What are Zhou and al-Sayyari doing? Reserve management and the Fed’s balance sheet operations
Paul Krugman noted that his discussion of the Fed’s recent policy steps was rather wonkish. If that is true, my comments will be off-the-wonkish charts.
I want to try to bring the actions of foreign central banks into Krugman’s discussion of recent Fed efforts to use the “asset” side of the Fed’s balance sheet to stabilize financial market conditions.
Krugman argues that the literature on sterilized central bank intervention in the foreign exchange market can be used to understand the Fed’s recent policy steps.
When the central bank sterilizes, the “liability” side of the central banks’ balance sheet doesn’t change. It doesn’t print any more money, or issue any more bills. That is what “sterilization” means.
What changes is the asset side of the central banks balance sheet. It might for example sell US Treasury bonds and buy German government bonds. That increases the supply of US Treasury bonds in the private market, and reduces the supply of German government bonds in the private market.
Krugman - building off the thought-provoking work of Mr. Waldman - argues that the Fed is effectively reducing the supply of “Agency MBS” in the market while increasing the supply of Treasuries in the market. It is trying to keep a surge in private demand for Treasuries (and a surge in private sales of Agency MBS) from pushing up the price of Treasuries and pushing down the price of Agencies (increasing the “spread” between the two bonds well beyond historical norms). He writes in today’s column:
Banks that might have raised cash by selling assets will be encouraged, instead, to borrow money from the Fed, using the assets as collateral. In a worst-case scenario, the Federal Reserve would find itself owning around $200 billion worth of mortgage-backed securities.
Krugman (with some support from Delong) then applies the standard critique of sterilized intervention to the Fed’s actions - namely, that just as central banks fx purchases and sales are too small to have much of an impact on the a multi-trillion dollar fx and bond market, the fed’s purchases of Agencies will be too small to have much of an impact. Krugman, from his blog:
… this is just like the way you analyze sterilized intervention in currencies. And the usual problem with such intervention applies: the financial markets are so huge that even big interventions tend to look like a drop in the bucket. If foreign exchange intervention works, it’s usually because of the “slap in the face” effect: the markets are getting hysterical, and intervention gives them a chance to come to their senses.
That piqued my interest, because I have long wondered if central bank actions - counting the actions of emerging market central banks — are still really “a drop in the bucket” in the foreign exchange and government bond markets.
The Fed and the ECB typically do intervene in small quantities - a few billion here, a few billion there - and for a short period of time. But the Fed and the ECB aren’t the only central banks active in the US and European government bond market. China has something like $60b to place in these markets every month right now.
That doesn’t strike me as a drop in the bucket.
If it put all $60b in Europe that might make difference on the euro/ dollar, and on bund yields.
Conversely, if it put all $60b in Agency MBS market and stopped its existing purchases of euros, it could provide a fair amount of support to that market - and perhaps for the dollar.
And it is a sustained flow. A change in China’s pattern of purchases and sales in the fx market today could signal a change in how it manages its $60b plus flow next month.
The fact that this is a sustained flow — and that there is no hard upper limit on how much foreign exchange China can accumulate — potentially differentiates China’s $60b a month from the $40b the Bank of England spent defending the pound back in the early 1990s.
I should note though the there is an enormous debate on how central bank actions impact the fx market. My sense is that most traders think central bank flows matter, and have an impact on price. And my sense is that most central bankers think that central bank flows don’t matter all that much, because private investors offset their impact.
This view, for example, informs Alan Greenspan’s view that a complete shift in China’s purchases from dollars to euros would not increase US bond yields by more than 50 bp (See the Age of Turbulence). That was back when China was adding a bit less than $660b to its portfolio a year. But it is typical of the dominant view in the Fed and perhaps the ECB as well. China’s sales don’t changing any of the “fundamentals” so private investors should be quite willing to buy what China wanted to sell without much of a change in price.
Nonetheless, if my characterization of the views of traders and central bankers is right, the gap between the view of the trading floor and the view in the central bank’s board room is potentially quite large.
