The availability of residential mortgage bonds in the United States has been shrinking. Private mortgage securitization markets are nonexistent since the financial crisis and the GSEs are not generating enough new supply. The reason of course is the lack of mortgage loan growth in the US. While corporate, consumer, and commercial real estate loan balances are rising, residential loans have stalled.
Source: FRB
On the supply side here are some reasons for the weakness in mortgage loan origination:
Scotiabank: - ... banks have been
more stringent with lending standards since they were forced to buy
back soured mortgages from Fannie Mae and Freddie Mac [putbacks] which
led to significant losses in 2012 and 2013. Then, last year, Fannie Mae Fannie Mae stopped guaranteeing mortgages
with down payments of 3% or less. Plus, in early January, the
Consumer Financial Protection Bureau implemented its
Ability-to-Repay and Qualified Mortgage Standards rules [see post] which tightened
regulation surrounding mortgage securitization. In a special section of
the Federal Reserve’s July Senior Loan Officer Survey, 36% of
respondents said their approval rate was lower than it would be without
the rule for those with lower credit scores
(less than 680), and 31% of
respondents replied that it was lower for those with higher credit
scores (greater than 680).
Add to that the recent increases in FHA mortgage insurance premiums (needed to replenish the FHA reserves - discussed here) for high LTV loans. Many potential first-time buyers with no ability to come up with sufficient down payment are shut out of the market.
At the same time the demand for mortgage loans has weakened, with the recent rate drop only impacting refi activity. Part of the reason is the rise in property prices over the past couple of years which also prices many first-time buyers out of the market.
Source: Source: Scotiabank
As a result of these trends, US mortgage bond market continues to shrink. The amount issued this year is on target to be the lowest since 2000.
2014 figure is based on annualized Q1-Q3 issuance (source: SIFMA)
To exacerbate the situation, over a quarter of outstanding MBS bonds has been permanently locked up on the balance sheet of the Federal Reserve. This takes a significant chunk of an already shrinking market out of private hands. And in spite of the securities purchases ending, the Fed will continue to buy MBS to compensate for prepayment amortization.
Staying with the theme of credit expansion, one area where growth has been more constrained lately is residential mortgage lending. And one potential culprit seems to be the Consumer Financial Protection Bureau's Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (ATR/QM rule). These rules are part of the Dodd-Frank Act. In their zeal to avoid a repeat of the subprime crisis, politicians and regulators have made it tougher (and more complex) for banks to lend to anyone who either doesn't have a stellar credit or seeking a loan that is larger than the "conforming" limit (a "jumbo" loan). Banks, particularly smaller ones, are reporting that the ATR/QM rule has negatively impacted the likelihood of them approving such a loan.
Here are the results of the latest survey from the Fed:
1. For loans that are of "conforming" size but with the borrower's FICO credit score below 680, almost 36% of all banks and half of small banks report "lower" or "somewhat lower" approval rate due to the rule.
2. For jumbo mortgages the number is a whopping 52% (and 57% for smaller banks).
But what's particularly surprising is that even for conforming loans the number is 31% for all banks and 44% for smaller banks. That means that even agency-eligible loans are impacted by ATR/QM.
Source: Board of Governors of the Federal Reserve System
As a result, it seems that at least in part, these new regulations have materially reduced production of new loans, completely halting net new agency (Fannie, Freddie) MBS securities issuance.
Source: Deutsche Bank
Deutsche Bank: - Since January 10, banks have had to originate loans according to the Consumer Financial Protection Bureau’s Ability-to-Repay (ATR) and Qualified Mortgage (QM) rules. The Fed survey suggests that the rules have put a drag on all mortgage production. Since the rules went into effect shortly before net production tumbled, the possibility of a link is worth considering.
The ATR/QM rule has certainly not helped the housing market and may have created a drag on the US GDP. Welcome to the world of unintended consequences.
As discussed earlier, the supply of mortgage-backed securities (MBS) continues to fall behind the potential demand - even with the Fed's taper in place. New issuance has steadily declined over the past year, with the Fed becoming an increasingly larger proportion of that market.
Source: SIFMA
Mortgage Daily: - For nine straight months, agency issuance of mortgage-backed securities has moved lower -- though there was an uptick at the Government National Mortgage Association. Securitization volume now stands at its lowest level in more than two-and-a-half years.
