Principal reductions: DeMarco vs Geithner

By Felix Salmon
July 31, 2012
Ed DeMarco vs Tim Geithner today, on the subject of principal reductions, and the fight is getting ugly.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

It’s Ed DeMarco vs Tim Geithner today, on the subject of principal reductions, and the fight is getting ugly.

DeMarco, in a letter to Congress, explains that the Treasury has put a lot of effort, and is willing to put very large sums of money, into something with the rather unwieldy name of HAMP PRA, where PRA stands for Principal Reduction Alternative. But here’s the thing: PRA is going to get nowhere unless Fannie and Freddie sign up for it. And they can’t sign up for it until DeMarco signs off on it. And DeMarco is refusing to sign off on it.

DeMarco has released a 15-page paper explaining his decision, although in reality his letter, and the reasoning in it, is much clearer. He basically says that principal reductions would be costly to implement, and he doesn’t have a lot of time for Treasury’s offer to pick up the tab: “although principal forgiveness may provide some financial benefit to Fannie Mae and Freddie Mac,” he writes, “it presents operational challenges for them and their servicers as well as a risk of loss to the taxpayer”. It’s not his job to worry about costs to Treasury or taxpayers generally — but he’s making it his job.

But the main thrust of DeMarco’s argument is less financial than it is moral. Any kind of principal-reduction strategy risks encouraging strategic default, and DeMarco hates the very idea of strategic default. And even if there’s no strategic default at all — and the letter from Geithner makes a very strong case that strategic defaults as a result of this plan would be de minimis — DeMarco still hates principal reductions on, well, principle:

Perhaps the greatest risk of the Enterprises’ allowing principal forgiveness is one with far more significant long-term consequences for mortgage credit availability. Fundamentally, principal forgiveness rewrites a contract in a way that other loan modification programs do not. Forgiving debt owed pursuant to a lawful, valid contract risks creating a longer-term view by investors that the mortgage contract is less secure than ever before. Longer-term, this view could lead to higher mortgage rates, a constriction in mortgage credit lending or both, outcomes that would be inconsistent with FHFA’s mandate to promote stability and liquidity in mortgage markets and access to mortgage credit.

This is a classic “parade of horribles” argument, and it’s not a particularly strong one, either. As Jared Bernstein says, “in unusual times, like the aftermath of the worst housing bubble implosion in decades with 30+% price declines, guess what? Write downs happen.”

But the weirdest thing about this argument is that the horribles aren’t particularly horrible. Higher mortgage rates? Um, fine: no one is exactly complaining that mortgage rates are too high right now. A constriction in mortgage credit lending? That’s fine too: it was too-lax credit lending that caused this whole problem in the first place. Both together? Even that’s fine: it would help bring homeownership rates down from their current too-high levels, and encourage more people to rent rather than own, creating a more flexible national labor force.

The fact is that while it’s imperative that we fix the problem of broken mortgages issued at the height of the credit bubble, the last thing we want to do is return to those days and tell anybody who wants a house that they can just go out and buy one, whatever their creditworthiness or cash position. If you want to make mortgages safer, you should ensure that homeowners have large amounts of positive equity in their homes. That means significant down-payment requirements for people buying houses. And it also means significant principal reductions for those who are currently underwater.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

I think DeMarco hinted at it, but didn’t get to say it explicitly:

Principal reduction rewards those who took out unaffordable mortgages, and hence increases future moral hazard. It really is that simple, and I know that Felix has thought about this.

Suppose for the sake of argument that we start to enter another bubble in five years. Oily fly-by-night mortgage brokers who last time around were selling everyone who couldn’t afford one dodgy loans will now be able to say to those potential borrowers sitting on the fence, unsure if the marketplace is overheated, “Well, buy now, and if we are indeed in a bubble and it pops, then the government will bail you out and force a principal reduction, reducing the amount you owe! If prices go up, you win by taking advantage of increased equity! It’s heads you win, tails you win, you’d be a fool to not buy now!

And of course, you would be a fool to not take advantage of that. The only way to be sure that this wouldn’t happen is if that person sitting on the fence can think about what happened to their neighbor down the street who is still far, far underwater from the subprime mortgage he took out in 2007 and is still paying on it because he’s in a non recourse state and thoughtful people like Ed DeMarco stopped principal reductions from being either offered or taken advantage of. DeMarco has the more convincing case that avoiding wholesale principal reduction is a better way to “make [future] mortgages safer”, and Felix’s last paragraph make no sense whatsoever when you realize that.

Posted by Strych09 | Report as abusive

The proposal was absolutely unfair to those who did not speculate and continue to make their payments, month after month and year after year. People should have to stick to their agreements.

The American taxpayer should not have to bail out poor business decisions.

Posted by stevedebi | Report as abusive

The regulator just stopped the scam that would be abused in the swing states before elections to buy votes!

Posted by StopNonsense | Report as abusive

Thank God there are at least a few people like Ed DeMarco in government.

