Does high yield actually cost you appreciation? I tested 2015–2026.

[David Walters]

A question came up in a recent thread here that I couldn't answer off the top of my head, so I went and ran it against metro-level data: do the highest-yield metros actually have the lowest long-run appreciation?

It's one of the most repeated rules in this business. There are cashflow markets and there are appreciation markets, and you pick your lane — high yield in the cheaper Midwest and South where values creep, or low yield on the expensive coasts where the equity compounds. High yield, low growth. Low yield, high growth. You can't have both.

I had data back to 2015, so I tested it properly. The answer wasn't what I expected — and I want to walk through it, including the part where my own prediction was wrong.

First, the trap, because it's why this question rarely gets answered honestly.

The obvious way to check is to sort markets by their yield today and compare against how much they've appreciated. Do that and you'll usually find the inverse you expected. But it's circular. Yield is rent over price. Appreciation is price going up. So any market that appreciated a lot mechanically shows a lower yield now — not because low yield caused the growth, but because the growth pushed price up and dragged yield down with it. You haven't found a relationship; you've rediscovered the definition of a fraction. I measured how strong that pull is: the correlation between a metro's appreciation and the change in its yield is −0.59. Strong, negative, and entirely an accounting effect.

To get a real answer you have to sort by entry yield — the yield you could actually have bought at, at a fixed point in the past — and measure appreciation forward from there. That breaks the circle. And you have to look at separate eras, because one boom can impersonate a permanent law.

Here's where I was wrong. I expected the naive cut to show a strong false inverse. It came back at +0.08 — basically zero. The mechanical echo is real but it's being canceled by something pulling the other way. Sort by entry yield instead and the relationship isn't negative at all. Higher-yield metros in 2021 appreciated more over the next five years, not less.

The rule's whole life cycle shows up when you split it into three regimes.

Before 2020 (2015–2019): the rule had a kernel of truth. Lowest-yield metros appreciated about 38% over five years; highest-yield about 28%. A real gap, if a modest one. The folklore described the pre-2020 world reasonably well.

The boom (2019–2022): it vanished. Every yield tier appreciated 35–40%. When everything rips, yield tells you nothing about appreciation.

The cooling (2022–2026): it reversed. The lowest-yield metros — many of them the 2021 boomtowns that had been bid up hardest — gave back the most, around 4%. The highest-yield metros kept gaining, around 12%.

One honest caveat on that last one: a good part of the reversal is mean-reversion. The low-yield boomtowns overshot in 2021 and corrected, which mechanically flatters the high-yield side. I'm not claiming high yield causes appreciation. I'm claiming the old rule broke. Over three regimes it was clearly true in one, gone in the second, inverted in the third.

The number that actually matters: total return.

Nobody buys yield or appreciation in isolation — you keep the combination. Folding the two together (gross, before financing and opex, so treat it as a direction not an IRR), the highest-yield tier had the highest gross total return in every regime. The appreciation high-yield markets gave up, where they gave any up, was always smaller than the extra yield they brought. The "I'd rather own a 6% with growth than a 12% losing residents" instinct is right about what matters — the demand under the rent — but in the actual data, the high-yield markets weren't the ones losing residents.

It holds inside metros too. Rank a single metro's ZIPs by yield and all of them appreciated about the same — 22 to 23% over the recent five years. Within a metro, yield and appreciation are simply decoupled. The high-yield ZIP down the road isn't quietly trading appreciation for income; it's getting both.

What I can and can't claim. This is blind to markets that have hollowed out so far they no longer carry a usable rent index — the genuinely dying places are invisible here, and that's exactly where the old tradeoff might still be alive. It's gross return, not net of leverage. And it's history, not a forecast — regimes turn. What's solid: across 2015–2026, the rule that high yield costs you appreciation held in one era of three, never held on total return, and disappears entirely inside metros.

The way I'd put it now is that yield is a symptom, not a predictor — downstream of risk perception, capital flows, how hard it is to build, and where demand is heading. When those move, yield moves, and so does appreciation, often alongside it rather than against it. The tradeoff was never a law; it was a description of one capital regime that the last five years rewrote.

The open question I keep landing on: the leading half of the demand picture — household formation, vacancy direction, absorption — is the part that should move before rent and price do. If you've watched a high-yield market quietly hollow out, what showed it first? The rent rolling over, or vacancy and absorption moving before the rent did? That's the signal I'm trying to figure out how to watch.

— David