Wednesday, July 22, 2009

Bernanke's exit strategy

As he laid it out in his op-ed in yesterday's WSJ, the Fed's exit strategy looks sensible. I've argued that the exit strategy shouldn't be all that difficult. It should start when it is clear that the economy has come out of the woods (which appears to be the case already), and when risky asset prices start moving higher (of which there is evidence in abundance: equity prices, commodity prices, energy prices, junk bond prices, even commercial mortgage-backed security prices). Stronger markets for risky assets are not only the signal to begin the exit strategy, they also provide the perfect environment, since the Fed needs to sell many hundreds of billions of risky assets to unwind their quantitative easing of last year.

Already the Fed has unwound some of the liquidity additions of last year, as the op-ed explained, but meanwhile they are still buying more Treasuries and MBS. The Fed's balance sheet hasn't shrunk by much so far.

Although the exit strategy is sensible and feasible, it has not yet been executed. As I argued a month or so ago, the Fed should have already started to withdraw liquidity, and every day that goes by that they don't, the inflationary pressures (stemming from an excess of dollars) accumulate. We can already see this happening: the dollar is down over 10% from its highs earlier this year; gold is up over 10%; commodities continue to rise; and the yield curve is very steep. And Bernanke takes pains to say the Fed is not likely to tighten anytime soon.

We are going to need to watch all of these indicators, as well as the changes in the Fed's balance sheet and the money supply numbers, very carefully. There's no reason yet to sound the alarm about hyperinflation, but inflation complacency is not where you want to be. The Treasury bond market cheered the exit strategy earlier this week, but I think that is a naive reaction. T-bond yields are likely to continue to drift higher until the Fed takes some positive action to shrink their balance sheet.

A tighter Fed does not imply higher bond yields, nor do higher bond yields pose a threat to the economy. The bond market actually loves tight money, since that keeps inflation low. In any event, it would be the most natural thing in the world for interest rates to rise as the economy improves, mainly because interest rates are still very low, in a way that makes sense only if the economy remains mired in a funk and deflationary pressures lurk under every rock. The worst thing would be for the Fed to delay tightening for too long, since that would set up an inflationary problem that would push bond yields sharply higher and require a huge Fed tightening and an eventual recession in a few years' time.

Full disclosure: I am long TBT and long TIP and TIPS as of the time of this writing.


Public Library said...

I am long Real Return, Commodities, EM Local/External, and Non$.

Bernanke is committed to low rates for a loooong time.

Eccono-monkey said...

Scott, Great post. Do you offer any personal consulting services or do you know anyone who shares your insights who offers such consulting?

Scott Grannis said...

Thanks. What sort of consulting?

__ said...

"the Fed needs to sell many hundreds of billions of risky assets to unwind their quantitative easing of last year"

I suspect this will be difficult and so won't be the primary means of tightening. Looking at (say) TICs data, there is little appetite for Agency debt, and the Fed effectively is the MBS market, now, given the percentage of primary issuance they're buying.

As ever, thank you for posting.

Scott Grannis said...

I'm not sure it would be that difficult. Bear in mind that Treasury is now routinely selling about $100 billion every month to finance the deficit, and despite this being public knowledge, bond yields are still quite low.

Eccono-monkey said...

I want to understand a variety of credit spread indicators and their leading qualities. I am looking for someone to dialog with to provide background context on parts of the WLI and the LEI. Also to bounce ideas around regarding difficult to research topics in macro econ. I see some interesting charts every week would like to get a second set of eyes to share their thoughts as well...

__ said...

There is a big difference in the appetite for MBS and Agencies and the demand for Treasuries. The official sector never really bought MBS, and largely stopped buying Agencies last year (indeed, the Agency/Treasury switch was the trade last year for reserve managers). It seems reasonable to assume that source of demand will not return in the near-term.

And I would be surprised to see the private sector suddenly eager to buy [Agency] MBS, but I could be wrong. I suspect the Fed will end up being the buy and hold investor they originally claimed to be.

Scott Grannis said...

Again I think you are being too pessimistic. The MBS sector is still larger than the Treasury sector. MBS spreads are still lower today than their long-term average.

I don't see any evidence here that the appetite for MBS relative to Treasuries has weakened. Both markets are supporting high levels of issuance and their relative spread is virtually unaffected. They're joined at the hip.

Agency spreads have come back to normal, so there is no sign there either of any shortage of buyers.

There are roughly $4.5 trillion of MBS out there, and $3.7 trillion of marketable Treasury bonds. Treasury is increasing the supply of Treasuries by 2.7% per month without having a big impact on prices. If the Fed sold $100 billion of MBS a month (2% of aggregate MBS supply) I don't see why prices should collapse.

Believe it or not, $100 billion is not a very big number as far as the bond market is concerned.

__ said...

I do have an idea of what notionals are outstanding. My point is that the Fed has bought a very high percentage of both [Agency] MBS and Agency debt issuance this year (and outside of Agency MBS, there has been very little private-label MBS issuance). I would also remind (though you probably need no reminder) that the Fed is trying to hold mortgage rates low (as close to 4% as possible, I would guess) because at those levels, the % of outstanding Agency mortgages that can be refinanced is very high (ignoring the 20% or so that are underwater, though the 125% mortgages takes care of that!). More to follow.

__ said...

I did write "More to follow", and here it is. According to Credit Suisse, Fed purchases of Agencies and MBS are about 250% of net issuance in those markets. That strikes me as a sufficiently high percentage* that, as I wrote previously, the Fed will not use sales of MBS and Agencies as a primary means of tightening policy (I'm certainly not suggesting prices will "collapse", as you wrote).

*It's even a meaningful percentage of the notional outstanding.

Regards, MW

Scott Grannis said...

MW: I still think that the impact of Fed purchases is determined by the overall size of the market, not net issuance. And I suppose I'm trying to give the Fed the benefit of the doubt. But in the end I agree that rates are going to have to rise, especially if the Fed keeps buying bonds and postpones the execution of its exit strategy. Bond yields are ultimately determined by inflation, period.

Scott Grannis said...

Eccono-monkey: I don't think I can be of any help with this, nor do I have any recommendations.