Is development picking up and will credit markets slow it?
As the housing market continues to slump, with the worst yet to come as the number of teaser-rate mortgages resetting to market interest rates balloons this autumn, there is growing concern in some corners of the seniors housing industry that this could have a negative impact on census and profitability. How negative is anyone’s guess, and it will have more to do with how long the current housing slump lasts and what its ultimate impact will be on the rest of the economy. Or will it?
There are two distinct issues that we are dealing with here. First, will the current housing market slump cause the elderly to delay their decision to sell and move into a retirement community (of any type) because they can’t get their price, or even a buyer? Second, is new seniors housing development rebounding to such an extent that the supply will surpass the potential slowing demand (see issue one), resulting in a slowdown or end of the four-year industry recovery? Add to this the credit crunch in the worldwide debt markets and you could have the perfect storm of decreasing credit availability, higher cost of debt when it is available, higher cost of existing floating rate debt on seniors housing properties and a slowing acquisition demand. Perhaps.
When a perfect storm occurs, as made famous in the book with that title by Sebastian Junger, not much is left in its path but death and destruction. But even a so-called perfect storm in the seniors housing market would not have quite the same impact. For one, the seniors housing market is extremely diverse with many different product types and price points. Second, about one-third of the product is need-driven as opposed to choice-driven. If skilled nursing beds are thrown into the total supply, that percentage increases to about 60% or higher for need-driven beds/units, depending on your assumptions on the number of assisted living and skilled nursing beds within CCRCs. Third, the fundamentals of the overall seniors housing business are quite strong right now, so it would take much more than a housing market “slump” to derail the market. Remember, for every subprime loan out there, there is an elderly woman or couple with a house purchased for $40,000 some 30 years ago, with no mortgage and a vastly higher current value, who has a lot of pricing flexibility when it comes to selling.
As mentioned, there are many different types of seniors housing properties, and they will all be impacted differently by the current external factors. Similarly, the residential housing market is not the same in Michigan as it is in California. When looking at the sales of existing homes in July, there was an overall decline of 0.2% nationwide, but the number of sales was actually up 1.0% in the Northeast and up 1.8% in the West. That is why we must be careful when looking at “national” trends of new construction in seniors housing. It all comes down to the local markets, and we have not heard of many local markets undergoing a new development boom that would harm overall occupancy rates nationally.
The early release of some preliminary statistics from the upcoming Seniors Housing Construction Report 2007 from ASHA and NIC apparently shows a startling increase of nearly 32% in new independent living, assisted living and dementia care units/beds under construction in the 75 largest metro markets in March 2007 compared with March 2006. The actual numbers (preliminarily) are an increase from 18,597 units/beds under construction in the year-ago period to 24,493 units/bed this year. If the IL units within CCRCs are stripped out, the change is from 10,371 to 14,126, according to a report from Stifel Nicolaus, or an increase of 36%. Also excluding CCRCs, the number of IL units under construction increased from 3,109 to 5,656, while the number of assisted living units/beds increased from 5,961 to 6,753.
We don’t mean to bore you with these details, but it is important to understand them in an historical context to try to determine if we have a looming problem. The peak of the seniors housing development market was in the 1997-1999 period, with the majority of the activity in the assisted living market, and we all know what happened in the following years. According to ASHA’s Seniors Housing Construction Report 1999, a total of about 48,000 assisted and independent living units were built in 1999, plus more than 17,000 units/beds in CCRCs. That was obviously more than the market could bear, but it was also 2.5 times the number today. Admittedly, at worst we are comparing apples to oranges, but at best Macintosh to Red Delicious, because the methodology in the two reports is different, but we are confident in the overall comparison and result.
Breaking it down further at the peak, there were about 30,000 assisted living units built in 1999, compared with just 6,700 today, and about 16,000 independent living units (excluding CCRCs) in 1999 compared with less than 6,000 today (also excluding those within CCRCs). So with a high degree of confidence we can say that today we are nowhere near the overbuilding dilemma of the last decade, not that anyone thought we were.
Now, let’s compare today’s market with the lows in new development, which would be in 2002 and 2003, according to the ASHA reports that came out in those years. In 2002, about 8,000 independent living units were built (30% higher than the level today) and about 3,500 assisted living units (50% lower than today), and if you include CCRCs in both sets of numbers, the total in 2002 was about 17,000 units compared with 24,500 in 2007. So the industry’s new development today is about 40% higher than the absolute bottom of the market when Sunrise Senior Living (NYSE: SRZ) and Holiday Retirement Corporation were about the only major companies continuing their development pipelines.
This does not, however, tell the full story, because the three years from 2004 through 2006 were relatively slow years with regard to new development. Again, the methodology is different, but there are no other viable statistics. The number of new units increased by 18% from 2003 to 2004, but were flat (-0.8%) in 2005 and then plunged by 20% in 2006. In fact, the number of new units in 2006 was lower than in 2003, one of the slowest years in the past decade, so the sharp increase in 2007 really represents a relatively small increase of 4% over both 2004 and 2005 development activity. No one knows why there was such a large percentage drop in 2006, and presumably the data collection is at its best today, so the numbers from several years ago may even understate the situation when compared to recent data. In any case, the development market is in a bit of a catch-up mode after six relatively lackluster years. The current data is based on the largest 75 metro areas, whereas in the past the data was from ASHA’s survey results based on the largest 90 to 110 operators. The NIC MAP data covers about 60% to 65% of the national market, while the ASHA data may reflect between 50% and 70% of the national market, which is why we think they are comparable enough in a general sense for trend analysis.
