Friday, December 5, 2008

Ben's still dancing...


At first I thought I misheard… or that some charlatan journalist was trying to make headlines.

But no. Ben did say he's thinking of buying long-term Treasuries. His speech last Monday is crystal clear: “[T]he Fed could purchase longer-term Treasury or agency securities on the open market in substantial quantities […]”

Treasuries! Of all things! It’s like going to your favorite pastry shop, everything is at a discount, and you get a diet coke!

What’s wrong?

What is wrong is that this is the wrong solution to our problem, which is three-fold:

First, there is a huge hole in the balance sheets of banks, because a large amount of the assets in their books are either illiquid or junk or illiquid because they are junk.

Second, as a result of the first problem, there has been a brutal disruption in the availability of credit, as financial institutions hoard cash out of fear of each other, of nonbank firms and of their own ability to meet future obligations.

Third, the economy is on freefall, with the latest payroll numbers attesting to this (I’m looking for the superlative of “abysmal”).

So given these three aspects, what are the Fed’s priorities and objectives and what instruments will it use to achieve them?

Let’s start with the second, which is, ironically, the easiest part. By now, the Fed has established an entire infrastructure of facilities to address shortages of liquidity—the TAFs, CPFFs and ABCPMMMFs of this world. These are designed to be temporary, since the underlying assumption (and prayer) is that financial conditions will normalize.

That’s all good, only that along with praying, Ben’d better do something too. Like, deal with the first part of the problem, which is to plug the hole of the banks. What is he doing on that front?

It’s probably the Nth time I will argue “very little.” Yes, we had the use of part of the TARP (by the Treasury) to recapitalize some banks. Yes, we had the rescues of the selected few (AIG, Citi etc), partly with Fed money, to avoid a repeat of the Lehmans aftershocks. Yes, we had the announcement of the TALF, which was certainly a step in the right direction. But it has been very (very) patchy.

Take the rescue of Citi. It was last-minute. The modalities (fate of creditors, shareholders, etc) were a work in progress. No rationale was offered behind the amount of assets carved out of the bank (with the Fed's and the Treasury's help), including whether it would be sufficient.

Don’t get me wrong, it was the right “idea,” i.e. carving out assets on one hand, and plugging a potential hole with new capital on the other. But why not apply this approach uniformly and across the board of (systemically important) institutions, and clean up the system once and for all? And at the same time, prepare for the failure of institutions that (given our limited resources) we simply cannot rescue?

So let’s go back now to Ben’s idea of potentially buying long-term Treasuries. Clearly Treasuries are not really “toxic” (at least not yet!) and, as such, they’re not the kind of asset banks would like to see carved out of their books. So the purchases would do nothing to address problem #1 and, by extension, #2.

But then again, Ben really meant it as a potential means to address #3. Indeed, in his own words “This approach might influence the yields on these securities, thus helping to spur aggregate demand” (my emphasis).

Even here, this is not the most efficient use of Fed cash: First, long-term Treasury yields are super low already, so they don’t exactly need support.

Secondly, if there is any fear among investors that yields will go up in the future, that’s because of uncertainty over how inflationary of the Fed's easing is going to be—an uncertainty that is building fast, I should add, and which is not really dealt with by promising to, urh, print money to buy Treasuries!

Third, yields of, say, mortgages or loans to companies are high largely because banks currently demand a sizeable premium for liquidity (i.e. they’d rather hold cash or liquid Treasuries than lock their money on a long-term loan). This is not going to go away no matter how many Treasuries the Fed buys--unless the problems at banks get fixed.

So I’ll go back to my perennial call for the Fed to step in and buy a large chunk of the "troubled" assets in banks’ books. Like in the case of Citi, this should come in conjunction with a potential “refill” of bank capital, if the carve-out results in a new hole.

Buying toxic assets is better than buying Treasuries for another reason: IF the Fed is successful in revitalizing the economy, long-term Treasury yields are bound to go up (as growth expectations rise) and their prices would go down. This will translate into a large capital loss for the Fed.

On the other hand, if it buys the troubled stuff, success in rejuvenating financial markets will likely bring the yields of these securities down—at the very least because the "liquidity premium" will go down. Translation: The Fed could actually make a gain on at least some of these “toxic” assets.

Finally, let me throw another bombshell here--food for thought, you might say. This may be the time to revisit the desirability of an inflation-targeting regime for the Fed, of which Ben is a famous fan. Why?

