2017 was another roller coaster of a year for REITs, exemplifying how short-term volatility is merely a noise in the long-term goal of producing mid to high single digit total returns. Our +5-7% projected 2017 total return for the MSCI US REIT Index (Bloomberg: RMS G) proved to be in the ballpark, as it finished the year with a total return of +5.1%. However, as shown in Figure 1, the path towards the end result was nowhere near a straight line. For example, the RMS had produced a year to date (or YTD) total return of +4.2% by February 27, only to fall to a YTD total return of -2.4% by March 12, a swing of over 600 basis points (or bps). There were nine such swings of over 400 bps throughout the year, creating a jagged rise to finish the year with the somewhat boring mid-single digit total return.
Adding to the "boring" tagline that could be associated with 2017 REIT performance was the unexpected surge in the S&P 500, which finished the year with a total return of +21.8%. As REITs underperformed the S&P 500 in 2016 as well, will 2018 be the third year in a row of REIT underperformance? For reference, there has been only one period in history (1997-1999) when the S&P 500 outperformed REITs for three consecutive years.
We don't believe this is the right question to ask. REITs provide a completely different risk-return profile than traditional equities. While they can produce returns of similar volatility (the best REIT was up over 50% in 2017 while the worst was down over 80%), the cash flow streams are more stable. Revenues may not double or triple as they can for tech companies in the S&P 500, but they also generally have a much smaller downside. As long as REITs can produce dividend growth above inflation with a low-risk profile, the sector will be an attractive long-term investment, marked by short-term changes in valuation.
As we head into 2018, we believe the risks associated with REIT dividends and dividend growth are near all-time lows. Balance sheet strength, portfolio quality, and payout ratios have never been better. One way to quantify the perceived risk of REITs is through the difference between the REIT dividend yield and the US 10-year Treasury yield, or the "spread". A higher spread would indicate that investors would like a bigger cushion over a "risk-free" investment due to a higher perceived risk, and vice versa. As of December 31, 2017, the dividend yield of the MSCI US REIT Index was 4.1%, which compared to a yield of 2.4% for the US 10-yr Treasury bond. The 170 bps spread between the two compares to the historical average of 110 bps.
Even with the low risk associated with REIT dividends today, we assume no change in the dividend yield spread in 2018 in each of our scenarios. In our base-case scenario where the US 10-yr Treasury yield finishes the year between 2.4% and 2.6% and REITs grow dividends by 5.7% (Citi estimate), the total return for REITs would be in the range of +5-10%.
In a bear-case scenario for REITs where the US 10-yr Treasury yield rises to 2.8-3.0%, the total return for REIT would be between -5% and +1%.
In a bull-case scenario where the US 10-yr Treasury yield finishes between 2.2% and 2.4%, REITs would produce a total return of +10-15%.
Our base case of +5-10% is supported by a net asset value (or NAV) approach as well. As shown in Figure 2, if we assume NAV growth of 5% through levered same-store net operating income growth of +3.7%, additional return of +1.6% from the reinvestment of free cash flow, and no change to capitalization rates (or "cap rates), the total return would be +9%. Even if cap rates were to rise 25 bps, the total return would be +4%.
In both cases, there is an assumption of no change to the NAV discount. This is similarly conservative given that REITs were trading at a 6% discount to NAV as of December 31, 2017, which compares to the historical average of a 1% premium.
Assumptions for our Projections
Underpinning our base-case projections are assumptions for an economy on strong footing across most metrics, low levels of new construction across most property types and stable funds flows.
Funds flows are particularly unpredictable. Over the past three years, flows have been negative for REITs, which is one of the reasons REITs have not been able to trade in line with their historical NAV premium or dividend yield spread. Our base case assumes that negative flows do not accelerate.
New construction has been below the historical average since 2007, and we don't expect that to change in 2018. While supply has increased steadily since the beginning of the cycle in 2009, developers have been more disciplined than in the past. Each time deliveries have exceeded demand, planned construction has slowed to levels that have allowed markets to return to equilibrium, in general. While there are some cities where a product type's supply may be ahead of demand, we assume that there is not a massive spurt of new speculative construction.
