Summerset Group Holdings - Cash generative compounding machine at attractive valuations !!
* The following investment pitch would be applicable to both Summerset and Ryman Healthcare. We have structured our investments in both these stocks as a single bet which we believe provides us the best Risk-Reward to play the following thesis. All figures in this note are in New Zealand Dollars (NZD).
Elevator Pitch:
Summerset (SUM) and Ryman (RYM) have a unique business model that is a mix of a healthcare operator, real estate developer and timeshare seller. This model helps it to create a Win-Win proposition for all stakeholders in the ecosystem and generate strong shareholder value (20%+ CAGR investor returns since their IPOs in 2011 and 1999 respectively). These compounding machines are now entering a demographic sweet spot that provides massive tailwinds for similar future value creation. SUM is currently trading at around 1.7X Price/ Book and I believe that the true adjusted Price/ Book is less than 1X, providing an attractive entry point for long-term shareholders.
Detailed Investment Rationale:
Compounding machines are those businesses that have the unique combination of quality, growth, and longevity. If you’re able to buy such businesses at a fair price, you can expect to do well over time. I think both RYM and SUM fits into the above definition well and provide an asymmetric risk-reward opportunity which is difficult to find at this stage of the bull market. I believe that this investment opportunity exists primarily because of the underappreciation of their unique business model and short-term concerns of most market participants that are clouding its long-term potential (Behavioural Edge).
There are very few business models such as ULCC’s (Ultra-low-cost carriers) or Value retailers that when executed in a disciplined manner deliver tremendous shareholder returns because of their highly cash generative model and ability to reinvest capital at attractive rates. I delved deeper into studying New Zealand based retirement operators because of the following amazing value creation statistic:
“RYM raised $25m in IPO at a valuation of $135M in 1999. The share count till today continues to be the same with no equity dilution. Since its IPO, the company has paid out $630 million of dividends and reinvested $2.9 Billion in its portfolio. The current market cap of the business is $5.15 Billion.”
The company’s business model, growth opportunity, and management culture were the key reasons for this success over the last 2 decades. I would like to detail my investment rationale by explaining that the opportunity for investing in both RYM and SUM is as attractive now as it was then. My key insights and variant perception on the important drivers are highlighted below and then followed by a detailed note:
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1.) Valuable Business Model - Creates free call options for real estate price appreciation:-
Important Insight - The ORA model creates a powerful growth flywheel only for operators with a strong development engine like RYM and SUM.
Key Variant Perception - Needs-based demand of 80+ year olds and their key priorities make the full continuum care model very resilient.
2.) Favourable Industry Structure - Long growth runway for integrated operators:-
Important Insights - Demographic sweet spot accelerates growth over the next 15 years; Economies of scale, Size, Location and Portfolio mix are the 3 key profitability drivers for retirement villages; Local network effects and economies of scale help in market share growth.
Key Variant Perception - ORA Model works best for operators with a long-term mindset/ culture; Transactional flexibility in the model protects large integrated operators in an economic downturn Vis-a-Vis marginal operators.
3.) Underappreciated Financials - Recurring and perpetual cash flow stream:-
Important Insight - Model is very tax efficient as fair value change in Investment property is not taxed.
Key Variant Perception - Book value growth of 15 to 20%+ can continue with accelerating build rate.
4.) Attractive Valuations - Book value is skewed by high discount rates and large float:-
Important Insights - SUM will derive significant valuation uplift from development engine and maturing of its portfolio. Both, RYM and SUM are trading at below historic averages on P/ B multiple.
Key Variant Perception - 85% of the assets on SUM’s balance sheet is a DCF model in which the net cash flows are discounted at 14-15%. Similarly, 68% of the liabilities is a costless float and not a true economic liability.
5.) Major Risks - Short-term risks remain, but long-term risks are overstated:-
Important Insights - Healthy market conditions with low unsold inventories. Potential short-term liquidity squeeze and CBRE revaluation.