And on this question, I tend to be sympathetic to the view of the trading floor -
That has one sort of surprising conclusion.
Central banks - globally — have not been helping during the recent “liquidity” crisis.
A few quick points.
First, emerging market central banks likely added over $150b to their reserves (a $1.8 Trillion annual pace) in January alone. China added $55b alone, adjusted for valuation. The state banks may have chipped in a few billion, or tens of billion more. China doesn’t export oil. The Saudis, who do export oil, added $18b. There are other oil exporters as well. Brazil added $5b. India added well over $5b. My methodology here isn’t complicated. I just add up the total that various big central banks disclose. If central banks are using state banks to hide the scale of the intervention, the real total could be higher.
That $150b doesn’t count sovereign funds. It is safe to assume that the big Gulf sovereign funds received another $10b. $100 a barrel oil and all.
That $160b in foreign asset growth is well in excess of the emerging world’s current account surplus. The oil exporters current account surplus is maybe $60b a month. China’s is around $30b right now (it might rise later in the year). The rest of the emerging world is likely in rough balance …
It consequently reflects ongoing capital inflows into the emerging world. Michael Pettis is right to be obsessed with the scale of these flows.
Private investors fleeing Agency MBS and US ABS have been fleeing into Treasuries. But they also have fled the dollar. Those investors with some remaining appetite for risk are betting on the emerging world.
Call it a flight from risk assets in the US to “risk” assets in the emerging world.
Dr. Hamilton argues that there also has been a flight from nominal assets to real assets - like commodities.
Now, the private funds going into the emerging world are in effect bought up by the emerging world’s central banks. Emerging market central banks don’t want a surge in private demand for emerging market financial assets to lead to a surge in the market price of their currency, so they intervene. That means they buy dollars. And those dollars in turn are recycled back into the US market.
And where are they going right now?
Well, by all appearances, into Treasuries, short-term debt issued directly by the federal Agencies and bank deposits. Not into Agency MBS.
In that way, a shift away from “risk” assets in the US toward “risk” assets in the emerging world leads to additional official demand for the safest of US assets precisely at a time when US private investors are also piling into safe US assets. Right now, it seems like Treasuries are the only asset that both private investors and central banks think is really safe.
In that way, the activities of foreign central banks are adding to the “liquidity” crunch in the US credit market.
And really hurting any investor who had bet on the persistence of historical spreads between Treasuries and Agencies …
To go back to my post on Friday, foreign central banks - who also have to manage the asset side of their rapidly growing balance sheets - could take actions that reinforce the Fed’s actions. The Fed is now effectively lending cash against Agency ABS collateral (through the TAF). Foreign central banks could join in, by buying Agencies. The asset side of their balance sheet is now quite large, even relative to the asset side of the Fed’s balance sheet (and it is growing faster). Felix notes:
…some of the trillions of dollars currently being held by foreign central banks could definitely come in handy right now. They’re sitting on nice capital gains on their Treasury holdings, thanks to the current flight to liquidity. It would be great if they started dumping those Treasuries and buying the bonds of Fannie and Freddie instead, at least for the time being.
Conversely, the world’s central banks could take actions that would tend to offset the Fed’s actions - basically buying all the Treasuries that the Fed is releasing into the market as it buys Agencies. As Rebel Economist noted in the comments here, reserve managers who bet on Agencies in the past are staring at mark to market losses, and could well be told to be a bit more conservative with their money.
That brings us back to Dr. Krugman’s theoretical discussion.
If you believe that large-scale sterilized intervention doesn’t matter — and the allocation of the emerging world’s massive reserves across different assets doesn’t matter — then none of this matters.
Market action simply reflects a reappraisal of the Agencies “fundamentals” in a deteriorating housing market. Central bank actions cannot change those fundamentals.
The US may need to bite the bullet and use public funds to recapitalize the GSEs to shore up global confidence in their ability to honor their obligations as it turns to the GSEs to support the mortgage market.