Finally, we do not forecast a recession in 2018. Before the passage of the new tax bill, the Federal Reserve (or Fed) had forecasted GDP growth of +2.5% for 2018, in line with the current estimate of +2.5% for 2017. Consensus estimates project job growth to slow slightly to 162,000 per month in 2018, down from 173,000 in 2017 (through November), bringing the unemployment rate down to 4.0% by the end of the year. If GDP were to grow upwards of 3%, long-term interest rates may rise to our bear-case scenario for REIT total returns. Similarly, lower GDP growth could anchor long-term rates, bringing about the bull-case scenario.
Tax Reform Effects on REITs
In general, the new tax bill is a positive for publicly-traded equity REITs.
First, the bill does not repeal any of the benefits currently available to REITs. Notably, there is no change to their business interest deductibility, deferred taxes on like-kind exchanges, or tax-exempt status.
Second, the reduction in corporate taxes for non-REITs should boost overall demand for commercial real estate across all property types. Furthermore, retail real estate should benefit from the expected rise in disposable income by consumers.
Third, the bill reduces the maximum tax rate that an individual will pay on the 'ordinary income' portion of REIT dividends. Averaging about 70% of the dividend historically, the ordinary income portion was historically taxed at an individual's top marginal rate, which was 39.6% in 2017. The bill proposes that REIT ordinary income be treated as 'pass-through income', which is subject to a 20% deduction. When combined with a new top bracket tax rate of 37%, the new maximum tax rate on REIT dividends is 29.6% ((1-20%)*37%). Therefore, the after-tax yield for REIT investors rises by almost 30 bps ((39.6%-29.6%)*70%*4.1%)! In theory, each of the forecasted total return scenarios presented above could support a 10-yr US Treasury yield 30 bps higher due to this change, creating an even larger cushion to our conservative outlook.
In addition, there are some components to the tax bill that will affect specific property types and geographic regions differently. The $10,000 limit for deduction of State And Local Taxes (or SALT) is a relative win for states and cities with lower taxes (e.g., Texas, Florida, Phoenix, Seattle, and Las Vegas) and a potential detriment to states and cities with high tax rates (e.g., New York City, California, Chicago, Washington D.C.). While corporations and people will not likely not move in the short-term, this could affect long-term employment and population growth trends. The new tax bill also limits the deductibility of mortgage interest to new loans less than $750,000 (versus $1 million prior) and requires a longer holding period to get an exemption for gains on the sale of a primary residence. Combining this with the lower deductibility for SALT (which includes property taxes), homeownership will likely decline further, driving demand to rent multifamily and single-family homes. We are overweight multifamily in the Chilton REIT Composite.
2018 Portfolio Positioning
In 2017, there was an extremely wide gap between the best and worst performing property types. In particular, the data center sector was up 29% for the year, while the shopping centers sector was down 11%. Thus, the REIT industry has grown in size and property types to the extent that making generalizations can lead investors to miss out on the unique attributes of each stock or property type.
We believe stock selection will be the key determinant of investment returns followed by relative property type weightings. Using our bottom-up stock selection methodology, we have positioned the Chilton REIT Composite with property type weights that vary significantly from the benchmark. Specifically, our largest overweights are regional malls, shopping centers, and data centers/cell towers. The top three underweights are healthcare, triple net, and industrial. The only difference from a year ago is the inclusion of shopping centers in place of office as an overweight.
Regional malls became a significant contrarian call early in 2017 due to the threat of e-commerce coupled with the high number of retail bankruptcies and store closures, especially department stores.
The retail evolution will have winners and losers. However, the perception that all regional malls would suffer the same fate resulted in a buying opportunity for extremely high-quality class A regional malls at prices significantly below NAV and at the lowest funds from operation (or FFO) multiples that we had witnessed since 2005 (excluding 2008-2009). It reminded us of the early 1980s when equity REITs were selling at similar discounts, resulting in many REITs becoming either buy-out targets or opting to sell properties at much higher prices than the public was willing to pay. GGP (NYSE: GGP) was one of them, creating value by selling malls at higher values than what was implied by its stock price.
