For real-time updates, do join our:
- Telegram Group: t.me/theinvestquest (or search “The InvestQuest” on Telegram)
- Whatsapp Broadcast: Just send “Hello IQ” to +65 8840 2520
- Facebook Page: The InvestQuest | Facebook
- LinkedIn Page: The InvestQuest | LinkedIn
This is Part 2 of our series on rising govt bond yields.
In Part 1, we had analyzed five past instances of short-term US Treasury yield spikes and its impact on the broader stock market. We ranked the sectors that performed from best to worst on average, and also covered the two criteria that typically results in positive stock returns, during a rising yield environment.
Article Summary
1) Prior yield spikes and corresponding S-REIT sector performance
2) S-REIT sector valuations relative to history
3) Which S-REITs are analysts most bullish about?
The InvestQuest View
Historically, S-REITs have underperformed the STI Index during periods of yield spikes, and outperformed the STI Index in the months after the yield spikes. For the current episode, the S-REIT sector has underperformed the STI Index by 18%, significantly more than past episodes.
This potentially opens up an opportunity for a convergence trade, which could materialize in the short-term if rate volatility subsides.
At the same time, we are cognizant that S-REIT valuations are not cheap from a historical standpoint and so, we retain a neutral longer-term view. Among the large-cap REITs, Ascendas REIT looks fairly interesting to us at this juncture.
Disclaimer: We currently hold shares of Ascendas REIT and may choose to accumulate more going forward. As always, our articles are meant as a record of our investment journey. It is not intended as investment advice and should not be relied upon as such. Do your own due diligence!
1) Prior yield spikes and corresponding S-REIT sector performance
Here are the prior short-term spikes in the US Treasury Yield
Some local investors may be concerned about rising yields, given the potential negative impact on the S-REIT sector. Let’s first see if there’s any truth to such thoughts by looking at historical short-term US Treasury (UST) yield spikes, and the corresponding S-REIT sector performance during these periods.
In the chart below, we plotted the 10-year US Treasury yields since 2001, compiling the dates where there had been a short-term spike in UST yields (circled in red).
During these yield spikes, what was the return of the STI Index & the FTSE ST REIT Index?
For these date ranges, we calculated the returns of the STI Index and FTSE ST REIT Index (see table below).
- On average, the STI Index returned 8.6% and the FTSE ST REIT Index returned 4.1% (rightmost column in below table), implying that S-REITs had underperformed by 4.5% on average. This result is in line with our expectations.
- During the most recent UST yield spike, S-REITs have underperformed the STI Index by 18.1%, which is much more than prior instances. This potentially opens up an opportunity for a convergence trade, especially if rate volatility stabilizes from here.
In the 6 months following these yield spikes, what was the return of the STI Index & the FTSE ST REIT Index?
AFTER short-term spikes in 10-year UST yields have occurred, S-REITs outperformed the broader STI Index by an average of 3.5% in the following 6-months. In the rightmost column of the table below, we can see that across the five date ranges, the FTSE ST REIT Index returned an average of 11.9% vs STI Index’s 8.4%.
No one knows for sure if US Treasury yields will continue rising. However, the magnitude of the recent yield spike is closing in to the largest spikes seen in the last 20 years, so we remain optimistic that rate volatility should subside soon, which could be a short-term positive for S-REITs.
2) S-REIT Sector Valuations relative to History
While S-REITs could potentially narrow its underperformance against the broader STI Index, we acknowledge that valuations for the sector aren’t cheap, so stock selection might be key from here.
S-REIT sector current pullback vs largest historical declines
The below chart shows the largest declines experienced by the FTSE ST REIT Index since 2007. The recent pullback has been relatively shallow in comparison, at less than half of the 2013 taper tantrum so far.
Historical Price-to-Book Ratio
Some investors buy into REITs for the underlying value of the investment properties, hence looking at P/B ratios might be relevant. At a current P/B of 1.10x, the S-REIT sector trades at a 13% premium above the historical 0.97x average (see chart below).
To be fair, this could be justified from a relative value perspective, since broader markets are also trading at a valuation premium given to the sustained low interest rate environment and the increase in market liquidity since pre-Covid.
Historical Trailing Dividend Yield
The S-REIT sector trailing dividend yield is currently at 4.1%, below the historical average of 6.1%. Do note that this is backward looking. In 2020, many REITs were impacted from Covid restrictions and some may have also withheld dividend distributions, so we should take the 4.1% figure with a pinch of salt.
However, we would imagine that even with some normalization to operations in 2021, the trailing dividend yield would likely be below the historical average of 6.1% (Dec-2007 to current) or 5.5% (Jan-2011 to current).
Historical Dividend Yield Spread over 10-year SG Govt Bond Yields
The dividend yield spread is a measure of how much investors are getting in higher returns as compensation for buying REITs (a riskier asset) versus buying the SG government bond (risk-free asset).
Dividend yield spread = Trailing Dividend yield – 10-year Singapore government bond yield.
The dividend yield spread is currently at okay levels of 2.8% versus its historical average of 3.9%, especially once you factor in the depressed dividends for 2020 mentioned earlier.