On the other hand, if central banks actions can impact the market, the kind of coordination that now exists between the Fed and the ECB may soon need to draw in the asset managers of the big emerging market central banks.
They are the new 800 pound gorillas of the fx market.
And the Treasury market.
And, I suspect, at least parts of the agency market.
Excellent comments Brad. I appreciate you turning this into a more game-theoretic analysis by considering the incentives of the various players involved . I believe that the analysis of Krugman et al. is oversimplified. Many players here are involved with differing interests and many variables changing such that a simple static analysis of sterilization is not in order. The global central banks are not just price takers.
Economists generally argue that transparency and credibility make for efficient outcomes. If they are right, perhaps the present state of “constructive ambiguity” about the status of the GSEs is the worst of all. Given that a US government commitment to let the GSEs go bust would probably not be credible, it might actually be best to give them an explicit guarantee. If so, I am sure that central banks would be very happy to stabilise the agency market, and spreads would narrow dramatically.
China’s Trade Surplus plunges 63 percent, imports from US surge 33 percent
http://ap.google.com/article/ALeqM5go_fvEZzwRs0sWLADbucTzdCsS8QD8VAIP8G1
February’s trade surplus was $8.6 billion, down from $23.7 billion in the year-earlier period, according to China’s customs bureau.
The data reflected slowing growth in exports to the United States and Europe while China’s still-robust economy is driving demand for imported energy, consumer goods and industrial equipment.
China’s imports in February surged 35 percent to $78.8 billion from the year-earlier period, according to the customs agency. The rate of export growth, meanwhile, plunged to 6.5 percent from January’s 26 percent.
Exports to the United States fell 5 percent in February to $16.4 billion, while imports of American goods jumped 33 percent to $6.1 billion.
The Fed isn’t targeting treasury/agency spreads directly. The objective of the TAF operation is to ameliorate the deterioration in LIBOR/fed funds spreads. It is targeting interbank liquidity directly - not the solvency or liquidity of the agencies. To do this, it’s getting money into the hands of those banks that are having difficulty in the unsecured interbank market, but have collateral that is acceptable at the Fed under TAF requirements (essentially the same as discount window requirements). It turns out that much of this acceptable collateral is agency related.
The TAF operation is obviously very different from FX intervention. It is not the role of foreign central banks to provide US dollar interbank liquidity to the market; nor is it their role to ‘help’ treasury/agency spreads via their FX operations. The macroeconomic implications of a central bank that uses its monetary base to fund FX reserves (e.g. China) is very different than that of a CB that uses it to fund domestic intermediation (e.g. the Fed). Accordingly the analysis of materiality may be quite different. Given the short term interbank liquidity conditions that are the target of TAF, the effect of a $ 200 billion intervention may be significant in terms of short term money market liquidity conditions. Given the FX conditions that are the target of China’s intervention, the effect of $ 1.5 trillion may be significant in term of these rates, and given the mostly term instruments that are involved in the asset allocation, the result of investing $ 1.5 trillion in US and other currencies may be significant in terms of its effect on term interest rates.
But the Fed is not targeting term interest rates with TAF. It is targeting the LIBOR/Fed funds spread. There may be some moderate and welcome derivative effect on terms yields, but this is not the primary objective of the TAF exercise.
Finally, the comparison of TAF sterilization and FX sterilization is interesting, but a footnote to the TAF operation. Sterilization is a normal central bank operation, whatever the asset activity is that results in its necessity. Whether it’s the Fed extending TAF funds, or PBOC buying FX, both banks must take deliberate offsetting steps in order to avoid inadvertently increasing the level of domestic bank reserves held with them. CBs must have tight control over their liability structures and ensure their asset activity doesn’t interfere with this. Hence - sterilization - whatever the source of its necessity. One exceptional aspect - should the Fed end up increasing the size of the TAF operation beyond the size of the Fed balance sheet itself (roughly $ 700 billion), it may have to start issuing sterilization bills like the PBOC in order to switch sterilization management from the asset side to the liability side - thereby increasing the size of the Fed balance sheet while still controlling the size of the monetary base.
