What’s Wrong with Health Care REITs?

Three reasons the REIT structure isn't working for health care facilities

Health care was once considered a defensive REIT sector, but that story was starting to crack even before Covid, and now their earnings are not coming back as fast as many have predicted. What’s going wrong?

In this post I want to lay out three fundamental concerns that go beyond the direct impacts of the pandemic. I won’t get into any individual REITs or their current valuations; these are just three uncomfortable questions in the REIT world that might benefit from an explicit definition and a wider conversation. I’m not a health care specialist, and I’m willing to have my mind changed. But let’s get them on the table.

1. The “gray wave” may be a mirage

This point is mostly about higher end senior housing operators like Sunrise or Brookdale. Financing these facilities is not the only thing health care REITs do, but it’s where most of their earnings growth is expected to come from.

And they’ve already been overbuilt more than once. But the North Star of every senior housing bull is the coming “gray wave” of Baby Boomers aging into the target demographic.

What is that demographic exactly? 80+ is one common definition, and of course it’s not a “mirage” that the annual number of Americans turning 80 is about to increase sharply. My problem is with the implied penetration rate of senior housing at any given age beyond that. Most industry forecasts seem to implicitly hold this variable constant, or sensitize it higher — but shouldn’t we be sensitizing it down?

Because there seems to be a consensus in the industry that the average move-in age and acuity have already been trending higher over time. And we know that affluent Boomers are the first generation that moved their own parents into these settings — and even when it was a positive experience, it’s made them better informed about how to put off the same decision for themselves. So why should we be so confident they won’t succeed?

I’m not denying that this is largely a need-based product, and there will be millions of new assisted living residents in any scenario. But there are also plenty of reasons to believe that this upcoming cohort, on average, will find ways to stay in their own homes for a bit longer — and that could be enough to flatten out much of the demographic bump that the industry is counting on.

This is not just about the steps that seniors can take proactively (like “aging in place” renovations) or what they can advocate for politically (like better home care reimbursement). It’s also the changes in the broader consumer tech landscape, right? For example, losing the ability to drive safely can be one last straw for staying at home, but how much have ride-sharing apps changed this dynamic? And what about all the forced adoption of delivery apps, FaceTime, etc. during lockdowns?

Now, there are also some differences from prior generations that cut the other way, like fewer non-working adult children living nearby. But I have never seen a forecast that goes into enough detail on all these factors to effectively net them out.

And when you dig into these industry forecasts (here is one good example) you’ll see that what they’re forecasting is not exactly what we should be focused on from the REIT side. For example, why would we use a single cutoff (75+, 80+, 85+) instead of brackets (75-80, 80-85) or annual cohorts? Why do we care at all about the penetration of total “senior housing,” rather than the individual facility types with very different target populations? And for that matter, why is the huge variance in penetration across markets still a “puzzle“?

Sometimes half the battle is just asking the right questions. So let me try to lay out my own view as explicitly as possible:

  1. The most relevant metric for a REIT investor would be a same-facility type, same-resident penetration rate over time. For example, what % of 84-year-olds with $X inflation-adjusted net worth are in an assisted living setting in year Y vs year Z? And this is a bare minimum — it doesn’t get into the ways that supply is creating more demand at the margin, or the profitability of that marginal resident. But it would be a place to start.
  2. My guess is that the trends on this basis would already look worse than the (flattish) pre-Covid long-term penetration trends that we’ve heard about in the industry. (Although they might vary widely by facility type.)
  3. My forecast is that the trends on this basis will start weakening even faster as Boomers enter the target demographic.

You could certainly argue for a different metric at the industry level, but from a REIT perspective you’d just be rearranging the problem. (For example, if the solution is going to be new formats or more urban locations, then we need to factor in more obsolescence risk in what REITs already own.)

Again, if there’s evidence that would change my mind about this, I’m eager to see it. But in the meantime, if I’m right about this demand “mirage”, what would the leading edge of it look like? I think it might look a lot like what we were already seeing pre-Covid: an increasingly frustrating game of kick-the-can, with non-recurring rent cuts starting to look more recurring, and the upward inflection in REIT forecasts always 6-18 months away — never more, never less.

2. No leverage with operators

REITs aren’t allowed to operate senior housing themselves, and in the REIT world we’ve been very focused on finding the “right” structure to work around this and align owner/operator incentives. First we lobbied for the 2007 RIDEA Act that allowed management contracts instead of leases, and since then we’ve gone through various iterations on how to structure these management contracts, restructure leases and JVs, and so on. On a parallel track, we’ve also gone through multiple sentiment swings about the merits of nationwide operator rollups (like BKD) in comparison to smaller regional managers and developers.

