{"title":"Impacts on REIT Stock Capital Structures, Equity Costs, and Market Liquidities of Being Included in ETF Managed Portfolios","authors":"Long Ma, Ronald W. Spahr, Mark A. Sunderman","doi":"10.1080/08965803.2023.2254039","DOIUrl":null,"url":null,"abstract":"AbstractIn recent years, especially as compared to mutual funds, exchange traded fund (ETF) markets have grown and advanced significantly compared to other financial asset classes because of relative advantages. We found that inclusion of real estate investment trusts (REITs) in ETF assets under management (AUM) positively impacts REIT capital structure (financial leverage), cost of equity capital, and stock market liquidity. As percentages of REIT outstanding shares included in ETF AUM increased, we found corresponding reductions in financial leverage (both book and market leverage), reduced costs of equity capital, and greater market liquidity. This should be of particular interest to REIT and ETF managers as well as REIT and ETF investors. Partially because regulatory statutes incentivize REITs to rely more heavily on external equity financing, REIT stocks included as ETF AUM showed greater reductions in leverage compared to non-REIT stocks also held as ETF AUM. Our results, including applying difference-in-differences models, were robust with respect to these findings, REIT type, and firm fixed effects.Keywords: REITETFAUMmarket liquiditycost of equity capitalfinancial leverage Disclosure StatementNo potential conflict of interest was reported by the authors.Notes1 We suggest that the reason REITs experience a significant increase in institutional ownership and in stock turnover on ETF inclusion as AUM is that ETF provide another and possibly stochastically superior way to own real estate assets and their associated advantages.2 The Investment Company Act of 1940 is an act of Congress that regulates investment funds, investment companies, and pass-through companies that include REITs and ETFs. It was passed as a United States Public Law (Pub.L. 76–768) on August 22, 1940, and is codified at 15 U.S.C. §§ 80a-1 – 80a-64. The act is enforced and regulated by the Securities and Exchange Commission (SEC), and defines the responsibilities and requirements of investment companies, including ETFs, and the requirements for any publicly traded investment product offerings such as open-end mutual funds, closed-end mutual funds, and unit investment trusts. The act primarily targets publicly traded retail investment products.3 Typically, ETFs hold assets in trust in their portfolios, technically not holding title to assets. ETFs are formed by an ETF manager (sponsor) filing a plan with the U.S. SEC to create an ETF. When approved, the sponsor forms an agreement with an authorized participant (AP), generally a market maker, specialist, or large institutional investor that is empowered to create and redeem ETF shares. Often, the AP and the sponsor are the same. The AP then borrows REIT stock shares from an institutional investor, often a pension fund, places those shares in a trust, and uses them to form ETF creation units (CU). CUs bundle stock, commonly 50,000 shares (one creation unit) of an ETF. Then, the trust provides fractionalized shares of the ETF (legal claims on the REIT shares held in the trust) to the authorized participant. Since these transactions are in-kind trades, securities traded for securities, there are no tax implications. When the authorized participant receives the ETF shares, they are sold in the open market and traded during trading hours, continuously, just like any stock shares. Thus, we used ETF AUM (assets under management) to denote an ETF’s control/management of shares.4 Financial leverage in our applications refers to the use of fixed income financing or debt in the capital structure. Financial leverage also may be affected by the use of leases and preferred stock or the use of any fixed income financing.5 Our sample of REITs held as ETF AUM includes both equity and mortgage REITs. Our rationale, verified by subsequent findings, for including both types is that from the REIT stockholders’, managers’, and ETF managers’ perspectives, benefits associated with REIT stocks being held as AUM in an ETF are essentially identical. We found that the type of assets held by a REIT is less important than benefits associated with REIT structures, associated tax advantages, and inherent advantages of ETFs.6 Table 1, “Sample Statistics” show a minimum value for ETF AUM of zero, indicating that our sample period included a time period before any REITs were included as ETF AUM. Also, an alternative methodology uses a dummy variable set to 1 after each REIT’s being first held by at least one ETF AUM, to examine the impact on each REITs’ leverage. However, results using dummies were not statistically significant; thus, these results are not reported.7 Other potential econometric model concerns include causality, inverse causality, endogeneity, and omitted variable biases. For example, omitted firm characters may affect both the capital structure and the likelihood of a REIT being included in ETF portfolios. Also, it is possible that ETFs select REITs for their portfolios that already have lower leverages. Other methodological concerns include controlling for omitted variable biases. We did this by including firm fixed effects to alleviate concerns regarding controlling for omitted variable biases; however, fixed effects coefficients were not statistically significant. We did not include industry fixed effects since all REITs belong to the same industry. We failed to find reasons to be overly worried about these potential econometric concerns being problematic relative to our results.8 The National Best Bid and Offer (NBBO) quote reported the highest bid price and lowest ask (offered) price in a security, sourced