Combined issuance of fixed-rate MBS at the Federal National Mortgage Association, the Federal Home Loan Mortgage Corp. and Ginnie Mae totaled 7 percent less in January than in December.
Securitizations have been lower each month since April 2013, has fallen by more than half compared to January 2013 and was the lowest since July 2011.
At the same time the Fed is to continue purchasing large amounts of this paper through the end of the year, albeit at a slower pace.
Some of the decline in new issuance has been due to the drop in mortgage refinancing - old securities don't amortize as quickly and fewer new securities are issued. But part of the reason for the lower volume of these securities remains the supply of mortgage loans. The total mortgage debt outstanding is barely growing - with a great deal of that growth coming from multi-family residence (apartment) financing.
Source: FRB
As a consequence, MBS paper outstanding has been shrinking since the peak reached back in 2007.
Source: SIFMA (includes CMBS)
Some of that of course is due to the collapse in private label MBS market, including sub-prime. But the amount of agency (government-backed) MBS has not grown much either.
While some would say that $9 trillion of MBS paper should be sufficient, one has to keep in mind that the US and the global economy is now substantially larger than it was in 2007. MBS, particularly agency bonds, are becoming a smaller portion of the overall capital markets, worsening the shortage of higher quality bonds (see story). Liquidity in that market has also suffered, with average daily trading volumes at the lowest levels in nearly a decade (see chart). These are some of the reasons the Fed should return the MBS market to the private sector by exiting its purchases as soon as possible.
While we've discussed some of the economic implications of the Fed's current policy, let's now take a quick look at the impact of QE on the overall mortgage bond market.
Here is a simple fact: the amount of mortgage-related securities in the US has been declining since 2008 - after reaching just over $9 trillion at the peak.
Source: SIFMA
The reason is simple. With a large portion of all mortgages funded via the bond markets, the ongoing decline in total mortgages outstanding results in smaller MBS balances. Of course as the population grows and more homes are built (albeit very slowly) this trend should reverse.
And now with these market dynamics as the backdrop, put the Fed into the mix. At it's current pace the Fed is taking about half a trillion of MBS securities out of the market. In fact the Fed is now removing more than 100% of the paper that is being issued. The supply of agency (Fannie and Freddie) MBS securities in the market is declining sharply as the Fed reduces the total "tradable float". According to Credit Suisse, without the Fed's anticipated taper in Q1, the demand for agency paper could outstrip the supply by $340bn in 2014, creating a liquidity problem.
CS: - Liquidity in the MBS market could come under pressure in the coming months due to
Fed’s settled purchases exceeding 100% of gross issuance of non-specified conventional
30-year pools. Tradable float in conventional 30-year MBS should decline
between 6% and 30% during the year, increasing the risk of a potential liquidity
disruption in the market under longer taper delay scenarios.
As a result some of the private participants, particularly banks, have been reducing their agency MBS holdings. The chart below shows the year-over-year changes in MBS holdings by commercial banks.
Here is what the conventional 30-year agency MBS float will look like under the taper vs. no-taper scenarios (chart below). Without the taper, the float in these bonds will decline by 30% from the October levels. These are dangerously low levels for what used to be one of the largest bond markets in the world.
Source: Credit Suisse
Taper therefore becomes essential in order for liquidity to stabilize and for more private market participants to begin returning to this market.
The recent increase in long-term rates is causing major changes in the mortgage markets. Here are some key trends:
1. Refinancing activity continued to decline through Q3. The proportion of mortgage applications for purchase vs. refi has doubled this year (and that's not because of higher demand for homes).
Source: DB
2. A number of lenders who focused on mortgage refinancing such as US Bank, Provident Funding, and Flagstar are struggling (although the largest banks such as Chase and Wells seem to be less affected). This may result in an increase in the number of riskier mortgages.
DB: - Lenders who specialized in refinancing transactions have experienced dramatic
loss of market share and either will have to become more competitive on rates
in growth sectors such as ARMs to regain market share or loosen credit
standards.
