Buildings Management Magazine & Expo



How to Measure Risk as Property Values Continue to Climb


A bank regulator once posed the following hypothetical scenario to us: “Suppose, in 2007, you had a property that was generating $6 million in cash. If capitalization rates were 6% for that type of asset, the property would be valued at $100 million.” Since lending was loose at the time, even a conservative banker might have been required to lend “just” $75 million on the property for a 75% loan-to-value (LTV) ratio. But what about today and the market’s current lending standards?

Ten years later, rent growth at the hypothetical property has been negligible, owing to the fact that it’s not in a 24-hour or tech-heavy city. So, cash flow remains at $6 million but the cap rate has fallen to 4%. Our conservative banker gets even more conservative and would provide financing amounting to just a 65% LTV ratio. That equates to a nearly $100 million loan for the same cash flow. Just how conservative has the banker been?

In spending the last five years helping the 50 largest US banks build their own default models, we at Trepp have learned that quantifying and accounting for that risk provides challenges. So, the bank regulator’s hypothetical scenario described above comes up often.

We won’t go into all the nuances and challenges that came up with banks building their own models, but we can talk about some of the ways we tried to account for periods of high liquidity and value run-up in our model building.

Let’s start with some background. It is widely accepted that almost every commercial real estate market in the United States has rebounded to (or close to) its 2007 peak. In some 24-hour markets, such as New York, Boston, and San Francisco, values can be 125% more than peak values. The laggards usually come from non-trophy markets where the rebound has been more tepid. But even in those markets, values are often at par or better than their pre-recession peaks.

Any model worth its salt will use future gross domestic product, unemployment, the consumer price index, and interest rates to project future commercial real estate values. However, they should also account for liquidity. How to do that with statistical backing rather than seat-of-the-pants judgment was our challenge. We knew that 2007 was “loose” and we knew 2011 was “illiquid” – but how does one define that with data?

Our solution was to create a liquidity score using the growth rate of the mortgage market that ranged from 0 for the lowest historical growth rate to 10 for the highest historical growth rate. If the mortgage market was growing at a rate midway between the two extremes, it would receive a score of 5. The data used is from the Federal Reserve’s Flow of Funds series and spans a 65-year period. We created separate scores for the multifamily and commercial mortgage markets.

This allowed us to further project what liquidity would be in the future if, for example, GDP dropped to -2% and unemployment increased to 7%. That would correspond to a tightening of liquidity and would result in a dampening of values upon a loan’s maturity. 

Another potential source of data – one more specific to CMBS – is the ratio of loans used for refinancing versus those used for acquisition. While not part of the models we developed to track mortgage market liquidity, it shows that during periods of high liquidity, the percentage and volume of acquisition loans is greater than during periods of lower liquidity.

We found that during the bubble years of 2006 and 2007, almost 50% of CMBS lending was done to fund property acquisitions.

During the post-crisis years of 2011-2013, the number hovered between 20% and 30%. The number started to creep up last year as it hit 32%. The modest uptick over the last six years matches our gut instinct: the mortgage market has indeed become more liquid, but we’re not close to the manic levels reached 10 years ago.

Screen Shot 2018-08-03 at 10.03.54 AM.png

Trepp, LLC, founded in 1979, is a leading provider of data, analytics, and technology solutions to the global securities and investment management industries. Trepp specifically serves three key sectors: structured finance, commercial real estate, and banking to help market participants meet their objectives for surveillance, credit risk management, and investment performance. Trusted by the industry for the accuracy of its proprietary data, Trepp provides clients with sophisticated, comprehensive models and analytics.


Travis Watson