flow of Funds from Operations (FFO) is often the preferred measure when analyzing returns. We choose not to use data on FFO because doing so would reduce our sample size substantially.

11 We use the sample from 2003 for regressions given that few REITs CEOs had made political contribution prior to 2003.

12 In the unreported analysis, we examine subsamples to measure the effect of a turnover in the State Governor and “purple” states (whose political leadership combines Democratic and Republican elements), which document a silent effect on REITs stock liquidity and risk taking, indicating that our results are purely driven by the political value not the effect of lobbying or rent seeking.

13 While the left wing vs. right wing distinction is not isolated to the U.S., identifying those beliefs in other countries is harder. In countries with more than two parties that may form a government, the intuition of the popular median voter model is not obviously relevant and, thus, a more intrusive research methodology may be necessary. Since competition for voters takes place on more dimensions, and government policies may be based on a coalition made up after the voting has finished, the beliefs associated with a party are not as stable. The United States political system also permits greater financial contributions from private individuals, unions or corporations, compared to most other countries and that observed behavior is less likely to be bound by administrative criteria. Government policies concerning financial support of political parties vary in other countries.

Additional information

Funding

Xiaoying Deng acknowledges the financial support of the National Natural Science Foundation of China (71703095).