3. While a larger number of buyers now prefer ARMs, the dynamic within the fixed rate universe is a greater demand for 30-year mortgages vs. 20 or 15. That's because the monthly payments on 30-year mortgages are lower (slower principal repayment) and buyers are looking for the cheapest solution.
DB: - As interest rates have risen and volume has dropped, the product mix has
shifted sharply ... 30-year mortgages are much more popular with homebuyers—more than 50%
of 30-year mortgages are used for purchase transactions but less than 20% of
shorter-term mortgages. As a consequence, the share of 15-year mortgages
fell from 20% in September to 17% in October as the share of 30-year lending
rose to 63% from 59%. Meanwhile, the ARM share has doubled to more than
5% since June as HARP’s share of lending has fallen to 3% from a high of 7%
this spring.
4. As a result, MBS bond markets are taking a hit in the form of lower volumes. The sharp decline in refinancing activity has reduced the need to issue new agency mortgage bonds. New issuance is the lowest in years.
Source: SIFMA (note: this includes CMBS but the bulk of the activity is agency MBS)
Similarly, trading volumes in MBS have dropped off to new lows.
Source: SIFMA
Here is a summary on US mortgage markets from Freddie Mac (who, just as Fannie Mae, has been issuing fewer bonds):
Frank Nothaft, Freddie Mac Chief Economist: -
With the close of 2013 will also come a major transition in the housing finance industry. For the first time since 2000, we're going to see the mortgage market dominated by purchase activity as the refinance share drops below 50 percent. And with mortgage rates rising, we're also going to see the home-sales gains as well as the impressive house price growth begin to moderate to more sustainable levels.
While yields on treasury notes and bonds have risen across the board in 2013, the jump in rates has been uneven. The 5-10-year rates - the "belly" of the curve - have increased materially more than other maturities.
Furthermore, the volatility of rates across the different maturities has also been exhibiting a similar pattern, with the yields in the belly of the curve becoming substantially more volatile.
But it hasn't always been this way. The chart below shows how the 7, 10, and 30-year volatility evolved over time.
There was an inflection period early this summer, when the 7-year yield volatility spiked above all the rest. What caused this adjustment? Some of this of course is the selloff related to the Fed's treasury holdings. With fewer purchases of certain bonds, the demand is expected to decline, pushing yields higher.
But there is another explanation. Back in June we discussed the so-called "convexity hedging" (see post). When rates began to rise, MBS durations extended, as mortgage refinancing slowed. And as rates kept increasing, higher coupon MBS became more vulnerable to extension risk. Those with a 4.75% mortgage could still refinance earlier in the summer, but the window on that mortgage closed quickly. MBS holders who saw no need to hedge in the past couple of years had to start shorting treasuries to match their increasing portfolio durations. And intermediate-term treasuries have been the choice hedging instrument. Note that a 30-year treasury is not a good hedge for a 30-year mortgage because the probability of homeowners holding on to their mortgage to maturity is quite low - a shorter instrument is therefore required.
The spike in MBS volatility early in the summer (chart below) increased hedging activity, disproportionately raising the volatility (and yields) of the belly of the treasury curve. This hedging is what created the inflection in the chart above.
NY Times: - In an appearance in Phoenix, Mr. Obama will endorse bipartisan efforts in the Senate to wind down the two companies and end their longtime implicit guarantee of a federal government bailout. That dread prospect, once thought improbable, was realized in the fall 2008 financial crisis; Fannie Mae and Freddie Mac, then bankrupt, were made conservators of the government at great cost to taxpayers, who only now are being repaid.
The president, according to administration officials, will make clear that he will only sign into law a measure that puts private investors primarily at risk for the two companies, which buy and guarantee many mortgages from banks to provide a continuing stream of money for lenders to provide to additional home buyers.
"Wind down the two companies and end their longtime implicit guarantee" is a hell of an undertaking, considering that transferring these agencies into private hands did not work so well the last time. What about simply getting rid of them? The problem is that the US banking system simply can not absorb the mortgage loan volumes currently generated in the US. And with the new Dodd Frank-based capital constraints on banks, adding massive mortgage portfolios to their balance sheets will be even more difficult. The only way to reduce this reliance on the federal government is to involve private investors in underwriting mortgage credit risk.
