If WE owned a pool of loans… would WE allow principal reductions?
It seems that there’s quite a bit of discussion lately about principal reductions. Are they the best possible medicine for homeowners, investors, the housing market and our economy as a whole? Or are they going to cause the global financial system to come crashing down around us, trigger quadrillions in counterparty payments, the reversal of the north and south poles, and possibly start World War III?
The debate should center on whether reducing principal will do two things… significantly reduce or even largely eliminate the potential for future default, and lead to increased consumer spending as homeowners feel something akin to the wealth effect and decide it’s safe to spend at least a little bit once again. Hey, look… I know it’s news to many, but when your government abandons you completely and publicly… it does tend to make you more thrifty.
There are a lot of people that seem to want to base the decision on the financial impact that principal reductions would likely have on those that own the loans being reduced. It’s not the right argument to have but I’m not naive enough to think I can stop those framing the debate in this way from doing so.
Well, the truth is… there’s simply too much involved to make a ‘YES’ or ‘NO’ answer on principal reductions meaningful.
So… first, let me use an oversimplified example in order to suggest what the answer isn’t:
The ABC Trust owns a 30- year fixed rate mortgage with a face value (“FV”) of $100,000. Whoever is servicing the loan grants a principal reduction of $40,000.
Question: How much will the investor lose as a result of the $40,000 principal reduction?
A. $40,000 B. Less then $40k C. More then $40k D. No way to know.
The correct answer is ‘D,’ there’s no way to know the answer with the information provided.
To begin with, individual loans are never valued, only the pool can be valued. There’s no way to calculate the value of a pool of loans after principal has been reduced without knowing the assumptions used to determine its value going forward.
So, there’s no point in getting in a discussion about principal reductions with someone who wants to use a single loan as an example… as in the following sentence: “Let’s say the loan amount is $400,000 and the principal gets reduced by $100k… that’s a lot better than the home going into foreclosure and selling for $200,000 or even less.”
If that’s how the discussion on principal reductions starts off, reach for your cell phone, explain that you downloaded the Dog Whistle ringtone… answer it, hang up… start apologizing for having to pick-up your incontinent Labrador who has run out of Depends while at your mother’s. Either that, or as sincerely as possible just reply with… “Yep, that’s true,” and then order another drink.
To really understand what happens when trying to determine the present and future value of a pool of loans within a mortgage-backed security, let’s put ourselves in the place of the investors… we’ll be investors together… and we’ll want to start when the pool was born… a primordial pool, if you will. Come with me… in the way-back machine…
OKAY, THE YEAR IS NOW 2005…
We’re sitting around in my back yard one day having a cold beverage… I tell you my wife wants to see Brokeback Mountain, but I’d rather break my back climbing a mountain. No one can believe how Dubya handled Katrina, and Jack Johnson’s “In Between Dreams,” is awesome.
Interest rates are still low, credit is flowing and homes are the place to invest. My Uncle Fester passed away recently and left me some dough. You just purchased, listed and sold a home in a brand new gated community still under construction, all in under four hours, and have a decent size pile of cash as a result.
So, I bring up an idea I’ve been tossing around lately… why don’t we invest in a pool of mortgages, set them up in a trust with a servicer, and create ourselves a decent little income stream. You don’t know too much about it, but you say…
“I might be interested… how would that work exactly?”
Funny you should ask… and here we go!
1. There’s a pool of loans that we’re looking to buy for our new trust… the DOER Trust 2012, which will end up being an RMBS, or Residential Mortgage-backed Security. Ultimately, to keep things simple, our pool will have 100 loans in it, and each will be paying a fixed rate of interest for 30 years.
But, in order to buy our 100 loans, we’ll need to determine how much we’re willing to pay for them… that is to say, how much they’re worth… to us. Then we’ll know how much to offer the CEO of Oceanwide Home Loans, Tangelo Godzilla, for the pool. He’ll want to negotiate, but we need to know how high we’re willing to go before we start haggling.
So, how do we determine how much we are willing to pay for the pool of 100 mortgages? It’s not hard… in fact, we’re going to take it step-by-step and keep it extra simple on purpose so we can see how the process works, before making it more complicated.
