Now Is Not The Time To Sell A Strategic Portfolio
Just a few years ago, owners and investors of seniors housing portfolios thought that there would be a new wave of buying interest from Australia, especially given changes in their pension system which would result in more funds available for investment than could be put to profitable use down under, not to mention the strength of the Australian dollar. Our lead story in the December 2007 issue was called “Aussies Buying Seniors Housing,” and featured two significant acquisitions by Australian companies. This included FKP Property Group’s (ASX: FKP) purchase of a portfolio at $216,500 per unit and The GPT Group’s (ASX: GPT) second purchase of a 95% interest in the remaining 15 properties managed by Benchmark Assisted Living for $268 million, four of which were leased facilities. GPT’s initial investment in the U.S. seniors housing industry came in late 2006 with its purchase of a 95% interest in 19 communities operated by Benchmark in the Northeast as well as a minority interest in the management company.
Just a short year ago, GPT’s plan was to use the Benchmark portfolio as its platform to significantly expand in the U.S. seniors housing industry. The marriage with Benchmark seemed to be the perfect fit because GPT usually has a long-term investment horizon, up to a 20-year holding period for real estate, and Benchmark’s CEO, Tom Grape, had to be growing weary of dealing with what seemed to be a constantly changing group of owners of his properties. He had built up quite an attractive geographic concentration in the Northeast and we assume he wanted to spend more of his time expanding on that strength instead of dealing with more recapitalizations of his properties.
GPT is a huge real estate holding company with assets across Australia in the retail, office, industrial and lodging markets. Oddly enough, the company does not own any seniors housing assets in Australia and this was only GPT’s second acquisition in the U.S. real estate market. This was a bit different from FKP’s foray into the U.S., because it was the largest owner of seniors housing in Australia, with a 15% market share, when it completed its first U.S. seniors housing acquisition a year ago. GPT was so confident about its expansion in the U.S. that it lured Kathy Sweeney away from Boston-based AEW Capital to become GPT’s managing director of U.S. seniors housing with the mandate to grow the business and, with her office in Benchmark’s headquarters, keep an eye on the $693 million investment already made. It was a portfolio she knew all too well, since AEW was an early investor in Benchmark’s properties. GPT was looking for a steady 7% cash-on-cash return, which certainly seemed attainable at the time given the markets Benchmark’s properties were in, the census and the relatively high rates they could command.
That was then and this is now, as the saying goes. Back home, GPT is suffering from the effects of the credit crisis and slowing world economy. The company’s first half of the year earnings were down significantly from a year ago, it has put about A$1.3 billion of Australian real estate assets up for sale and its share price has plunged more than 60% in a year. Sounds like the typical company in the U.S. Even though the U.S. seniors housing portfolio represents just 2% of GPT’s assets, it is now looking at possibly shedding this portfolio “over time” to redeploy its capital back home. Timing is everything, and in our humble opinion, we believe that GPT is making a mistake with its Benchmark portfolio.
We can’t blame GPT for entering the U.S. market right at its peak, since the prospects were still bright in late 2006, even though they were beginning to dim with the second Benchmark purchase a year later. Unless you are desperate, you don’t buy high and sell low. Even though we are really not at a low at this point in time, the credit freeze is certainly pointing the direction downward, and optimism about anything is fading with each 100-point drop in the Dow.
The Benchmark investment represents just 2% of GPT’s A$14 billion portfolio, it is still profitable with properties located in good markets, occupancy appears to have stabilized and our bet is that cash flow next year will be higher than it is now. Combine this with what is going in the credit markets, with only a few buyers who could consummate a deal of this size at a price anywhere near what GPT paid, and there is no rational reason to explore a sale at this time. Publicly, that is all they have said they are doing, and that they may do it over time. But to what end? Although our opinion has not been sought, we would advise GPT to wait.