from among all available exchanges or trading venues. The NBBO, therefore, represents the tightest composite bid-ask spread in a security.9 This is similar to Titman and Wessels (Citation1988) and Lipson and Mortal (Citation2009) for non-REIT stocks.10 Fama and French (Citation2002) and Lipson and Mortal (Citation2009) also suggested using these control variables.11 We included both market value financial leverage (Dt/Vt) and book value financial leverage (Dt/At) in our analysis. Theoretically, market values should be more relevant since these values represent the current market structure of the firm; however, book values better represent the value of capital originally inputted in the firm.12 Again, the matched sample uses propensity score nearest neighbor matching with replacement based on total assets in previous year of ETF first inclusion. In addition, we matched treatment group firms with control firms by matching REITs of the same type (Equity REIT, Mortgage REIT, or Hybrid).13 This is similar to the approach used by Lin et al. (Citation2018).14 We ensured validity and consistency of the DID analysis by requiring that differences between the treatment and control group were consistent over time, controlling for parallel trends.15 These two measures are detailed in the Methodology, Data, Relevant Literature, and Preliminary Results section.16 Our conclusions regarding the longer windows may be somewhat tentative since longer windows may be affected by other endogenous and/or exogenous factors that may impact returns and volatility. However, homogeneous trends across all three windows were observed; thus, we are confident that our reported results are robust.17 Definitions of the three spread measures are detailed in the Methodology, Data, Relevant Literature, and Preliminary Results section.18 Table 7, Panels A, B, and C found consistent, statistically significant differences for all three market liquidity measures before and after each event date (month 0). Speadbefore¯ is the mean of spreads in periods (–12m, 0m), (–24m, 0m), and (–36, 0m) prior to month (0m), the event date and Speadafter¯ is the mean of spreads in periods (+1m, +12m), (+1m, +24m), and (+1m, +36m) subsequent to each event date. Δ Spread is the prior/postperiod differences in spreads, ranging from 23% to 46%, and was statistically negative in all cases, providing further evidence that stock market liquidity is increased subsequent to each REITs being held in ETF AUM.","PeriodicalId":51567,"journal":{"name":"Journal of Real Estate Research","volume":"25 1","pages":"0"},"PeriodicalIF":1.2000,"publicationDate":"2023-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Real Estate Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1080/08965803.2023.2254039","RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"BUSINESS, FINANCE","Score":null,"Total":0}
引用次数: 0
Abstract
AbstractIn recent years, especially as compared to mutual funds, exchange traded fund (ETF) markets have grown and advanced significantly compared to other financial asset classes because of relative advantages. We found that inclusion of real estate investment trusts (REITs) in ETF assets under management (AUM) positively impacts REIT capital structure (financial leverage), cost of equity capital, and stock market liquidity. As percentages of REIT outstanding shares included in ETF AUM increased, we found corresponding reductions in financial leverage (both book and market leverage), reduced costs of equity capital, and greater market liquidity. This should be of particular interest to REIT and ETF managers as well as REIT and ETF investors. Partially because regulatory statutes incentivize REITs to rely more heavily on external equity financing, REIT stocks included as ETF AUM showed greater reductions in leverage compared to non-REIT stocks also held as ETF AUM. Our results, including applying difference-in-differences models, were robust with respect to these findings, REIT type, and firm fixed effects.Keywords: REITETFAUMmarket liquiditycost of equity capitalfinancial leverage Disclosure StatementNo potential conflict of interest was reported by the authors.Notes1 We suggest that the reason REITs experience a significant increase in institutional ownership and in stock turnover on ETF inclusion as AUM is that ETF provide another and possibly stochastically superior way to own real estate assets and their associated advantages.2 The Investment Company Act of 1940 is an act of Congress that regulates investment funds, investment companies, and pass-through companies that include REITs and ETFs. It was passed as a United States Public Law (Pub.L. 76–768) on August 22, 1940, and is codified at 15 U.S.C. §§ 80a-1 – 80a-64. The act is enforced and regulated by the Securities and Exchange Commission (SEC), and defines the responsibilities and requirements of investment companies, including ETFs, and the requirements for any publicly traded investment product offerings such as open-end mutual funds, closed-end mutual funds, and unit investment trusts. The act primarily targets publicly traded retail investment products.3 Typically, ETFs hold assets in trust in their portfolios, technically not holding title to assets. ETFs are formed by an ETF manager (sponsor) filing a plan with the U.S. SEC to create an ETF. When approved, the sponsor forms an agreement with an authorized participant (AP), generally a market maker, specialist, or large institutional investor that is empowered to create and redeem ETF shares. Often, the AP and the sponsor are the same. The AP then borrows REIT stock shares from an institutional investor, often a pension fund, places those shares in a trust, and uses them to form ETF creation units (CU). CUs bundle stock, commonly 50,000 shares (one creation unit) of an ETF. Then, the trust provides fractionalized