Currently the bulk of the default risk on mortgages is borne by the taxpayer (via Fannie, Freddie, FHA, etc.). Investors only assume the mortgage prepayment risk - via agency MBS securities (although they effectively pay the government some fees to take mortgage default risk). Do investors have any appetite for a security that allows them to take on credit risk as well - and get compensated in the process?
Prior to the financial crisis, the non-agency MBS business was fairly active. However given the artificially low mortgage rates (driven by implicit government guarantees for the GSEs and their low funding cost), the only way investors could make money was in the non-traditional mortgage space (sub-prime, Alt-A, etc.) - with the help from the rating agencies of course. We all know how that turned out. As a result, the non-agency residential securitization business has virtually disappeared, leaving the responsibility for such securitization (and risk) to the taxpayer - see charts below for illustration.
Source: SIFMA
Now, under pressure from the federal government, Freddie Mac is trying out an experiment. The goal is to create a security to allow investors to take on mortgage default risk. The current structure (shown below) is offering subordinated first-loss tranches with no principal guarantees. Defaults are defined as 180+ day delinquency or short-sale, REO sale, etc. Once there is an event of default, the investor's principal is reduced by a predefined amount (severity). Effectively Freddie is buying a CDS protection on a tranche of its portfolio from investors at a predetermined recovery level. The investor knows exactly what the losses will be based purely on the number of defaults - which is easier to project and model. Only class M1 and M2 (below) will be sold to investors - although Freddie would probably love to sell the B-H (equity tranche) as well.
One of the industry’s largest and most diversified reference pools
Freddie Mac holds the senior risk, which is unfunded and not issued
Senior mezzanine and junior mezzanine notes, which are not guaranteed by Freddie Mac, are sold to investors
Freddie Mac may retain a first-loss piece
STACR notes have a 10-year final maturity
The notes are paid monthly principal similar to a senior/subordinate, private label residential mortgage backed securities structure
Losses based on credit events in the reference pool are allocated to the Notes in reverse order of seniority, and reduce the balance of such Notes
Source: Freddie Mac
If this works, it will be an important first step in getting private investors engaged in this business. It will also create price discovery to allow the agencies to properly price the so-called guarantee fee (the insurance premium the agencies charge for MBS principal protection).
Barclays Research: - This structure protects the taxpayer from certain risks by transferring them to the
private market, provides the GSEs a way to achieve true risk-based pricing of the
guarantee fee and is likely to be appealing to a range of political camps. Importantly, it
achieves all of these without affecting the agency MBS market.
Pulling the government "guarantee" plug from the GSEs at this stage will destroy the housing market - potentially plunging the nation into another recession. To do it gradually however will take years in order for the private sector to absorb some of this risk - potentially using structures like these for a portion of it. It's not clear how large such a program can get, particularly if house price appreciation slows. It is however a certainty that the more of the mortgage risk is shifted to the private sector, the higher the mortgage rates will get.
Furthermore, if the guarantee fees become linked to the pricing of such securities, the mortgage market will become highly procyclical. When investors become concerned about the housing market, mortgage rates will rise (due to higher cost of the guarantee), putting more downward pressure on the housing market and so on, potentially temporarily freezing the mortgage market altogether.
These are enormous challenges to overcome and there is quite a bit of skepticism about the new proposal from the Obama administration. Nevertheless, this Freddie Mac structure (which got very little press so far) is the first step.
This was the gist of Bernanke's statement on July 10th.
MNI: - "We've said that we will not raise interest rates at least until unemployment hits 6.5% as long as inflation is well behaved," but, he stressed, "again, as I've said before, that 6.5% is a threshold not a trigger. There will not be an automatic increase in interest rates when unemployment hits 6.5%."
Rather, Bernanke said, "given weakness in the labor market - the fact the unemployment rate probably understates the weakness of the labor market - and given where inflation is, I would suspect it may be well some time after we hit 6.5% before rates reach any significant level.
"So again, the overall message is accommodation," the central banker said. "There is some prospective, gradual, possible change in the mix of instruments - but that shouldn't be confused with the overall thrush of the policy, which is highly accommodative.
