Analyzing the Popular Proposals for Mortgage Principal Writedowns, Part I
This is the first in a series of articles intended to review the matter of principal reductions for “underwater” residential mortgages. In a few installments, we will discuss:
• The numbers quoted in the proposal as stated in the media
• The issues surrounding the stated numbers
• The costs to the Borrowers
• The costs to the Taxpayers
• The costs to the Banks (and eventually the Taxpayers) and Timing
• The “fairness” of the proposal
• The most recent proposal from banks to a “finite number of current borrowers”
• The ultimate solution to the decline in home values
Today, we're going to focus on the first two points. Enjoy reading and please leave your thoughts in the comments section, as we know this is a rather contentious debate.
Recently there has been a lot of talk from the likes of #OccupyWallSreet, Elizabeth Warren in an ABC News segment and Martin Feldstein in a New York Times Op-Ed piece about the residential housing sector and the “need for principal reductions”. They (the pundits and politicians) toss this “solution” out as a panacea for the national residential market that will cure the continued decline in home values. The thinking then turns to the magical notion that increasing home prices and increasing sales volume will result in a decline in the unemployment rate and the national economy will be reinvigorated.
Well, we don’t buy what they are trying to sell. This will get a bit long winded, but let us examine some of the “facts” they introduce and try to see what this really means in dollar terms for borrowers and banks.
The numbers quoted in the proposal as stated in the media
As the New York Times Op-Ed piece states, we could easily write these mortgages down to save borrowers a ton of money ($350 billion!) and the banks will only need to eat half that as the government (read: tax payers) will take on the rest. Maybe the largest issue with this principal reduction scheme is when people try to throw out numbers and rely on the sheer size of the numbers to impress people in to believing that this will make a magical impact on their monthly expenses. Let’s dive in to the nitty-gritty of these calculations before we get in to all the other reasons this is a horrible idea.
First, we need to get some statistics to base our assumptions on. We pulled ours from the US Census Bureau and a particular Saxo Bank Capital Markets study from August 2009 that we think does a god job of tackling a large issue and giving some solid data to the reader.
The US housing market is made up of approximately 160 million households living in somewhere between 115 and 130 million housing units. Roughly 90 million of these housing units are of the single-family residence (SFR) type and approximately 80 million are occupied full time (primary residences). Of these 80 million housing units approximately 27 million are owned free and clear of debt while the other 53 million are encumbered by a mortgage of some sort. The total value of residential real estate in the US is approximately $19.3 trillion. This total housing stock value breaks down as $8.7 trillion in equity and $10.6 trillion in outstanding mortgage debt. If 53 million housing units have $10.6 trillion in debt that equates to an average of $200,000 in debt per housing unit.
The issues surrounding the stated numbers
Given these statistics we can start to make some very general assumptions. It is important to note that these assumptions will be imperfect as you would need much more accurate micro level statistics to create a true representation of the markets we are discussing, however, for the purposes of this article these numbers will suffice to have the conversation. Additionally, using the same simplistic math as the people backing this principal reduction scheme is kind of fun as it shows how wrong they can be when using their own facts.
These articles and politicians cite that 15 million homeowners are underwater on their mortgages, having a loan to value (LTV) of greater than 100%. 15 million sounds like a lot, but it is 9% of the total households and 19% of the primary residences, a minority whichever way you slice and dice it. Of these 15 million underwater borrowers, 11 million have LTVs greater than 110% and nearly 7.5 million have LTVs greater than 130%. This means that $3 trillion in outstanding mortgages have LTVs greater than 100%, $2.2 trillion greater than 110% and $1.5 trillion greater than 130%. So roughly 10% of homeowners made some really poor choices by jumping in to the market at the top of the valuation bubble and didn’t have adequate (or any) equity to invest as their down payment.
The grand principal reduction idea suggests that we take the most underwater borrowers (those that made the worst purchases with the least to put down) and we should have their principal reduced to an LTV of 110% in exchange for granting personal recourse of the loans (meaning they can be sued for the amount of the mortgage they don’t repay, in contrast to the current system wherein the bank forecloses and sells your home). The real sales pitch here is that “If everyone eligible participated, the one-time cost would be under $350 billion.” A quick calculation shows that an average reduction of 16% (all the loans with LTVs greater than 110% reduced to 110%) to the 11 million underwater mortgages will indeed total $352 billion. The caveat being what we stated in a previous paragraph, where these are incredibly generic assumptions as most of these underwater mortgages do not have a $200,000 balance. A significantly larger balance, probably on the order of an approximate $300,000 average, with a huge proportion being even larger than that when figuring that the largest valuation declines were in higher valuation states such as California, New York, Arizona, Nevada and Florida, is to be expected. Should the average underwater loan amount be $400,000 this would immediately double the costs to $700 billion.
Are you starting to see why the principal reduction scheme won’t function nearly as the back of napkin math would have you believe? It’s dependent on overly simplified averages, not the accurate micro level numbers.
Later this week we'll delve into the further issues mentioned at the beginning of this post so stay tuned!