Key Variant Perception - Overbuild risk is minimal. Demand might outpace supply meaningfully if penetration rates increase.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Business Model of RYM and SUM
RYM and SUM are New Zealand based listed retirement village operators. There are 6 major village operators (Ryman, Metlife, Bupa, Oceania, Summerset, Arvida) and all operators except Bupa are listed on the stock exchanges. The NZ retirement and care home sector are well regulated and have been able to become more mainstream over the last decade. This has resulted in increased popularity and therefore higher penetration rate among the older population.
The industry operates with a creative ORA (Occupational right agreement) system in which the residents buy a right to occupy a unit indefinitely without actually taking title to the underlying real estate (similar to a vacation timeshare). When the resident vacates the unit (usually upon death), the operator repurchases the occupation right at a cost less a deferred management fee (DMF) that varies among different operators. The DMF is an annual, non-cash charge measured as a % of the original purchase price.
The cumulative DMF is capped at 25% (over 5 years) of the purchase price at SUM and 20% at RYM (over 4 years). Most other operators charge a 30% DMF over a 3-5 year period. The deferred management fee (DMF) becomes payable only when the operator re-sells the unit. So if a resident buys an occupation right for $400,000 and vacates the unit after five years, the resident will receive $300,000 cash with the operator retaining the difference before remarketing the unit at a higher price (assuming real estate values have increased).
Effectively the operator is a real estate developer that finances development with customer deposits and gets to resell each development every five to seven years in perpetuity. This ORA model enables the operators to generate significant cash flows and profit from the real estate price appreciation without getting their capital locked.
The occupation right is treated as an interest-free loan (occupancy advance on the balance sheet) rather than a sale. Consequently, no revenue is recognized (and no income tax is payable) when an occupation right is sold or even when resold after the first cycle. Instead, the company recognizes changes in fair value of its real estate produced by the development uplift in the first cycle and real estate price appreciation in the subsequent cycles. The DMF is recognized as non-cash revenue. The deferred management fee is amortized over the expected period of tenure. As a result, the income statement does not truly represent the underlying economics.
One of the main attractions for residents at an RYM/ SUM village is the operator's ability to offer a full continuum of care for its residents (Retirement Village -> Rest Home Care -> Hospital Care -> Dementia Care). This allows an elderly couple in which one of them needs the care to stay together in the retirement village. As they continue to get older, they can have a peace of mind that the operator has all the facilities that they would need until death. Only 60% of the villages in the industry have a care facility within them whereas both RYM and SUM have 100% of their villages with care facilities. A lot of the care services are critical in nature and hence attracts needs based demand.
Residents pay a nominal weekly fee to access on-site services on a cost-recovery basis. RYM charges around 129$’s/ week and SUM charges around 110$’s/ week. This allows the residents to use all the community centers such as Cafe, Spa, Sym, Salon, Theatre, Billiards room, Recreation center etc.
To understand the reasons for the success of the ORA model, it is important to understand the profile of the residents of these retirement villages. The average age of entry for a resident in an RYM/ SUM village is 80+, with the breakup as follows: Independent RUs (82+), Serviced & Care Apartments (86+) and Villas (79). The average tenure is less than 5 years for independent units, 2-3 years for serviced and care apartments and 5 years for Villas.
The independent houses sell for around 75% of the median house prices in that area and serviced apartments at 46% of the median house prices. For example, the median House price in Auckland is 1.2M and an independent unit in the retirement village in Auckland will cost 0.85M. This allows the elderly to monetize 25%+ real estate value while they sell their home and move to a retirement village. Even when the price differential is not meaningful, the elderly downgrade from a 3 Bedroom home to a 2 Bedroom unit or so to release some equity for their remaining years, thereby making the transaction similar to a reverse mortgage.
Similarly, I believe there is an inherent risk pooling mechanism in these villages just as in insurance. While the average tenure of a resident is 5 years for an operator, the distribution curve varies from 1 year to 10+ years. After the initial 4 or 5 years, there is no DMF fee that is applicable. The priorities for a 80+ year old who is in the last 5-8 years of his life is not exactly to profit from the real estate price appreciation. They are happy to trade that away for good quality of care in an affectionate community.
