The Dead Zone

Usually my public notes are more about tenants and operators, but this one is very REIT-centric. I think the problem I'm describing is one that every REIT-dedicated investor has some version of in their head, but we tend to discuss it vague and circular ways. By giving it a scary name and laying it out as explicitly as possible, I hope to break some of that logjam.

So if you're not a dedicated REIT investor, this note will explain a lot of the recurring discussions in REITland that might otherwise seem like odd rituals. And if you're a participant in those rituals, I hope you'll reply and tell me what I’m missing, or what your personal "dead zone" parameters look like.

What It Is

A REIT is in the dead zone when they meet all three of the following criteria:

  1. not growing FFO/share (on a trailing basis)
  2. trading at a material NAV discount (let's say 15-30%)
  3. at or above their leverage targets
  4. It's common for REITs to make a brief stop in the zone, and many have just done so during Covid. But the longer you stay there, the more difficult it is to leave. It almost doesn't seem to matter how you got there. I won't pick on any individual REITs in this note, but there are a few that have been stuck in the dead zone for almost a decade.

How Not to Get Out

There are two common strategies, which I'll call "upgrading" and "arbitrage." The first is usually favored by management, the second by institutional shareholders. Whatever their merits outside the zone, neither one seems very effective for breaking out of it. But why not?

Upgrading

The standard management strategy is some variation on the following:

  1. We can improve our portfolio on a leverage-neutral basis by selling the bottom [20–30%] to buy higher quality assets
  2. Of course this won't be accretive in the near term, but it doesn't have to be too dilutive (e.g. if we leg into it by buying first) and it will jumpstart our growth in year 2 or 3
  3. In the meantime, it should improve our same-store metrics and lower our recurring capex, and be no worse than NAV-neutral
  4. The market will "recognize" the higher portfolio quality, underwrite the future growth, and re-rate us before it arrives — which will give us a better cost of capital to grow further

If you come back a year later — or two, or three — what you'll often find is that they've executed on the first three steps, but the fourth one hasn’t happened. And frequently they'll have found more non-core assets to sell, or something more value-add to buy, or some other way to push the goal posts back another year. What's going wrong?

This is where it usually gets hand-wavy. But let’s lay out three kinds of things that could be going wrong, with an example of each.

The first is that the offsetting frictional costs of portfolio turnover are higher than we think. This is not just about the fees and other one-time items that are added back to adjusted earnings. For example, the greatest transaction cost may be crossing the bid-ask spread on both sides — especially because the plan is (almost by definition) to be a liquidity taker, and sell out of favor assets to buy more crowded ones. So if we mark our NAVs to the transaction price, rather than the cover bid or the midpoint, then we’re missing a little bit of NAV erosion with every transaction.

The second possibility is an attribution error in these assumptions about what the market wants. The details vary a little by property type — but for example, if you're looking at peers with a higher quality portfolio, higher forecast earnings growth, and a higher multiple, then the causal link between those qualities may not be as simple as it looks. For example, these higher valuations may be a reward for a long and stable track record of prior growth.

The third area to think about is signaling and credibility, especially for a REIT that’s tried the match-funded “upgrade” strategy before. How does the market know you’re done this time? And if there’s likely to be another upgrade in five years, with all the same frictional costs again, shouldn’t that factor into today’s valuation?

Arbitrage

The standard investor recommendation is a lot simpler. Instead of selling assets to buy better assets, you can just sell assets to buy back your stock. You'll be locking in NAV accretion, and there's zero execution risk in buying your own portfolio. If you sell average or better assets instead of the worst ones, this may even be accretive to FFO. And the market will have an easier time re-rating you for real value creation today than promised value creation in the future.

It may be obvious from that description that I have a buy-side background, and I'm more partial to this strategy than the one above. It's just math, right?

Well, there are some technical objections that I don't want to dismiss completely — about taxes, for example — but at the level of NAV discounts we're talking about, they rarely add up to much of an offset. So yes, the NAV creation is just math, and for many investors that's enough. Even a static discount to a higher NAV is still a higher stock price, right?