** is tempted to retract all prior complimentary remarks offered in re: FS **

Posted by MrRFox | Report as abusive

Stycho9 wrote: “The only way to be sure that this wouldn’t happen is if that person sitting on the fence can think about what happened to their neighbor down the street who is still far, far underwater from the subprime mortgage he took out in 2007″

Every contract has (at least) two sides – you’re correct in that it wouldn’t discourage people from taking out mortgages but it WOULD prevent people from offering them in the first place. “Oily fly-by-night mortgage brokers” (as you put it) need financing from somewhere and it would persuade those stumping up the money for the mortgages (banks, etc.) to be a bit more careful with their money.

Posted by NickFFF | Report as abusive

Only a small percentage of households are seriously underwater. A third are not homeowners. A third of the homeowners don’t have a mortgage. Only a third of mortgages are underwater. And only a third of those are seriously underwater. Plenty of angst for that 5% to 10% of households that find themselves in a serious financial bind, but are they any worse off than:

* That 5% to 10% (or more) who lost jobs in the recession and haven’t been able to find a new one.

* That 5% to 10% who graduated college but can’t find a decent job.

* That 5% to 10% who suffered heavy equity losses in their home but didn’t have the foresight to leverage the investment to the hilt.

* That 5% to 10% who continues to struggle paycheck to paycheck and couldn’t even begin to pay for a house.

There are plenty of people hurting in this economy, plenty of people who NEED help. Those upper-middle-class homeowners who are underwater on their mortgages are the loudest, but they aren’t the most deserving and they aren’t the most desperate. I hear claims that bailing them out will somehow reinvigorate the economy — but those claims don’t hold water. The numbers simply don’t work.

You might consider what the economy DOES need to recover? I would argue for lower housing prices, lower healthcare costs, and a more equitable distribution of wealth. But look first to the needs of the kids in their 20s and 30s, not to those age 40 and up (like myself). Growth will come from the youth.

Posted by TFF | Report as abusive

@TFF: +1

Posted by Curmudgeon | Report as abusive

I don’t know what the right call is here. But when I read this, see Counterparties promoting a call by Krugman to fire DeMarco, and a prominently posted article from WaPo on the 370,000 jobs that would be created (ugh!), I have to call BS. I recognize that you’re under no obligation to be unbiased in your opinions, but sometimes you just go so far around the bend that it makes me want to not believe a word you write.

Posted by Curmudgeon | Report as abusive

DeMarco’s stance is beneficial to the Obama’s reelection campaign. For the reasons well explained in the above comments – particularly TFF’s – sending billions of taxpayers’ dollars to principle reductions will generate a net ill will that the President can’t really afford. Talking as if you wanted to do something like this, but not actually doing it is probably the political sweet spot. If actually done it would take about 10 milliseconds for stories and ads to appear about the absurdities and injustices of the allocation of national resources. And they’d mostly be true and they’d be easy to understand, like: you’re such a moron for not leveraging yourself to the hilt that you deserve to help payoff your neighbor’s home. Maybe he’ll invite you out on his boat someday – but probably not.

Posted by Eric377 | Report as abusive

NickFFF, I seems to me you perhaps don’t get the “originate to distribute” model that banks use with mortgages nowadays. Original lenders don’t care a whit about responsible underwriting. The risk isn’t on their books, it’s been shifted out to investors via the securitization process.

I haven’t seen any evidence at all, anywhere, that mortgage brokers have any input at all into the degree of due diligence exercised by lenders during the underwriting process. Mortgage brokers want their commission, and they couldn’t care less what happens to the borrower after a closing.

You’re of course correct that “every contract has (at least) two sides”, but from a behavioral finance perspective, the only one public policy such as the one proposed here can (or should?) influence greatly is the borrower’s side. They have to resist the urge to take out a mortgage they can’t afford.

The fear of being underwater for ten or more years would be a great motivator, and would be effective in counterbalancing the sales pitches from shifty unscrupulous mortgage brokers who want to close deals at any cost. If we take away that fear by allowing principal reduction, and thus the future expectation that declines in real estate values will be met by government intervention, then we create or contribute to the catalyst for the next bubble.

That is only one reason of many that widespread principal reduction is a bad idea, and why I’ve never understood why Felix is so attached to it.

Posted by Strych09 | Report as abusive

Posted this down below, but is applicable to this topic and would appreciate critiques.

When a house is seriously underwater, there are four ways out:

(1) Homeowner defaults, leading to foreclosure. Bank recovers what it can, typically a fraction of FMV. Homeowner’s credit rating is dinged for 7(?) years, but (because the foreclosure process takes months) gets to live rent free for a while.

(2) Short sale. Bank receives FMV for the property, without the legal costs. Homeowner’s credit isn’t quite as badly dinged.

(3) Principal reduction. Bank receives a promise of FMV on the property in the future (bird in bush, not in hand). Homeowner retains the property (and any future gains) with no damage to credit score.

(4) Homeowner stays in the home, continues paying the mortgage, and eventually is above water again. For a property that has lost 50%, that might take a decade.