The primary considerations of the elderly while choosing a retirement village are security, companionship, lifestyle, amenities, maintenance free living etc. The catchment area for most retirement villages are the communities that are within a small 5-7 KM radius of the village. Most people move into villages in which they already have friends. Most people in the industry say that peace of mind for the elderly in knowing their weekly fee commitments will not rise significantly is more important than the 20%/25%/30% deferred management fee cut that applies to the capital that would be returned to their offsprings post their death. Even the young are happy to see their older parents have a good quality of life. Hence, moving to a retirement village is a more emotional transaction for the elderly than just a mere financial transaction.
Hence, the model ensures that the operators and their shareholders are happy with the float and their ability to profit from the real estate price appreciation, while the residents are happy with their quality of life. Similarly, Society gets better off as younger families move into established communities once the elderly move into nearby retirement villages. Thus, it is a Win-Win solution for all stakeholders.
Let’s understand the underlying economics of the model through a sample. An average retirement village from an integrated operator like RYM/ SUM in Auckland will require something like 200 Million $’s for land acquisition, construction of retirement units (RUs), serviced apartments (SAs), Community centers etc. Usually, around 35 Million is required for building community centers. The RUs and SAs get sold at a 20 to 25% development margin, thereby helping the developers net back the 200 Million of invested capital through ORA sales. During strong real estate markets, the real estate price appreciation during the construction period of 3-5 years allows a higher margin and helps developers earn cash surplus.
Hence, during the first cycle, the developers get back their entire development capital and sometimes a higher cash surplus. In the next ORA sales cycle in 5-7 years, assuming there is a 3% house price inflation every year, the developer with give back 150 Million (25% DMF cut) to the first cycle residents and receive 231 Million from the second cycle sales, leading to a 80 Mn+ incremental cash flow that is spread out.
With similar assumptions of 25% DMF fee, 3% house price inflation and a 5-year sales cycle, the float will grow to 311 Million in 15 years. The maintenance CAPEX for refurbishment after every sales cycle is less than 3% of ORA income and there is a village level upgrade CAPEX during the 12th-15th year. This enables the village operator to earn 12.5 Million in DMF (4% net yield post maintenance) and 2-4 Million from care profits every year as the villages mature. Hence, with almost no real equity getting locked, the operator will continue to receive strong cash flows that grows with the underlying real estate value throughout the village lifecycle. Thus the development in the retirement sector can generate high ROIC if the operator is able to sell ORA’s to incoming residents and generate a costless float with every sales cycle. The villages will continue to generate aged care fees, deferred management fees and real estate capital gains in perpetuity.
This powerful flywheel is more effective for larger operators with a strong development engine as they can continue to use the float to fund newer working capital intensive developments and diversify the risk of a float drawdown. Retirement village development and operations require two very different skill sets. Developers need to scout for the right land, create good designs and construct efficiently while the operators need to ensure that the village community is happy and vibrant, enabling continuous resales in every cycle. Care facilities continue to be an important selling point for most integrated operators. Amongst all the operators, only RYM and SUM have both a well-oiled development and operating engine.
Usually, good operators like RYM/ SUM build care facilities with 40% more spare capacity and this allows them to tap other patients from the community as well. RYM especially has large reputed care facilities. The learning curve for operating specialized care facilities like Dementia is not easily replicable. For the large integrated operators, the ability of care facilities to draw ORA sales allow them to cross-subsidize it if necessary. Also, demand for needs-based care tends to be resilient and help operators to perform well irrespective of economic conditions as RYM’s track record has shown.
The combination of growing float and increased development allows RYM and SUM to compound at an attractive rate. For example, RYM currently has a gross occupancy advance or float of $ 2.7 Billion ($ 1.1 Billion in 2012) and the management expects it to increase to 5.2 Billion by March 2022, allowing it to fund the majority of its growth CAPEX during this period. This powerful flywheel will continue to compound as long as there is strong demand.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Favourable NZ retirement industry structure for RYM/ SUM
In the current disruptive world led by new technologies, there are very few trends or needs that can be projected over a two-decade period. One clear trend that can be predicted with meaningful certainty is the dramatic aging of the population in developed countries over the next 1-2 decades. Similarly, the need for care and compassion in your later stages of life is not easy to disrupt. The core underlying business drivers for RYM/ SUM will continue to be the same even 15 years from now.
