Reflections of a slow day

No economic data was released today. Treasuries rallied quite hard ending the day with the 10 year up just over a point, pushing it's yield back down into the 3.16% area. Forgive my imprecision, it's late and I'm not near my Bloomberg...I'm working off memory here.

There has been some decent volatility in the government bond market recently. Part of the market knows the Fed wants to keep rates artificially low to help consumers and businesses have access to cheap credit, thereby stimulating the economy. Another part of the market knows however that the Fed doesn't have pockets deep enough to pull this off by themselves. Adding to the confusion is that nobody knows what the Fed will do next...will they boost the $300 billion Treasury purchase program? Will they do nothing and watch the 10-year Treasury slowly creep up toward 4.00%...killing their hopes of a refi-wave induced by artificially low mortgage rates?

Nobody has a clear idea at this point and the various market forces are moving against each other like techtonic plates...creating friction.

The Fed doesn't seem to be very aggressive in their buying lately. Their open market operations have been tepid. Going on memory I recall that many times the Primaries will offer $36 billion or so in product and the Fed might take $6 or $7 Billion of it. The Fed gave some lip service to buying less if the "green shoots" of economic growth progressed into a recovery. I think they'll be forced back into the market before long.

What to buy?

The question we keep getting is "what bonds look good?" The answer we keep giving is "none look good...but some look OK."

Most banks are expecting inflation to come back with a vengeance in response to the current level of government spending. They've got good reason to be worried.

Because the primary fear is inflation, the primary response is to stay short. Inlation kills the value of a stream of fixed payments....bonds. The guy that goes long right now to get yield is the same guy that is going to get crushed when rates rise on inflation concerns.

Most everything that banks like to buy looks terrible...especially from the viewpoint of a buyer that is paying north of 2.00% to bring deposits in the front door. Where do you go with 2.00% money? 1 year bullets at 0.67%? That's a nice way to lose 133 basis points.

3 year non-call 1 year paper at a 2.16% and pick up 16 bps? Hardly worth the effort.

MBS?

I began writing to you quite some time ago about the widening of spreads on mortgage backed product. This began waaaaayyyy back when the hedge funds started to get in trouble. The bad assets they held generated losses, which led to writedowns, which led to margin calls. They couldn't sell the junky stuff at that big a loss so they sell the good stuff that is highly liquid...Agency MBS were among the highest quality bonds they owned.

The worse the market got, the more margin calls came in and the more MBS they had to sell. Fear began to drive the markets and pretty soon there was so much selling of MBS to meet margin calls that we saw spreads that historically run at 125 bps over Treasuries start to blow out to 250 bps, then 275, then 300, at the highs there were in the 315 bps range.

Everything I wrote at the time preached mean reversion...when the turmoil subsides spreads will go back to a more normal range and those that bought at +300 will enjoy massive spread tightening. NOTE: Spread tightening is caused by prices rising, thereby pushing yields (spreads) lower. So the guy that bought a bond at $100-4 back in October is now sitting on a price in the 104 range. 4 points of profit largely due to spread tightening.

What about MBS now?

It's difficult to find and MBS under $103. With premiums this high I can't overemphasize the importance of structure. A 30 yr 5.00% MBS and a 10 yr 5.00% MBS are two completely different creatures when it comes to prepay risk.

The 30 year borrower has tons of incentive to refinance his mortgage as a large percentage of his monthly payment is comprised of interest. Those 30 year pools can generate some terrifying prepay speeds...which will kill your yield if you bought it at a premium.

The 10 year borrower by contrast has a much smaller percentage of his monthly payment that is interest...so he is less senstive to movements in mortgage rates. Now consider that he is already in the shortest mortgage vehicle available right now and you can see that this pool doesn't have near the incentive to prepay as does the 30 year pool.

10 year MBS have prepay histories that are slow and steady. This is "structure". If you have to buy at big premiums, at least buy a structure that will provide as much protection as possible. The trick is that 10-year MBS paper is difficult to find currently. I saw at most...maybe $4 million of new production 10 year paper for May. I'm sure there is more out there but it's getting put away quickly.

In summary, 10 yr MBS are very attractive compared to most other product we're seeing right now (keep in mind this is written from the perspective of a bank investment portfolio manager...other investors have other needs and constraints that change the view of what's attractive...or even permissible).

That ends a rambling look back at today.