Saturday, March 08, 2008

Return of Capital

Investors are more worried about return of capital these days, and not as much with return on capital. This most starkly evidenced by the nominal yield on 2-year Treasury Inflation Protected bonds is currently negative, about -0.72%. This means that buyers of this bond are happy to accept negative real returns over the next two years. Obviously investors would only accept such a situation during a time when fear is at extreme levels.

But investors won't accept negative real returns forever. And the fear levels have created some very easy opportunities to outperform Treasury yields. No, I'm not talking about MBIA's 14% surplus note or sub-prime closed-end second lien bullshit. Take Ginnie Mae (GNMA) mortgage-backed securities (MBS). GNMA securities are backed by the full-faith and credit of the U.S. Government, so the credit quality is exactly the same as a Treasury bond. The cash flow for these securities is equal to the cash flow that is paid by borrowers, so when a borrower sends in $1 of interest, bond holders get that $1 of interest. If they pay $1 of principal, $1 of principal is repaid to bond holders (well, not exactly, but close enough). If any borrower defaults, Ginnie Mae buys the loan out of the pool, returning 100% of the principal amount to bond holders.

As of Thursday close, GNMA securities with a 6% coupon based on 30-year fixed-rate mortgage loans were trading about $101.5. What your yield would be depends on how quickly the bond repays principal. Since principal is repaid at $100, faster repayments would lower your return.

So let's look at how fast prepayments might occur. At the height of the refi wave in 2004, GNMA 6% mortgages prepaid principal at a 44% annualized clip. If that prepayment rate reoccurred over the next 12 months, assuming a zero reinvestment rate, investors would earn 3.72%. Hey! That might actually keep up with inflation!

Of course, the mortgage market is quite different today versus 2004. First of all, 30-year fixed mortgage rates fell as low as 5.45%. Today we're still stuck just over 6%. Second, mortgage credit was very easy to come by. Today? Its the worst mortgage credit environment in generations. So the odds of seeing 44% prepayments seems unlikely. In fact, prepayment rates over the last 6 months have ranged from 9.5 to 15.6%, with an average of 12.9%. If prepayment rates stayed at 12.9% over the next 12 months, the GNMA investor would earn 5.19%. Don't look now, but that yield level is higher than 5-year Goldman Sachs, Wachovia, or Bank of America paper. By the way, if you are feeling frisky, the yields on Fannie Mae or Freddie Mac guaranteed MBS are even higher.

Of course, I say you'd "earn" 3.72% or 5.19% in yield, but what about price return? Indeed, MBS prices have suffered particularly over the last 6 weeks amidst horrible technicals. Leveraged MBS players are seeing their allowed margins reduced. Mortgage originators sold like mad in late February to try to get ahead of more expensive Freddie Mac and Fannie Mae pricing. Banks are not adding to positions, if not selling themselves. MBS prices fell about 1/2 of a point Thursday(while Treasuries rose more than 1/2 of a point) after the story hit that Thornburg had been served with a default notice. They recovered a bit on Friday.

On the positive side, supply technicals are excellent. You may have heard that housing turnover is extremely low, and bank credit is extremely tight. There won't be a flood of mortgage supply for the market to swallow.

So yes, perhaps the technicals of MBS are difficult right now, but the fundamentals are excellent. Maybe the demand for MBS will come in the form of money managers and retail investors who finally tire of Treasury yields below 2%. How long it will take before this happens is any one's guess. But if you are going to be a fundamental investor, you can't be overly concerned with waiting for a bottom. Similar to what happened in municipals, eventually buyers come in to buy when the value proposition becomes easy to understand. I think MBS have reached this point.


Anonymous said...

Return **of** capital, not return **on** capital?

Didn't that gramps guy write that several months back?

Accrued Interest said...

I believe I responded with "When the 2-year is below 2%, its obviously about return of capital."

Anonymous said...

So how does a quasi retail
investor buy GNMA's
is there a ETF?

Anonymous said...

The problem with MBS is that time scale for return on capital can vary from less than 1 year to more than five years.

If you buy GNMA today, you might get all your capital back in 2009 or 2018.

Mentioning these risks might not be such a bad idea.

gramps said...

Hello again AI...