2. First we’ll take the loans all the way out to 30 years and see what the total amount of money we’d would be if every loan paid as agreed for the entire 360 months. Since they’re all 30 year fixed rate loans at 5%, it’s easy to do the math.
Each loan principal is right around $100,000… and at 5%… we’ll receive $93,256 in interest payments, according to my mortgage calculator. We’ll have 100 loans, so that’s $10 million, and interest payments will come in at $9,325,600.
3. Now, obviously we can’t really expect all of our 100 loans to pay as agreed for 360 months, and there are four major factors, and a slew of miscellaneous other factors or costs that will impact how much money we actually collect on our loans:
- Early payoffs from refinancing – Not many people that take out a 30-year mortgage keep it for the whole 30 years. Some people sell their home and move into another, others stay put, but take advantage of fluctuating interest rates and refinance when rates are lower than 5%, or for whatever reason at whatever the rate.
- Accelerated payments – Some people pay off the loan early by making 13 payments each year, or when they received an inheritance from their Aunt Betsy when she passed on… or whatever.
- Loan defaults – Some people, over a 30-year period, will run into serious financial difficulties for one reason or another… maybe due to a divorce, an injury or illness, perhaps because the plant closed unexpectedly… or it could be a gambling problem… anything could happen over a thirty year time frame. When life happens, some will be forced to default on their loans and we’ll have to foreclose and resell the house, which brings us to the fourth factor that can hurt our returns…
- Loss severity ratios – Loss severity is another way of saying how much will we lose one each home when a borrower defaults and we have to take back the home and resell it. How severe do we think that loss will be… or, if there will be a loss at all.
- Servicing costs, trustee fees, legal fees, insurance, taxes, etc. We’ll want to hire a servicer to handle the payments and loan defaults, we’ll need to pay a trustee and lawyer or two, plus whatever other miscellaneous costs are involved.
4. So, absent a working crystal ball, how will we know what amounts to factor in for each of these areas, I hear you ask.
No problem, I reply… I’ll call a ratings agency, they’ll have lots of data on stuff like this for loans all over the country, and to reduce our risk, we will want our 100 loans to be geographically spread out around the country. That way, if the Florida housing market goes into a slump, we’ll be okay because loans in New England and Texas will be okay.
The ratings agency will have a plethora of data on home loans all over the country, and how they perform over various periods of time, and based on borrower demographics, down payments and FICO score, and degree of underwriting scrutiny.
But, it’s important to realize that the data will only be… well… data.
There will be ranges and other uncertainties and to some degree, we’ll have to decide what assumptions about the future we’re comfortable using in our valuation. We can get “expert” help from our guy at Morgan Stanley, to be sure… but no one is truly “expert” at knowing what the future holds, so at the end of the day, we’ll have to make the call.
5. So, starting from our calculations of how much our loans would produce if all paid as agreed over 30 years, we make assumptions as to how much the different factors will impact that amount, and start deducting.
6. The first thing we see from the historical data on home loan performance is that very few people end up keeping their loan for the entire 30-year term. In fact we see that in some parts of the country, the average 30-year fixed rate loan performs for just five years before being paid off either by refinance or prepayment, while in other areas of the country, the average lifespan of such a loan is seven years.
7. We make a decision to figure on a 10-year lifespan for our 100 loans, knowing that some will only make it for three years, while others might still be performing for 15 years or even longer. Ten years feels about right to both of us, so we start using 120 months, instead of 360 in our cash flow calculations.
And because we’re now only expecting 10 years of payments, we won’t have to forecast as many defaults going forward, because over 30 years the chances of a life event causing serious financial difficulties is much greater than over a 10-year period. (This is why long term rates are always higher than short-term rates… the idea is to be repaid before a life event strikes the borrower, and the longer the term, the higher the risk.)
8. Now, as to the issue of forecasting prepayment speeds, we recognize that as interest rates go down… prepayments go up as borrowers take advantage of the lower rates and refinance their loans. And conversely, as rates rise, the expected speed of prepayments goes down. So, we make some assumptions as to whether we believe interest rates are going in the near term and plug those assumptions into our valuation model.
9. As to loan defaults, we see that the historical average over the last 20 years is one number, while the average over the last 10 years is another… and over the last five the number changes again.