As explained before, the average age of a new resident is 75+. Currently, there are around 300K people in NZ who are over 75 years old. This population bucket is expected to grow to 1 Million over the next 25 years. In fact, the annual growth in the 75+ population is expected to increase from the current 10K/ year to 15K/ year over next 5 years and reach 20K/ year over next 10 years, providing a very favorable tailwind for the NZ retirement sector. The birth rates tripled during the baby boomer generation. In Australia (Ryman made a strong entry into this market recently), the opportunity size is even bigger. The total number of 75+ year olds in Australia are 6 times that of NZ and Victoria state in which Ryman made its Australia entry alone being 1.5 times that of NZ.
The baby boomers who were born following World War 2 would start turning 75 in 2021. Birth rates were actually very low during the 1930s which means that the industry had been fighting a demographic headwind in recent years. Baby boomers are the wealthiest generation and they have the highest homeownership rate. Both of these directly correlate with higher retirement unit penetration rate.
As of 2016, the total number of retirement villages in NZ was around 300 and the total retirement units were 30500 and care beds were 37000. The average resident rate in a retirement unit is 1.3 approximately. The penetration rate of retirement villages among the 75+ population was 6.8% in 1998 and it has grown to over 12.4% by 2016. As the retirement villages have continued to increase in popularity, the penetration rate has grown by 30 bps every year and has accelerated to 40 bps over the last few years. The penetration rates in more urban areas like Auckland have started touching 15%+ and 20%+ in some prosperous sub-regions of Auckland. It is important to note that over 30% of the 75+ population growth in NZ is expected to come from Auckland alone.
In terms of the NZ retirement village market structure - 53% of the market is with the big corporates, 32% with small private players and 16% with not for profit operators. The Top 6 players contribute around 44% of the total market. In terms of sources of planned supply, big corporates contribute 54%, small private operators 41% and not for profit players 5%. In the care beds segment, the industry is far more fragmented. Only 1/3rd of care beds is with the top 6 corporates.
The profitability of the retirement village is dependent on 3 factors - Location, Portfolio Mix, and Size. The larger urban villages in good neighborhoods with the right mix of care and independent units perform the best. RYM is the gold standard here and they got the model right early on. Even though care facilities in isolation are less profitable, the presence of care facilities allows the retirement unit to sell at better rates. SUM is trying to replicate RYM with more large urban development compared with its historical portfolio of regional villages.
The industry’s profit pool is continuously getting consolidated among the top operators. Most smaller operators build villages with usually 50 to 100 units and lack the amenities provided by larger integrated operators including good care facilities. The larger villages have over 300 units and have better economies of scale. In key real estate markets such as Auckland and Christchurch, the market share of larger developers is high while the smaller players concentrate on non-urban regions. With small catchment areas, there is definitely local network effects in this business and it is very rare to see a new development that is close enough to another successful village unless there is huge demand growth.
Short term operators are better off playing in the real estate market in which the cash conversion is much easier and there is no recurring servicing need. For subscale operators, the cash flow would be negative on the operating side as the organized players compete on this price aggressively. Also, float becomes valuable only when there is a strong development pipeline. It takes 5 to 8 years from purchase of land to getting back their capital outlay. Banks are not ready to lend to smaller developers. Long gestation projects make retirement villages unattractive to short-term players.
Similarly, the construction timeline for smaller operators is much longer than that for larger operators like RYM and SUM which have a strong development engine. Working capital demand peaks in the year 2 to 3 of the project as the community centers and care centers are built first to generate interest for the development. The care facility mix varies among the top players with companies like Arvida and Oceania skewed more towards care with >70% facilities and companies like Metlife have less than 20%. RYM has the rough mix of 50% each between care and retirement. SUM is moving towards RYM from its current 33% care mix. Over the last 5 years, there has been a 2.28X growth in care beds compared with 1.72X growth in retirement units at SUM.
The industry is running woefully short for the upcoming care bed demand. The care fees have a government subsidy and the unorganized players have been hit hard by the low profitability in the sector. RYM’s management claims that for intensive care units like Dementia or Memory Care, the number of beds has shrunk over the last decade as supply has left the market because of lower profitability. It is not easy to replicate the care units as they require a long learning curve.