I stopped posting here a month ago because things at the office have been beyond hectic-- but since my ears are ringing (from an alias being used :) Even though I stopped posting, I have still been reading the posts of others.

We are now getting to the point I feared we would reach if the Jim Cramer crowd got there way (screeching hysterically "The Fed **MUST** ease, they have no idea how bad it is out there!!").

As everyone knew last summer, the world had become leveraged to just an absurd degree. Homeowners had 90% LTV, 95% LTV or higher (they were leveraged 10 to 1, 20 to 1 or higher). It was not unusual for many "conservative" hedge funds to be leveraged 20 to 1, and the less talented by a lot more.

Home prices meanwhile had reached historically unprecedented multiples of income -- we are talking 6-7 standard deviations above the trend. Borrowers were struggling to reach even the 5% down easy underwriting criteria. You never know what will happen, but odds favored a pretty significant pullback.

At 20 to 1 leverage, you have 5% equity-- meaning a 5% move against you and you are Chapter 7. Even during boom times, a 5% pullback is hardly unusual. At 20-1 leverage, failure is a question of when, not if.

Academics have pointed out many times that the Depression was made worse because money just seized up. The lender of last resort (the Fed) didnt step in.

This is an incomplete analysis. Every loan has two players: a borrower and a lender. If you don't want money to seize up, people have to be willing to borrow (seemingly NEVER a problem in the U.S.), but also people have to be willing to lend.

If you have only 5% equity, that means your lender is likely to get stuck with a 2-3% loss during a "standard" market pullback -- and a heck of a lot bigger loss if there is a real pullback. You don't need a PhD from the Princeton economics dept to know lenders are loss averse.

So the party ended in August. Lenders quickly moved to cover themselves against (further) loss by demanding historically standard underwriting standards. 20% down (5 to 1 leverage) became non negotiable for new lending.

Hedge funds started going belly up as prime brokers also demanded reduced leverage / bigger haircuts. Banks were forced to delever at the same time -- meaning there was less money to lend.

Pick whatever you think the base money supply was. I am going with $1 trillion to make the math easy. $1T money supply times 20-1 leverage equals $20 trillion credit. Overnight, $1T money supply became $5 trillion credit -- a 75% contraction.

Lets be blunt: there is absolutely NOTHING the fed is able to do to counteract that, short of running massive levels of inflation. 5.25% Fed funds can be lowered to zero percent, and it won't even scratch the paint.

We watched this scenario play out in Japan 15-20 years ago. But there is one major difference between the U.S. and Japan -- our net savings rate is zero.

When times are tough, if you have savings in the bank (the proverbial rainy day fund), you can always "lend" to yourself and ride out the storm. With no rainy day fund, you must pay whatever price the market wants to get money to tide yourself over.

One surefire way to make sure that money is NOT available is to insist that a lender give you the money at below the rate of inflation.

Plenty of people have criticized CPI as absurdly under-representing the rise in the cost of living. But even if you accept CPI as truthful - it is higher than the yield on the 10yr Treasury.

You are desperate for funds so that you can avoid a fire sale delevering-- and you think putting rates below inflation is going to encourage people to lend?

One year ago, CPI was running 3.2% which was above the Fed's "comfort zone". Bernanke assured everyone that as the economy slowed, inflation would come down to the Fed's target area. Well, one year later and CPI is 4.1% YoY (PPI is over 7%). Bernanke was just flat out wrong.

Anyone who looks at the 1970s can tell you that inflation can be quite high even in terrible economic conditions. There is absolutely no guarantee that a slowing economy will tame inflation for the Fed.

Not to mention, the Fed is predicting the economy will "recover" in the second half... not sure I buy that, but if you accept the Fed's own prediction, a recovering economy will not slow inflation.

The Fed eased to stupid levels in 2001-2005 in a fruitless effort to reinflate the dot-com bubble. Even at 1%, they could not make un-economic companies somehow be economically viable. Today, even if the Fed lowers rates to 1%, they cannot make un-economic housing be economically viable.

The Fed is just doubling down on a losing trade. See LTCM if you are not sure how this will end.

Or ask Jim Rogers, Warren Buffet, Jullian Robertson, or a host of other successful investors who are all selling dollars. To pay for the Fed's mistake, the U.S. will need to run onerous levels of inflation and/or raise taxes to unsustainable levels (and we won't bring up that entitlements are a $40 trillion unfunded additional liability). The U.S. has lived way beyond its means for a long time -- the bill is now coming due.

