---
title: "How to value a REIT: FFO, P/FFO and the yield spread"
description: "A step-by-step walk from a REIT's income statement to the three questions that price it: FFO and P/FFO, yield against the 10-year, the balance sheet."
date: 2026-09-18
updated: 2026-09-18
tags: [reits, valuation, basics]
author: "Allen Yan"
image: https://sweetvaluelab.com/blog/img/255e84e8f9c616d3.png
lang: en
canonical: https://sweetvaluelab.com/blog/how-to-value-a-reit
---

Open Prologis on any stock screener and the first thing you see is a **P/E of 30**. Thirty times earnings for a landlord whose entire business is collecting rent on warehouses. That sounds like a lot. Then you notice the dividend yield is 3.1%. The payout ratio on that same screener says 94%. The whole picture looks like a company paying out everything it earns at a price that already assumes a great deal.

Every one of those numbers is correct, and every one of them is the wrong number to look at. This guide walks through why, one step at a time, and ends with a way to read any REIT in about five minutes. It uses Prologis (PLD) throughout, on the 2026-09-18 close. One real company beats a page of definitions, and every figure here comes straight off the [REIT board](https://sweetvaluelab.com/tools/reits), so you can follow along on [PLD's card](https://sweetvaluelab.com/tools/reits/pld) and then do the same for any other name.

If you have read [the bank guide](https://sweetvaluelab.com/blog/how-to-value-a-bank), the shape will feel familiar. An industry whose accounts do not mean what they usually mean, one number that fixes that, and a short list of questions that follow from it.

## Step 1: Understand what a REIT is before you value it

A real estate investment trust is a company that owns buildings, collects rent, and, in the part that matters here, is required by law to pay out at least 90% of its taxable income as dividends. In exchange it pays no corporate income tax. That one rule explains almost everything odd about REIT numbers:

- **There are almost no retained earnings.** A normal company grows by keeping profits and reinvesting them. A REIT cannot; it grows by raising new money, selling shares and borrowing, to buy more buildings.
- **The dividend is the product.** Most of the return a REIT shareholder gets arrives as cash every quarter. So the dividend, and whether it can be maintained, sits at the centre of the analysis.
- **Debt is normal.** Buildings are bought with mortgages. A REIT with no debt would be leaving money on the table; a REIT with too much is one bad year from cutting the dividend. Where the line sits is a question we will come back to.

Hold those three in mind. They are why the next step is necessary.

## Step 2: Why earnings lie: the depreciation problem

The accounting rules are the problem. A company that buys a building must **depreciate** it: write its value down on the books a little every year, and running that write-down through the income statement as an expense, until the building is carried at zero. For a machine that wears out in ten years this is sensible. For a warehouse in a port city, it is fiction.

![A building's book value falling to zero over thirty years while its market value rises](https://sweetvaluelab.com/blog/img/ad3b6264ae3facc8.svg "The accountant's book and the market's book for the same building. The dashed line is what the income statement charges every year; the solid line is roughly what happens to well-located property. The gap between them is a cost that never left the bank account.")

Look at what that does to Prologis's income statement. In the trailing twelve months it charged **$2.74 billion** of depreciation and amortization against **$4.21 billion** of net income. Nearly forty percent of its "cost" of doing business was a bookkeeping entry for buildings that, by the market's estimate, gained value over the same year. No cash left; no warehouse got smaller.

So when you divide the share price by earnings, you are dividing by a number that has been reduced by a charge that is not real. The P/E of 30 is not measuring what you think it is measuring. Worse, it compares badly across REITs: two landlords with identical buildings and identical rents will show different earnings depending on when they bought and how their accountants schedule the write-down.

## Step 3: Fix the earnings by hand: FFO

The industry's answer, standardised by Nareit in the 1990s, is **funds from operations (FFO)**. The idea is simple: start from net income and put the depreciation back.

> **FFO = net income + depreciation & amortization − gains on property sales**

The last term matters too. A building sold for a profit is not recurring rent, so the full definition removes it (and adds back impairments, the opposite case). Our board uses the **simplified form, net income plus depreciation**, because the quarterly statements we compute from do not break out gains as a separate line. Most quarters that changes nothing; when a REIT has a big sale or a write-down in the window, the card says so instead of pretending. Keep that caveat in mind and move on.

![Prologis net income plus depreciation equals FFO; less dividends equals what was retained](https://sweetvaluelab.com/blog/img/604864d692d36882.svg "Prologis, trailing twelve months to June 2026, in billions: net income $4.21B, add back $2.74B of depreciation, FFO $6.95B. Dividends took $4.04B of that; $2.90B stayed in the business.")

Now do the two divisions yourself:

- **FFO per share** = $6.95B ÷ 958 million diluted shares = **$7.25**
- **P/FFO** = $134.84 ÷ $7.25 = **18.6×**

The P/E of 30 has become a P/FFO of 18.6. Same company, same price, same day. And the "94% payout ratio" the screener showed you, dividends over earnings, is really **58% of FFO** ($4.22 of dividends per share over $7.25 of FFO). The dividend is not eating everything; it is taking a little over half of the cash the buildings throw off, which is normal for the sector.