There are significant economies of scale on both the development side and operating side of the business. The industry’s development engine is skewed very much towards RYM and SUM. For example, the existing stock market share (Retirement Units/ Care Beds) of RYM is 18.5%/ 8.8% and SUM is 10.8%/ 2% respectively, but in terms of planned supply 33% of the incremental market, supply is coming from these two companies alone. SUM is the 5th largest in the industry on existing stock terms but is the 2nd largest in terms of the development pipeline. In the NZ market, both RYM and SUM have over 2500 to 2600 units in pipeline compared with the total industry pipeline of around 15500.
The value creation happens only with a strong development engine that can enable a useful deployment of the float. In terms of 1 year forward build rate, SUM and RYM have a development pipeline of around 14% of their existing portfolio. MET has 5%, while OCE and ARV have 3%. All the other 3 primarily work on brownfield expansions and acquisitions for growth and don’t have a strong development engine. SUM and RYM are the only companies that have since inception focused 100% on pure greenfield retirement village development, allowing them to accumulate best development practices in their firms.
SUM and RYM have a land bank that is >60% of their existing portfolio size (SUM has a bigger development engine because of a small existing base and strong ramp up in build rate in recent years). On a pure greenfield development basis, Oceania, Arvida, and Metlife have a land bank that is less than 5% of their existing portfolio. Hence, these companies tend to have higher dividend payout ratios.
SUM Landbank - 2670 RU’s and 412 care beds.
Current Portfolio - 3000 RU’s and 748 care beds. 4400 residents.
Build rate - 450 units.
22 Completed Villages.
RYM Landbank - 14 new villages.
Current Portfolio - 5250 RU’s and 4000 care beds. 1100 residents.
Build rate - 781 units.
32 Completed Villages.
Even at such a large base, RYM wants to double its size every 5 years. They want to complete 2 new villages in NZ and 2 in Australia every year going forward. SUM has a good development execution engine. It has grown its development portfolio at 18.5% CAGR since inception in 1999, 12% in the last 10 years and 13.6% since its IPO in 2011. SUM with its increased financial capacity has started to focus on larger urban villages like RYM. For example, 44% of SUM’s incremental supply will come from Auckland and around 23% from Christchurch.
RYM has had a strong entry into Australia with its Weary Dunlop Retirement Village in Wheelers Hill. Ryman’s first Victorian village, was extremely well received, with the management reporting that sales of the units were the fastest they had ever experienced. The integrated approach to retirement villages and aged care are not common in the Australian market. Based on the success of Weary Dunlop and a strong waiting list for Brandon Park (an upcoming Victorian village set to open in 2018), it appears that Ryman’s unique offering is an excellent fit for the market. SUM has informed that they would also like to get into the Australian market in the near future as the quality of competitors there is much lower and size of the opportunity bigger.
The ORA model is ideally well suited for the larger operators. It provides them economies of scale as they reuse designs, procure furniture and fittings at a cheaper rate etc. The scale has also allowed them to invest in technological activities such as the myRyman app to serve the elderly better. For example, as SUM’s build rate started to move from 122 in 2011 to 303 in 2015 and 450 now, its development margins have scaled from 6% to 20%. When you operate 20+ villages, you also have the opportunity to benchmark operations, centralize functions and enable better staff learning.
Compare this with the fact that an estimated 65% of the villages are over 20 years old and 35%+ villages are over 30 years old in the industry. A lot of these villages haven’t invested back into further developments and look weary, making them less attractive when compared with the bigger villages with more amenities offered by the larger integrated developers. Unlike a normal real estate transaction, in which, you sell the product only once, the ORA model requires you to sell the units every 5 years. This ensures that the marginal capacity gets hurt badly.
There are 35K retirement units and all these units would need to be sold over a 5-7 year cycle repeatedly. So, in an economic downturn, operators with premium assets in good neighborhoods will continue to generate demand and it is those villages which are at the margins that get affected. Also, unlike a real estate transaction which happens only once and so you need to ensure a good price, in an ORA model, you can have one sales cycle in a bad environment with a lower price and then re-adjust your price in the next cycle. This transactional flexibility helps the industry to adjust pricing to generate demand and not have an independent unit idle for a long time.