The worst outcome for the banks, by far, is #1. Their recovery, after costs, is a fraction of FMV. But this is frequently also the worst outcome for homeowners. Most cannot buy again without a new mortgage — and qualifying for a new mortgage in the wake of a default is surely difficult! So they rent for several years, repair their credit, and hope that prices/borrowing costs haven’t risen too much by the time they are ready to buy again.

The banks don’t see a heck of a lot of difference between #2 and #3. They get FMV in both cases, and they don’t REALLY want to keep that reduced mortgage. But they are very afraid of the “moral hazard” implied by #3. If they offer widespread principal reductions, then everybody with an underwater loan will want one. Right? There’s really no downside to the homeowner, if it is a free choice between #3 and #4. So the banks are doing everything they can to make principal reductions available ONLY to those who would otherwise end up in foreclosure. (Including telling homeowners that they can’t even apply for a reduction unless they are already in default.)

Does the game theory aspect make sense? Option #3 is best for homeowners, with (depending on your situation) #2 and #4 coming in second best. Option #4 is best for banks, with #2 and #3 coming in a distant second.

So banks threaten #1 (worst or second worst for everybody) to keep as many borrowers as they can in #4 (best for the banks). Homeowners largely cooperate, because they really don’t want to default and because (given rents these days) they aren’t THAT much worse off in #4 than in #1. At least not if they have the income to manage it.

The more people who willingly default, the more likely the banks are to accept #3 as a viable alternative. When they eventually come around, we’ll see a final wave of writedowns with willing participation from the banks.

But from what I’ve seen, calls for principal reduction typically do not adequately address the “game theory” aspects described above. Until they do so, they are doomed to failure.

Posted by TFF | Report as abusive

Also note that the cash flow on a current underwater loan is very high relative to FMV. Suppose the borrower bought for $500k, borrowed $400k at 6.5%, and the house is now worth $300k. The annual payments on the original mortgage come to $30k, or 10% of the present FMV. If the bank can collect on the underwater loan for five years, and subsequently unload it REO at $200k, then they are better off than they would be with an immediate $300k short sale.

Pushing defaults down the road is very profitable for banks, even when it leads to a lower residual value.

Posted by TFF | Report as abusive

@TFF – your #3 option, principal reduction, the bank gets a performing loan with a face amount of current FMV. The bank gets no future appreciation in FMV, AIUI. It’s a straight-out gift of the underwater portion -

and not just to the borrower. The equally big gift-getter is the bank holding a Second Mortgage on the property, if any. A virtually worthless Second can be made instantly valuable by this partial forgiveness on the First. Big banks hold a lot of Seconds – they weren’t securitized like Firsts were.

Posted by MrRFox | Report as abusive

Sorry, MrRFox, was poorly worded on #3. You put it more clearly. What I meant is that the bank keeps a mortgage, rather than getting the cash outright as in #2.

And yes, I wasn’t even getting into the First/Second dilemma. Coupon payments flow to both First/Second proportionately. Principal payments flow to the First before the Second. That is an irreparable asymmetry, with the “fair solution” depending strongly on the time-to-default.

A second mortgage on a heavily underwater home is effectively the interest portion of a STRIP. Values very differently if you expect a default in two years vs. five years.

Posted by TFF | Report as abusive

TFF, great comment, I would just add that “shadow inventory” adds a wrinkle to your game theoretic analysis of point (1).

I’m in California, which is a non recourse state, and it’s not at all uncommon for banks to fail to foreclose on a property so they can avoid marking that mortgage loan to market (i.e., write it down). I have many acquaintances who decided to strategically default and the lender still hasn’t foreclosed or even scheduled the trustee sale 28 or 29 months after the filing of the notice of default.

And of course, even after foreclosure, any number of banks simply hold those properties off the market in order to avoid lowering the overall price level by “flooding the market” with REOs. That’s why cities like San Bernardino are considering using eminent domain to seize underwater mortgages and restructure them for lenders, because the lenders aren’t doing it themselves. That’s why some cities are passing laws obligating banks to maintain properties and do things like lawn maintenance and mosquito abatement, because properties aren’t turning over. And of course, Bank of America is actually bulldozing perfectly livable homes in some areas in order to avoid having to pay for these responsibilities and to reduce inventory in order to prop up prices.

The effect of this and Felix’s proposed widespread principal reduction scheme is to hurt the people who were responsible, waited on the sidelines during the bubble for prices to return to the reasonable/trend level, and now still can’t buy at an affordable price because normal market action can’t take place due to artificial price supports.

Posted by Strych09 | Report as abusive

There are alternative rules for HAMP loan modifications with and without principal reduction. Under principal reduction there are additional alternatives where the lender can reverse the order of certain steps. And any of these sets of rules will find a way to lower the payment to 31% of income.

For a while I’ve had an Unofficial HAMP loan modification calculator up at that can calculate these various rules. It is totally free, and the source code is available under the GNU public license.

While a tool like this can not answer the fundamental questions, it can be used to gain some experience with how the various rules might operate.

Posted by DrPaulBrewer | Report as abusive