What growth rates belong in an SFR model when rents are running 1.3 percent

Question

I've been reading pro formas instead of buying anything, and the same pattern shows up in nearly all of them. Rent grows 3 percent a year, expenses grow 2 percent, and by year five the model produces a number that makes the deal work. Take the rent line to 1.5 and push expenses to 4 and the same house is a slow bleed.

ATTOM's read on 2026 was that potential yields fell in about 55 percent of counties because prices outran rents, and single-family rents were up 1.3 percent year over year early in the year. Against that, insurance and property taxes in a lot of places have been climbing much faster than 2 percent, and taxes often reset off the price you just paid.

So there's a real split here. One camp says a five year model with growth in it is just a story you tell yourself, and you should underwrite year one cash flow, buy only if it works flat, and treat any growth as upside you didn't pay for. The other camp says a flat model is also a forecast, and a wrong one, because a 30 year fixed payment against rents that rise even slowly is the actual mechanism of the strategy, and refusing to model it means you never buy anything and you sit out the compounding.

There's a middle position where you grow rents slowly and expenses fast, which produces ugly numbers but at least matches the direction things have been moving.

I don't have a view worth defending yet. What do you actually put in the cells?

luis_carrera · Answer

I've built the same model at four different growth assumptions and the honest result is that the assumption picks the answer, so I've stopped treating the model as a decision tool. I run it flat to see if the house survives, then I run it at rent 2 and expenses 4 to see when the survival ends. If the flat version is negative I don't need the other tabs.

jessreyes12 · Answer

Question I should probably know the answer to. When people say expenses grow 4 percent, is that the whole expense line or just insurance and taxes? Because my management fee is a percentage of rent so it grows with rent automatically, and my mortgage payment doesn't grow at all. Feels like only part of the column is actually moving.

omar_haddad · Answer

@compass you've got it, and it matters more than the growth debate. On a typical house the payment is over half the outflow and it's frozen for 30 years, taxes and insurance move on their own schedule, and repairs move with the trades. The line nobody models properly is capex, because it doesn't inflate smoothly, it arrives. A 14k roof in year six is not 2 percent a year, it's zero for five years and then a number bigger than your annual cash flow. I put capex in as dated events with real dollar amounts instead of a percentage, and the model got a lot less flattering and a lot more useful.

stephen_ok9 · Answer

Coming at this from where to put money rather than which house, the growth assumption is doing double duty and that's the part that bothers me. It's setting cash flow and it's setting exit value, because most of these models exit on a cap rate applied to year five NOI. So a small change in the rent cell moves the terminal value by a lot, and that terminal value is usually most of the return. If I could only fix one cell it would be the exit cap, not the rent growth.

Rena Guide · Answer

A few definitions for anyone reading this cold, since the thread is moving fast.

A pro forma is a projection of income and expenses over several years. Every number past the current rent roll and the current tax bill is an assumption, and the person who wrote it chose those assumptions.

Growth rates are the annual percentages applied to rent and to expenses. They compound, so a difference that looks small in year one gets large by year five. Rent growth in single-family rentals has been positive and cooling, around 1.3 percent year over year in early 2026, which is well below the 3 percent that appears in many models by habit.

Capex means capital expenditures, the big replacements like a roof, a furnace, or a sewer line. These are lumpy rather than annual, which is why @harrow models them as dated events.

The exit cap rate @meridian mentions is the assumed capitalization rate a future buyer pays, used to convert projected income into a sale price. Because it divides into the income, a small change produces a large change in value.

Two practical points. Property taxes frequently reassess after a sale, and both the timing and the method are set by state and county, so confirm the treatment with the local assessor rather than carrying the seller's bill forward. Insurance is a quote, not an estimate, so get a bindable number in writing from a carrier before you rely on it. Neither of those is a growth rate question. They are both knowable today, and pinning them down shrinks how much the growth cells have to carry.

jordan_ellis2 · Answer

Voting flat, and part of why is that a flat model is the only one I can defend to somebody else. If I'm asking a partner for money and my case rests on rents rising 3 percent, the whole conversation becomes about whether rents rise 3 percent. If it works at today's rent and today's tax bill, there's nothing to argue about. Six months of research and that's about the only underwriting rule I've actually kept.

What growth rates belong in an SFR model when rents are running 1.3 percent

6 replies