Accrued Interest said...

Anon: There are lots of mutual funds which specialize in MBS. Vanguard has a GNMA only fund.

Anon: Yes your cash flow timing is uncertain. But would I rather earn a 5-somthing yield for a 2-5 year time period or a locked in 2.4% for 5 years? Because the later is what you get in Treasuries. Does that make sense?

Gramps: I thought you had left us. I think its clear that Treasury yields cannot remain below the inflation rate. I mean, almost the entire coupon stack is below CPI. I think that's why the 10-year has struggled to get below 3.5%, don't you? I think the Fed is doing the right things (broadly at least) by ensuring liquidity, but its clear their actions today will have inflationary consequences down the line.

gramps said...

This is admittedly very simplistic, but here is how I would divide up the housing market in the U.S.:

40% don't care what interest rates are. The rich have a mtge for tax or estate planning reasons only. The elderly have generally paid off their mtges. The fiscally conservative "millionaire next door" types have very manageable mtges. And you have renters.

45% are going to be hurting, but they will manage through. These people got a little ahead of themselves, but not crazily so. Their discretionary spending will probably be negligable for a few years, but they will muddle through. Better known as the middle class

15% are totally screwed. These people got over-leveraged and/or represent homes that never should have been built. Half of these people made honest mistakes, half are con-artists who realized they would get the upside and the bank would get any downside (a free call). Its very tough to sort out which is which -- but these loans are dead no matter what the Fed does.

In other words, the sky is **NOT** falling for the vast majority of people, but we are in for some lean times. Hardly surprising after we lived above our means for decades, and hardly justification to slash Fed Funds 225bp and counting.

I do agree with you that the long term consequence of easy FF will be inflation. I think inflation will be significant, and that's why I object to the policy.

Inflation falls disproportionately on the elderly and the middle class. Rich people can re-allocate assets to mitigate the effect (abroad if necessary). Middle class people tend to have fewer investment options / less sophistication. The elderly rely on a fixed income to meet their expenses. Higher inflation is going to devastate the middle class and much of the elderly population.

That's a very very big cost -- so I think the Fed needs a very very big benefit to justify running inflation. Just as with Japan, I don't see much benefit at all. Maybe you postpone a little suffering (at a cost) -- but we won't avoid it.

The 15% are still screwed. The middle class is trading a few years of fairly lean times (a recession might be 3-4 years?) in exchange for delaying retirement 15-20 years (because their savings will be decimated and they will need to work longer). That's a bad trade. The elderly (including much of the baby boomers) are earning less on their savings, but their costs are still going up. This means they have to dip into principal-- so next year they are getting a smaller interest rate on a smaller principal amount, requiring a bigger principal withdrawal... it quickly spirals.

The Fed's policy may or may not buy some people a little time -- but I don't see it actually fixing anything. A very low return.

The costs will be very high, and disproportionately born by the middle class and elderly. The government's tax base gets destroyed, and it becomes even more beholden to, and controlled by, the rich.

This is a recipe for creating a banana republic, not a strong democracy.

gramps said...

I should probably have mentioned that a large percent of that 15% screwed category is vacation property and/or flip property. Either way, property that was never going to be occupied.

As for inflation, few people really think the problem through. Plenty of people have long term investment horizons: saving for your child's college, saving for retirement, retirees investing so they don't outlive their assets. A 25 year horizon is a good rough estimate.

Read this link on another blog about inflation over 25 years.

In 25 years, even if inflation were at the top end of the Fed's "comfort zone", one thousand dollars becomes $603. You have to earn $397 return, after taxes, just to stay still. Inflation has been running well above the Fed's supposed target for years -- and there is a good chance it will stay way above that for many years. With oil over $100, there's a decent chance it will go way above the target.

Using last year's CPI inflation (4.1%), your same $1000 in savings becomes about $350... You have to make $650 after taxes just to break even.

There are always some lucky and some skilled people, but history shows most people have a tough time achieving anywhere near those returns. In other words, the Fed's policy is going to hurt most people.

Anonymous said...

FNMA might be the next to go according to Barrons...

Here is the link

Anonymous said...