At the end of 2007, GPT’s 34-property Benchmark investment had an overall occupancy rate of 91.4% (we don’t know about the additional nine properties that Benchmark manages). By the end of June 2008, occupancy had dropped by nearly 300 basis points to 88.5%. That news surprised the U.S. investment market when GPT started talking about some issues with its U.S. portfolio, which was partially responsible for seniors housing stock price declines last June. Apparently, Benchmark held firm in its markets and did not start price discounting until sometime after March of this year. When they joined their competitors, occupancy started to rise, and while we don’t know where the bottom was, we assume it was below the 88.5% mark. In an interesting tidbit, while Benchmark’s occupancy was more than 100 basis points below the Boston MSA average, its revenue per occupied room was above the average, which could mean many things, including more services, higher acuity of residents or anything else. The point is that there were some mixed signals when being compared with the local market, and while occupancy is important, for the long term there are other crucial factors as well.
In the case of GPT’s 34 Benchmark facilities, eight are 100% Alzheimer’s, and these eight had an average occupancy rate as of June 30 of 93.4%, compared with 87.7% for the rest of the facilities. Obviously, this can be the result of location, age, design and a host of other factors, but in general, well-run Alzheimer’s programs tend to have higher occupancy rates, especially in a market environment like this one. Benchmark has taken notice and has begun to increase its Alzheimer’s units in some Connecticut and Massachusetts facilities. These “value added” projects are some of the highest return projects being implemented at Benchmark and will contribute to what we believe will be a higher cash-flowing company next year. In one facility, a new Alzheimer’s wing opened last October and occupancy now averages 95%.
Getting to valuation, which is admittedly a difficult task in this market, especially for a large portfolio, our records show that GPT paid approximately $693 million for a 95% interest in its 30 Benchmark properties and the four leased properties, and a 20% interest in the management company, the market value of which is clearly speculative. Based on current occupancy and average rates, we estimate annualized revenues to be about $180 million and with GPT’s reported EBITDAR margin of 27.4% at June 30, we derive about $49 million of cash flow before rent payments on the four leased properties. There may be some expenses in that number which may not be applicable or relevant to a buyer, so after rounding up for that and deducting an estimated lease payment, we come to EBITDA of about $46 million. Using a 6.5% cap rate, close to the Health Care REIT (NYSE: HCN) deal for the Sunrise Senior Living (NYSE: SRZ) properties HCN is buying from Arcapita, we derive a theoretical value for GPT’s 95% interest of about $670 million, or 3% less than its original purchase price.
The Sunrise portfolio had a 94% average occupancy rate and we guess would have to be considered of a higher quality than the Benchmark properties. The HCN price, assuming the deal closes as planned, comes to about $343,000 per unit, compared with a theoretical value of $239,000 per unit for the Benchmark properties, which is understating it a bit because of the four leased buildings. At a 7% cap rate, GPT’s investment would fall to $624 million, or 10% below the original price. Finding a buyer today at a 6.5% cap rate would be difficult, to say the least, and even 7% or higher would not generate any kind of a buying frenzy. The only buyer that would make sense in this market is a REIT, possibly taking advantage of the new regulations.
To fit with GPT’s potential phased withdrawal from the U.S. market, it would make more sense to sell a portion of the assets first, and a REIT may want to take that step as well. The upside for GPT is that they would most likely achieve a higher price down the road in the second sale and it would put competitive pressure on that first buyer to come out on top in the bidding for the second sale, assuming all went well with the first group. Remember, for every 100 basis point increase in occupancy, the value of GPT’s Benchmark portfolio will increase by about $20 million, all things considered equal. The increase will even be more pronounced when price discounting ends and valued added services are expanded. The flip side of this is that for every 20 basis point increase in cap rates, the value declines by about $20 million. The increase from what is the $64 question.
We know GPT is going to do whatever it needs to do to bolster its share price and strengthen its balance sheet at home. It is just that a sale of the Benchmark portfolio at this time would seem to have little added value to the parent company given it is such a small part of the total. And, we might add, Benchmark is not in a distressed situation. While we can understand why GPT may want to exit a market (both the U.S. and seniors housing) where they have little knowledge and presence, if they can get their domestic books in order, now would seem to be a better time to buy in the U.S. than sell, especially in the seniors housing market. So give it a year and see if occupancy and cash flow have increased and then make the decision, but don’t sell at a potential loss in this market.