Even though the Q&A came across quite dovish, JPMorgan's analysts are convinced that the Chairman was referring to the Fed Funds target rate only. The Fed is giving itself room to keep the overnight rates low even if the unemployment rate dips below 6.5%. When it comes to securities purchases on the other hand, according to JPMorgan, the pace of purchases will begin declining after the September FOMC meeting.
JPMorgan: - We heard those
comments as dovish on the outlook for interest rates, but doing
little to counter the notion that tapering will occur relatively
soon. In some ways his comments echoed those from
earlier in summer and spring: distinguishing the asset purchase
policy from the zero rate policy, and emphasizing that
decisions on one policy won’t necessarily impact decisions on
the other policy. However, this week his remarks went further,
by laying out the reasons why rates won’t rise dramatically
in the medium term.
...
Nothing in this week’s developments led us to question our
view that tapering in September is coming, conditional on the
data cooperating.
Here is JPMorgan's forecast for the Fed's securities purchases over the next 12 months.
Mortgage backed securities (MBS) have sold off sharply over the past month as fixed income markets face the new reality of rising rates.
Source: Mortgage News Daily
But unlike most other fixed income securities, MBS duration tends to increase with yield. That's because higher MBS yields typically mean higher mortgage rates and lower mortgage refinancing activity (which we have already seen). Pools of mortgages backing many MBS, particularly loans with lower coupon, will experience slower prepayment speeds going forward. Slower refinancing extends the effective duration of the bonds (illustration below).
If you have a 30-year mortgage at 3.6%, it is now less likely you will refinance any time soon (mortgage rates today are above 4%). One reason your mortgage is not treated as a full-term 30-year note is the probability that you will sell your house, thus terminating the note. There is also some probability that in the future, mortgage rates will drop below 3.6% again, providing you another opportunity to refinance. The expected average life of your 3.6% mortgage and the security that is backed by your loan has therefore been extended (from you potentially refinancing in the next six months to you selling your house in say 5 years). And if rates rise further, prepayments will slow even more. The chart below shows the prepayment speed (PSA) for just such a security over the past month as well as the expected prepayment speeds going forward. It also shows what happens if rates increase by another half a percent.
Source: CS
What does this mean for investors who hold MBS or actual pools of mortgage loans? They own securities that become riskier (more sensitive to rates) as rates rise. These portfolios have what's often called a "negative convexity" risk profile. To combat rising durations of their portfolios and therefore higher exposure to rates, many investors now have to short longer dated treasuries or rate swaps. And until recently many MBS investors haven't been doing much hedging because the hedge (the treasury short positions) has consistently lost them money. Now many are jumping in - all at the same time - to put the hedges on. And that hedging is putting downward pressure on treasuries (upward pressure on yields).
Bloomberg: - As rates increase, the expected average lives of mortgage bonds and loan-servicing contracts extend as potential refinancing drops, leaving holders more vulnerable to losses from rising rates. Investors then may seek to pare the duration risk or rebalance existing hedges by selling longer-dated Treasury securities, mortgage bonds or transacting in interest-rate swaps or options on those contracts, sending yields even higher and spreads wider.
...
One measure of duration of agency mortgage bonds rose to 4.8 years last week from 3.7 years in April, Barclays index data show. The duration of Fannie Mae (FNMA)’s 3.5 percent debt, which is now 6.2 years, would rise to 7.8 years if rates rise 1 percentage point, according to Bloomberg’s prepayment model.
This mortgage-driven rise in rates happened before - in 1994 and again in 2003. The act of shorting (selling) treasuries increased yields, raising MBS durations and forcing more treasury selling. Some have referred to this market dynamic as the "convexity hedging spiral".
The 1994 convexity hedging impact
Of course this is not 1994 (or 2003 for that matter.) At least for the time being, the Fed is still buying massive amounts of both MBS and treasuries. And all that buying limits the impact of convexity hedging. Treasury sellers that want to hedge portfolios always find one large buyer with "deep pockets". Furthermore, all the MBS held by the Fed is not getting hedged. With no requirement to mark the book to markets, the Fed is fine taking losses on that portfolio.
Bloomberg: - “The actual convexity hedging flows will be less when rates rise this time than it was in the past,” Dominic Konstam, the global head of interest-rates research at Deutsche Bank ... The hedging “was massive in 2003, and we won’t see a repeat of that. With the Fed holding so much of the mortgage paper, it really knocks down the amount of mortgage hedging needed when yields rise.”
