Fed’s MBS exit surprises some with muted rates vol

Shrinking of huge portfolio led to predictions of vol jump that – so far – has not appeared

For years, mortgage-backed securities (MBSs) were the tail that wagged the US rates dog. When rates moved up or down, hedging activity from investors noticeably moved the market and raised volatility.

Their impact lessened when the Federal Reserve built up a $1.8 trillion portfolio of agency MBSs after 2008. With that programme winding down and the bonds increasingly returning to the hands of private investors, some market participants believed MBS hedging was set to become a big rates vol motor again. It’s a view that is increasingly being challenged.

“When those bonds go back they’re not going to end up like before – in the hands of active hedgers. They’re going to end up in the hands of passive hedgers, so it won’t necessarily affect the vol market. And even if it does, in terms of where rates are, we’re quite far from levels at which they would need to be buying vol and hedging it,” says a rates volatility trader at one large hedge fund.

According to this view, the MBS market is a different beast than it was 10 years ago, when the Fed became a major player. Fannie Mae and Freddie Mac are no longer the giant buyers and hedgers of the bonds they once were. Commercial banks are now much bigger MBS investors and they benefit in part from different accounting treatment. More broadly, a long period of low and less volatile rates has made MBS hedging less of a necessity. Any hedging also has a smaller impact on an expanded rates market.

At least, that’s how it looks now. “A lot of people thought that when the Fed stepped back, vol was going to move higher,” says an MBS strategist at a large bank. “It didn’t, because the Fed has gotten out of the mortgage market in a very methodical way.”

The argument that the Fed’s exit should increase rates volatility mainly goes as follows. Unlike many other MBS investors, the central bank does not hedge its holdings of the bonds. As it lets increasing amounts of its MBS holdings mature without reinvesting, private investors should pick up the bonds the Fed no longer wants and put on hedges using a variety of rates products, pushing up volatility in the process.

Some still see that potential. Kokou Agbo-Bloua, who leads a global team of financial engineers focused on trade ideas, implementation and hedging at Societe Generale, predicts a rise in hedging compared with “where we come from”. He says: “There was almost no one doing it because the Fed held the bulk of it [MBSs].” The Fed held 28% of the US MBS market before it started unwinding the portfolio (figure 1).

There isn’t that much hedging that goes on in the mortgage market anymore 

Walter Schmidt, FTN Financial

Others are also cautious. “There isn’t that much hedging that goes on in the mortgage market anymore,” says Walter Schmidt, who manages the mortgage strategies group at FTN Financial. “Now the Fed’s out, that might change. That certainly is a possibility.”

The question is: will any rise in hedging be large enough to affect volatility? 

Probably not, the naysayers argue: any growth in MBS hedging will be limited by the post-2008 changes in the market.

Recent data supports this view. For example, in 2018 – with the Fed’s exit well under way – average implied volatility on one-year options for 10-year US dollar interest rate swaps was still below the levels of recent years (figure 2). 

Vicious circle

MBS investors face a number of risks to their holdings. Most importantly, they are negatively convex – roughly speaking, when rates fall, the underlying mortgages will be repaid before maturity, and when rates rise, the duration of the bonds extends. 

Convexity is highest when interest rates fall to levels that incentivise many homeowners to refinance their mortgages. This means investors are holding an instrument with an above-market coupon, but the shorter expected lifespan means they may have to reinvest earlier in a lower-yielding market. If rates rise, the opposite occurs.

Investors can hedge these risks by using interest rate instruments such as Treasuries, futures or swaptions. If large enough, this hedging activity can lead to a feedback loop that increases hedging needs and raises volatility. 

Here is one example of how this works. The risk of early repayments rose sharply in May 2003 as yields on 10-year Treasuries fell. That higher risk affected Fannie’s and Freddie’s MBS portfolios by shortening their duration – a measure of a bond’s sensitivity to moves in interest rates.

To offset the shorter duration, the two organisations loaded up on Treasuries, pushing prices higher, yields lower and encouraging further early repayments by homeowners – creating a vicious circle in which hedging activity exacerbated the movement in rates.

Then, in June 2003, Treasury yields suddenly started going up. MBS investors – Fannie Mae and Freddie Mac chief among them – had to sell these long-dated interest rate hedges, accelerating the rise in yields and pushing up volatility once again.    

“Something you get out of [this hedging] is excess volatility in the fixed income market, essentially,” says Philippe Mueller, finance professor at Warwick Business School.

The products also have volatility risk from the embedded option sold to mortgage holders allowing them to prepay. If volatility rises, there is more uncertainty about the timing of cash flows, so investors demand better coupons and cause the existing MBS to decline in price. 

