The Truth About IRR and Leverage in Multifamily

Two numbers get quoted more than any others when a deal is marketed: the projected IRR and the leverage.

Both are useful, and both are routinely read backwards. A higher IRR is treated as a better deal, and more leverage is treated as more return. Neither is reliably true, because both numbers are downstream of assumptions that the headline figure hides.

Here is how to read each one honestly.

IRR is a question, not an answer

Internal rate of return is the compounding annual return an investment has to earn to get from the starting equity to the ending proceeds over the hold. That makes it the most complete single return metric, and also the easiest to be misled by.

The wrong question is "should a higher projected IRR attract me and a lower one deter me?"

The right question is "how is this IRR achieved, and do I believe the assumptions embedded in it?" The number is a summary of everything upstream: rent growth, exit cap rate, hold period, and leverage. Change any of those and the IRR moves, without the deal getting one dollar better.

Consider a real pattern from 2021. A sponsor markets a 15% gross IRR on a three to five year hold. Instead, they sell after fifteen months into a frothy market and post something closer to a 100% IRR.

The return looks spectacular, but almost none of the outperformance came from the business plan. It came from timing, and the compressed hold shifts a disproportionate share of the profit to the sponsor's promote rather than to the limited partners. The IRR did its job as a number and lied about the deal.

This is why IRR should never travel alone. Read it beside the equity multiple, which is time-blind and cannot be inflated by pulling cash forward. Some investors go further and set multiple-based hurdles with splits that adjust for the hold period, precisely so a fast flip does not hand the sponsor an outsized promote for a bet on timing.

Leverage cuts both ways

The second number people misread is leverage. More debt is assumed to mean more return, and it does, right up until it does not.

The concept that governs this is the loan constant: annual debt service divided by the loan amount, or the all-in yearly cost of a dollar of debt. Compare it to the property's cap rate:

• Positive leverage exists when the cap rate is higher than the loan constant. The property yields more than the debt costs, so adding leverage lifts the levered return. This is the normal, healthy case.
• Negative leverage exists when the cap rate is lower than the loan constant. The debt costs more than the property yields, so every additional dollar of leverage drags the return down.

You can see the same thing in the cash-on-cash return: under positive leverage it sits above the cap rate, and under negative leverage it falls below it.

When negative leverage is still rational

Negative leverage is not automatically a mistake. Buyers accept it deliberately in three situations, each of which is a specific bet:

1. Projected NOI growth. The going-in yield is below the loan constant today, but the business plan lifts NOI enough that the yield on cost clears it within a year or two. You are underwriting to the stabilized picture, not the day-one one.
2. Anticipated falling rates. The buyer expects to refinance into cheaper debt, flipping the relationship. This is a rate forecast, and it should be stated as one rather than buried in the model.
3. Low or no leverage on trophy assets. Institutional buyers of the best assets sometimes accept a thin going-in spread, or buy unlevered entirely, because they are underwriting durability and long-term appreciation, not a levered yield.

The common thread is that in each case the negative leverage is a known, defended assumption, not an accident.

That is the whole discipline here. A strong IRR and heavy leverage are not the deal. They are the outputs of the assumptions that are the deal, and the assumptions are what you actually underwrite.

See the assumptions, not just the number

The point of a real underwriting is to make the assumptions behind the IRR visible and testable, so a headline number cannot hide a timing bet or a leverage trap.

That is what MultiScreen is built to do. It shows how the return moves when you change the exit cap rate, the hold period, or the debt terms, so you are evaluating the assumptions rather than trusting the summary.

Stress-test your assumptions free →