Path to prosperity

In its outlook statement, SIA mentioned that the recovery in international air travel is slower than expected and industry experts have continued to revise their projections downward. A full recovery could take 2-4 years according to industry forecasts. For SIA, its own forecasts are that passenger capacity would not reach 50% of pre-Covid levels by the end of FY21 (Mar 2021) and will continue to re-assess its fleet size and network. For the end of 2QFY21, SIA projects that its passenger capacity would be c.7% of pre-Covid levels, a >7x jump from the current levels of <1% of pre-Covid. 

SIA appears more positive on cargo recovery as the reopening of economies and manufacturing plants could imply a greater need for air freight. It is also looking to deploy cargo-only passenger flights when justified. SIA currently has 7 freighters and 33 passenger aircraft that are deployed on cargo-only services. 

When the recovery comes possibly in 2022, SIA will also have to bear the extra maintenance costs of getting their parked aircraft back to air-worthy shape. It currently has 119 aircraft parked at Changi Airport and 29 in Alice Springs. I also do hope that by 2022, most of SIA's high fuel hedges have worn off and SIA can start getting better margins. 

On the bright side, SIA has raised $8.8bn from the rights issue and secured other lines of financing worth c.$2.2bn since Mar20 and this could put it in a better shape vis-a-vis competitors who may not have the backing of a strong parent. We have also seen other airlines like AirAsia having going concern issues and this could be a slight positive when the recovery does come for surviving airlines. 

Overall, I believe that this set of results is more negative than expected and could lead to a further softening of SIA share price. 

I believe the decline would have been even worse if Ascott had compared it on a same-store basis since there was the merger between Ascott and Ascendas Hospitality that would have boosted the distributable income. During 1H2020, 21 of Ascott's properties were temporarily closed due to the pandemic. Since then, 12 properties have reopened and 7 more are scheduled to reopen in 3Q20. 

Operationally, we also saw the benefit of having master leases or management contracts with minimum rents. From a gross profit standpoint, management contracts experienced the worst decline of -61% while master leases actually grew 51%, although partially contributed by the Ascendas merger. 

Ascott has further highlighted that it has a highly diversified set of revenue streams from a geographical perspective and this could help with the recovery. As mentioned in a previous article, exposure to non-Singapore countries allows Ascott to ride on the push for domestic tourism compared to its SG-focused peers. The higher APAC exposure also allows it to ride on the strong response to Covid that APAC countries have had relative to Europe and the US. 

I expect Ascott's hotel-focused peers to be even worse off as serviced residences tend to have a more stable occupancy due to longer stays. Additionally, in its largest country, Japan, Ascott actually has residential properties that it is renting out, which provides further stability to income. 

At its current price of 90cents, Ascott provides an annualized yield of 2.3% (taking 1.05Scts multiplied by 2). In reality, the yield could be around 2.5% assuming Ascott pays out the amounts retained. At such low yields, I do not believe that investing in Ascott shares provide good risk-return ratio. In fact, investing in their perpetuals might be a better choice since it has seniority over common shares and provides a higher yield at 3.07%. 

Yesterday both Keppel DC REIT (KDCREIT) and Mapletree Industrial Trust (MINT) released their results for the quarter ending Jun2020. Both REITs have performed strongly after their end-Mar lows due to their exposure to data centres. MINT has performed more strongly even though it only has a 32% exposure (by AUM) to data centres vs KDCREIT's 100% exposure. I think this could be due to the more visible growth pipeline of MINT from its sponsor's stakes in the US data centres and also redevelopment opportunities at its existing non-DC assets. 

KDCREIT reported a 1H2020 DPU of 4.375 Scts which as a 13.6% improvement yoy. This was due to new acquisitions of SGP4 and DC1 in 2019 as well as the latest acquisition of Kelsterbach DC in May 20. Based on the 1H2020 closing price of $2.54, this represents a 3.44% annualised distribution yield. As all of KDCREIT's tenants are in the data centre business, they were not subject to any shutdowns and hence did not require any rental reliefs/support from KDCREIT. KDCREIT was thus able to payout 100% of its distributable income. 