What will happen once the Fed ends its program however is anyone's guess.
We've received a number of emails pointing to what looks like a slowdown in lending by US-chartered banks. The amount of loans and leases on balance sheets of US banks has stopped growing.
As in 2009 and 2011, some people are upset to see record levels of bank excess reserves that are not being turned into loans. These are deposits at the Fed earning 25bp and people are asking why banks are not lending more of this capital out.
But what exactly caused the loan growth on banks' balance sheets to stall? More precisely, what types of loan balances are no longer growing? It turns out that while commercial and industrial loans continue to grow - in fact accelerating - the growth in retail mortgage balances has stalled. And that's the explanation for the flat-lining of the overall loan balances (the first chart above).
Fixed maturity mortgages on US banks' balance sheets (source: FRB; not seasonally adjusted)
"Aha", some economists would say. Banks are not extending as much credit in the mortgage space as they should, which is slowing down the economy. Those evil zombie banks...
The real answer however has to do with the wonderful world of "shadow banking". Why would banks want to keep all these mortgages on their books when they can blow them out to Freddie and Fannie, who in turn sell them to the market in the form of agency MBS (mortgage backed securities). And who are the buyers? The usual suspects of course - insurance firms, mutual funds, etc., as well as the biggest buyer of them all - the Fed. In fact the holdings of MBS on Fed's balance sheet just hit a record. Mortgages are simply making their way from banks' balance sheets onto the Fed's balance sheet in the form of MBS.
Source: FRB (unit = USD million)
The data from Freddie and Fannie confirms this trend, with the first quarter of this year showing the largest net MBS issuance volume in two years. As much as people don't like to think about it this way, Freddie and Fannie are the biggest "shadow banks" around.
The last time we had a spike in MBS issuance in early 2011, mortgage balances at banks declined, as banks did some "spring cleaning".
But what about loans that are not Freddie and Fannie eligible? Those should still be sitting on banks' balance sheets, right? Not exactly. The private side of shadow banking is now kicking into gear, particularly in the so-called jumbo loans (mortgages too large to qualify for the GSEs).
Inside Mortgage Finance: - The private-label market is "showing new signs of life," according to Standard & Poor’s, which predicted that banks are likely to increase their securitization of jumbo mortgages. In a report released late last week, S&P projected $14 billion in non-agency jumbo MBS in 2013. Redwood alone set a goal of issuing $7 billion in non-agency MBS this year and is on pace to exceed that volume, helped by a pending $425 million deal, its sixth of the year. PennyMac Mortgage Investment Trust is also aiming to issue a non-agency jumbo MBS in the Redwood mold in the third quarter of 2013.
The demand for fixed income product has manifested itself in the so-called "private-label" MBS, allowing banks to securitize mortgages that don't qualify for Freddie and Fannie. Once again, it's not about lending less - which is how some economists are reading the first chart above. It's about originating product, collecting fees, and then selling into the hot securitization market - public or private. And taking those loans off the balance sheet creates room to do it all over again (the "recycling" of capital.) As one banker put it, "I want to be in the origination and fee business, not in long-term warehousing ..."
After this post in early December discussing the possibility that MBS, particularly the 30yr FNMA had become a "crowded trade", we've received numerous e-mails arguing that based on the Fed's recent actions, agency MBS has more upside. Furthermore, the US consumer got so used to mortgage rates constantly moving lower, the reversal (discussed in that post) of that trend seemed unfathomable to many. But that is in fact what happened as the 30yr mortgage rate stopped declining.
Source: Bankrate.com
This reversal in mortgage rates was in fact driven by the sell-off in long-dated agency MBS, which many argued would not happen this quickly - yet here we are.
Source: Mortgage News Daily
There are a number of reasons for the sell-off and the recent (mild) rise in the 30y conventional mortgage rates:
1. Dealers have built up a massive inventory of this paper, making it a bit more vulnerable to a correction.
2. Treasuries have sold off materially since early December (about 35bp yield increase on the 10y note), dragging MBS with them.
