The Development and Maturing of the US REIT Sector

Abstract

The US Real Estate Investment Trust (REIT) market is the oldest and most established REIT market in the world. While REITs were formally created by Congress in 1960 their history actually goes back to the nineteenth century and Massachusetts. Massachusetts State law in the mid nineteenth century effectively precluded the use of a corporate structure in real estate investment. This was because corporations were prohibited from owning real estate if it was not an integral component of the business of the firm. In response to these state laws what became known as the Massachusetts Trust was created to provide an effective vehicle through which real estate investment activity could be carried out. These trusts, as would REITs a century later, provided a tax transparent investment vehicle. The tax status of these structures was however removed in 1935 by the US Supreme Court. Ironically, only 5 years later the 1940 Investment Company Act introduced similar tax transparency vehicles for equity investment in the form of mutual funds. However, it took a further 20 years of lobbying on the part of the real estate industry before a similar vehicle was introduced in the context of the real estate market.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

The US Real Estate Investment Trust (REIT) market is the oldest and most established REIT market in the world. While REITs were formally created by Congress in 1960 their history actually goes back to the nineteenth century and Massachusetts. Massachusetts State law in the mid nineteenth century effectively precluded the use of a corporate structure in real estate investment. This was because corporations were prohibited from owning real estate if it was not an integral component of the business of the firm. In response to these state laws what became known as the Massachusetts Trust was created to provide an effective vehicle through which real estate investment activity could be carried out. These trusts, as would REITs a century later, provided a tax transparent investment vehicle. The tax status of these structures was however removed in 1935 by the US Supreme Court. Ironically, only 5 years later the 1940 Investment Company Act introduced similar tax transparency vehicles for equity investment in the form of mutual funds. However, it took a further 20 years of lobbying on the part of the real estate industry before a similar vehicle was introduced in the context of the real estate market.

The restrictions under which US REITs operate have changed only slightly over the last half century. The two main restrictions are that 75 % of assets and income must be derived from real estate activities and that the REIT is legally obliged to pay out 90 % of its taxable income in dividends. Other restrictions relate to the ownership structure of the REIT. At present no more than 50 % of the REIT shares can be held directly or indirectly by any group of five or fewer investors. As long as these restrictions are complied with then the REIT may deduct dividends from its corporation tax liability. US REITs do however differ from those in most other jurisdictions in that they are not limited to the private real estate market, rather they can also invest in real estate based debt. In any analysis of the US sector this distinction is important due to the quite different characteristics of the two main sectors. Mortgage REITs frequently move in line with bank stocks due to their underlying asset base. The distinction is also important in that while Equity REITs dominate the listed REIT sector today both in terms of the number and market capitalisation, this was not true in the 1960s and 1970s when they generally made up less than 50 % of the sectors market capitalisation.

An ongoing issue with the US REIT sector is that while legislation was introduced in 1960, with the first REIT being introduced the following year, it is really only since the early 1990s that the market has developed and grown both in terms of trading volumes and also market capitalisation. Two key problems limited the development of the sector during the first 20 years of the sector. REITs were constrained as they were only permitted to own real estate and could not operate or manage it. This meant that REITs were required to find third parties and that the trusts did not have direct control of the portfolio investment decisions undertaken, with the possibility of agency conflicts arising. Furthermore, provisions of the US tax code also distorted the real estate market by making real estate investment tax shelter orientated. A taxpayer using high debt levels and aggressive depression schedules could take interest and depreciation deductions that significantly reduced their taxable income. In many cases these deductions led to ‘paper losses’ that were used to shelter a taxpayer’s other income. As Congress designed REITs specifically to create a taxable income on a regular basis the structure did not permit REITs to pass losses through to shareholders. These two issues were effectively addressed by the 1986 Tax Reform Act. This piece of legislation changed the real estate investment landscape in two respects. It reduced the potential for real estate investment to generate tax shelter opportunities by limiting the deductibility of interest, lengthening depreciation periods and restricting the use of ‘passive losses’. Furthermore, the legislation permitted REITs to operate and manage most types of income producing commercial properties.