shares of the ETF (legal claims on the REIT shares held in the trust) to the authorized participant. Since these transactions are in-kind trades, securities traded for securities, there are no tax implications. When the authorized participant receives the ETF shares, they are sold in the open market and traded during trading hours, continuously, just like any stock shares. Thus, we used ETF AUM (assets under management) to denote an ETF’s control/management of shares.4 Financial leverage in our applications refers to the use of fixed income financing or debt in the capital structure. Financial leverage also may be affected by the use of leases and preferred stock or the use of any fixed income financing.5 Our sample of REITs held as ETF AUM includes both equity and mortgage REITs. Our rationale, verified by subsequent findings, for including both types is that from the REIT stockholders’, managers’, and ETF managers’ perspectives, benefits associated with REIT stocks being held as AUM in an ETF are essentially identical. We found that the type of assets held by a REIT is less important than benefits associated with REIT structures, associated tax advantages, and inherent advantages of ETFs.6 Table 1, “Sample Statistics” show a minimum value for ETF AUM of zero, indicating that our sample period included a time period before any REITs were included as ETF AUM. Also, an alternative methodology uses a dummy variable set to 1 after each REIT’s being first held by at least one ETF AUM, to examine the impact on each REITs’ leverage. However, results using dummies were not statistically significant; thus, these results are not reported.7 Other potential econometric model concerns include causality, inverse causality, endogeneity, and omitted variable biases. For example, omitted firm characters may affect both the capital structure and the likelihood of a REIT being included in ETF portfolios. Also, it is possible that ETFs select REITs for their portfolios that already have lower leverages. Other methodological concerns include controlling for omitted variable biases. We did this by including firm fixed effects to alleviate concerns regarding controlling for omitted variable biases; however, fixed effects coefficients were not statistically significant. We did not include industry fixed effects since all REITs belong to the same industry. We failed to find reasons to be overly worried about these potential econometric concerns being problematic relative to our results.8 The National Best Bid and Offer (NBBO) quote reported the highest bid price and lowest ask (offered) price in a security, sourced from among all available exchanges or trading venues. The NBBO, therefore, represents the tightest composite bid-ask spread in a security.9 This is similar to Titman and Wessels (Citation1988) and Lipson and Mortal (Citation2009) for non-REIT stocks.10 Fama and French (Citation2002) and Lipson and Mortal (Citation2009) also suggested using these control variables.11 We included both market value financial leverage (Dt/Vt) and book value financial leverage (Dt/At) in our analysis. Theoretically, market values should be more relevant since these values represent the current market structure of the firm; however, book values better represent the value of capital originally inputted in the firm.12 Again, the matched sample uses propensity score nearest neighbor matching with replacement based on total assets in previous year of ETF first inclusion. In addition, we matched treatment group firms with control firms by matching REITs of the same type (Equity REIT, Mortgage REIT, or Hybrid).13 This is similar to the approach used by Lin et al. (Citation2018).14 We ensured validity and consistency of the DID analysis by requiring that differences between the treatment and control group were consistent over time, controlling for parallel trends.15 These two measures are detailed in the Methodology, Data, Relevant Literature, and Preliminary Results section.16 Our conclusions regarding the longer windows may be somewhat tentative since longer windows may be affected by other endogenous and/or exogenous factors that may impact returns and volatility. However, homogeneous trends across all three windows were observed; thus, we are confident that our reported results are robust.17 Definitions of the three spread measures are detailed in the Methodology, Data, Relevant Literature, and Preliminary Results section.18 Table 7, Panels A, B, and C found consistent, statistically significant differences for all three market liquidity measures before and after each event date (month 0). Speadbefore¯ is the mean of spreads in periods (–12m, 0m), (–24m, 0m), and (–36, 0m) prior to month (0m), the event date and Speadafter¯ is the mean of spreads in periods (+1m, +12m), (+1m, +24m), and (+1m, +36m) subsequent to each event date. Δ Spread is the prior/postperiod differences in spreads, ranging from 23% to 46%, and was statistically negative in all cases, providing further evidence that stock market liquidity is increased subsequent to each REITs being held in ETF AUM.
期刊介绍:
The American Real Estate Society (ARES), founded in 1985, is an association of real estate thought leaders. Members are drawn from academia and the profession at large, both in the United States and internationally. The Society is dedicated to producing and disseminating knowledge related to real estate decision making and the functioning of real estate markets. The objectives of the American Real Estate Society are to encourage research and promote education in real estate, improve communication and exchange of information in real estate and allied matters among college/university faculty and practicing professionals, and facilitate the association of academic, practicing professional, and research persons in the area of real estate.