But the question of whether it's a credible signal for re-rating is even more slippery than with the upgrade strategy above. The first answer from REIT investors might be that it only "doesn't work" in this respect because it rarely ever happens — or when it does happen, it's a minor and reluctant add-on module to the "upgrade" strategy above. REIT management teams have been so reluctant to fully implement this strategy that we just don't have much of a sample.

The obvious follow-up would be "why are you still voting for these Boards then?" but that would take us into a separate discussion about corporate governance. What I'm focused on here is just the market response. Are there any concerns with this arbitrage strategy that should offset a re-rating towards NAV?

The one that I increasingly come back to is just scale. It's getting more difficult every year for smaller REITs to stay on investors' radar screens, to compete with private real estate for executive talent, and so on. So this aversion to "shrinking the company" on the management side is one that I'm getting more sympathetic to. At a certain point you're either guiding the company towards a sale (which would bring us back to the governance debate) or you're not.

Got Any Better Ideas?

Yes, but you might not like them:

  1. Increase target leverage, and just buy assets without selling them. Not a "temporary" increase that you're going to work back down, because that creates a new valuation overhang. Just take your target leverage up permanently by a turn or two. You're only paying half the transaction costs, you're gaining economies of scale, and the added volatility in your stock might actually get more investors to pay attention to it
  2. Match fund investments at a positive spread — e.g. by investing in riskier assets, mezz debt, secondary markets, new property types…
    (One thing I would not include here is development; in general I would love to see more ground-up development from REITs, but as a strategy for getting out of the dead zone, it just doesn’t work fast enough.)
  3. Rip off the Band-Aid — If you really have to do a lot of dilutive non-core asset sales, it's better to do it all at once in a spinoff, portfolio sale, or at least a quick process. Take the earnings hit now and start growing.

Yes, I know about Modigliani-Miller and the other textbook finance objections to these proposals. I know they sound gimmicky and short-term oriented. I know I'm also contradicting some sound instincts on the real estate side, like taking time to get the best price on each disposition.

But the “upgrade” and “arbitrage” plans above are smuggling in some of their own assumptions about market structure that almost certainly don't hold. And remember, we're not talking about the average REIT here. Often the REITs that should worry most about leverage and complexity (like heavy developers) are not the ones that get stuck in the dead zone to begin with.

So this is what I meant at the beginning about jogging the debate forward. I will readily admit that my suggestions are a little more trading-oriented than investment-oriented. But they’re based on the simplest trading principle there is (“earnings move stocks”) and in a situation with no good options, simplicity should count for a lot.

How to Stay Out in the First Place

  1. Trade your stock. Out of the three components of the dead zone, the NAV discount is the one that management has the least control over. The stock market is a strange place, and getting stranger. You could do everything right and be in the wrong Reddit post tomorrow.

    But that's all the more reason that every REIT should have an ATM plan and buyback plan in place all the time, and be ready to buy at a 10-15% NAV discount or sell at a 10-15% NAV premium. Every REIT. All the time.

    I am actually not a stickler for NAV in that +/-10% range, and with a compelling use of proceeds, I'm often fine with REITs issuing equity even at a modest discount. The point is not to impose a rigid formula. It's just that by nibbling at the stock on both ends, a management team is demonstrating that it's always on their menu. And there is simply no other way to credibly send that signal.
  2. Create some leverage room now. If you’re trading at or above NAV, you can do that by overequitizing your new investments or your current portfolio. If you’re trading at a discount, you can just increase your stated leverage target without using the extra room. (Seriously.)
  3. Manage for stable FFO growth over high growth. This is the part that I don’t want to be true, but I can’t really get around it. Another way of describing the dead zone would be that there’s sort of a discontinuity around flat earnings growth, right, where all this other negative feedback starts to kick in. So a REIT with an internal forecast for lumpy 6% annualized FFO growth over the next five years might be well-served to tune it down to 5% with a lower variance.

    Of course overequitizing is one way to do that, and the best way is probably to keep reinvesting in the tail of a long and well-staggered development pipeline. But another way might be to run the “upgrade” playbook even when you’re already growing earnings — and this is one reason I tend to be less critical of those match-funded rotations outside the zone than inside them.

Sign up for updates

No Thanks

Great! Check your inbox and click the link to confirm your subscription.