However, despite the changes in the tax code, the late 1980s did not see a major expansion in the REIT sector. In part this was due to the fact that strong market conditions financing and investment from domestic and international institutions remained strong. Therefore, the need for new sources of capital in the form of REITs remained subdued. However, in common with a number of real estate markets across the globe many US markets witnessed severe downturns in the early 1990s with falls in real estate values of over 30 % observed in some markets. While in common with other markets this downturn was due in part to over-supply in many markets and worsening economic conditions impacting occupational demand, the US also had the problem of the Savings & Loan crisis. The combination of poor underlying market conditions together with the drying up of traditional sources of capital meant that many participants in the US real estate market were forced to look for alternatives.

The other major factor behind the growth of US REITs in the early 1990s was the development of the UPREIT (Umbrella Partnership REIT) structure. The UPREIT structure effectively comprises of two vehicles; a REIT and an operating partnership. The rationale behind this structure is that it allows those investors who properties are being placed into the REIT, to obtain units in the operating partnership on a tax deferred basis. Only if the units in the partnership are swapped for REIT shares are they subject to tax. Therefore, existing investors can effectively defer capital gains tax liabilities. The actual underlying properties are owned by the operating partnership, in which the REIT owns units. Therefore, the REIT only indirectly owns the properties in its portfolio.

The Omnibus Budget Reconciliation Act 1993 also played a major role in the growth of the sector and increased the attractiveness of the sector to Pension Funds. The legislation allowed greater flexibility concerning the 5/50 ownership rule. As long as the REIT had a minimum of 100 shareholders then pension funds could treat each contributor to the fund as an individual investor in the REIT. Two further key changes occurred in the late 1990s with the REIT Simplification Act of 1997 and the 1999 REIT Modernisation Act. The first piece of legislation contained a number of changes to the rules regulating REITs, including the elimination of the rule that required REITs not to earn more than 30 % of gross income on the sale of assets not held as long-term investments. The major change that occurred with the 1999 Act was that the minimum dividend payout was reduced from 95 % to 90 %. This came into effect in 2001.

The growth in the sector following the early 1990s crash and the introduction of the UPREIT structure can be can be clearly seen in Figs. 5.1, 5.2, 5.3, and 5.4. Figure 5.1 details the number of REIT Initial Public Offerings (IPOs). In 1993 and 1994 alone 95 REIT IPOs took place. As Fig. 5.2 illustrates the majority of these IPOs were Equity REITs, resulting in a structural shift in the sector, with the majority of listed REITs now being Equity vehicles not Mortgage or Hybrid REITs. This dramatic increase in the number of listed REITs alone accounted for much of the increase in the overall market capitalisation of the sector shown in Fig. 5.3, which also demonstrates how small the sector was prior to the early 1990s and Fig. 5.4 demonstrates the changing composition of the sector in percentage terms. This is a highly important factor in the analysis of the US REIT sector and will be returned to later in the chapter. The reduction in the number of listed REITs in recent years is primarily due to consolidation within the industry, with a large amount of M&A activity taking place.

Fig. 5.1

Number of US REIT IPOs

Fig. 5.2

Number of US traded REITs

Fig. 5.3

Market capitalisation of US REIT sector (US $m)

Fig. 5.4

Percentage composition of US traded REITs by market capitalisation

2 Investment Dynamics

The performance of the US REIT sector in recent years has attracted a large amount of attention. As can be seen from Table 5.1 REITs substantially out-performed the overall Equity market from 2000 through to the onset of the financial crisis, with an average quarterly return in excess of 5 % in comparison to 0.63 %, in addition to a lower risk measure. However, as will be addressed in the following section, there are a number of factors that occurred during this specific period that affected these figures. In addition, as can be seen from Table 5.1 the relative performance and the correlations between the three assets are not stable. To a large extent some of these shifts are due to the structural changes that have occurred within the REIT sector.

Table 5.1 REIT performance

One of the main arguments in relation to REITs relates to their perceived diversification benefits. These benefits can be viewed from both real estate and capital market portfolio perspectives. A number of recent papers have shown that REITs still provide sufficient diversification opportunities to mixed-asset portfolios that already contain an allocation in the private real estate market. Feldman (2003) finds that both public and private real estate have a place in a mixed-asset framework. Mueller and Mueller (2003) extend the analysis of Feldman (2003) to examine the impact of private and public real estate on the mixed-asset portfolio for various holding periods for the last 5–25 years up to 2002. The authors find that for the full sample period the inclusion of private real estate as measured by the NCREIF index, either appraisal based or de-smoothed, led to improvements in the performance of the efficient frontier at the lower risk levels, while REITs provided improvements to the entire frontier.

