The impact of the stapled security structure on the quality of financial disclosure: Evidence from Australian Real Estate Investment Trusts and Listed Infrastructure Funds

Abstract

This paper is the first study to explore whether the stapled structure influences firms’ activities in earnings management (EM). Trusts and firms under stapled securities are exposed to various managerial opportunities and activities that can provide the flexibility of using EM approaches. Therefore, the stapled structure is expected to induce increased EM behavior and signal a lower level of financial disclosure quality than the unstapled structure. This empirical research analyzes a panel dataset that contains information of Australian REITs (A-REITs) and Listed Infrastructure Funds (LIFs) from the year of 2000–2017. Evidence shows that stapled A-REITs and LIFs use a greater magnitude of EM approaches than unstapled entities. The results imply that the stapled security structure may signal lower-quality of financial disclosure for firms than the unstapled security structure. This study provides additional insight into the understanding of how the security structure may impact firms’ financial disclosure behavior.

Introduction

Enhancing transparency through improving the security structure has become an increasingly important field in finance and accounting research. The present study aims to contribute to the literature by providing additional evidence of how the choice of security structure impacts publicly traded trusts’ and firms’ earnings management (EM) activities. In this study, EM is defined as the approaches used by managers to influence the disclosure of financial information in their preferred style.

A company is legally allowed to contractually bind two or more securities into a single security for public trading on the stock exchange of various countries. The countries allow publicly trading of bound securities include Australia, Singapore, Hong Kong and Malaysia. This structure of bound security is known as the stapled structure. This structure has been widely adopted for securities of banks, Real Estate Investment Trusts (REITs) and Listed Infrastructure Funds (LIFs) in the above countries (Brown, 2005, Davis, 2009). Fig. 1 shows the structure of a stapled security by using the example of a stapled Australian REIT (A-REIT). The ordinary shares of the asset management company are bound together with the units of the trust when the stapled A-REIT is publicly traded on Australian Securities Exchange (ASX).

Listed (i.e., publicly traded) trusts like REITs and LIFs are required to be passively managed and must comply with restrictive regulations in most countries, such as U.S. and Australia. However, the business activities of asset management companies are not restricted by these regulations, and the asset management companies need to pay corporate tax. A stapling structure can provide the opportunity for some listed trusts like REITs and LIFs to carry out more transactions with a stapled asset management company. The listed trusts can potentially engage with more types of business activities indirectly. These business activities can be restricted by the regulations for unstapled trusts. Consequently, the stapled structure is expected to provide more flexibility for managers to engage in EM activities than unstapled ones. Additionally, managers of listed stapled REITs or LIFs tend to utilize EM activities to cover that the entity of listed trusts may circumvent the regulation regime by using the stapled entity of an asset management company. Thus, the stapled structure indirectly facilitates business activities that can be restrictive to the entity of trusts. This may be one of the reasons that the managers of REITs or LIFs prefer the stapled to the unstapled security structure, especially before the outbreak of 2007 global financial crisis.

Listed trusts like REITs and LIFs in Australia, Singapore and Malaysia were increasingly stapled to the shares of asset management companies before the outbreak of the 2007 global financial crisis (Newell, 2010). After the global financial crisis, a restrictive financial disclosure environment was observed and imposed on the markets (Liang and Dong, 2018). The new requirement may reduce the flexibility of EM activities for stapled securities. Consequently, the stapled structure has been gradually replaced by unstapled structure after the global financial crisis (Davis, 2012). The reduced proportion of stapled REITs and LIFs, as observed after the global financial crisis, is consistent with the expected outcome of enhanced regulation environment and the following hypothesis: The stapled structure for REITs and LIFs provides more flexibility and the need of using EM than the unstapled structure.

REIT (complemented with LIF) markets are ideal for the present study for the following two reasons. One is that, in countries allowing for stapled structure, REITs and LIFs both play an important role among publicly traded securities with regard to the value of their market capitalization and the number of observations. The other is that analyzing REITs and LIFs help control for heterogeneity across different industries in equity markets because REITs and LIFs have unified operating structures and an industry-specific organizational form (Anglin et al., 2012).

The present study focuses on Australian REIT complemented with LIF markets because the number and market capitalization of A-REITs are the second largest in the world, following the United States’ (U.S.) REIT market. The stapled structure for REIT and LIF is not allowed in other countries, such as U.S., Canada and the United Kingdom (U.K.) because the structure provides potential tax arbitrage opportunities for asset management companies (Corry, 1980, McCall, 2000, Wern, 1999). The equivalent of stapled structure in U.S. is the paired-share structure that was outlawed by the U.S. Congress in 1984. However, a few paired-share REITs and LIFs that were created before 1984 have been permitted to operate. Paired-share REITs and LIFs are essentially different from externally managed REITs and LIFs in U.S. Under the paired-share structure, the trust entity of (REITs or LIFs) is not subject to corporate tax. These trusts must comply with the specific regulations for REITs and LIFs. However, an asset management company bound to a trust is subject to corporate tax and does not need to comply with regulations for these trusts. For externally managed REITs and LIFs, all of their business activity should comply with the regulations for REITs and LIFs, and they are exempt from corporate tax.

Outside U.S., a few countries, for example, Australia, employ the dividend imputation tax system. This system allows a firm to attach paid corporate tax as credits to dividend when dividend (together with the tax credits) is distributed to shareholders. Shareholders can thus use the attached tax credits to offset their individual income tax. The stapled structure for Australian REITs and LIFs becomes attractive to both domestic and international investors, due to the benefit from tax credits, preferred pre-tax income and potentially reduced cost of capital. Consequently, Australia hosts the largest publicly traded stapled REIT market among other global financial markets.

This research conducts empirical tests using a panel dataset containing information on the accounting and business performance of A-REITs and LIFs from the year of 2000 to 2017. The study finds that stapled REITs and LIFs tend to use a greater magnitude of EM activities than unstapled ones. The results imply that stapled listed trusts, for example REITs and LIFs, are likely to have lower quality of financial disclosure on earnings than unstapled securities.

This paper contributes to the literature and practice on the following six aspects. It is the first to explore and provide empirical evidence on the influence of security and managerial structures on earnings management. Secondly, the study sheds light on how the stapled structure may impair the quality of financial disclosure of earnings. The results on potential low quality of financial disclosure may imply the problem of asymmetric information between principals and agents in the markets for stapled listed trusts. Thirdly, the findings can facilitate investors to improve their interpretation of reported financial information for listed trusts of REITs and LIFs. Fourthly, this research provides additional explanations on the phenomena of shifting structure (stapled vis-à-vis unstapled) for REITs and LIFs before and after the 2007 global financial crisis. Fifthly, the implications on findings can possibly be applied to other types of stapled securities, on which similar regulatory requirement on pass-through income and restrictive asset components is imposed. Overall, the results of the study can assist regulators and policy makers to design regulations for equity markets and to improve transparency in capital markets.

The rest of the paper is organized as follows. The next section reviews the literature and discusses how the different security structures (stapled or unstapled) induce different magnitudes of EM activities. The methodology section estimates EM measurement and develops testing models, followed by the section of data description and empirical findings. Concluding remarks are provided in the last section.