Just like how investors with long-term thinking can outperform the majority of the market, I believe management cultures with long-term thinking can do better than competition overtime especially in a business like ORA sales in which the brand value really matters. Both SUM and RYM continue to be the best rated in terms of customer care. They have been taking decisions that will boost their long-term brand value. The core culture is of care, especially for RYM. Unlike other operators, both these businesses don’t do hikes to weekly prices more than the NZ superannuation increase. This is specifically very important to elderly who depend on pensions. They give back the customer deposits within 6 months, unlike other operators who might take a long time till the resale happens. They also do a lot of work to keep their communities lively with continuous engagements and focus on even small things like good uniform, tasty food etc. RYM has a slogan called, “Good enough for Mum” which basically means that they wouldn’t undertake any village initiative unless it passes the test of they doing the same if their Moms live in the village.
Hence, I believe that RYM and SUM have the best business model to make the most of the upcoming demographic sweet spot. There will be no dearth of growth opportunity for these businesses over the next 15 years and the only limiting factor would be internal execution skills.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Strong Financials
RYM and SUM have both delivered strong financials since their IPO. SUM has grown its underlying profit at 44% CAGR, net operating cash flow at 26% CAGR and book value/ share at 21% CAGR since its IPO in 2011 Similarly, RYM has grown its book value per share at 21.54% CAGR, Operating Profits at 12.5% CAGR, Operating Cash Flows at 20.5% CAGR since IPO in 1999.
As discussed before, RYM and SUM have 4 sources of earnings:
- Management/ Weekly Fees (a kind of rent paid by the residents)
- Care Fees (paid mainly by the residents in the care center)
- Development Margin (on the initial sales of occupancy rights)
- Resale Margin (subsequent resale of occupancy rights)
The Care and Management fees are bond like in their predictability and are recurring in nature. The average development margins are expected to be between 20% and 25% for both RYM and SUM, depending on the real estate market cycle.
Underlying Profitability = Reported PAT - Fair Value movement of Investment Property + Realized gain on resales + Realized development margin.
In terms of Balance sheet structure, it is important to note the size of the value of Investment property (which is a CBRE valuation estimate) and the Resident Loan (Float):
SUM:
Total Assets on BS = 1.93 Billion $’s.
Value of Investment Property = 1.81 Billion $’s (94% of assets), out of which 1.6 Bn is CBRE valuation estimates for retirement villages and 0.15 Bn is the land at fair value.
Total Liabilities = 1.3 Billion $’s.
Resident Loans = 0.87 Billion $’s. (67% of liabilities).
Net Equity = 627 Million $’s.
Net gearing (net debt/ net debt + equity) of 32.5%.
RYM:
Total Assets on BS = 5.28 Billion $’s.
Value of Investment Property = 4 Billion $’s (76% of assets).
Total Liabilities = 3.48 Billion $’s.
Resident Loans = 2.28 Billion $’s. (65.5% of liabilities).
Net Equity = 1.8 Billion $’s.
Net gearing (net debt/ net debt + equity) of 34%.
In Ryman’s case, PP&E which equals about 1 Billion $’s in assets also includes care centers, community centers, and land bank. RYM has operated with a 25%+ gearing levels for a long time. Both these companies use debt only for new village development and not to support any existing assets. For example, the 302 Million $ in net debt in SUM is lower than the capital required for its new construction and land bank.
On the P&L statement, it is the fair value movement in Investment property that dominates. From a tax perspective, the fair value movement is not taxed and hence the net tax rate for SUM was 0.11% and RYM was 1.65% in 2016. This also leads to Deferred tax liabilities of 12 Mn on SUM’s balance sheet and 72 Million on RYM’s balance sheet that are not true economic liabilities. The high float ensures that the real interest outgo is minimal when compared with the total liabilities as seen below,
SUM - Net finance cost of 9Mn on total liabilities of 1.16 Bn (0.75%)
RYM - Net finance cost of 10.6Mn on total liabilities of 3.29 Bn (0.32%)
In terms of resales of existing units, the operators report the embedded profitability or resales bank value. It represents the extent to which the current market price exceeds the price paid by the current resident for the unit’s occupancy rights. SUM had a strong embedded value of 419 Million (275 of resales gains and 145 of DMF) or 140K/ Retirement Unit and RYM had an embedded value of 775 Million $’s that would get unlocked during the next sales cycle of these units.