For MINT, it reported a 1QFY20/21 DPU of 2.87 Scts which was 7.4% lower yoy due to rental rebates extended to tenants due to Covid-19 as well as holding back tax-exempt income of S$7.1m to mitigate the impact of mandated rental reliefs. According to MINT, if such income was not withheld, DPU would have been 3.19 Scts, representing a 0.09 Scts increase yoy. The amount held back was about 10% of its 70.6m distributable income. 

In my opinion, MINT continues to be held back by its legacy portfolio of flatted factories and older business parks as shown in the negative rental reversions as well as lower new rents achieved during the period. However, investors could possibly see this as 'land bank' to be redeveloped into more future-ready assets. Given MINT's management's strong execution track record, there could certainly be more redevelopment opportunities like 30A Kallang Place and Kolam Ayer. 

It is hard to overstate the attractiveness of the data centre asset class during the Covid-19 period where businesses have been forced to digitize and people have been working from home. More importantly from a real estate perspective, data centre leases tend to be long (KDCREIT's WALE of 7.4 years) and provide a steady stream of income so long as service level agreements are met. 

At their current trading levels, it appears that both KDCREIT and MINT are in a virtuous cycle as their trading yields (3-4%) are much lower than data centre cap rates (5-7%). This would allow them to make accretive acquisitions even with a greater proportion of equity funding. In turn, this growth potential can further drive up the share price and make it even easier for either to make accretive acquisitions. 

Investor optimism has certainly showed up in their strong price growth and I believe that investors are pricing in additional acquisitions. Therein lies a certain amount of risk if the REITs are unable to find assets to acquire and meet investor expectations. 

The pace of leasing has been slower as marketing and property visits have been postponed. CCT also mentioned that it has seen more renewals coming in as tenants want to minimize capital expenditure. Both CCT and KREIT cited that their high tenant quality was important in keeping occupancies high and allowing them to collect rents on time. There have also been no significant requests to downsize. 

During this period, CCT highlighted flex space operators as being more affected as their members defer or waive memberships. This is usually highlighted as the downsize of flex space operators as they incur long term rent expenses but with short term membership revenues. On the bright side (for CCT), the WeWork lease at 21 Collyer Quay remains on track and there has been no indication of a withdrawal. 

KREIT cited rental reliefs as a reason for weaker property income. Zooming out a little, we expect NPI margins to compress due to higher cleaning and digitizing expenses incurred by office landlords in making their properties suitable for a return to work. While this is somewhat already expected, the investing community is still waiting on more specific guidance on how much earnings would weaken (similar to what the hospitality players have disclosed recently). 

In their outlook statements, both KREIT and CCT mentioned that the impact is still hard to assess and they would continue to be prudent in distributions. 

As generally known, demand for office space follows economic cycles and it is up to developers and urban planners to manage the supply of office space. While demand is expected to decline, Covid-19 has also resulted in construction and renovation works being pushed back and thus reducing new supply in the next few years. Furthermore, developers could also take the coming few years of weak demand to undertake redevelopments and asset enhancements at their properties. Examples of this are AXA Tower and Keppel Tower. 

Overall while the impact from Covid is still limited to the first-degree impact from rental reliefs, I think that we can see higher vacancies and even negative rental reversions in the coming years as leases expire. Office demand will not collapse as fast as hospitality and landlords generally remain fairly positive about the long term potential of office. The slide below from Keppel REIT sums the discussion up quite aptly. 

Ascott issued a profit warning at the start of the week citing that income available for distribution for 1H2020 could decline by 55-65% yoy while DPU would decline 65-75% yoy. The larger decline in DPU was probably due to the divestment gains from Ascott Raffles Place in 1H2019. Declines are attributed to the Covid-19 pandemic which have led to a 44% yoy decline in international tourist arrivals according to the World Tourism Organization from Jan-Apr2020. On a full-year basis, the decline is expected to be 58-78% yoy . 

Ascott REIT continues to earn income from its master leases and long stay clients whose demand is less affected by the pandemic. Additionally, as mentioned in an earlier post, the gradual opening of certain countries' domestic tourism could be a slight boost to occupancy as the REIT is geographically diverse and present in countries with some sort of domestic tourism (vis-a-vis Singapore). 

On 17 Jul 20, ARAUSHT announced that the pandemic had resulted in a significant decline in gross revenues such that it expects to report a net property loss and no distributable income for 1H2020. This is probably the first time we see a REIT or BT having no distributable income! ARAUSHT is a high-beta play on the US hospitality industry as it does not have a master lease structure and net property income fully depends on hotel demand. 