Very well informed anonymous poster –
thanks, very fair points. Wouldn’t though the effect of the Fed’s efforts to bring down the LIBOR/ fed funds spread still be an increase in the relative supply of Treasuries v Agencies, as the Fed takes in Agencies through the TAF and then sells (or otherwise releases to the market) Treasuries to avoid growing its overall balance sheet? It may not be the goal, but it could still be the result.
The speculation the the Fed may need to expand its balance sheet and sterilize through bill issuance not just bond sales is interesting. that in some sense would combine with the ongoing growth of the (fx) balance sheets of various central banks as they absorb dollars to avoid fx appreciation, and the overall effect would be an enormous expansion of the asset side of central banks balance sheet globally, with much of the asset expansion offset by increased sterilization.
During most financial crises in emerging economies, central bank balance sheets did expand to support the domestic financial system, so such an expansion wouldn’t be unprecedented. I am not sure tho if the fed has ever done this in past episodes of distress.
Agreed, the market float of treasuries would increase as a result of the required sterilization. And the TAF acts as insurance of sorts against the possibility of distressed outright selling of the collateral that is being posted under TAF by those otherwise experiencing funding difficulties. This would help spreads moderately at the margin, relative to the alternative. So there is a connection and a correlation. But again the primary objective of the Fed is to help LIBOR spreads and interbank funding conditions, rather than rein in term agency/treasury spreads.
RE: Krugman’s “FX drop in the bucket.” I used to think the same way before. What does it matter if CBs intervene? Even China with its $1.5T in reserve holdings pales in comparison to the $3T transacted daily in FX markets.
I’ve updated my thinking on the effect of CB intervention being pronounced because maintenance of FX stocks matters more than flows represented by FX turnover. It’s those who buy for keeps, not those who merely buy and sell, that we must watch.
Krugman is quoting as saying: “In a worst-case scenario, the Federal Reserve would find itself owning around $200 billion worth of mortgage-backed securities.” Hedge funds and others are writing off MBSs. Could the FRB find itself writing off $200 billion worth of MBSs?
“The Fed sold outright $10B of Treasury securities today at the same time as it offered $15B in exchange for mortgage-backed securities under the new program (at a low interest rate than in traditional repos against MBS collateral). The net cash injection was small, but the composition of securities on bank balance sheets changed markedly, as illiquid subprime securities were exchanged for liquid Treasuries.”
Yes, the Federal Reserve is taking increasingly suspect subprime collateral destroying the monetary value of the US Dollar. The US taxpayers are being taken to the cleaners again by the Bernanke banksters.
anan poster
The libor funds spread has been relativily stable since the TAF was introduced? Seems to me TAF is not about interbank lending at this point; it is collateral parking in the hopes you can horizon push and reduce the severity of eventual marks balanced of course against regulatory capital requirements. Ted spreads have widened out but seems that is a derivative, as opposed to proximate cause, of the sytemic solvency issues. Not sure why the TAFY is viewed as so creative as the only difference here is the anonymity it provides in essence. Great so now we have the central bank putting its balance shset at risk for banks we know nothhing about and collateral which we know nothing about. There has been a lot of talk of a long standing provision whcih allows the Centbank to buiy any collateral in cases of national emergency etc..Waldman’s post is the most attune to this fact. The FT dipped its toe in the collateral issue a few weeks ago, but was half hearted in its pursuit of this entire “solution.”
The TAFY WAS about the interbank market, but that was a useful idiot for the underlying solvency rot that it is now being used to mask. Can;t wait for act III
one more things….
TAFY assumes that it is liquidity and nothing more that is at the heart of the asset reprice. That assumption alone is highly questionable.
unable to follow the plot as the wonkishness index went off the top right of the graph paper - i was driven to google ‘tulip mania’, as a way to recover perspective. i find that the dutch government of the day defused the severity of the seventeenth century tulip mania bust, somewhat, simply by declaring all contracts gambling contracts and thus unenforceable at law.
you probably know by now that i think in pictures, rather than numbers or statistics. i am dreaming these nights about helicopters emblazoned with the red star, scattering confetti as promised, but returning to base laden with tulips. any interpreters of dreams, out there ?