Factoring in the stringent underwriting standards that CEO Tangelo Godzilla promised that Oceanwide has in place, and the fact that our loans are all full-doc, fixed rate, to borrowers with over a 700 FICOs, and with Loan-to-Value ratios of 80%… we base our projected defaults on historical averages, plus or minus a factor that reflects our views of what’s ahead, and we enter our assumptions about defaults by year over the ten year period.
10. Next up will be our forecasts of loss severity ratios, or in other words how much do we expect to lose when a borrower defaults and we have to foreclose and resell the home. In that event, we’ll have foreclosure costs, legal fees, property preservation costs, filing fees, real estate commissions, sales price depreciation/appreciation, et al.
We’ll want to forecast the percentage of defaults by year. And whether we forecast a loss severity ratio of 100%, more than 100%, or less than 100%, and this is the sort of data that will be attached to the rating for our RMBS… AAA, AA, A, A- et al.
Perhaps we think some homes will experience a 50% loss severity… others might be less… and if residential real estate gets hot in a given area… in some instances we could conceivably end up selling the homes for more than the amount we originally paid.
The determining factor for forecasting loss severity involves what the banks refer to as REO Hold Time, which is the number of weeks or even months it takes to foreclose, once the borrower stops making payments, and when the home has been re-sold to a new borrower, in which case we’ll receive our principal back, but no more interest payments.
The reason for this is that, when you’re talking about vacant homes… time is not on the investor’s side. Houses don’t produce money while they sit empty. In fact, they can end up costing a lot of money depending on how long they remain on the market unoccupied… which is why it’s entirely possible that after a certain point… our loss severity could exceed 100%… meaning a given home could end up costing us quite a bit more than we paid for it.
11. It’s important to note that loss severity is the BIG Kahuna, when it comes to destroying the accuracy of our forecasts used in calculating the valuation of our pool of loans, meaning that it has the largest impact of any of the factors.
The reason is simple… all of the other factors involve fewer zeros… they’re forecasts are in terms of months, or lost interest payments… but, if we have a $100,000 loan that defaults… and it then takes us two years to evict the borrower, during which time he or she made no payments… and then we have to put $30,000 to bring it back to code… only to find that the housing market has softened and as a result, even though we lower the price every few months… it doesn’t sell for over two years.
And when it does, it goes for $50,000… but after commissions, legal, repairs, taxes, et al… and it shouldn’t be hard to see that we might end up having a loss severity ration of 200%… losing $100,000 over the amount paid for the loan.
Just consider… how many monthly mortgage payments of roughly $500 get drained from our forecasts of future cash flows by a $100,000 in loss severity from even one property.
12. So… these pools are initially valued based on fully amortizing all 100 loans and then subtracting in line with the assumptions discussed above, along with the various fees.
Once we’ve done that, using our assumptions and fully amortized 10-year period, we’ll need to conduct what is referred to as a Net Present Value (“NPV”) analysis, the outcome of which will be the Present Value (“PV”) of the future cash flows going out 120 months. The PV is the amount we’d pay today in exchange for the 10-years of mortgage payments from the loans.
Some people find it easy to understand NPV by thinking of it as “the time value of money.”
Here’s the question one would answer using an NPV formula: Would you rather have $1,000 today, or $2,000 in ten years? To know which would be a greater amount, you need to know the Net Present Value today… of $2,000 in ten years. And to do that calculation, you need to determine a “discount rate,” which can be thought of by answering the question…
Where could I invest the $1,000 today… and what interest rate (“discount rate”) could I expect to earn were I to invest for 10 years? By determining all that… how much would I have in 10 years at that rate of interest?
When I was in grad school, the professors used to tell us to use the rate of interest available on the 10-years U.S. Treasury bond, because it’s an investment that is considered as safe as cash. However, as investors we might choose to use our Internal Rate of Return (“IRR”), or some other metric that we’re comfortable using.
13. Okay, so once we’ve done our NPV calculation, we’ve got a figure that represents the PV… Present Value… or, again the amount we’d pay today for the future cash flow of the pool of loans… BASED ON OUR ASSUMPTIONS for defaults, prepayments, and loss severity… minus the other associated costs for servicing, legal, taxes, insurance et al.
