by Calculated Risk on 4/20/2008 10:45:00 PM

On Friday I posted a video of an interview of Professor Roubini on Canadian TV. It is well worth watching.

On Saturday, I posted a few comments on why I thought Professor Roubini might be a little too pessimistic. I gave three reasons: 1) I believe starts of single family homes built for sale has finally fallen below the current level of new home sales (note: I'll have more on the housing starts vs. new home sales issue soon.) 2) I think we may be a little further along in the write down process than Roubini, and 3) I felt Roubini might be overestimating the number of homeowners that "walk away'.

Clearly we agree on more points than we disagree, and I hold Professor Roubini in the highest regard (for those that don't know, Roubini is very well respected among his peers).

Today Roubini wrote: The worst is ahead of us rather than behind us in terms of the housing recession and its economic and financial implications. Here is an excerpt on the write downs:

[M]y most recent estimates have been that credit losses on mortgages could be as high as $1 trillion and total credit losses for the financial system could be as high as $1.7 including all the other losses (commercial real estate loans, credit cards, auto loans, student loans, leveraged loans, industrial and commercial loans, corporate bonds, muni bonds, losses on credit default swaps). How many of these losses are borne by banks (I meant both commercial and investment banks in my use of the term “banks”) depends on the allocation of these impaired assets among banks and non-banks.

The argument for a trillion dollar of losses on mortgages alone is based on the following three parameters (two of which an undisputed while a third is more subject to uncertainty. First, let’s conservatively assume that home prices fall about 20% rather than 30% so that only 16 million households are underwater; this assumption is not very controversial as most now would agree that a cumulative fall in home prices of 20% is a floor, not a ceiling to such price deflation. Second, lets assume – as Goldman Sachs does – that a foreclosed unit causes a loss of 50 cents on a dollar of mortgage for the lender as, in addition to the fall in the home price one has to add the large legal and other foreclosure costs including loss of rent on empty properties, risk of the property being vandalized and cost of maintaining an empty property before resale. Third, lets assume – and this is more controversial – that 50% percent of households who are underwater eventually walk away or are foreclosed. Then, since the average US mortgage is $250k total losses from borrowers walking away from their homes are $1 trillion. Goldman Sachs agrees with me on two parameters (20% fall in home prices and 50% loss on a mortgages) but more conservatively assumes that only 20-25% of underwater home owners will walk away. In this case mortgage losses would be “only” $500 billion. But home prices may likely fall more than 20% and with a 30% fall in home prices 21 million households (40% of the 51 million with a mortgage) would be underwater. So, there is certainly uncertainty on how many underwater households will walk away but given the recent evidence of subprime but also near prime and prime borrowers walking away even before they are foreclosed one can be pessimistic on this.

This uncertainty is why Tanta and I have been begging for better data on how many homeowners are actually resorting to ruthless default. Tanta wrote a great primer for the media: Let's Talk about Walking Away (hint to the media!!!)

I certainly agree Roubini's scenario is possible. Last December, I wrote:

If every upside down homeowner resorted to "jingle mail" (mailing the keys to the lender), the losses for the lenders could be staggering. Assuming a 15% total price decline, and a 50% average loss per mortgage, the losses for lenders and investors would be about $1 trillion. Assuming a 30% price decline, the losses would be over $2 trillion.

Not every upside down homeowner will use jingle mail, but if prices drop 30%, the losses for the lenders and investors might well be over $1 trillion.

Although possible, I felt this was somewhat the worst case.

On recourse loans and 'walking away', Roubini argues:

I have for a while argued that in the US mortgages are de-facto, if not always de-jure, non recourse. Indeed, even in states where mortgages (or refinanced ones) are de jure recourse loans these mortgages become de facto non-recourse as the legal cost for lenders to pursue such legal action against jingle mail borrowers can be massive.

Tanta commented on this, and generally agreed with Roubini:

Back in my day working for a servicer, we never went after a borrower unless we thought the borrower defrauded us, willfully junked the property, or something like that. If it was just a nasty RE downturn, it rarely even made economic sense to do judicial FCs just to get a judgment the borrower was unlikely to able to pay. You could save so much time and money doing a non-judicial FC (if the state allowed it) that it was worth skipping the deficiency.

But notice the "willfully junked the property" phrase - aren't these the homeowners that we are talking about when we say someone will "walk away"? Aren't these the solvent homeowners who can make the payment, but decide not to simply because they are underwater? This is one of the great uncertainties, or as I wrote last year:

One of the greatest fears for lenders (and investors in mortgage backed securities) is that it will become socially acceptable for upside down middle class Americans to walk away from their homes.

This is a critical issue, and hopefully someone will provide some research on the number of homeowners actually walking away.