It is time to turn from statistics to current market specifics. Most of the publicly traded seniors housing company stocks dropped in the second quarter of this year, and some quite significantly. One rationale given was that the growing weakness in the housing market would eventually make its way into the seniors housing market, with the elderly delaying a move out of their homes until better times. The problem with that analysis is that three of the six publicly traded companies are primarily in the need-driven assisted living business, where move-ins are not usually timed to one’s advantage; rather, they are caused by an episode of some sort. Consequently, a weak housing market should have a relatively small impact on those companies. So the market meltdown was either an overreaction or an acknowledgement that some seniors housing stocks got a bit ahead of themselves from a valuation perspective. It was probably both.
Some of the second quarter results scared investors because almost every company reported second quarter occupancy that was lower than the year-ago quarter or the previous quarter. The one exception was Emeritus Assisted Living (AMEX: ESC), which reported an occupancy increase from 84.6% in the year-ago quarter to 85.9% in the current quarter. The big decline came from Assisted Living Concepts (NYSE: ALC), which had a 350 basis point decrease from the year-ago quarter and a 280 basis point sequential decrease from the first quarter of this year. But this is like comparing apples to oranges, because the company has been exiting the Medicaid waiver business, causing an intentional, and presumably temporary, drop in occupancy. So despite the occupancy decline, the private pay census has moved up from 71.2% to 76.7% of total census in one year, and it should be well over 80% next year if all goes by plan.
ALC was followed by Sunrise Senior Living, with a 220 basis point decline in same community occupancy, the result of flat move-ins but a 6% increase in move-outs when compared with the year-ago quarter. Much of this was reversed in July alone, when move-ins were up more than 5% and move-outs were down almost 6% from July 2006. More importantly, same community independent living units, the product most vulnerable to a housing market downturn, are at 94% occupancy, which the company considers to be “full” occupancy. Brookdale Senior Living (NYSE: BKD) seemed to have the most explaining to do with regard to its CCRC business and an unexpected decline in entrance fee revenues, which is most likely related to the housing market, but these communities represent less than 5% of its facilities. Overall occupancy fell just 10 basis points from the first quarter of this year, and while up from the year-ago quarter, that comparison is not relevant because of large acquisitions since then.
Private companies don’t usually disclose occupancy levels, but we did learn that Atria Senior Living is currently at a record occupancy level and has increased its occupancy by 150 basis points since May. Of its 124 facilities, about 15 are at 100% occupancy with a waiting list and another 45 to 50 are at 95% occupancy or better. Inquiries and tours at facilities are both up, but one large independent living community has slipped from 100% occupancy to just under 96%, with the weak condo market in its location the most likely culprit. Overall, the company is very comfortable in the current market environment.
Another large private company, Merrill Gardens, is still running at a 95% occupancy rate for its portfolio of properties (excluding those still in fill-up). In one acquisition that closed at the end of 2005, which included six communities with 980 units in northern California, the company has pretty much met its targeted goal of increasing occupancy while increasing rates on occupied units and bringing formerly vacant units to market rates. From the time negotiations began to when the transaction finally closed, occupancy had dropped from close to 90% to about 86%. Today, the six-property portfolio is at 94% occupancy with an average margin for the group of about 45% before management fee.
With staffing efficiencies, rent increases and the census increase, we estimate that the company has increased the net cash flow of the acquired communities by at least $4 million in less than two years. Even though most of this portfolio is independent living, they have not yet been impacted by the problems in the housing market, and in their markets there is not much chance that new development will be a threat. Management believes that a prolonged housing slump may have a 1% to 2% impact on census, but what is key to understand is that it usually represents just a delay, and when the housing market turns back, you get an extra bump in census from those who delayed a move combined with others just starting the process.
If there is concern about future overbuilding, that may have been put to rest this past month as lenders are reacting to the liquidity problems in the debt markets by tightening up their lending terms, raising borrowing costs and generally being a bit more conservative with loan requests. As an example, if permanent financing lenders now insist on a lower loan-to-value than has been the case in the past few years, the construction lenders will decrease their loan per development, because the construction lenders want to make sure that the permanent mortgage available to the developer will be large enough to take them out. Assuming the current credit crisis stays with us for a while, this will put a crimp on future development no matter how you look at it. The reality, however, is that new development often benefits the existing communities in the local market, which then become the low-cost producer because of the current high development costs.
The current healthy industry environment may become even healthier because an impact can already be seen in the acquisition market. Prices, cap rates and return expectations will be rationalized with the events of this summer, and that is good for the industry, especially those operators that want to make strategic acquisitions. The sellers are the ones who will have to change their expectations, but they will still be better off today than in eight of the past 10 years. Waiting for the next rally to sell, however, may be a long wait.