30 years later, GGP was (and still is) trading well below the value of its properties, prompting Brookfield Property Partners (NYSE:BPY) to make an unsolicited bid for the whole company. At about the same time, activist investors started showing up as major shareholders of two of the other leading owners of Class A malls, Macerich (NYSE:MAC) and Taubman (NYSE:TCO). As of December 13, 2017, the stock prices for the three names were up an average of 32% from the lows in 2017!
Furthermore, Unibail-Rodamco (OTCPK:UNRDY), one of the largest mall owners in Europe, agreed to purchase Westfield Corporation (OTCPK:WFGPY), an owner of primarily US Class A malls, in a deal worth $25 billion. Based upon our analysis, all of the publicly traded Class A regional mall REITs would need to see stock prices expand significantly to approach the valuation implied by the proposed Westfield transaction.
As shown in Figure 3, we began the year with an underweight to shopping centers but finished the year with a significant overweight. The catalyst that caused us to increase the allocation was the announcement in June that Amazon (NASDAQ:AMZN) would be acquiring Whole Foods, which caused a significant pullback in shopping center REIT prices.
In an environment when most investors were questioning the legitimacy of brick-and-mortar real estate, the Amazon announcement confirmed our thinking that physical locations are necessary for the efficient delivery of groceries. From the Whole Foods announcement to December 14, 2017, shopping center REITs in the Chilton REIT Composite produced an average total return of +11.0%, which compared to the MSCI US REIT Index at +0.1%. Still, price to FFO multiples for the group averaged 13.6x as of December 14, 2017, the lowest level since 2005 (excluding 2008-2009), which compared to the MSCI US REIT Index at 15.7x. As a result, we would not be surprised to see activism and M&A emerge within the sector in 2018.
Retailers of all stripes are discovering that "omnichannel" retailing, blending of physical stores, e-commerce, and outlets, is the best model for the future. Admittedly, many shopping centers are vulnerable to store closings and tenant bankruptcies as the transition to omnichannel occurs, but we believe the outlook for grocery-anchored shopping centers is far better than for power centers with large boxes and tenants without a clear strategy to get customers into the store.
The Chilton REIT Composite contains shopping center REITs with management teams that are ahead of the changing trends in retail. As such, they have been upgrading portfolio quality and securing tenants that fit the new theme of live, work, and play. Many locations are suitable for densification with other end uses such as residential and office that should improve the predictability of cash flow generated and, ultimately, better price/FFO multiples. Balance sheets for the shopping center REITs in our composite are characterized by debt to EBITDA (earnings before interest, taxes, depreciation, and amortization) ratios of 6x or less, the lowest we have witnessed in years. Coupled with retained cash flow, these REITs have plenty of financial flexibility to fund the elevated capital expenditures that are necessary to modernize and repurpose their properties.
Data Centers and Cell Towers
Our third significant overweight, data centers/cell towers, reflects how important these structures have become in the evolution of the internet and the consumption of data on mobile devices. Data centers stand to benefit as enterprise customers realize savings by outsourcing IT (information technology) to external data center operators and demand continues to ramp for applications with "cloud" providers such as Amazon, Google (NASDAQ:GOOGL), and Microsoft (NASDAQ:MSFT).
We believe cell towers will benefit from the densification of the networks required to support the explosion of data. Coupled with the future requirements for the connectivity of millions of devices ("internet of things", which includes driverless cars), the sector should enjoy high predictability of growth for the foreseeable future. Adding support to cell towers is the increasing acceptance of this property type by dedicated REIT investors and the recent decision by Vanguard, the largest REIT investor in the US, to incorporate cell towers, timber, and infrastructure REITs in the index it uses for its funds in 2018.
We remain confident that our methodology will continue to lead us to ideas that give us the best opportunity to produce total returns above the benchmark in 2018 and beyond.
Disclosure: I am/we are long GGP, MAC, WSR, WRI, ROIC, KRG, REG, UE. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.