I think we’re gradually converging on some better answers, but I’m starting to understand why the progress has been so slow. This is just a very operating-intensive business, where management talent is scarce and it’s difficult to scale it over too large a portfolio. And there is more value being created on the operating side (and less on the real estate side) than REIT investors would like to believe. So in that environment — a permanent shortage of talent, a permanent glut of capital — the risk-adjusted returns should always flow towards talent, right?

And there are many ways for that to happen in the terms of a lease or management contract, but ultimately these execs can also just work for a different operating company, or start their own. So most REIT/operator relationships are just inherently more one-sided than they appear. Even the largest operators have never been very well capitalized, and we’ve seen that the REITs will cut their rent over and over again to keep them afloat — just as the “personal guarantees” at the smallest operators aren’t always called in. And we can’t just blame this on REIT management teams for not driving a harder bargain; to some extent it’s built into the structure of the industry.

With leases for higher-acuity settings like nursing homes or hospitals, some of the mechanisms are different but the principle is similar. As value flows into ancillary businesses like home care or ASCs, they are more likely to be outsourced, sold, merged — whatever it takes to separate them from the real estate and get them out of reach of legacy landlords.

Again, the details are complicated, but the basic principles are not. If we haven’t seen them from the REIT side, I think it’s partly because we haven’t wanted to see them.

3. Structural exposure to rising wages

I know that some readers would offer a more cynical version of the incentive problem above — in which the key “talent” is not for better and more efficient operations, but rather for overbilling Medicare, exploiting workers, or skimping on patient care. For our purposes here I’d prefer to focus on the positive story, because I know there are also a lot of good operators who are trying to provide the best care possible in an impossible system. And a regulatory critique of the industry is way beyond the scope of this note.

But one thing we can’t ignore is that health care REIT investors have effectively been betting for a long time that below-inflation wage growth for hourly health care workers would last indefinitely. And that bet is finally coming due.

In contrast to #1, this point is a bit less focused on private pay senior housing — where it does seem like there’s pricing power to push at least some of these rising wages through into rents — and more on higher acuity facilities that depend on government reimbursement, like nursing homes or hospitals.

And let’s be as clear as possible. Some of the current labor crunch certainly relates to Covid surges. Some of it may be anti-competitive behavior at the staffing agencies, although that’s a rich accusation coming from hospitals in particular. Some of it may still relate to school schedules and other lingering Covid disruption.

But at this point it’s getting silly to treat “extra” temp staffing costs as a completely independent variable from base wage growth, or as an excuse not to lay down clear forecasts for base wage growth on a per FTE basis. The dust has cleared enough to see that the permanent landscape has changed. Again, if we aren’t seeing that in REITland, it may be because we don’t want to see it.

And the most important change relates to compensating wage differentials, a subject that’s still not getting enough attention anywhere. Whatever your forecast for the overall labor market, I don’t think anyone would deny that front line health care jobs have gotten more difficult relative to other employment options than they were before Covid, and that they’re going to stay that way for the foreseeable future. And all else being equal, if a job is more difficult relative to other options, it will have to pay more.

By the way, this effect could be even greater within health care. For example, we talked in the last section about how operators can move better-reimbursed or more profitable revenue into other settings, like hospitals moving outpatient surgery to ASCs. But have we considered that an ASC or a home care agency is also an easier place to work than a hospital? And that this gap may have grown exponentially during Covid, as ERs fill up with angry anti-vaxxers?

Again, this is the bet we signed up for. All those long-term leases with tight coverage and inflation-plus rent bumps were a bet on squeezing labor. I’m not saying that was the only factor, but it’s absolutely part of the bet that REITs were making. And I’m sorry if it feels uncomfortable to phrase it that bluntly, but I’m not moralizing or trying to lecture anyone. It is what it is. REIT investors underwrote a different labor market than the one we’re in. Who’s going to pay for that?

Not the operators, as we covered in point #2 above. If you think it’s ultimately the government, then in many cases you’re probably right. But we’re seeing right now how slowly those gears can turn. And there is no reason not to expect more permanent rent cuts along the way — because while there’s a strong public interest in keeping these facilities open, there is no public interest in preserving REIT dividend yields.

Don’t these REITs own anything else?

Yes. Mostly medical office buildings. And the risks to MOBs — rising rates, retailization, telemedicine — would be worth a separate public note at some point. But right now I might be arguing with myself. I don’t hear anyone getting excited lately about the dedicated medical office REITs (a whole other sad story), and at the diversified health care REITs, investors have treated MOBs more like ballast.

There are some other exposures (life science, international investments) that I generally feel better about. And of course my three points above are not binary doomsday arguments. There are ways for REITs to invest around them, and there are some REITs doing it better than others. But at the sector level, even one of the three could be a serious long-term problem.

Subscribe to The Drawing Board

Sign up now to get access to the library of members-only issues.
Jamie Larson