These are findings supported by those of Lee and Stevenson (2005), who examine the consistency of REITs within a mixed-asset framework. The results show that the benefits that REITs provide tend to increase as the investment horizon is extended, indicating that REITs may be more attractive to investors with longer holding periods. This increased attractiveness over longer holding periods may also be due to the linkages between REITs and the private real estate market increasing with the use of longer horizons. Therefore, over these longer investment periods, REITs may be displaying more of the diversification qualities of the direct market, further enhancing their diversification qualities. In addition, the diversification benefits of REITs appear to come from both its return enhancement and risk reduction benefits. In the low risk/return portfolios the allocations obtained in the return enhancement tests are larger than those when examining risk reduction. This trend however, reverses as one along the efficient frontier. This would indicate that as an investor moves along the frontier the rationale behind the inclusion of REITs alters, with increasing emphasis being placed on the assets risk reduction qualities rather than its return enhancing capabilities. This trend is in part probably due to the low correlations relative to both asset classes. While REITs have lower correlations with the general equity indices examined, this is not at the expense of increased coefficients with regard to bonds. REITs therefore effectively sit between the broad equity and fixed-income sectors, with both risk and return measures in-between stocks and bonds. This enables REITs to appear return enhancing to bonds, without the same degree of increased risk that would be seen with stocks, and also risk reducing to stocks. In the case of stocks they provide diversification benefits due to their relative low risk measures and correlation coefficients, without the same level of return sacrifice that would occur if funds were switched into the fixed-income market. It can be argued that perhaps much of these benefits result from the mandatory high dividend payout that gives REITs fixed-income like characteristics.

Any analysis of performance in the US REIT sector has to take into account the structural changes in the market. The combination of those factors detailed in the previous section concerning the 1986 Tax Reform Act, the impact of the early 1990s cycle and the introduction of the UPREIT heralded a new era for the REIT sector. Many empirical studies have shown that the characteristics of the sector changed quite substantially in subsequent years. For example, Glascock et al. (2000) found that that from 1972 to 1991, REITs were segmented from the broader equity market, but were integrated from 1992. In addition, the authors find that prior to 1992 the returns of both Equity and Mortgage REITs behaved in a fashion more similar to the fixed income market, but that the Equity REIT sector acted more like stocks post 1992. Clayton and MacKinnon (2001) note that the correlation of REIT returns with stocks and bonds underwent a structural change in the 1990s, with the sensitivity of REIT returns to large-cap stocks declining over time, while that with small-cap stocks increased. In addition, recent papers such as Cotter and Stevenson (2006) and Case et al. (2012) have highlighted the general instability and time-varying nature in the relationship between REITs and the broader equity markets.

3 Market Maturity

While REITs were established in 1960 the main growth has been since the early to mid-1990s. Not only has this obviously affected the size of the sector but it has also led to changes in the investment dynamics of REITs. Clayton and MacKinnon (2001) argue that their finding relating to the changing relationships with the broader equity markets can at least in part be attributed to the growing maturity of the REIT sector. This growing maturity can be illustrated by the level of trading in REIT shares. In 1993 SNL Financial estimates that the average daily aggregate volume in the sector was three million shares. By 2005 this figure had increased to over 40 million. A well observed phenomenon in the financial economic literature is that volume is a key determinant of volatility. As can be seen from Fig. 5.5 daily volatility in Equity REITs has increased substantially over the last decade. Of interest however are results from a working paper by Cotter and Stevenson (2008) that shows that while volume is a significant determinant of REIT volatility, the sector is half as sensitive to changes in volume as the stock market generally.

