A Simple and Accurate Approach to Forecasting Cap Rates

June 25, 2025

“While ratios (price to earnings, debt-to-equity) are commonplace in most value discussions, there is perhaps no other industry that relies on such a singular metric as real estate does on capitalization rates."

- Matt Larriva and Peter Linneman, from the research paper discussed today

A surprisingly simple ratio - total mortgage debt outstanding divided by GDP - can predict commercial and residential real estate cap rates with startling precision. When mortgage debt grows faster than GDP, cap rates compress. If the inverse happens, they expand. The study that uncovered the relationship between mortgage debt fund flows and cap rates illustrates how demand for mortgage debt is a proxy for “the amount of capital seeking real estate,” and below we will look at why this ratio works, how well it predicts cap rates, and a novel idea about what America could do with it.

The study's model - which uses fund flows into/out of mortgage debt plus historical cap rates (for the asset class in question) and U.S. unemployment - found that when mortgage debt grows 100 bps faster (slower) than GDP, cap rates fall (rise) by 22 and 65 basis points for multifamily and office properties respectively. In other words if debt grows by 10% relative to GDP, cap rates stand to compress by 220 to 650 basis points.

Said another way, a period of optimism on real estate will drive demand for mortgage debt as a percent of GDP, which “should accompany a period of cap rate compression.” Conversely, when capital leaves for other assets, we can expect “a period of high opportunity cost of real estate and a devaluation of the asset class.” Mortgage fund flows is the key determinant in the model. Notably, in the model unemployment carries minimal predictive weight—a full percentage point drop in unemployment moves cap rates just one to three basis points.

This approach, by researchers Matt Larriva, who today runs research and data at Brookfield, and lauded researcher and occasional iconoclast Peter Linneman, provides robust short-term predictive power.1 

This model is exceptionally good at predicting one-quarter-ahead forecasts for multifamily cap rates (0.96 peak R-squared, for you math nerds) and office (0.90). One year out, that accurate predictive power drops to 0.31 and 0.17, which is to say the model can explain some of what will happen a year from now but not a lot. Practically speaking this model is best used for cases when you want to forecast short-term cap rates, perhaps if you are considering a refinancing vs. a near-term exit.

Haystack readers of course know how important cap rates are and their importance stems from having so much embedded information in a single metric. They respond to idiosyncratic risk related to a property (e.g. expectations for revenues and expenses) and macro-economic factors, like the availability and cost of financing and the cost of capital for equity providers. Because they are subject to so many disparate influences cap rates are notoriously hard to forecast, which is why not many people have tried and why this research is particularly important.

The myriad metrics that influence cap rates do not move in tandem. However, the model sees them as if they were tied to each other with rubber bands - they drift apart and together within a natural relationship. Over time mortgage debt flows reflect the collective direction and magnitude of all the factors driving cap rates up or down. Using GDP in the denominator indexes the overall sentiment about real estate to changes in general economic growth.

Things get really interesting when you extrapolate from the core finding that changes in mortgage fund flows strongly impacts cap rates and thus the value of real estate. That suggests if mortgage fund flows can be manipulated, even just at the margin, cap rates can be influenced by policymakers. In fact real estate values are already (inadvertently) being influenced through GSE lending policy and Federal Reserve purchases and sales of mortgage backed securities. If formalized, it is possible U.S. real estate valuations could be “targeted, with precision,” the authors note.

Sounds crazy? Consider the systemic benefits: reduced volatility, enhanced predictability, fewer boom-bust cycles that have repeatedly destabilized the broader economy. Zooming out, unexpected and severe changes in real estate valuations have contributed to more than one recent macroeconomic crisis. A transparent real estate value-targeting strategy - like the Fed’s approach to interest rates and money supply - around real estate valuations would give policymakers more tools to keep the economy stable and growth-oriented, and would likely reduce risk in the asset class to boot.

This idea feels more urgent when you consider commercial real estate is worth $22.5 trillion as of 4Q232 and it’s the fourth largest asset class after equities, residential real estate and U.S. Treasuries. From that perspective it feels like we might be playing with fire having it subject only to market forces like it is today.

Back to the study, the model's credibility rests on rigorous methodology: this cap rate forecasting was robustly accurate both in-sample (i.e. on a data set from which the model was derived) and, critically, out-of-sample, meaning it had high predictive power. Also, these researchers used real transactional cap rate data from Green Street, not appraisal-based cap rate estimates, on which most other cap rate-related research relies.

And a final hat tip, of the three ingredients of their model - mortgage fund flows indexed to GDP, unemployment and asset-type-specific historical cap rates - there’s an “impressive ability to forecast cap rates when two out of three of the input variables are not explicitly concerned with the target series.” Things are rarely that elegant.

The Rake

Three good articles.

  • Demand Outpaces Supply by Over 10% in Desert/Mountains Region - RealPage

    Data shows the Desert/Mountains region achieved positive apartment fundamentals with demand exceeding the 77,000 units delivered by 11%, driving occupancy stabilization despite 5% inventory growth—the second-highest nationally behind the Carolinas.

  • Co-Warehousing Is Reshaping the Industrial Market - CRE Daily

    Co-warehousing demand is driving sub-100,000 SF industrial properties to 4.1% vacancy rates while facilities over 250,000 SF exceed 10%, with institutional players like BKM Capital and Kayne Anderson launching platforms.

  • The New Playbook for B and C Class Office Space - GlobeSt.

    Organizations with long-term location requirements are positioned to capitalize on B and C class office distress, with creative financing structures including installment purchases, TIC arrangements, and partnership stakes.

The Harvesters

Someone making real estate interesting. They don't pay us for this, unfortunately.

What: Working in some of the most dense American cities, L+M develops, constructs and manages primarily workforce and affordable residential product, with 55,000 units completed or underway, valued at $20 billion. The company has a fund management arm and an affiliate that develops mixed-income communities.

The Sparkle: Sixth Street, an established investment firm with more than $100B of real estate AUM, recently made a strategically significant (although undisclosed) investment that will capitalize L+M’s growth and expansion into new markets. The transaction is notable given the high-rate, high-construction-cost environment: It signals a serious investor’s inclination to vertically integrate with a developer, not just partner as an LP in specific investments. It’s a big propco seeking opco exposure. Maybe it’s too hard to find good deals otherwise? Related article is here.

Thank You To Our Sponsors

Adler & Stachenfeld LLP has grown to become one of the top real estate law practices in NYC.

Since its founding, RERI has provided funding for over 320 research papers and has helped create a body of scholarly research on topics that are timely and of interest to institutional real estate investors.

Editor’s Note: The Real Estate Haystack believes in sharing valuable information. If you enjoyed this week's newsletter, subscribe for regular delivery and forward it to a friend or colleague who might find it useful. It's a quick and easy way to spread the word.


Reply

or to participate.