The fed is trying to rescue the banks. It will continue to use the TAF (T is for temporary, but never mind), increase it, and accepting toxic collateral from the banks. If that starts to work, then why continue lowering the fed funds rate? I agree with those who say inflation is a serious problem. And low yields stimulate speculators in commodities making the inflation problem more difficult and prolonged.

So if you are a speculator, would you rather be in commodities or GNMAs?

Anonymous said...

How can you have debt levels slashed by 50% and deflation?

I don't think that's possible.

What am I missing?

Dr. Dan said...


Good set of commentary. Could you pls let us know what is going to be the prognosis for the future ?

I agree this is a Japan like situation to an extend and now we dont have savings. So are we going to be destitude ?

tell us what you think ...

Robert said...


Thank you for posting. Great insights. I linked your commentary to a few other people today b/c it was so interesting, especially the inflation break-evens.

Darth Toll said...


Agree with 95% of what you said. Let me just add that the real purpose of the Fed slashing rates isn't to necessarily prop up the RE market. That ship has done sailed. Obviously they want to steepen the yield curve to allow banks to make more money on fewer transactions, but this is only part of the story.

The other part of the story is one of obfuscation and zombification. The Fed (via the TAF) wants to allow zombie banks to exist in spite of wrecked balance sheets. This allows the Wall St and banking elite (pigmen) to solidify their personal positions and properly hedge and gives the impression (real or not) that the Fed and/or government will step in and ultimately monetize MBS directly. I refer to recent stories about WAMU golden parachutes in spite of terrible performance and numerous cynical calls by PIMCO to monetize MBS, thereby shifting the monstrous debt burden to taxpayers.

The shroud of the dark side has fallen.

Apolitical said...


I think you're missing something significant with:

"On the positive side, supply technicals are excellent. You may have heard that housing turnover is extremely low, and bank credit is extremely tight. There won't be a flood of mortgage supply for the market to swallow."

I agree that there's no new supply coming down but banks are antsy in the pantsy to get MBS off their balance sheet as soon as the stuff upticks. All the CDOs they're holding, where possible, are going to liquidate (that's where the best value proposition is for the super senior holder), taking resi supply from bank balance sheets into the market

Anonymous said...

Exactly right -

The demand for MBS is sinking faster than the supply. Same is true for the U.S. dollar money supply/demand.


Accrued Interest said...

MBS are going to fly today after this new Fed lending facility.

Turley M Muller said...

I also suspect sometimes when there is a divergence that speculators are shorting TBA forwards which drive rates up, prices down, thus rates on mortgage loans in crease. Higher rates translate into more closed loans and lenders can end up being way long and will need to sell more forward. Which allows shorts to exit their positions.

Wednesday 12th is settlement for 30yr conventional making Monday 48hr notification, thus last week originators would be doing heavy trading to square positions by pairing off or rolling coverage. This can invite all kinds of speculation, especially from hedge funds. Betting that lenders aren't delivering into a specific coupon, which they will buy-back creates the opportunity to buy that TBA and hold it hostage. Crazy stuff can happen. Yet. I don't think the credit aspect is an issue, rather just market noise that gets exacerbated at settlement time.

Josh said...

abx indexes barely budged

mbs did not 'fly' today

Turley M Muller said...

MBS flew pretty good. 10yr note dropped more than a 1pt and conventional MBS picked up a couple ticks. Relative to the market, MBS performed well as spreads tightened. Good call

Accrued Interest said...

Josh... Josh... Josh... I usually don't even reply to posts like yours but I'm making an exception because of how frustrated I am with people completely misunderstanding the term MBS and yet acting like they indeed know everything. Remember that scene from Raiders of the Lost Ark? When Belloq telling Indy to "Sit down before you fall down?" That's what you need to do before posting. Sit down. And read a book about bonds. Then you can post comments again.

First of all... almost the entire ABX stack was higher today. Some by a point or more. That's a pretty good day.

Second, the ABX is made up of home equity loans. It has absolutely nothing to do with the Agency MBS market. Nothing at all. And when traders generically say "MBS" they mean Agency MBS.

Everyone please remember the Accrued Interest comment rule. Contribute meaningfully to the conversation or don't bother.

Roxana said...

It won't truly have success, I believe so.