RYM and SUM both have their dividend policy fixed on the underlying profit basis which excludes the change in investment value. RYM pays out over 50% of its underlying profits and SUM pays out around 30% as it is still in high growth mode. On a net dividend payout ratio, RYM has operated at around 30% since IPO and SUM at around 10% since IPO.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Attractive Valuations
Since the P&L statement doesn’t truly reflect the economics of the business, I prefer valuing these businesses on the book basis. While I don’t calculate the exact value of the business, it is not difficult to see that at the current price, the stocks look attractive enough for a long-term investor. The book value can continue to compound at such attractive rates going forward. It is important to understand the key drivers of the book value.
I believe that the book is undervalued for two reasons: the conservatism of CBRE valuation and float liability accounting on the balance sheet. The entire industry (all the 5 listed players and every private transaction) in the NZ retirement village sector is valued by CBRE and they have the best database of transactions across the industry. The valuation of Investment property which makes up over 85% of the balance sheet assets for SUM or 75% for RYM is a valuation estimate for their villages provided by CBRE every 6 months once. CBRE has a 15-year transaction history on all villages across the market.
Hence, it is important to understand the inputs that go into CBRE valuation. CBRE does a cash flow analysis of the villages and the valuation estimate is the output of a 20-year discounted cash flow (DCF) model. The major inputs for the DCF are the retirement unit sale price, projected price inflation, discount rate, entry age of residents, average stay span, maintenance CAPEX, idle time between ORA sales etc. The weekly fees and operating expenses are projected to grow at similar rates and are netted out without contributing meaningfully to the valuation. CBRE estimates a 3% maintenance CapEx and includes them into its models. Everyone in the industry acknowledges that CBRE’s cash flow models and the real economic flows are very similar.
CBRE does valuation at the village level and then sums it up to get to the overall fair value of Investment property that is reported on the company’s balance sheet. SUM provides the valuation assumptions used by CBRE at the village level and RYM provides the aggregate assumptions used by CBRE for its Auckland portfolio, Australia portfolio and others separately.
For SUM, CBRE uses a housing price growth assumption of 0 to 2.5% growth over the next 4 years and 3.25% over the long term. For RYM, it uses a growth assumption of 1 to 3% in next 4 years and 3.4% over the long term. To provide some perspective, the average unit price growth has been around 6.5% CAGR since 1992. The more interesting piece is the discount rates which CBRE uses for its DCF model. For SUM, the village level cash flows are discounted at a 14.5 to 15% rate and for RYM, the village level cash flows are discounted at a 12.5% rate for its Auckland projects and 14% for others. I believe that the discount rate at the portfolio level looks very high for these businesses and there is enough margin of safety built into these 20-year DCF models because of CBRE’s conservativeness.
The float which is not really an interest-bearing liability also depresses the book value meaningfully. So, 67% of the liability is not truly economic and 85% of the assets are discounted at 14%+ rates. This combination will ensure that the book values will continue to compound at a healthy rate. Even if we consider that 60% of the costless 5-year float of resident loans as “no liability”, the SUM stock is trading at <1X book value. The float only continues to grow with newer projects.
There are also various valuation uplifts for a company with a good development engine like SUM. As the new villages move from development to operations, the IP value gets readjusted. Similarly, once the assets mature over the first cycle (5-7 years), there is a big improvement in pricing. CBRE also reduces the discount rates to 8% as the villages mature over 10 to 15 years. The value of IP also doesn’t provide any value for the new profitable development projects that SUM can add overtime to its pipeline.
On Price/ Book trading multiples, the current P/B, peak P/B and trough P/B for Ryman is 2.8X, 5X (2007 and 2014) and 1.5X (2008) respectively. Similarly, for Summerset, it is 1.7X, 3.1X (2014) and 1.4X (2011) respectively. As we can see, both these stocks are currently trading at below their average Price/ Book. I believe that a lot of the real estate price risks are more than factored into the current price. The current dividend yield for SUM is 1.7% and for RYM is 1.9%. Also, SUM is trading at around 16X underlying profit and RYM is trading at around 25X underlying profit.