Our PV number, by the way, will be greater than the FV, or Face Value, of the loans, which is simply an acknowledgement that interest is being paid on the loan amounts.
Still with me? If not… go over it again slowly… you can do this.
14. Now, we’re ready to get in the car and head on up to Calabasas to Oceanwide Home Loans to meet with Tangelo Godzilla, and make an offer on the pool of 100 loans.
Oceanwide’s sales people are going to meet with us in the conference room and tell us that the RMBS market is extremely competitive… and how all of Oceanwide’s offerings are over-subscribed. Tangelo himself may even come in for a few minutes and act like we’re old friends, offer us something expensive to drink… you know the drill.
He’ll tell us all about how he built the company from the ground up… and there’ll probably be one of those photos of the CEO with the President of the United States… the kind you get by paying $5ka plate at a political fundraiser.
We don’t get sucked in by any of this… we’ve already done our own detailed analysis, made our assumptions… and we tell them we’re prepared, based on that analysis, to offer 105% of Face Value for the pool of 100 loans. But, Tangelo says he can’t make that deal… he’s got too many investors clamoring for his pools of loans with his Best in Class underwriting system… if we want the pool we’ll need to pay 106% of the FV.
We consider our calculations that determined the PV and decide we have a little wiggle room, so we agree. We sign the paperwork, write the check… and we’re the proud new owners of a pool of 100 residential mortgages from all over the country… underwritten by Oceanwide, who even offers to service our loans at a lower cost that we were quoted by Aurora… so we make the deal.
We’ve bought our pool of loans for 106% of Face Value, which we refer to as having paid a six percent premium.
As we’re leaving, we happen to notice a sign in Tangelo’s expansive offices. It says…
Underwater All Over the World… Oceanwide Home Loans.
But, neither one of us gives it a second thought.
“He must have a boat,” you theorize.
“Yeah,” I reply. “Or maybe he’s a big time scuba diver.”
“True,” you say, turning on the radio.
We drive home happy as clams because we’ve just invested in the next ten years of income, and are pretty pleased with ourselves at how we handled things.
“We’re just getting started on our way to being real estate tycoons,” I say while driving.
THE NEXT QUARTER…
At the end of each quarter we reevaluate the pool of loans. We receive the asset list from our servicer representative, Linda D. Manhattan, and that’s how we find out which of our loans are performing as agreed, and whether any have prepaid or defaulted.
And using the up-to-date data on the report, each quarter we go through the same valuation exercise we did to arrive at the PV of the future cash flows, over and over… each time to determine value of the pool of loans.
But now, since we have real investment experience data, when we catch something being out of sync with the assumptions from the forecasts we plugged into our original valuation model… we decide whether it’s only one quarter or whether we should change our assumptions to reflect the actual data in the forecasts.
In fact, there are a variety of reasons having nothing to do with Oceanwide’s quarterly report that we might decide to change our original or even our updated forecasts and assumptions, including…
- If the availability of credit were to tighten significantly – The credit markets tightening significantly is going to reduce prepayments to lower levels than previously thought, which on one side of the coin would increase our future cash flows because we’ll be getting payments for a longer period of time. And, on the flip side of that scenario is that the longer the loans are performing, the greater the chance that a life event is going to impair a borrower’s ability to repay… which could dictate increasing our forecasts of default percentage assumptions.
- If credit loosens, and interest rates rise, that will impact your prepayment assumptions too.
- As the quarters pass, if we start to see that the actual number of defaults has started trending higher than originally assumed, or if loss severity is coming in higher than forecasts, we’d change the assumptions used in the value calculation used, which would impact the pool’s value.
- If we’re in an environment of rising home values, then we might reduce default assumptions, and conversely if home prices start to fall, we may assume that the number of defaults will increase, as compared with past assumptions.
- If in stable housing market, REOs sell in certain time frames… but in a distressed market REO hold time increases along with loss severities.
