Investment View
Dexus remains the most heavily discounted AREIT despite the changing conditions for office operators and easing macro headwinds. News coming out of the US seems to be driving fear in the Australian market despite the landscape being vastly different.
Occupancy levels are trending back towards pre-Covid levels. Whilst a degree of Work from Home (WFH) will remain, Australia is seeing a strong drive for its workforce to return to the office. A similar trend to what has been seen in EU and the Middle East.
We believe underperformance from DXS is largely being influenced by the fear being generated by US office data. The US has seen a dramatic shift away from its work in office culture, predominantly in Tech (the #1 contributor to leasing activity pre-COVID). This has led a dramatic lift in vacancy rates causing significant discounts in asset transactions and greater urban decay from foreclosures. It’s important to note that the leverage of US REITs is significantly higher than Australian leading to a greater impact from rising interest costs.
DXS trades at a ~36% discount to its book value, which we continue to view as largely overdone. Proposed market transactions indicate that property values have only come down ~9% from their December 2022 values. We note that the proposed transactions are older, capital intense assets and so are unlikely to represent the equivalent discount to the higher quality assets DXS holds.
In our view, RBA has put through its last rate hike for the cycle, which should provide a tailwind for DXS as bond yields continue to fall. The market is starting to ascribe greater value to the peak rate sensitive names, and we believe DXS will fit the theme for this rally. The stock has begun to move, starting to outperform the REITs index for the first time since February 2023.
There is still a degree of anxiety within the market around asset valuations, occupancy headwinds and balance sheet conditions but the data suggests that this storm is clearing. We have been waiting for a clearer indication that valuations aren’t falling as significantly as what’s being priced in (which we are seeing), occupancy’s improving, (specifically within A grade and premium assets) and the gearing is well below target with a strong level of hedging.
We upgrade our rating to Buy. It’s only a matter of time before the market realises that the pricing of Dexus is far too dramatic, whilst ascribing close to zero value to the funds management segment of the business.
Asset Valuations
We are gaining confidence that the ‘collapse’ in asset values is being largely overestimated by the market. Information on proposed transactions suggest that the market has largely only felt a ~9% decrease from December 2022 asset values. Noting that these assets have come from Mirvac (MGR) and Dexus, both who flagged to the market that they were looking to offset older, more capital intense assets to feed their ‘flight to quality’ thematic.
Whether or not the discount on the assets is truly reflective of the entire asset base is yet to be completely known but given the low occupancy and higher cap rates of the assets being sold, we are under the impression that asset values will follow a case-by-case basis (with emphasis on asset quality). The sale of 60 Margaret Street (A grade) being flagged to be closer to flat discount on December values being evidence for our view.
Three of the transactions are pre-2000’s acquisitions with lower-than-ideal occupancy rates, with the exception being 367 Collins. It fits the theme of the flight to quality thematic and less so highlights a point of balance sheet necessity (which would be a larger contributor to a property value collapse).
Figure 1: Proposed Office building transactions
The current share price implies a ~1.8% expansion of cap rates (when excluding value for the funds business) which is difficult for us to see happen over the next 12 months especially given the backdrop of peaking rates and expectations that higher grade assets hold strong. We’d expect group cap rates to expand somewhere around 0.75% (-16% to asset valuations) off the back of where the market is trending but even this may be overdone.
We think that the market is overestimating how much portfolio values will decline. We expect that the June portfolio valuation statistics will demonstrate a smaller than expected decline and this may come as a catalyst for a positive reaction in the share price. Further transactions will play a role in re-enforcing this.
Multiple
Dexus currently trades at a Price to funds from operations (P/FFO) of 11.8x, a ~3.4x turn discount to its 5-year average. Expectations of a collapse in asset valuations has the market ascribing little-to-no value to the funds management business which will see a significant uptick post the Collimate capital acquisition. We expect that the P/FFO multiple disconnect should close now that rates are expected to have peaked and occupancy fears are being quashed.
DXS continues to lead the AREIT sector in its discount to its book value. At ~36% we largely think this is overdone. A portion of it bakes in the decline in asset valuations but we expect it to be far less than what the Price to Book Value (P/BV) multiple is currently representing.
Whilst conditions aren’t crystal clear over the next 12 months, we think that the market is currently pricing in a worst-case scenario, and that the multiple suggests that DXS will be at the forefront of excessive asset valuation write-downs.
Australian Offices Aren't US Offices
Looking at the global landscape, it’s easy to get caught up in all the negative information coming out of the US. Office occupancies aren’t seeing improvement, valuations in some cities have dramatically collapsed and the WFH thematic seems to be much more stubborn than expected. What’s a reality for the US, isn’t quite the same for the Australian market.
Where we see the key difference are:
- Work from home trends taking longer to return to normal: The US is demonstrating a far slower return to pre-pandemic work conditions versus the rest of world. Europe and the Middle East have seen a significant pickup in employees returning to office and the same is now being seen for Australia.
The technology dominance of the US market vs Australia highlights a significant contributor to the differentiation in office utilisations. US Technology leasing activity saw a ~53% reduction from 2019, falling from its number 1 to now number 2 behind banking and finance. Real estate, Media, Food and Beverage and Education also saw significant falls of +50%.
