Why one ZIP out-yields the next one over: it's the price, not the rent
A while back I posted that most of the variation in rental yield sits inside metros, not between them. Pool every ZIP with both a home-value and a rent index and about two-thirds of the spread in gross yield is within-metro; only about a third is the gap between metros. Morgan asked the follow-up that stuck with me: fine, but why? When one ZIP prints 7% gross and the ZIP a few miles over prints 4%, what is actually different?
Only two possible answers, because yield is annual rent over home value. A ZIP out-yields its neighbor either because its rents run strong for the metro (numerator) or its home values run cheap (denominator). So I decomposed it on a fresh May-2026 snapshot.
The baseline held: across 8,346 ZIPs in 155 metros, within-metro variance came back 67% vs 33% between, the same 2-to-1 split. Then the decomposition, splitting the within-metro yield variance into a rent share and a value share that add to 100%: the value denominator accounts for essentially all of it, and the rent numerator's contribution is actually slightly negative. Within a metro, a high-yield ZIP is high-yield because the houses are cheaper, not because the rents are stronger.
Memphis makes it concrete. In ZIP 38106 the typical home is worth about $58,500 against roughly $11,100 of annual rent — a 19% gross yield. Across town in 38120 the typical home runs about $427,600 against roughly $16,800 of annual rent — a 3.9% yield. Same metro, same rental market. The home values differ more than sevenfold; the rents differ barely 1.5x. The whole yield gap is the price gap. Rents inside a metro are fairly compressed; home values fan out enormously ZIP to ZIP, and yield inherits the fan.
It isn't one metro. Across the 87 metros with at least twenty scored ZIPs, 85 are value-driven — the median metro's value share is around 105%. It shows up in cheap markets and expensive ones: Jacksonville runs 134%, and even Boston, with yields squeezed into a 3.1%-to-5.6% band, comes in at 128%. Two metros genuinely invert, and I'll name them for honesty: in the New York-Newark-Jersey City area the internal yield spread is mostly a rent story (value share about 39%), and Cape Coral-Fort Myers is even more rent-driven. So it's a strong central tendency with real exceptions, not a universal law.
I also checked it wasn't just a cooling-market snapshot. On a balanced panel — the same 3,419 ZIPs in the same 51 metros, each May from 2023 to 2026 — the within-metro share went 60.7, 60.6, 61.9, 64.1. Never below 60, drifting up. Durable.
The practical read: when you screen for yield and a ZIP pops inside an otherwise ordinary metro — a Memphis 38106 versus a 38120 — the odds are overwhelming you're looking at a cheap-basis ZIP, not a strong-rent one. Not disqualifying; plenty of good cash flow lives on a cheap basis. But it moves the question. A yield built on strong rents is a demand signal. A yield built on a cheap price is a signal about why the house is cheap, and the yield number can't answer that. And to be clear, this is a read on housing-market structure — price versus rent — not a call to buy 38106 or avoid 38120, and not a statement about anyone who lives there.
When a high within-metro yield shows up in your own markets, do you treat it as a cheap-basis flag and go dig into why the price is low, or are you seeing rent strength drive it in a way this metro-level pooling would miss?
David