When it was listed together with Eagle Hospitality Trust in 2019, ARAUSHT was seen as a safer play due to the sponsor's brand name and presence of more cornerstone investors. However, in a situation like this, I feel that Eagle's structure would have provided more comfort to investors as they had master lease income (notwithstanding the sponsor's default). 

SIAEC announced 1Q21 business updates and mentioned that flights handled in the quarter was ~13% of pre-Covid levels. On the costs side, the impact was partially offset by the Job Support Scheme given by the Singapore government; management and BOD had also taken pay cuts. Revenue was down 54% as flights handled declined 87% yoy and SIAEC reported an operating loss of $8.6m vs $17.7m profit in 1Q20. Without the JSS, its loss would have been $36.7m. Management also mentioned that the pick up in June was not material and outlook for the MRO business will be challenging. 

Looking ahead to the other hospitality REITs, we are likely to see similar trends where REITs with master leases experience a smaller decline in income while those like ARAUSHT can see their distributable income wiped out (or close to being wiped out). I continue to believe that players with overseas exposure will be more cushioned against the revenue declines (ceteris paribus) as larger countries do have substantial domestic travel demand. 

Flights handled is also a good forward indicator of hospitality revenues as Singapore's hospitality receipts are almost entirely dependent on foreign tourists. A side revenue that could gradually taper off is the use of hotels for 'Stay home notice' and 'Quarantine' of travellers that recently entered Singapore. The hope now is that this quarter would represent an earnings trough as it covered the periods worst hit from Apr-Jun. Looking forward to see what results are like in the coming weeks!

0.94 new ESR REIT units for every 1.0 Sabana units held. This is an all-stock deal with no cash consideration to be paid out (unlike the previous ESR-Viva deal). ESR units would be issued at an indicative price of $0.401 (vs last traded at $0.39) while this implies a $0.377 price (vs last traded $0.36) for Sabana units. 

1. DPU and NAV accretion - According to the pro-forma, this would be DPU accretive to ESR REIT shareholders by 3.5% while Sabana shareholders would enjoy a 12.9% accretion. What is slightly positive to unitholders is that any income retained would have to be paid out before the effective date of completion. In the past quarter, DPU was affected due to amounts retained for Covid-related purposes. 

The higher accretion by Sabana could also be a function of the management fees being 60% paid out in units vs 100% cash previously as the pro-forma assumes that Sabana's fee structure follows that of ESR's. Looking at the fee structure, the base fees are both 0.5% of the deposited property. There could be some savings from the performance fee perspective as ESR has a high watermark of 6 cents DPU (following which perf fees is 0.25% of the increase) which it is unlikely to hit in the next few years vs Sabana's 0.5% of NPI if DPU grows by 10%. ESR also has a lower trustee fee of 0.1% of deposited property vs 0.25% for Sabana.

One point to note is also that the enlarged REIT would not be maintaining its Shariah compliance so this could save additional costs. According to Sabana management, they have been reviewing the Shariah compliance over the years as it appears that the cost of compliance has outweighed the benefits from having an additional pool of investors. 

DPU and NAV accretion (Source: ESR REIT)

 

2. Shift in portfolio exposure - The merger will allow the enlarged ESR REIT to have greater exposure in high-specs and logistics properties. I think this is a positive as high-spec properties have enjoyed strong demand due to their future-ready characteristics while logistics properties have been the flavour of the year given the increased need for logistics services during lockdowns. 

3. Operational synergies - The REIT Manager posits that the merger could bring about operational synergies from a marketing and leasing perspective. There could also be cost savings due to consolidation of contracts and stronger bargaining power as a result of becoming a larger entity. While I agree with this in theory, in practice, we have yet to see major cost synergies from the previous ESR-Viva merger. This could be due to maintenance contracts not expiring. Also in relative terms, Sabana increases the estimated GFA by about 27%, so not that sure how much incremental benefit ESR can extract out of the merger. 

4. Size and cost of capital - To me this is one of the strongest rationales for the merger. As a larger and more established entity, ESR REIT would be able to secure more competitive rates from both the banks and the market. Evidence of this is already provided when it managed to refinance Sabana REIT's debts and reduce the cost of debt from 3.8% to 2.5%. 

