Well.... (1) Fed doesn't care about nominal interest rate - they care about real interest rate. If real rate went from 1% to 4% then they may have to face a tough choice of another round - but chances are that increase is driven by better economic data. In which case, mission accomplished. No need for more purchases. (2) They can - but they can also accomplish liquidity withdrawal via reverse repo. They can also raise fed funds rate. (3) They can hold it to maturity and withdraw liquidity via the steps I outlined in (2). There no reason for their to have to face the dilemma of a large balancesheet or taking losses. That said, there will surely be unforeseen side effects to all of this - but I don't think it'll be simple mechanical issues you raised.
Thanks to sjfan and everyone for your replies. sjfan - Re (2) and your proposal they the Fed can accomplish liquidity withdrawal via reverse repo as opposed to selling off their bond holdings. Is this quite feasible in the size required ie. $1-$2 Trillion? Do you know how big their reverse repo operations typically are, say, on a month by month basis? Would they likely conduct reverse repo on an overnight, weekly or monthly rolling basis? My understanding of the repo market is this involves lending our their UST holdings for a few weeks at a time to bond dealers, and in return paying out interest on the funds which they withdraw. Then repeating such reverse repos whenever it matures. Presumably the interest they would pay out to dealers would be 'close' to the Fed Funds rate, and assuming a positive yield curve the Fed would not lose out because the higher coupons received on the bond holdings less the repo rate payments would be greater than zero (i.e. favourable carry). Sounds simple enough, but is there anything that could go wrong here? Due to the huge size of the repo operations could they drive up the repo rate to such a level so that even without selling off their positions it becomes a loss making exercise?
The Fed actually wanted to use reverse repo to mop up QE1 before QE2 was on the table. They went far enough to conduct operations tests to make sure their systems and their counterparties' systems were prepared for it. In general, the repo market is extremely deep. It's an extremely important funding market for all the primary dealers as well as money markets. And, for the reason you specified, the treasury doesn't take a cash flow loss unless the repo rate is > the yield on the bonds (which is of course possible - consider all the short dated tsys they bought at low yields - but this is a problem the Fed would LOVE to have). As far as driving up the repo rate is concerned - this is exactly the point. If the Fed drives up the repo rate high enough to dampen the demand for short term money, then they've accomplished the monetary contraction. They can stop their reverse repo operations. Also, it's worth noting that central banks DO take losses; The usual theory is that the losses tax payers sustain via central bank operations is offset by economic growth that those operations generated. Whether this is true or not is surely difficult to ascertain and not particularly relevant to a practical discussion of the bond market. Anyway, once again - betting on unintended consequences in all of this is not a bad one; But it's not the obvious stuff.
Hell, yeah, CBs do take losses... What was that H1 number from the SNB? Smth like Sfr 14bn, which is pretty chunky for a small country like Switzerland. On the Fed, there's other tools available to them, if need be. They have been tentatively experimenting with reverse repos tgt specifically at the money mkt fund community, where a lot of the cash ends up. And, in summary, I am with sjfan here... Anything can happen, in theory, including runaway inflation etc. However, to assume that this is a foregone conclusion here and now is rather foolhardy.
whole move fibs wow! ES 1586-665: 61.8% fib is 1234.50 ZB 136.31-114.06: 61.8% fib is 122.29 look at the high in ES and low in ZB pro style! ZB should bounce from here. both hit same day rarely happens on daily continuation chart. fibs are usually a shit daily trade and those who use them usuallly fail but on the whole move from 2008-2010 in ES and this whole year in Zb well that's another story...
WTF????!!!!! Another 3 handle day! stopped out again! down $ buy 123.14 take em to 24 then buy em all the way down from 122.29 ! done thank you! this literally is fucking impossible!
it's tough to be long on a day like this. You know mortgage relays convexity flow might come in at 3 to dump the market on the extension.
Well at the risk of blowing my own trumpet I did say the Bernanke QE2 event would be the time to sell (page 368). Unfortunately trading is not just about being right, so those put options I loaded up on...well I was a bit of a pussy there and took the easy profits, way too early. However, it looks ugly I don't think the sell off is over by any stretch. ZN, ZB, GBL all broke recent supports yesterday and overnight JGBs took out their lows in style. Short term at least, I cannot imagine this market rallying much, longs are trapped here. The question is, whether we see 4% yields by the end of this week or by the end of December?
One theory I have that might explain some of the bond weakness is the possibility that China will revalue its currency soon. Inflation is coming out tomorrow, rumoured to be 5% plus, food inflation already in double digits. So if they know their going to take a hit on their UST holdings, say on a 5-10% one-off revaluation, better to offload some in advance. Just a thought.
5%+.... That's really really doubtful. I'll take the other side of that bet any day. If that's the case, 10Y TIPS breakevens should be flying... they are not. Also - there's no China dumping. Overnight asia flow shows the Japs selling on weakness (they've been buying on strength). Looks like a bout of thin volume volatility to me - probably take us higher in yield to 3.3-3.5 in 10y yield before year end.