FAST FORWARD – 2012
Our current housing market is distressed because of:
- Extreme credit tightening, government is only lender
- High and persistent unemployment
- Increasing negative equity
And all of this will cause us to change our default assumptions used in valuation, and we will see our loss severities rise. We could sell the homes we’ve had to repossess… and recognize the loss… or we could just keep the houses and assume real estate prices will return by the time we sell and therefore not recognize the entire loss.
15. Meanwhile all the rest of our loans are still paying as agreed, but you should start to see that in a market such as today’s perfect storm, eventually the defaults and severity of losses are going to eat away at the performing loans and hence your future cash flow. At some point, in reality, you could end up with a trust of all houses and no cash flows, but houses aren’t income generating so you’d either have to liquidate them… or find an alternative use for your portfolio of houses, such as renting them out.
Obviously, however… most bond investors are not looking to be property managers, in fact they wanted loans to be geographically spread out to reduce market risk, which means managing rental properties all over the country isn’t something desirable to a bond investor. To do that… you need “boots on the ground.”
NOW LET’S ADD SOMETHING… HOW ABOUT SOME LEVERAGE?
16. Here’s the really interesting thing… if we were bankers, then we probably borrowed money to buy this pool of loans in the first place, and now as defaults rise and as cash flows are falling our creditors may want more collateral for the now much riskier loan… or they may force us into bankruptcy and liquidate our position that way.
In real life, all sorts of pension plans, insurance companies, and sovereign wealth plans lent the money to hedge funds to buy the pools of loans, and now want their money back… with interest.
17. But, they may not want to force us into bankruptcy because they view the market as being so bad that they won’t recover much by going that route, therefore they decide to wait it out and leave us alone. Or, they could sue us to repurchase the loans because they claim we didn’t “rep and warranty” properly, and lied about underwriting. In other words, sue because it wasn’t what they thought they were buying.
18. But, we also could still be paying the weighted average payments to the investor who loaned us the money to buy the pool, using some combination of what’s referred to as “over-collateralization” (i.e. Oceanwide gave us 104 loans instead of 100), private mortgage insurance from PMI/AMBAC/MBIA, or credit enhancement in form of extra cash in the pool, or in some instances, maybe some very limited ability to swap out good loans for bad.
19. OR, WHAT IF WE COULD BORROW UNLIMITED AMOUNTS FROM THE DISCOUNT WINDOW AT THE FEDERAL RESERVE AT 0%?
Of course, we could never even hope to do that because we’d have to be a Bank Holding Company, and there’s no way the Fed would ever make every company in the country a BHC… that wouldn’t be safe… it could put our banking system at risk if the money couldn’t be repaid and the collateral was worthless… oh wait… hang on… now that I think about it… that’s exactly what the Fed did, didn’t it?
20. Now do you see where the $16 trillion went?
But… if we’re not getting the money from the payments on the mortgages we originally bought, because most or all have defaulted… but we’re still paying the investors who loaned us the money by using borrowed money from the Fed… what’s going to happen when… I mean… aren’t we blowing a giant bubble and one day it has to pop?
What are we hoping for… that someday demand for our assets… the ones for which there is no market today… will one day return and they’ll be worth billions once again? Could that happen?
Assets that re-inflate themselves? Re-inflating assets?
Sure, okay… why not?
Okay, so do you see how accounting for the value in a pool of loans works, at least fundamentally?
If so, then you should see why it is that we really can’t know whether granting a principal reduction on some number of loans in a given pool will result in a loss… or not. We won’t know what the current set of assumptions are that are being employed in the calculations that are done to value the pools assets now, so we won’t know whether writing down principal creates a loss, because depending on those assumptions… it could create a gain.
The reason to write down principal on a mortgage is to change the future outcome of the entire pool of loans and even other pools of loans, by stabilizing the housing markets… by stopping the free fall in prices… by stopping foreclosures, which at this point won’t stop on their own until they’ve destroyed our nations citizenry essentially in its entirety. They’ve already been allowed to go too far, and like a forest fire that burned too long before the fire trucks arrived… left alone there’s nothing to stop it… it’ll burn until it runs out of forest.
If you and I actually were investors in our own pool of loans… and here we are in 2012… if we’re honest, ethical, responsible people… we’ve long since changed the assumptions being used to value the pool’s future cash flows such that our expectations are very low. Our assumptions should be assuming ongoing unprecedented percentages of defaults and long-since unrecoverable loss severities.