All in all, it is not difficult to understand that paying 1X adjusted Price/ Book for a business that can continue to compound at high teens is attractive enough for a long-term investor.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Major Risks
I believe that there are some serious short-term risks to these businesses if the real estate price drops meaningfully. For example, the book value would be adjusted by CBRE to adjust for the new prices. The decrease in float will increase the liquidity issues. The cash flows from just pure operations would not be able to balance the float reduction and hence the gearing ratios would increase. For example, the Care Fees/ Development Fees for RYM was 228 Mn/ 61 Mn and for SUM was 57 Mn/ 28.5 Mn. The lower proportion of care fees will ensure higher volatility in their economic results, especially for SUM.
Despite these, the long-term risks wouldn’t be meaningful as the ORA model provides transactional flexibility to get back higher prices once markets improve even if prices drop temporarily. Also, a majority of the demand is from needs based and hence demand would be resilient as shown by Ryman’s performance in 2008. Even in the last half year when the real estate transactions had dropped by 20%, RYM was able to report a 6% increase in its transaction volumes.
The long-term risk is high only if you believe that real estate prices would be on a meaningful structural downtrend for more than a decade. I am no expert on NZ real estate, but it is not difficult to understand that for a small island which has concentrated urban clusters that continue to attract a lot of immigration, the value of real estate should more or less track inflation. To check if there is any massive bubble, I tried to normalize interest rates from the current 5.1% to 7.2% and even then only around 41% of the median household income will go towards interest payments which is not very different from the 35% of median income that was prevalent 20 years back when interest rates were 9.9%.
There is no overbuild risk in the NZ retirement sector. The market continues to be very tight. From the JLL sector numbers, even with an assumption of no increase in penetration rate, the market can absorb 1600 incremental retirement units per year. This number goes up to 3500 units per year within the next decade. The current supply pipeline will be in shortage even if the penetration rate increases at 30bps. There is no visible oversupply in the market except for 1 or 2 sub-pockets of Auckland. The big demand uplift from demographic sweet spot will help the sector.
The inventory across the retirement village industry continues to be very low. SUM sold 660 ORA’s in last 12 months (400 new sales, 260 resales). The uncontracted stock availability for SUM is 2.2% and the total resales stock availability for RYM is 0.8%. These levels are among the lowest the industry has witnessed. Similarly, the care occupancy at RYM is at 97% and over 88% across the industry. According to SUM’s management team, a below average village in their portfolio has around 13 people on the waitlist and a good performing village has as many as 100 people on the waitlist. All indicators point to strong demand in the near to medium term.
We did have a chance to speak with both the management teams and industry people. We couldn’t find obvious red flags in either of these businesses. RYM has had a stable management for 2 decades. The founding family continues to be active in the business at the board level and that provides a sense of long-term orientation which is a big positive. Insiders own 18.3% of the business. Meanwhile, SUM has a young professional management team and board with a limited track record. The overall insider ownership is only around 3.35%, but the company has been using ESOPs as an important management incentive tool. The shareholding is widespread among professional investors.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Conclusion: As discussed before, RYM and SUM are compounding machines that offer long-term investors quality, growth, and longevity at an attractive price. It is very clear that Ryman is the thought leader in this business and the one that has the best culture and long-term track record. Summerset is the challenger with a good team that is trying to replicate Ryman and become a credible no:2 in the market. Some of the premium for Ryman is definitely well deserved.
The reason for structuring the trade on both these businesses instead of selecting one is that we like the quality of RYM, but the delta, margin levers and valuation differential of SUM makes it equally attractive as well. So we are playing this as a combo bet and would be dynamic in adjusting our position sizing going forward based on any incremental information.
Disclaimer: We are invested in both the stocks mentioned above and thus have vested interests in the same. We may or may not inform the readers of this platform about our future actions on this investment position. Our accumulation level on both these stocks is 5 to 10% below current market price.