3. Some institutional investors are preparing for the Fed's eventual exit by unwinding their MBS holdings.
Reuters: - The PIMCO Total Return Fund, the world's largest bond fund run by Bill Gross, decreased its mortgage holdings to its lowest level since mid-2011, ahead of the prospect of higher interest rates and emerging inflationary pressures.
4. The Fed's purchases have been increasingly focused away from the 30yr FNMA, which they probably view as overpriced, and more on the GNMA and the shorter maturity FNMA (such as 15yr) bonds.
Source: JPMorgan ("Conv." stands for "conventional")
That's one of the reasons the 15yr mortgage rate has not moved up as much as the conventional 30yr.
Going forward, the direction of agency MBS paper is less clear, particularly given the tremendous dependence on the Fed who will be growing its balance sheet to unprecedented levels. The upcoming US sequestration cuts could in fact push treasuries higher (by slowing economic growth), with MBS following. On the other hand institutions will certainly become more cautious on their MBS holdings, given the increased rate risk.
Some MBS settlements have now been reflected on the Fed's balance sheet, as agency paper holdings increase.
Bank reserves are also gradually moving up, though still down for the year. So far the growth in reserves has been underwhelming.
Bank reserves (soucre: FRB)
With the newly announced treasury purchases, this should pick up steam. And the settlement schedule will be much less "lumpy" than agency MBS.
One thing worth mentioning here is that the US treasury will be borrowing $45bn a month effectively interest free. That's because the Fed passes interest income back to the Treasury (less its own expenses) via earnings distribution once a year. This certainly helps reduce pressure on Washington to cut spending quickly - the can will be kicked down the road once more.
Per David Schawel's suggestion, here is the the settlement schedule for the Fed's MBS purchases since late August. It's quite "lumpy". Assuming nothing changes, we should see a $129bn increase in the Fed's balance sheet due to these settlements. Of course there will be more purchases as well as paydowns of existing securities this month.
For the first time in since 1991 the amount of corporate paper outstanding in the US is higher than the amount of mortgage related securities.
Source: SIFMA
There are a couple of reasons for this:
1. Net corporate issuance has continued at a good pace even after the financial crisis (see this post for example).
2. Consumer deleveraging after the financial crisis has been reducing the amount of mortgages outstanding. At some point however, these balances will begin to stabilize (driven by demographics) and mortgage related securities outstanding will begin to climb again.
Total mortgages outstanding in the US (source: Mortgage News Daily)
Primary dealers net holdings of MBS (mostly agency) hit another record. Employing a hole in the Volcker Rule (see discussion) dealers are making a bet that the Fed will be there to take this paper off their hands. What if the dealers are wrong?
Source: NY Fed
As discussed earlier (see this post), the Fed isn't jumping to buy large amounts of MBS at this juncture. The central bank's balance sheet as well as bank reserves and the monetary base have plateaued. Furthermore the headline employment numbers (with the exception of construction) show a marked improvement (see this post from Lee Adler). Both surveys of US employment are on the right path.
Source: ISI Group (also the US Labor Department)
What happens if the Fed goes into a long-term holding pattern due to these improved indicators and the dealers start selling their inventory? Mortgage rates would rise and refinancing would slow. Seems that "smart money" is not betting on the big refi wave continuing - which means they are not convinced mortgage rates (and MBS yields) will keep moving lower.
JPMorgan: - Many potential entrants to
the mortgage industry fear that a sell-off would
crush volumes, leaving originators to fight over a
dramatically smaller loan market, and struggling to
deal with large fixed costs. Indeed, roughly 70% of
all originations are refinances currently; a big
enough sell-off would make originators compete for
the remaining 30% of purchase borrowers. Thus,
despite the Fed’s assurances of “low for long”, we
have not seen a large influx of private equity, hedge
fund, or other capital into mortgage origination,
partly out of fear that a sell-off would take volumes
down significantly, saddling them with a large
infrastructure and fixed costs.
Have the long-term mortgage rates (15 and 30y) bottomed here? The on-the-run 30y FNMA bond is off the highs but it has had quite a rally this year (yields have declined). Is it possible that MBS may have become a crowded trade? If the Fed continues to stay put in the next few weeks, we may indeed see a correction. And given high dealer inventories, it could be quite sharp.