Fig. 5.5

Rolling fifty day standard deviation

From the mid-1990s onwards there has been a marked increase in investor awareness. As Chan et al. (1998) note, institutional investment increased substantially in the 1990s. This had a very quick impact in relation to the number of analysts following the sector (Wang et al. 1995) and a reduction in the bid-ask spread of REIT shares (Below et al. 1996; Bhasin et al. 1997). In addition, a number of studies have highlighted the positive relationship between the increase in institutional share ownership and REIT performance (e.g. Wang et al. 1995; Chan et al. 1998; Downs 1998; Ling and Naranjo 2003). Ling and Naranjo (2006) specifically consider flows from dedicated REIT mutual funds, finding evidence of REIT performance significantly impacting upon future capital flows. This finding would also be supportive of the momentum profits observed in studies such as Chui et al. (2003) and Hung and Glascock (2008, 2010). Whilst the process of increased ownership started in the early 1990s, it continued subsequently and the inclusion of REITs in the mainstream Standard & Poor’s indices in 2001 was a further step in increased investor acceptance of the sector. Indeed this increased mainstream equity market investment may, in what can appear initially a counter intuitive argument, have contributed to the changing relationship between REITs and large-cap stocks. It is frequently cited that REITs have had a lower correlation with equities in recent years, as the figures in Table 5.1 support. However, it may be that while the headline relationship has fallen, the influence of the broader stock market is actually greater than at any other time.

As Fig. 5.6 illustrates in the period 1998–2000 while the general stock market was rapidly rising during the technology/dot.com boom, REITs not only underperformed, but they fell in value. This was despite strong underlying economic conditions and robust performance in the private real estate market. Furthermore, operationally REITs were reporting impressive results. In 1998 and 1999 Funds from Operations for listed REITs rose by 14.70 % and 9.50 % respectively. In contrast REIT returns were –18.82 % and –7.38 %. Since 2000 REIT performance has been vastly improved. In the 3 years from January 2000 Equity REITs produced a total return in excess of 50 %, at the same time the stock market fell in value. Furthermore, the operational performance of REITs was subdued. In 2001 REIT dividends only rose by 3.80 % while Funds from Operations fell by 2.20 %. It would appear that certainly for the period 1998–2003 REITs effectively acted as a defensive counter cyclical equity sector. They were highly unattractive in the late 1990s boom in comparison to high growth tech stocks. However, following the technology crash in 2000 REITs became seen as a highly attractive sector. Their high level of asset backing, high dividends and following 2 years of underperformance relatively cheap pricing made them attractive for equity managers looking for income driven sectors. The strong performance since 2000, together with factors such as the inclusion in the S&P indices, has further encouraged increased investment due to investor momentum and herding behaviour. The result is increased flow of funds into the sector, leading to demand driven price pressure.

Fig. 5.6

US capital market performance: 1995–2004

This impact can also be observed if one examines the relative investment behaviour of the REIT sub-sectors. US REITs are generally highly focused as Figs. 5.7 and 5.8 illustrate. Most REITs have underlying portfolios concentrated in a single property type. Furthermore, many REITs are also geographically focused, with few having national portfolios. The arguments in favour of a focused strategy are that firstly the portfolio managers should have a better understanding and knowledge of specialist markets and sectors. This is of particular relevance in the context of such a large market as the US as it reduces the number of markets for which information and market analysis is required.¹ In addition, standard financial theory would argue that firms shouldn’t diversify, but rather allow investors to do it for themselves. While this normally refers to conglomerates the same argument can be advanced for firms such as REITs. However, this second argument is dependent on sub-sectors behaving different and effectively tracking their underlying markets. While earlier studies such as Mueller and Laposa (1996) and Chen and Peiser (1999) show strong relationships between focused REITs and the underlying specialist market segment, other papers have found diverging results. Young (2000), for example, shows that the correlations between sector focused REITs has tended to increase. Recent work by Chong et al. (2012) also illustrates this using a Dynamic Conditional Correlation framework. This can be illustrated in Fig. 5.9 which shows simple rolling correlations between Office REITs and other sectors. There is a clear trend that shows that over the last decade the sub-sectors have converged in terms of their return behaviour. This would indicate that the markets are increasingly viewing REITs more as an overall sector and that less attention is placed on the characteristics of individual REITs or sectors.

Fig. 5.7

Specialisation of US equity REITs (November 2012)

Fig. 5.8

Composition of US equity REITs by market capitalisation

Fig. 5.9

Fifty day rolling correlations: office REITs

4 Conclusion

The development of the US REIT market has been a major factor in the growth of the sector globally. However, it is vital in any analysis of the US market to appreciate the changing investment dynamics of the sector and its continued growing maturity. The market has developed significantly in the last decade and a half and this has resulted in the asset becoming more of a mainstream investment sector. This has had the consequence that the dynamics and investment characteristics have also altered.