negative convexity , questa è una parola dei vekki tempi
cmq sul ft fanno un interessante paragone con lo scivolone del 2003
More on maximum negative convexity
Posted by
Tracy Alloway on Jun 08 10:01. Wading into the Treasuries sell-off and negative convexity debate this Monday is Deutsche Bank.
Their central premise: you can’t explain the recent US government bond sell-off by convexity hedging alone.
First a recap — Mortgage-backed securities, or MBS, are said to have
negative convexity since they tend not to rise in price as much as a normal bond as interest rates decrease. When interest rates are very low, as they are now in the US, homeowners often take advantage of the rates by refinancing and paying off their old mortgages. MBS owners therefore find their bonds are repaid faster than expected. To offset that, the MBS investors usually buy longer-dated assets such as US Treasuries or longer-dated US dollar swaps. When interest rates start to rise the converse is true — MBS investors start selling.
As an example, in 2003 there was a wave of mortgage refinancing in the US as 30-year mortgage rates dipped to
5.2 per cent. When rates started rising in June 2003, MBS investors were suddenly forced to sell longer-dated assets.
According to Tom Fant at Minyanville, the rush to hedge meant the benchmark 10-year Treasury rate increased to 4.62 per cent by the end of August, from 3.11 per cent at the start of the summer.
But, per Deutsche Bank’s thesis, the hedging needs of the mortgage universe are now much lower than in the past. That’s down to a series of structural changes including the conservervatorship of the two US GSEs, Fannie and Freddie, and the series of bank mergers which have taken place in the system. Those mergers, according to DB, have effectively resulted in more balanced mortgage exposure for the banks — and therefore less need for convexity hedging.
So what did cause the sell-off then? Two things really — profit-taking by certain money managers and a good ole’ return to risk, according to DB:
Outright selling by index managers and primary dealers is more likely to be the cause of the sell off rather than negative convexity.
-
Money managers, some of whom were overweight mortgages by over 20% versus the index has a reason to sell. They have large gains due to the bond market rally over the last year and the phenomenal tightening of the mortgage basis due to Fed buying. The Figure 3 shows the Libor OAS of current coupon MBS since the beginning of 2007. They have been tighter than ever in history. Also, the portion that they are overweight has also been extending in terms of duration. The temptation to sell and take profits is very strong.
- The primary dealer position in MBS has dropped from $70b to $55b during the last three weeks of May. S
ince Fed has been buying the mortgage rolls and prepayments are relatively slower than most expectations, the dealers had built up a very large MBS position. It is also possible that some of the mortgages sold by the money managers ended up as dealer inventory.
As the bond market started to sell off, dealers cleared their position. . . .
…
At the start of the week, in the absence of Treasury supply and of convexity-hedging events, the level of Treasury yields was mostly driven by the changing perception of the macro outlook and changing levels of risk aversion. . . . The recent sell-off in the market bears out the risk of overshooting in the future. Starting around April 21, the normalization trade began with substantial declines in LIBOR rates, the start of the accelerated sell-off in Treasuries, the massive tightening of investment-grade corporate, CMBS, and agency spreads, and the reverse of the flight-to-quality resulting in the decline in the dollar. While the move in swap rates was nearly as large as the 2003 episode, it happened with only light mortgage convexity hedging flows.
But importantly, according to DB, that’s not to say there won’t be a convexity hedging event
eventually.
In fact, according to some commentators we are already at the point of ‘
maximum negative convexity‘, meaning if rates rise much from here, the duration of MBS portfolios will extend by a massive amount — investors will have to sell longer-dated things like Treasuries or swaps.
Here’s DB on what could spark such an event:
At some point the stock market may react negatively to further selloff in yields given the negative impact of higher financing costs on the economy but recent market action suggests that we are not there yet. The UST sell-off was mainly driven by higher mortgage rates and wider CMBS spreads following a series of downgrades by S&P.
Should the mortgage current coupon sell-off towards 4.90%, UST could well suffer from another convexity hedging event.
Related links:
The Fed’s asset-liability mismatch - FT Alphaville
Convexed - FT Alphaville
Is the Fed losing control? - FT Alphaville