This risk can be managed by delta hedging with swaps or using swaptions. Many investors, however, hadn’t been hedging this risk due to the persistent low rates vol environment, partially due to the tidal wave of swaptions flow from Taiwan’s Formosa bonds over the years. 

“There’s a huge push in the bank [not to] spend money hedging convexity if you don’t need to. Why buy options and eat the time decay when there’s no movement in vol?” says the large bank’s MBS strategist.

But with an increasing proportion of MBS instruments back in private hands, the theory is that hedging will increase from current levels. The rates environment is also more fluid – the Fed is raising rates, but it’s not necessarily a one-way bet. Societe Generale’s Agbo-Bloua says that with the Democratic party taking control of the US House of Representatives, the US may cease its aggressive fiscal easing through tax reforms. This means rates may be at their peak, opening up the potential for cuts in the coming years and therefore more two-way rates volatility. 

Agbo-Bloua says Societe Generale is starting to see some evidence of MBS hedging in the swaptions market: “There’s demand for swaptions from those who are buying the MBSs, but there’s also a general sense that the realised volatility of rates has picked up. So owning rates vol is something that has become more interesting than the period where rates were fairly subdued and not going anywhere.”

Brakes on hedging

Before the financial crisis, much of the MBS-related volatility in the US interest rate market came from Fannie and Freddie. Together, they held more than 20% of the MBS market in the early 2000s and roughly 17% by the end of 2008. 

During the financial crisis, however, the US government took control of the beleaguered organisations, severely curtailing their activity. The two now hold only roughly 3% of the US MBS market, according to data from the third quarter of 2018.

The former behemoths were replaced by the Fed – which, unlike them, does not hedge its MBS holdings. Like Fannie and Freddie before it, the central bank bought up the most negatively convex MBSs. Its portfolio still makes up 26% of the MBS market, meaning the remaining securities have an overall lower risk of early repayments.

But even as the Fed moves away from the market, many argue MBS hedging is unlikely to return to its pre-crisis levels.

That is partly because commercial banks have picked up a growing share of MBS supply – their holdings rose from 19% of the market in 2008 to 31% in the third quarter of 2018, according to calculations by Morgan Stanley. Commercial banks do not typically hedge their MBS portfolios with other rates products. For example, FTN Financial’s Schmidt says many banks also treat certain MBS holdings as available-for-sale in their accounts. That means the changes in fair value don’t affect the profit-and-loss statement, which reduces the motivation to hedge.

The past decade of consistently low interest rates has also resulted in a market where hedging is less necessary. MBSs composed of low-interest mortgages show less sensitivity to declining rates and the risk of early mortgage prepayments than those sold before the crisis, when mortgage rates were higher. Morgan Stanley reckons only 6% of US mortgage holders have home loans with rates that are at least 25 basis points higher than the prevailing mortgage rate – they are the homeowners most likely to refinance if rates went lower.

According to the bank’s estimates, less than 14% of outstanding US MBSs are being hedged today, compared with nearly 20% in 2008 (figure 3). 

The MBS strategist at the large bank says rising US interest rates will not necessarily lead to more hedging, given the gradual nature of the rises: “If rates move by 100bp in one day, then I need to buy options. If rates move by 100bp over 10 years, then I don’t need to buy options. I just manage my duration on the way up or the way down.”

Even if hedging does increase as MBSs accumulate in private hands, it may have less of an impact on the rates market than in the past, says Jay Bacow, executive director of agency mortgage research at Morgan Stanley. The value of hedged MBSs equalled 8% of the Treasury market in the third quarter of 2018, down from 28% in 2008, according to research by the bank. This reduction is explained less by a decline in hedging than by a tripling in the size of the Treasury market over the period.

Any rise in hedging will probably come from bigger purchases of MBSs. Investors surveyed by Bank of America Merrill Lynch see money managers as “the marginal buyer” of the bonds in 2019. 

“The bulk of the [MBS] supply during the first half of 2019 … will most likely need to be absorbed by money managers,” Satish Mansukhani, an analyst at the bank, wrote in a December research note. One reason, the analyst said, is money managers’ underweight position in MBSs, as they now own a smaller share of the market than before the crisis, despite significant growth in bond fund assets under management since then.

As with banks, though, money managers’ hedging needs are limited by their ability to offset MBSs with positively convex bonds. These are more traditional bonds, whose duration extends when interest rates fall, rather than contracts as is the case with MBSs.

Lastly, the Fed’s departure from the market is unlikely to lead to a spike in rates volatility for the simple reason that the central bank is exiting slowly and signposting its next steps. On September 20, 2017, the Fed announced that the following month it would start letting $4 billion of its MBS holdings run off each month. It then increased that amount every few months, reaching $20 billion in October 2018.

Editing by Olesya Dmitracova and Lukas Becker