4) S-REITs that look interesting to us
Overall, we are neutral on the S-REIT sector. While we think that S-REITs might outperform in the short-term if rate volatility subsides (as mentioned in Section 2), we are also cognizant that the sector isn’t cheap on traditional valuation metrics (as seen in Section 3).
S-REITs ranked by implied upside to analyst consensus target price
For investors looking for REIT-specific ideas, we have listed all the S-REITs with market caps of over SGD 1bn in the table below. They are sorted by the implied upside to the analyst consensus target price (rightmost column in orange).
Ascendas REIT looks interesting to us
Ascendas REIT looks quite interesting in our view, since it had recently completed an equity fundraising in Nov-Dec 2020. This was done via a preferential offering and private placement, with new shares being issued at S$2.96 and S$$3.026 respectively. So, you can potentially at buy below these prices which is great.
In addition, the intention to use half of the recently raised capital to purchase a portfolio of European data centres will likely sit well with investors and help support the REIT’s valuations moving forward. For instance, other large-cap Industrial S-REIT peers with more data centre exposure such as Mapletree Industrial Trust and Keppel DC REIT are trading at 1.6x and 2.3x book value respectively, a steep premium to Ascendas REIT’s 1.3x book.
The InvestQuest View
Historically, S-REITs have underperformed the STI Index during periods of yield spikes, and outperformed the STI Index in the months after the yield spikes. For the current episode, the S-REIT sector has underperformed the STI Index by 18%, significantly more than past episodes.
This potentially opens up an opportunity for a convergence trade, which could materialize in the short-term if rate volatility subsides.
At the same time, we are cognizant that S-REIT valuations are not cheap from a historical standpoint and so, we retain a neutral longer-term view. Among the large-cap REITs, Ascendas REIT looks fairly interesting to us at this juncture.
Hi InvestQuest,
Was wondering if you have any thoughts on the relatively shorter land leases (and hence more significant lease decay) for Singapore industrial/logistics properties? Understand that industrial land tenures in Singapore are on 20, 30 and 60 year leases, and that Reits with exposure to these properties include Ascendas, MLT and MIT. Do you factor in the shorter land leases into your analysis or do you give greater weight to other metrics? Thanks!
Hi AlpacaInvestments,
Thank you for your question. Shorter land leases should definitely be taken into consideration when assessing REITs as investments.
Some things to consider:
– Singapore REITs (S-REITs) trade at higher yields than REITs in Japan, USA. There are a few factors behind this, but the shorter land lease for properties in Singapore is one reason.
– Industrial S-REITs trade at higher yields than S-REITs in other sub-sectors.
The difficult thing about land lease tenures is that I don’t think there’s a super easy way to assess the impact without spending considerable effort on a model. In my view, there are three approaches to take:
– Simply eyeball each REIT’s land lease to expiry against the forward yields.
– Use Bala’s Table as a proxy when comparing REITs with similar operational outlook & gearing. https://www.clc.gov.sg/docs/default-source/commentaries/balas-table.pdf
– Build a model to come up with a valuation. For each property held by the REIT, either assume that the land lease ends at the expected expiry date or assume that JTC allows renewal of the land lease (the amount needed to renew can be found on the JTC website I believe). Sum all the projected cash flows from these properties, discount it backwards.
Note that ALL REITs pay dividends out of their capital (unlike other dividend-yielding stocks like banks). What this means:
– For banks, dividends are paid out as a % of net income
– For REITs, dividends are paid out as a % of “distributable income” (net income + depreciation added back). What this means is that instead of setting aside money to compensate for declining asset values, REITs are paying out that sum as dividends. Because REITs work this way, they will either need to retain more cash (but in SG they need to distribute at least 90% of DI) or they will need to raise capital from the markets. Obviously, REITs with shorter land leases are more likely to raise capital from the markets.
Thanks for raising this issue. A lot of industrial REITs have been making acquisitions abroad for this very reason (the short land lease for props w/in SG).
Hi InvestQuest,
Thanks for the detailed reply!
While Industrial S-REITs are trading at higher yields than S-REITs in other sub sectors, I think the issue of lease decay is still largely ignored by investors, which has resulted in some mispricing. If we look at forward yields of Industrial S-REITs vs Retail S-REITs, it is about 5+% vs 4+% respectively. In my opinion, REIT investors largely ignore the effects of lease decay, and prefer to focus on dividend yield. In the case of Industrial vs Retail REITs, remaining lease tenures for the Singapore properties of Industrial REITs are ~30 years, whereas Retail REITs own properties with lease tenures of at least ~70 years. I don’t think this is justified by the 1% yield spread, unless we are expecting some massive outperformance of industrial rents against retail rents.
However, I don’t think there’s a way to bet against this, as the inflated share prices of Industrial S-REITs actually results in a positive feedback loop for the them (similar to how Tesla’s high share price allowed them to raise capital at a lower cost of equity). In this case, Industrial S-REITs are trading at higher P/B ratios than other S-REIT sub sectors, which makes it cheaper for managers to raise equity, and concurrently more likely for acquisitions to be yield accretive. Existing shareholders are happy as long as acquisitions are yield accretive, as they are not diluted.
Just my thoughts on the issue of leasehold properties 🙂