Commercial real estate is the big dog in our economy, and he is growling. Not biting yet, but give him time. Unless we back away slowly and don’t let him see our fear, chances are good that he will take a chunk out of our collective hide.
As a good friend first pointed out on the Kondratyev discussion group at Colorado State University (Longwaves group) ten years ago, every modern economic collapse seems to culminate in a commercial real estate debacle.
He lives in the heartland, and knows whereof he speaks, with a client base in the commercial real estate market. Back in the 1990’s, he shared with us his calendar of crises, first derived in the 1970’s when he had way too much time on his hands and easy access to an excellent library of economic history.
Why do I bring this up now? So you know that not everyone predicting a commercial real estate collapse is just now waking up to it. I also wanted to walk you through how it will happen, with an example.
My Longwaves buddy was telling us in 1999 that he wasn’t expecting a commercial real estate washout until late 2009 or 2010, even though I thought it might happen as early as 2000/2001. When it came, he told us to expect it to be a three-generation sized wave.
This article I saw a couple of days ago looks at the data for loans in Commercial Mortgage-Backed Securities (indirectly derived from the Realpoint, LLC group I wrote about earlier).
After hearing all the scary numbers of residential mortgages hitting 20%, 30% or even higher default rates (depending on the sector and cohort group you’re looking at), 8.08% of a loan portfolio in trouble, heading to 9% next month, might not sound terrible. But compare the loans in workout today to how many there were a year ago. I sit up and notice whenever the bad news in a business segment grows by a factor of ten.
It’s even more significant that the news got that much worse in only twelve months when you consider that commercial real estate is the slowest mover of all the common investment classes. Like a train wreck, it actually takes a long time to happen.
The real problem with commercial real estate lending is that the assets are naturally levered even before you borrow money to buy them. This operating leverage never shows up in LTV calculations, but you better believe it’s there.
The owner of commercial real estate is, by definition, owning it for the purpose of renting it out. They have to pay taxes, maintenance, etc., whether they are getting zero dollars in rent per square foot or a thousand. If you think of the cash flows, you realize that those maintenance and other fixed ownership costs are just like the debt service on super-senior debt. If the building owner wants to keep their building, those costs are right up there with (or ahead of) first mortgage debt in priority, and often consist of well over half the gross receipts from the properties.
By my way of thinking, that means commercial real estate starts out with more than 1:1 leverage, even before it gets any debt put on it. Even a “safe” 70% LTV mortgage can eat up so much of the cash flows from the property that 10:1 or higher operating leverage is quite common once you look at all the moving parts.
That’s why every equity REIT analysis includes substantial attention to occupancy rates and trends in new rental rates, since a major tenant leaving is often enough to take a building from solid performance to the cusp of default on its loans. Add in the fact that most commercial mortgages aren’t self-amortizing like residential loans, and defaults become a predictable event at the end of the mortgage term, when refinancing is required.
In the early 1990’s, there were sales of some real estate at what sounded like (and were) giveaway prices. Owners simply wanted to get out from under the burden of their fixed costs. Mitsubishi walked away from their ownership of Rockefeller Center in the 1990’s, a deal valued at over a billion dollars when they bought 80% of that landmark commercial real estate complex in 1985.
The good news, if you could call it that, is that this history of dramatic declines prevented the Rating Agencies from getting too giddy, and CMBS were never securitized with incredibly thin subordination like residential mortgages.
The typical CMBS deal was subject to 30% subordination for a AAA, even during the height of the debt bubble. Compare that to subprime residential MBS bonds with 20% subordination, or option-ARM MBS deals sporting AAA ratings for 90% of the balance (10% subordination).
Having said that, almost any commercial property in America could not be refinanced today without substantial additional equity (down payment). That’s in spite of the fact that rates have come down on mortgages on prime properties.
Real estate owners can’t refinance today because lenders have no interest in adding layers of subordinated mortgages. Even if the first mortgage was 70% of the value of the property, lots of real estate investors managed to get second, third and even fourth layers of financing on their properties in the go-go years
It was common, in fact, to see 85% or 90% total financing associated with commercial real estate purchases in 2005 and 2006. Then there was the issue of rental expectations. I recall seeing analyses where the “worst case” shown in the pitch book was that rents would only rise at 2% to 3% per year.
Now there are stories circulating of mall owners letting their anchor tenants (big name stores that draw traffic to the mall) renew their leases at zero dollars rent. These owners know they’ll lose all their small store tenants along the arms of the mall if the big names go dark, so they take the pain in the hopes they can survive.
To add to the pain, potential real estate investors demand “cap rates”, or return on equity, more like 8% or 10%, rather than the 4% or 5% they accepted a few years ago when they were confident that prices would go up and make their investment a good one. Now it’s cash-on-cash return, or no deal.
Let’s run through some simplified numbers as an example.
First, we’ll start with a building that had gross rents of $20 million a year in 2005 when a buyer was looking at getting involved. Assume $10 million in fixed costs, leaving $10 million to support debt service and pay a return to the investor.
