Why Do REITs Trade Below NAV, and What Can AI Miss?
Many REITs and SOCIMIs trade for less than their published net asset value, for reasons that can include real company risk, sector conditions, or limited market visibility, and an AI tool researching the gap can miss the difference.
In short
A REIT is a real-estate investment company that typically owns or finances income-producing property; depending on the jurisdiction, it must generally distribute a large share of its qualifying income to shareholders. A SOCIMI is the Spanish legal and tax structure for a listed real-estate investment company, broadly comparable to a REIT. Both are assessed on earnings, cash flow, dividends and debt, and on net asset value. Plenty of REITs and SOCIMIs trade on the stock market for less than their published NAV. Some investors now use an AI tool as part of researching why, though AI is only one part of getting to a real answer.
Key facts
- NAV is published a few times a year on a set schedule while the share price moves every day; its calculation usually draws on property valuations, though the valuation method, appraisal frequency and reported NAV measure vary between companies.
- A discount, the share price sitting below NAV per share, is common across listed real estate and does not by itself say anything specific about one company.
- The reasons behind a discount can include genuine company risk, sector conditions, limited share liquidity, weak investor demand, or insufficient market visibility, and these look very different once you dig in.
- An AI tool can usually confirm that a discount exists, but the reasoning behind one specific company's discount is much harder for it to get right.
A simple example of a NAV discount
Imagine a property company owns buildings worth 120 million euros and owes 40 million euros in debt and other obligations. Subtract one from the other and its NAV comes to roughly 80 million euros. Divide that by 10 million shares outstanding and NAV per share works out to 8 euros. If the stock actually trades at 6 euros, the company is sitting at a 25 percent discount to NAV. That 25 percent gap is the whole subject of this article: why it exists, what it might mean, and where an AI tool researching it can go wrong.
Why REITs and SOCIMIs can trade below NAV
NAV comes from an appraiser's estimate of what the properties would sell for, updated every few months. A share price is formed through market trading, which may be highly active or relatively thin depending on the company, and it reflects investors' current expectations and risk assessments, including things an appraisal does not capture: how easily the properties could actually be sold, how much debt is coming due, and whether the market trusts the appraisal in the first place. A discount is what is left when the market's live opinion runs below the appraiser's periodic one.
Sector factors versus company-specific factors
Some of what drives a discount has nothing to do with one company. Interest rates, a nervous property market, financing that has gotten harder to arrange, or general uncertainty about where real estate values are heading affect nearly every REIT and SOCIMI in a country or sector at once. Against that common sector backdrop, company-specific factors help explain why otherwise similar businesses may trade at very different discounts: how strong the actual portfolio is, whether occupancy is holding up, when the debt comes due, whether dividends have been paid reliably, whether management has bought back stock, and whether investors trust the way the company values its own properties. The two rarely separate cleanly, sector mood and company specifics interact, but two companies in the same sector can still carry very different discounts.
Does a discount mean the shares are cheap?
Sometimes. And sometimes the discount is the market being right. It can mean the market has not caught up with a company that is doing fine. It can just as easily mean the market has spotted something real: high debt relative to the portfolio, falling property values in that segment, refinancing coming due at a worse rate, thin trading volume in the stock itself, weak governance, an uncertain dividend, heavy concentration in one property type or one country, or simple doubt that the appraised value would actually be achieved in a sale. Telling these apart takes the company's own numbers, one name at a time.
What AI may miss when explaining the discount
Ask an AI tool why a specific REIT or SOCIMI trades at a discount, and it may reach for the sector level explanation because that is what is widely written about: rates, sentiment, the broader property cycle. The company-specific reasoning, the debt maturity schedule, the occupancy trend, management's own view of the gap, may be spread across investor presentations, results releases, and call transcripts. Even when that material is public, it can be difficult for an AI tool to retrieve, connect, and prioritize correctly, buried in PDFs, inconsistently indexed, or simply not selected during its research. That is not the same as saying the discount exists because the company communicated poorly. It means an AI answer researched in isolation can flatten a real, company-specific situation into a generic paragraph that would fit almost any name in the sector.
Comparing several names at once
The version of this test that tells you something is the comparative one. Take four or five listed real estate names, ask an AI tool why each trades at its current discount, and lay the answers next to each other. If the four differ, they are probably picking up something real about each portfolio. If they read the same, with the same rates and the same sentiment in every one, the sector story is doing all the work and nothing company-specific has reached the answer. Worth knowing before you lean on any of it.
The bottom line
The discount itself tells you almost nothing. The reason behind it tells you everything, and the reason is company-specific. Anything a general AI answer says about rates and sentiment is the part that applies to every name in the sector; what it says about this company's debt maturities, occupancy and dividend record is the part worth reading, and how little of that appears is a finding in itself.
This article is for general educational purposes and does not constitute investment, legal, or tax advice.
MIRA is a source-grounded IR assistant that answers investor questions from approved company information the issuer controls; what that means in practice is explained here: What a source-grounded IR assistant does, and why now
If you want to see what MIRA would show for your own portfolio and debt position, book a MIRA demo.
About the author: Nataly Usuga is in Business Development at Finvictum, where she works with IR teams, including at listed real estate companies, on how their story gets found, or misunderstood, by AI before anyone ever talks to them. Connect on LinkedIn.