In theory, I believe that this is possible, the true extent of the savings is hard to estimate without further disclosure from the REIT. The lower rates are probably also a function of the lower base rates in 2020  and then slightly offset by the longer debt tenor and the lack of encumbrances. Best is if we know the interest margin pre and post refinancing. 

On the equity front, I am excited that the larger free-float market cap promised by the merger could possibly lead to inclusion in the EPRA NAREIT Index. Index inclusion has been a strong driver of REIT share prices in 2019 and I think the benefits of a re-rating cannot be overemphasized. For the longest time, ESR has had trouble being included as its largest shareholder, Mr Tong Jinquan, did not want to reduce his holdings in the REIT. While this was a testament to his confidence in the REIT, this also prevented the REIT from accessing a greater pool of investors that comes with index inclusion. With the completion expected to be in Nov20, the earliest that ESR can be included (assuming it hits the threshold) is probably going to be in 1Q21. 

Generally quite positive on the merger as I believe the larger platform can bring about most of the benefits that ESR purports it can. Looking at prices, there has been a divergence in the performance of larger market cap REITs vs small/mid cap ones and pushing itself to become a large market cap REIT is a step in the right direction. 

One big problem ESR always had post-Viva merger was the high gearing. As of 1H20, its gearing is 41.8% and in the merger announcement, it mentioned that its pro-forma gearing would be 41.7%. Which is weird if I look at the sources and uses since it appears that ESR is funding the deal 54% equity 46% debt. Implicitly I would have expected gearing to go up. Either way, the higher MAS gearing limit of 50% probably gave management sufficient comfort that they had some buffer if valuations come down even more. Pre-merger ESR also did a revaluation of its assets and thankfully the decline was <2%. 

Overall my stance remains that both sets of shareholders should vote for the deal as the larger platform would help. For shareholders that do not approve (or are unable to continue holding the REIT due to the lack of shariah compliance), they should sell their shares. 

SPHREIT released its 3QFY20 Business Update this evening. This comes after it shifted to half-yearly reporting; the link to the presentation can be found here. Would like to highlight a few things to take note of:

SPHREIT declared a DPU of 0.5 Scts for the quarter representing an increase over 0.3 Scts declared in the last quarter. However, it did not disclose its distributable income hence we are unable to accurately calculate what the current payout ratio is. If we take last quarter's results, this represents a payout ratio of 33%, up from 20% previously. 

In all likelihood, 3Q distributable income would have been significantly worse off than 2Q as the quarter covered Mar-May which were the months which felt the most impact from Covid-related mall closures. Additionally, SPHREIT also announced that they were giving rental waivers for tenants. Therefore, the reduced DI would mean that payout ratio is probably more than 33%. 

My original thesis was that prime district malls in Orchard//City area like Paragon would suffer substantially more than suburban retail due to the target audience. Apart from having a bigger residential catchment, suburban malls typically tend to have a larger proportion of 'essential' tenants like supermarkets and F&B. Whereas prime district malls tend to have a larger percentage of high fashion and targets discretionary and tourist spend. Therefore I expected suburban malls to be substantially more resilient. 

Looking at the visitor traffic figures, Paragon experienced a 58% decline in traffic while Clementi Mall's traffic declined 53%. Would have thought that the difference would be much bigger. My guess would probably be that pre-Covid Paragon was not overcrowded whereas Clementi Mall is frequently packed to the brim. Hence with a higher base there was much more room for traffic to fall at Clementi Mall. 

SPHREIT reported a committed occupancy of 98.8% across its malls in Singapore and Australia. Importantly, its 2 largest revenue contributors had occupancies of >99%! I think this is impressive as it probably implies only 1-2 shop lots are vacant. Going forward both Paragon and Clementi Mall have 28%/22% of lease renewals in FY21 and this will be a critical point to watch as leases do take time to expire. I also note that SPHREIT did not disclose its rental reversions in the presentation pack. We are therefore unsure if the high occupancies are a function of cutting rents. 

In the down-cycle that is coming, I believe that landlords will do well to shore up occupancy even if it is at the expense of rental rates. Unlike rental rates, occupancy is binary, either you are getting income from that shop lot or you are not. However, do hope that the REITs can be transparent in sharing such figures so that unitholders are well-informed.