The first mortgage would have been underwritten to a maximum 70% LTV, and that would have needed around 1.55 times debt service coverage. Since we don’t know yet what the “V”, or Value is, we’ll just take the $10 million and divide by 1.55 to come up with a mortgage payment of $6.45 million per year. Since the 10-year Treasury rate wobbled around 5% for the middle part of 2005, we’ll pick 7% as their first mortgage rate, suggesting a first mortgage of $92.1 million, or a theoretical value for the property around $131.5 million.
We’ll see once we run the rest of the numbers whether that investor would pay as much as $131 million for our building.
The second mortgage might cover the “slice” of the financing from 70.01% LTV on up to 80% LTV, and would have priced around swaps plus 400 basis points or so at the time, or 8.40% for our rough calculations. That lender would insist on minimum coverage of 1.35 times. Looking back at the $10 million net income, the 1.35 coverage point implied $7.407 million available for mortgage debt service. Subtracting $6.45 million leaves $957,000 per year to service the second lien.
That $957,00 could pay 8.4% interest on $11.4 million, so with this slice of financing, the potential buyer has gotten up to $92.1 + $11.4, or $103.5 million.
Next we’ll add another “B note” to finance the value slice from just under 80 LTV up to 85 to 90 LTV. For this kind of paper, my not completely reliable memory is that spreads were in the 600 to 800 over swaps region, and 1.15x coverage. We’ll call it an 11% interest rate, and note that 1.15x coverage gives us $10,000,000/1.15 = $8,695,000 = $6,450,000 +$957,000 + X, so X is $1,288,000. At 11% interest rate, $1,288,000 covers $11.7 million in principal.
Question: Having gotten $92.1 + 11.4 + 11.7 = $115.2 million in financing, how much was the remaining $1.3 million a year worth to an equity investor?
Answer: $26 million at a 5% “cap rate.”
Once you take out a few million for fees associated with the various levels of financing and the legal and other costs in the purchase, you still have a building an investor was willing to pay $138 million for. By the way, the lenders for those first, second and third mortgages were pretty happy, too, because that $92.1 million was only holding a 66.7% LTV mortgage, the second lien holder was looking at his LTV extending from that 66.7% level up to 75% LTV. Even the B note third lien holder was happy, because their exposure was from 75 LTV up to 83.5%….
Fast forward to 2009/2010.
Assuming the landlord has kept all of his tenants, and he hasn’t had to give lower lease rates, we still have to deal with operating expenses (e.g. heating, cooling and custodial) and taxes. They must have increased by 10% or so over the four and half years, so now the building is throwing off $9 million a year after fixed costs, and the B note is in trouble, but still performing. At roughly $7.5 million a year, the first and second liens are still OK, but certainly hoping to see higher rents come in soon, with refinancing only five years away.
Unfortunately, all building owners aren’t as lucky as our imaginary building’s owners, so they are probably dealing with lower rent and higher expenses. On top of that, plenty of them got even more than our imaginary building’s 83.5% financing.
Enough of those B notes and even second liens are now defaulting (10% or more) that no one wants them, and the CDO market that financed them in 2005 is dead, dead, dead.
Today, a prime loan might be available for just under 6%, but it will probably take 1.6x coverage ($5.625 million available for debt service from that $9 million net income). At a mortgage rate of 6%, the $92.1 million first mortgage could be refinanced and cover $1.5 million in fees, for a total financing of $93.75 million.
But the remaining $3.375 million in annual cash flow would only be worth $34 million to a buyer that wants 10% cash-on-cash return. The price of the property has dropped to $125 million. Now the first mortgage lender has a problem. They won’t lend more than 65% LTV in this market, and the $125 million value makes their mortgage more than 70% LTV.
We go back a re-work the numbers. If the borrower/buyer of the property comes up with more cash, the loan can be dropped to $85 million. At 6%, that loan will need $5.1 million per year in interest payments. That leaves $3.8 million per year in cash flow, nominally worth $38 million. But if $38 million is 35% of the value of the building, then the building is only worth $108.5 million, so the 65 LTV loan has to be for less (and so on)….
Eventually, if the buyer/owner really wants the building, they’ll accept less than a 10% return on $40 million or so invested, and settle for borrowing $70 million or so in financing, effectively re-pricing this $138 million building at around $110 million. That’s a lot of dough to come up with to hold onto a building the owners bought during the high times with $38 million in cash, plus financing.
But the reality could be much worse. We assumed that rents were still holding at their 2005 levels.
What does this all mean to the current bondholders in the securitizations? If they hold the B note, or paper backed by a pool of B notes, they are already in big trouble. If they hold the first mortgage, they’re sweating, because a 15% drop in rents puts the owner in the position of defaulting even on the first mortgage.
Needless to say, if the building only throws off $17 million a year in rents and has fixed costs of $11 million, there is NO WAY it will sell for enough to cover even the $92 million first mortgage it is carrying now.
This scenario will be playing out in office parks and malls near you for the next couple of years. Real estate bankruptcy lawyers are going to be busy.