Here is the same correction across eight REITs on the board, to see how much it changes the picture:

![P/E and P/FFO for eight REITs: the P/E bars scatter, the P/FFO bars sit in a band](https://sweetvaluelab.com/blog/img/bd0f6592fdc09857.svg "The grey bars are P/E, the blue bars P/FFO, on the same day. Welltower's P/E of 121 and Realty Income's 41 collapse to 44 and 14 once depreciation is put back. P/E is a lottery across REITs; P/FFO is a comparison.")

Welltower at 121 times earnings looks absurd; at 44 times FFO it is merely the most expensive multiple on the board, for reasons we will get to. Realty Income at 41 times earnings looks like a growth stock; at 14 times FFO it is a slow, steady net-lease landlord priced like one. The point of Step 3 is that **you cannot see a REIT until you have made this correction.**

> **Try it:** on any REIT's card, the head shows the close, P/FFO and FFO per share side by side; the "Where the cash goes" panel is the bridge above, drawn for that company.

## Step 4: Is 18.6× a lot? Two comparisons that actually work

A multiple is a price for growth, so on its own 18.6× means nothing. Two comparisons make it mean something, and a third one that people reach for first does not.

**The one that does not work: comparing across property types.** Data-center and healthcare REITs trade in the twenties and forties; net-lease and office REITs in the low teens. Those gaps are mostly about growth, lease length and how much capital each type needs, not mispricing. Comparing Welltower's 44× with Realty Income's 14× tells you they are different businesses, which you knew.

**The first one that works: against growth.** The board's first chart puts every REIT's P/FFO against the growth in its FFO per share over three years. A landlord compounding FFO per share at 9% a year deserves a higher multiple than one whose FFO has been flat, and the market gives it one.

![P/FFO against three-year FFO per share growth for every REIT on the board](https://sweetvaluelab.com/blog/img/df175bff572adb68.svg "Every REIT on the board: FFO-per-share growth across, P/FFO up, sized by market cap, coloured by property type. The dashed line is the board's median multiple. Up and to the left is paying for growth that has not shown up; down and to the right is growth the market is not paying for. Either deserves a reason.")

Prologis sits at 18.6× with a 9% three-year growth rate: a bit above the board's median multiple for a bit above the board's median growth, which is roughly what you would expect. Welltower's 44× is attached to 20% growth, the senior-housing recovery, which is the reason it can carry that multiple. The chart does not tell you whether either is right; it tells you what the market is paying for, so that your next question is specific.

**The second one that works: against its own history.** For each of the last twenty quarters, take the quarter-end price and divide by that quarter's own trailing FFO per share. That gives twenty readings of the multiple for the same company, and a median and a middle half to compare today against.

![Prologis quarter-end P/FFO over five years with the interquartile band](https://sweetvaluelab.com/blog/img/10407eaa27dd4633.svg "Prologis, twenty quarter-end P/FFO readings. The shaded band is the middle half of those quarters, the dashed line the median. Today's 18.6× is 8% below its own median and lower than three-quarters of its own recent history.")

Prologis has averaged **20.3×** over five years, with the middle half of quarters between 18.2× and 22.1×. Today's 18.6× is **8% below its own median**. That is a much more useful sentence than "18.6×" alone: this is a company at the low end of what it has usually been paid. Whether the reason is a slower warehouse market or something else is the question to ask next, and it is now the right question.

> **Try it:** the board's "vs. 5y median" column is this comparison for every row; sort by it and the REITs furthest below their own history rise to the top. Each card draws the full five years.

## Step 5: Read the dividend against the bond it competes with

A REIT dividend is an income stream. So is a Treasury bond. The same investors buy both: pension funds, insurers, retirees. They compare them. For most of the last thirty years REITs have yielded *more* than the 10-year Treasury, and the extra is what owning buildings costs against the government's promise.

![Dividend yields of several REITs and the board median against the 10-year Treasury yield](https://sweetvaluelab.com/blog/img/706cb4b27a13b710.svg "Dividend yields on 2026-09-18 against the 10-year Treasury at 5.01%, the dashed line. The number after each bar is the spread in percentage points. Only the net-lease names yield more than the bond; the board's median, and Prologis, yield less.")

On that day the board's median yield was **4.4%** against a 10-year at **5.0%**, a spread of **−0.6 points**. Read literally: as a group, these REITs paid less than the government bond. That is unusual, and it tells you something about what the market is buying. Prologis at 3.1%, nearly two points under the bond, is not being bought for income at all. It is being bought for the 9% growth in Step 4. VICI, at 7.6%, is the opposite: a casino landlord with 30-year leases, bought for the cheque.

Neither is a mistake. But the yield has to make sense. A REIT yielding less than a bond needs its growth to show up, and one yielding much more is telling you the market has a worry about the payout. Which brings us to the last step.