Primary dealer holdings of mortgage-backed securities was at an all-time high last week, as the dealers take advantage of the Fed's MBS purchases. Because of the public's push for increased transparency from the Fed, it is generally not that difficult to determine which securities the Fed will be buying (see discussion). And that means easy pickings for the dealers who "front run" the central bank.
USD mil (Source: NY Fed)
So how is it that securities dealers can load up on MBS, given the impending Volcker Rule? The new regulations have a small exception to the prop trading activities restriction. US treasury and agency securities will be permitted. And MBS securities in the chart above are mostly those issued by Fannie and Freddie. It's another example of regulatory driven market distortion. Moreover, these distortions are only going to get worse, as nations such as Canada and Japan are looking to obtain Volcker Rule exceptions for their government debt. A shift into mortgage and sovereign paper and out of corporate paper is the (not so) unintended consequence of the new regulations. This ultimately hurts middle market companies, as US legislators seem to prefer to see the dealers front run the Fed on MBS rather than holding middle market corporate bonds.
Of course private equity firms are loving this. With banks getting out of corporate bonds, private equity firms expand their lending business (a form of "shadow banking"), charging much higher rates. Some readers have questioned that this is actually taking place. Well, here is a direct quote from GSO/Blackstone:
Reuters: - "We really want to thank Mr. Volcker. That rule is a little bit like the Employment Act for GSO and what we do. And it kind of took our competitor prop desk and kind of put them off to the side, so that helps as well," Bennett Goodman, GSO's co-founder and a senior managing director at Blackstone told the Bank of America Merrill Lynch banking and financial services conference on Tuesday.
...
"The business model being adopted by the banks is they want to be syndicators of risk. They don't want to extend their balance sheet to provide capital to a mid-market single-B rated company, which is usually our target audience. And as a consequence, we want to own that risk."
US mortgage prepayment speeds have accelerated to the highest level since 2004 recently.
Source: JPMorgan
Most assume that this is all coming from recent mortgages with low loan-to-value ratios. As rates decline, those who took out a mortgage in 2010 for example are now refinancing it. But there is a bit more to the story. If one looks for example at the 5%, 30-year FNMA pool (these are loans paying roughly 5.5% interest on average), a different picture emerges. The pre-2009 "vintage" mortgage prepayment speed for these high coupon mortgages is higher. The chart below shows CPR (prepayment rate) by mortgage origination year.
The short-term prepayment forecast from Credit Suisse looks similar. It means that the older mortgages with homes that are more likely to be "under water" are actually refinancing faster.
What's going on here? The answer has to do with the Home Affordable Refinance Program (HARP). It's a little venture run by the US Treasury and HUD. HARP was set up to help borrowers who are current on their payments but with homes that have dropped in value so much that it makes them ineligible for bank refinancing. The FHA has its own version of HARP as well. Here are the requirements (from HARP):
The mortgage must be owned or guaranteed by Freddie Mac or Fannie Mae.
The mortgage must have been sold to Fannie Mae or Freddie Mac on or before May 31, 2009.
The mortgage cannot have been refinanced under HARP previously unless it is a Fannie Mae loan that was refinanced under HARP from March-May, 2009.
The current loan-to-value (LTV) ratio must be greater than 80% ["under water" mortgage ineligible for standard financing].
The borrower must be current on the mortgage at the time of the refinance, with a good payment history in the past 12 months.
So that's how 2009 cutoff comes into the picture. Those who took out a 5.5% mortgage (for example) after 2009 and could refinance, already did - that's why the refi speed after 2009 for these mortgages is lower. But those in the pre-2009 bucket are refinancing via HARP. According to JPMorgan, under the second HARP program, a million borrowers have refinanced their mortgages in 10 months. The program is expected to be in place until the end of 2013. The combination of HARP and the Fed's MBS purchases keeping rates low, mortgage refinancing is expected to stay elevated next year. And of course mortgage originators should do quite well in this environment. In fact banks are boosting staff levels to deal with the refi wave.
JPMorgan: - Sustained by HARP 2.0, and then QE3, [the refi wave] is
poised to last well into 2013 and eclipse the
prior record in duration, though not in
magnitude. Extended periods of low rates
compel originators to expand mortgage banking
capacity to take advantage of the business
opportunities.