Because uncertainty about future inflation is rising, as people get iffy about the expansion of the Fed’s balance sheet (which, by the way, is (still) non-inflationary for the reason I mentioned here). So this would be precisely the point where an explicit commitment by the Fed to a medium-term inflation target would help reign in inflation expectations.

Ultimately, what has been lacking all along is communication. Communication of the Fed's (and the Treasury's) accurate understanding of the size of the problem, their priorities and end-objectives, and the tools that will be employed to achieve them. We got one step forward last week, but Ben has now switched to foxtrot.

No offence, Ben, but it doesn’t look good! So why don’t you stick with the forward steps and let us do the dancing!


Glossary: troubled assets, cash hoarding, liquidity premium, communication, foxtrot.


3 comments:

Anonymous said...

Have you recreated Bernanke / Gertler / Gilcrest's models? If so I'ld love to see a post. Especially if it includes CDSs / CDOs / SIVs. I believe the later explains where thier analysis falls apart - which explains their policy moves.

The models would have to include a class of investment instruments that require capital infusions proportional to deviations from historical growth.

Additionallly, they would have to include the amount these instruments were leveraged PLUS the expected principal / interest payments required by the asset holders.

Matt Chanoff said...

Since you sound like an intelligent proponent of the "buy toxic assets" crowd, let me ask you an honest question, or actually two: how many are there and what should we pay for them? I'm concerned about the first question because it seems to me we've got a moving target - changing conditions in the housing market equal changing toxicity in the derivative assets. This is probably also becoming true of credit card, auto loan, college loan, and other debt, and the derivatives derived from it. So how much do we buy, and how do we know when we've bought enough? The second issue, price, seems to me a problem from a bunch of different standpoints. First, efficiency - if Treasury is the only buyer in the market, how do you find a market price? If you reverse auction, you'll end up fielding offers at prices at which the bank making the offer won't be insolvent - why would they go below that? That means healthier banks will offer the lowest prices, and sicker banks will look for some other kind of bail-out or go under. The resulting shake-out may have nothing to do with which banks are more or less important to the economy, so it seems to me an extremely inefficient way to get credit flowing. However, if you DON'T reverse auction, you'll have to set arbitrary prices (no market plus no auction means no intrinsically meaningful prices). How then, do you avoid the price setting process becoming politicized? One price will wipe out business as usual, a lower price will also wipe out equity holders, a lower one uncollateralized debt holders, and so on down the line. Who does the government close the door on? Isn't the answer, whoever has the pull? Also, assuming that the government isn't in a position to buy ALL toxic debt, how do you create a price signal for the rest of the market? Private investors will be aware of the process - what incentive will they have for accepting the government price as a given? I'd appreciate your comments on these issues.

Thanks,

Matt

Chevelle said...

Matt, I’m glad to see you’ve been reading M&A!… Or maybe this is more “great minds think alike”, in which case you can find a post with similar concerns here (http://modelsagents.blogspot.com/2008/09/ask-not-what-you-can-do-for-your.html).

These are not easy questions. But here are a few thoughts.

On the numbers, true, this is going to be a moving target so long as the “real” economy is still looking for a bottom. If I had to guess, a good starting point would be the banks’ so-called “level 3” assets—assets which are valued in banks’ balance sheets not based on the price in an active market, but based on the banks’ own valuations. The FT reported recently that, for the big financial firms, these stood at $610bn in Q3 (http://www.ft.com/cms/s/0/0627def0-c6f8-11dd-97a5-000077b07658.html?dbk). The numbers I’ve seen elsewhere are larger (by about a factor of two), but they include smaller banks, many of which are already in negative equity territory.

On the price to pay: Two broad options, as you suggest:
High price (in which case lower needs for capital refill); or low price (in which case a big hole in the banks’ books and a bigger need for a capital refill). From a fiscal/taxpayer perspective, once “you” have decided that the government will run the show, ie buy the assets AND recapitalize, it doesn’t matter which one you pick: The government bears the total cost either way. Personally, I would prefer the latter: It does less to distort the price discovery process, and it gives a better upside to the government, both when it comes to the assets purchased and the equity invested in the banks.

Finally: The government will have to draw a line on which banks to save and which ones will have to go. Resources are limited. Only that I hope that next time they let a bank fail, they are much better prepared than they were for Lehmans!