- US office overbuild: New US office builds have fallen dramatically over the past 4 years due to an oversupply of offices built prior to the shift to work from home. Office volumes under construction are down ~40% (since pre-COVID) and the US market is also undergoing its most rapid decline in office inventory as office to residential conversions gain momentum.
Whilst Australia does have a minor excess supply releasing to the market, it’s nowhere near the excess that the US is seeing. Average vacancy rates in the US (~20%) are almost double what we are seeing in Australia (~10%) and represent a much clearer cultural shift with a workforce that has greater expectations for remote work.
- Higher leverage: Typically, the US office operators are willing to carry far higher leverage which, given the higher rate environment, comes as a far greater risk. Gearing for Australian REIT’s typically falls within the 25%-40% range whilst its typically 50%+ for US operators. Examples being Boston Props (BXP.US) at 60%, Vornado Realty Trust (VNO.US) at 51% and Corporate Office Prop (OFC.US).
Leverage is a key metric to look at as it will play a significant role in determining the necessity for these office operators to transact assets at a discount. The lower leverage of the Australian operators implies that its less likely we see transactions feel the weight of balance sheet urgency, leading to more power within pricing discussions.
- US urban decay: Offices in the US are in urban decay (and some of CBD zones of major cities like San Francisco (CA)). Old buildings aren’t gaining any interest and are largely being avoided by the market. The interest in investing capital to redevelop these office buildings is fleeting and there is much greater interest in conversion (but largely on newer stock). Key examples being last year’s Jamison Realty’s LA Asset and the GFP Real Estate’ upcoming 25 Waterstreet Asset (Manhattan, NY).
Two notable examples being two towers in Houston (TX) which went to foreclosure before being sold (at a loss of 80-90% to Commercial Mortgage-backed security holders). Houston, Dallas, and San Francisco are seeing a considerably greater urban decay which has been exacerbated by the WFH culture change.
San Francisco has been hit significantly hard due to its tech dominance. Union Bank Plaza office tower was recently sold for a significant ~75% discount to its 2020 listing price. The discounted sale wasn’t driven by debt but more so by an old, capital intense asset that a bank needed to divest, in a city where vacancy rates aren’t seeing (and likely won’t) a return to pre-Covid levels.
Difference in the multiple: US Office Operators are trading far closer to their book values versus Australia. The average price to book discount in the US is ~5%, whilst Australian office REITs are closer to a ~30% discount. In our view, we see greater risks on US asset valuations vs Australia’s given the high vacancy rates and little evidence of a return to office.
Summary: We believe the days of DXS being viewed in the same backdrop as US office REITs is over. The business is much stronger operationally and the Australian market is significant structural difference to the US. We are starting to see this come through in the resilience of Dexus’ share price, but we think that this gap has room to expand as Australian offices see a greater return-to-normal.
Figure 2: DXS share price moving away from trading in line with US REITs. Clear operational benefits from Australian domestic market.
Balance Sheet
DXS continues to take a conservative approach to its balance sheet management. Current gearing is ~25.6%, which is well below its 30-40% target range and the hedging is ~85%.
The gearing has plenty of headroom in the case of expanding cap rates, even at the extreme end at a ~1.8% expansion of cap rates, the gearing still only falls to the mid-point of the target range. We expect with a ~0.75% expansion of cap rates that the gearing will only lift toward the low point of the target range. Even under an expansion of cap rates up to 1%, we don’t foresee the need to sell assets for balance sheet deleveraging.
The balance sheet allows for a considerable amount of flexibility on the dividend especially given gearing assumptions don’t include the capture of asset recycling (~A$700m) which is expected to come through in the next 12 months. The dividend yield is 6.4% at a payout ratio of ~75% (target 75-80% of FFO), allowing it to shift if necessary. We think the 51cps DPS is sustainable for FY23 and FY24.
Figure 3: Cap rate expansions impact on gearing. Sandstone scenarios
Figure 4: Hedging and debt maturity at high end of AREITs.
Figure 5: Gearing and cost of debt at the lower end of AREITs
Risks to Investment View
Investment Thesis
We believe conditions have started to shift for Dexus. Office operators are continuing to see improved momentum post-pandemic and the market appears to be slow in shifting valuation expectations off the back of recent data.
Cap rates are expanding at a far slower rate what the market is fearing. We are unlikely to see cap rates getting anywhere near the ~1.8% expansion rise that the share price is implying. Office occupancy is improving and even with an increase in hybrid work, the flight to quality theme will be a tailwind for Dexus’ assets vs other office landlords.
In our view, the market is likely to begin to differentiate between US office conditions and Australian office conditions. We believe Australian office operators are in a much better position in this point in the cycle vs US peers.
At a 36% discount to book value or ~9% mark-to market discount and a sustainable dividend of 51cps implying a dividend yield of ~6.4%, in our view the stock looks cheap.
We are upgrade our rating to BUY. We expect some fear will remain around the office market but it’s only a matter of time before the market realises that Dexus’ discount does not reflect reality.
Figure 6: P/FFO at a near 2SD discount
Figure 7: P/NTA
Figure 8: DIV yield
Figure 9: Discount to book value and FFO growth don’t reflect current operating conditions