> **Try it:** the strip across the top of the board carries the 10-year, the median yield and the spread every day. The "Spread" column in the table is the same number per REIT.

## Step 6: Can the balance sheet carry the dividend?

A company that pays out most of what it earns and borrows to grow is only as safe as its balance sheet. Three numbers answer most of the question, and the board colours each one against thresholds it prints:

| Number | What it asks | How to compute it | Green / amber / red |
|---|---|---|---|
| **FFO payout** | How much cushion the dividend has | dividends per share ÷ FFO per share | ≤70% / 70–85% / >85% |
| **Net debt / EBITDA** | How many years of operating cash the debt represents | (total debt − cash) ÷ (operating income + depreciation) | ≤5× / 5–7× / >7× |
| **Interest cover** | How far rents or rates can move before the dividend is reached | EBITDA ÷ interest expense | ≥4× / 2.5–4× / <2.5× |

For Prologis: dividends of $4.22 over FFO of $7.25 is a **58% payout**; net debt of $34.7B over EBITDA of $6.27B is **5.5×**; EBITDA over $1.05B of interest is **6.0×** cover.

![Prologis FFO payout, net debt to EBITDA and interest cover plotted against the board's thresholds](https://sweetvaluelab.com/blog/img/c5c519b19e6e86e9.svg "Prologis against the three thresholds. The dot is the company; the tinted segments are green, amber and red. A comfortable payout, leverage in the amber band that is normal for a growing landlord, and coverage well into green.")

Read together: the dividend takes about half of FFO, the debt is moderate for a REIT that is still building, and interest is covered six times over. Nothing here threatens the payout.

Now look at what the same three numbers say elsewhere on the board. Crown Castle pays out **105%** of its FFO, more than it makes, with net debt at **8.4×** EBITDA: two lights on, and a dividend that the numbers say cannot stay where it is. Extra Space covers its interest only **1.5×**. Realty Income, the steadiest name in the sector, carries **7.1×** leverage, over the line. None of these is a verdict. Each is a reason to open the card and read why.

Each threshold has a point past which the board turns a **light** on: payout above 85%, leverage above 7×, coverage below 2.5×. A light is on or off, with a fixed sentence when it is on, and there is no combined score. A score would be a grade wearing a disguise; a lit light is a reason to read further.

One more thing worth knowing. The data provider occasionally files a REIT's debt on the wrong line. Realty Income's latest quarter showed $2.8 billion of debt against $30 billion in reality, a balance sheet that could not have paid its own interest bill. The board checks for exactly that. Trailing interest divided by debt above 12% means the balance sheet is wrong, so it falls back to the newest quarter that passes, with a note on the card. A confident wrong number is worse than a blank.

## Step 7: Put it together: five minutes with any REIT

Here is the whole method as a checklist. Open a REIT's card on the board and go down it in order.

1. **What property type is it?** Industrial, net lease, healthcare, towers, data centers, residential. This sets the range of multiples you should expect and the growth you should expect with them. Do not compare across types without a reason.
2. **What is the P/FFO, and what does FFO per share look like over eight quarters?** Ignore the P/E. Look at the "Quarter by quarter" sparkline: is FFO per share rising, flat, or falling?
3. **Is the multiple above or below its own five-year median, and by how much?** Below its median with rising FFO is a different case from above its median with falling FFO.
4. **What is the yield, and what is the spread to the 10-year?** Under the bond means you are paying for growth, so go back to step 2 and check it is there. Well over the bond means the market is unsure about the payout, so go to step 5.
5. **Payout, leverage, coverage.** Green across is a dividend that can survive a bad year. Any light on is the first thing to explain before anything else matters.
6. **Read the data-quality notes.** If a quarter was dropped, if the debt line was rejected, if FFO went negative on a write-down, the card says so. The number you are looking at is only as good as what went into it.

That is the method. It takes longer to read than to do.

## What the board cannot tell you

A few things the numbers above leave out, so you know where to look next:

- **AFFO.** Companies report "adjusted FFO", which subtracts the recurring capital spending needed to keep buildings leased: new roofs, tenant improvements, leasing commissions. It is a better measure of distributable cash, and the true payout ratio is against AFFO. It is not on the board because the statements do not carry it; the company's own supplemental does.
- **Gains and impairments.** Our simplified FFO includes them. Look at American Tower. Its "+66%" FFO growth this year is a rebound from a large write-down the year before, rather than a business growing at 66%. Sun Communities shows no P/FFO because a loss on a sale made its trailing FFO negative. The card notes both.
- **Cap rates and NAV.** Professional REIT analysts value the buildings directly, then compare that with the price. Net operating income divided by a market cap rate gives a property value; less debt gives a net asset value per share. That needs property-level data the filings summarise away. The board's "EV yield" (EBITDA over enterprise value) is a rough cousin.
- **Lease length, tenant quality, occupancy.** All in the supplemental, none in the income statement.

None of this is advice. A multiple below a REIT's own median, or a light that is on, describes two numbers on a given day. The board is there so that the next question you ask is the right one, and after this guide you know which one that is.