By writing down principal, the default assumptions could be reduced along with the loss severities and therefore the value of the un-named pool of loans should increase… there should be no loss in many pools. And if there is a loss reported, it’s cause isn’t the reduction in principal, it’s the pool’s trustee who’s allowing assumptions to be used that are nothing more than pipe dreams.
In Gretchen Morgansen’s article about Ed DeMarco of the FHFA yesterday, she points out that writing down principal makes it more likely that a borrower would be able to pay the second, and she characterizes that as a bailout for those banks. Others discuss the situation in similar terms.
It’s classic Fannie Mae… it’s classic banker-think… It’s exactly what’s been done to Greece. It’s what they tried to do to Iceland. And it’s not the case.
For that to be the case, one would have to assume that sans principal reduction, me the borrower would have paid the first mortgage… and that’s the point… I would not have paid the first or the second. By writing down my loan, while it may make it possible for me to pay the second, it’s not the point. What’s it’s going to do, most importantly, is stop me from walking away and paying no one, not the first or the second.
It’s really amazing to me, but obviously there are a lot of people out there who are not only financially successful today but they’ve always been so… many are writers of blogs, others are journalists at traditional media outlets… and some run the FHFA and the Treasury… and they claim not to know the impact of principal reductions. They can’t study it, as it has no precedent… so they don’t know how to think about it.
And that’s precisely why where here today economically speaking. They couldn’t imagine foreclosures could get this far out of hand, because they can’t imagine themselves in foreclosure under any circumstances. And if they can’t establish it historically or through a study, since they can’t envision it happening to them, they are driving us directly off the edge of the Grand Canyon.
For anyone who is unsure about what will happen if you don’t take action to stop what’s in free fall… stop the foreclosures… it will tip… no one will have control… people will lose all hope… and a hundred thousand will walk away from their home in Phoenix over a matter of weeks… and there’ll be no talking to them… no asking them to return… it’ll happen so fast I won’t be surprised if Bernanke doesn’t even know its happened for weeks afterward when it shows up on some chart as an blip in the Southwest Sector… or whatever.
It reminds me of an article posted by Yves Smith on Naked Capitalism maybe six months ago. She was responding to a study that showed how short a period of time many Americans were reporting they could live were they tom lose their jobs. She admitted that it had taken her by surprise, she just couldn’t believe that many people were living lives that tenuous.
I wrote about the same study that day, only I was surprised that the numbers were as high as they were, not as low. I couldn’t believe how many reported that they could make it five months after losing a job. I thought about Yves being shocked by it… considered that my parents would undoubtedly be shocked as she was too.
It must be awesome to have lived a life that far removed from the world that worries about money every hour of every day. We have 50 million on food stamps, and half of them are working poor. Money is all they think about.
But you don’t have to go that far… I promise you, and I know this to be a fact… there are a million fathers in America… if not more… that in the last year have wondered whether their life insurance policy would pay off for their families after their suicide… or something very close. And lest you think I exaggerate… I chose the number because it’s the smallest possible one… no way it’s less than a million… it’s likely many more.
Not sure you can survey to verify it empirically however. So, I guess most won’t ever know how that thought feels inside, which means they’ll never know how the people who they see each day actually think and feel.
When times are okay, these people in policy positions aren’t making many happy, but in today’s crisis environment, they are about to allow the nation to burn to the ground from within… without even knowing that they are the ones placing us on a perpetual precipice. It’s been burning for several years now… they can’t see the flames… they can’t smell the charred flesh… but I can.
It’s the people, they are burning from within, and by the time DeMarco and others like him see what’s happening millions will fall into heaps of ashes. And then what? Will he say he didn’t know… that he’s sorry… he thought his degrees made him smart enough to see anything, and yet he couldn’t even smell the nation as it burned.
Do what you will… I can barely care anymore without crying or screaming… and in the end, I’ll fight to protect my own, I suppose… but to those who don’t know and who are allowing this crisis to continue even one more day… may your God forgive you, and take mercy on your soul.
P.S. If you liked this…
You’ll love Part 2… An Insider’s View of an Actual RMBS Securitization at Mandelman U.