.
The trouble with trying to assess the effect of the TAF is that Fed funds is essentially an overnight rate, whereas the LIBOR rate that it is usually being compared to is for a term of one or three months. As such the “LIBOR / Fed funds” spread is more affected by the outlook for the short term interest rate than the transmission of Fed policy through the money market.
“Opportunistic investors are pulling back from Asian property because they see more scope for picking up distressed assets in the United States and Europe and loans are harder to get in Japan, one of their favorite markets…”I think a lot of investors will return to home markets… Some will try to buy distressed core and refinance it. They could make good returns.”…” http://www.reuters.com/article/reutersEdge/idUSHKG11463220080310
Doesn’t the TAF, which increases bank holdings of treasuries and reduces bank holdings of mortgages, also reduce bank capital and reserve requirements relative to total bank assets? It would seem to be a very effective way to relieve the current bank capital constraints on the credit markets, and reduce the LIBOR/Fed funds spread.
Guest DK — interesting point. I am not sure how much capital banks have to hold against agency guaranteed mortgages so I am not sure what the answer is.
Of course, rumors of trouble at a not small broker dealer probably would tend to widen all financial sector spreads, so there are also forces working in the other direction.
“As such the “LIBOR / Fed funds” spread is more affected by the outlook for the short term interest rate than the transmission of Fed policy through the money market.”
This won’t hold up when examining recent spread history and the TAF effect in terms of either fed funds, fed funds futures, or term fed funds.
“It would seem to be a very effective way to relieve the current bank capital constraints on the credit markets, and reduce the LIBOR/Fed funds spread.”
Interesting point. But I doubt very much that TAF financing of this collateral affects capital requirements against it. This is financing - not a sale. The banks still own it. TAF is specifically intended to relieve liquidity pressures and not to address capital or other solvency issues, or put the Fed in a position of “bailing out” banks from a capital perspective via the TAF facility per se.
The TAF does not decrease banks’ holdings of mortgages. The Fed is not buying them, they are lending against them, and the mortgages stay on the banks’ balance sheets. The banks end up with a secured liability to the Fed, instead of something like an ABCP liability to a money market fund. The money market fund now holds t-bills, and the Fed holds less t-bills and more repo (secured loan against mortgage collateral). The operation nevertheless narrows mortgage spreads because the Fed has provided a backup method of financing holding them.
Good. Now all you have to do is persuade the sheiks and the Chinese to spend a lot of money buying mortgage-backed securities which haven’t got a market value. How are they going to be priced? Or do you assume that a big buyer will make a market where none now exists?
From the Michael Pettis link,
“In her speech she (Wu Xiaoling, formerly deputy governor of the People’s Bank of China) also warned that hot money inflows will keep pouring into the country because of the China’s “rosy economic outlook and its appreciating currency.”
“…Most of the action is of the wrong kind.
http://piaohaoreport.sampasite.com/default.htm
There is a big spike in coal prices adding to Chinese manufacturer’s problems, prices have hit $300/ton due to flood related production problems in Australia.
Production and loading facilities are likely to be impaired for about two months.
http://news.smh.com.au/coking-coal-prices-soaring-on-supply-woe/20080307-1xsh.html
Gordon — There are still prices for MBS backed by an Agency guarantee, and that is all I am asking the big central banks to buy. It is the MBS without an Agency guarantee that lacks a market. but I certainly would expect a lot of agency demand from the world’s central banks to effective create a market for that debt, and bring spreads in.
The Fed is not buying anything they are trying desperately to set off a daisy chain reaction witha soaking wet fuse. UIs it me or is it ironic that while treasury sec is out bloaviating about banks raising capital the Fed is launcyhing the Blackhawks. So we now have another temporary facility right in time for the Prime Brokers to report earnings next week. Let’s say again, this is not a liquidity crisis, it is a solvency issue. The Fed is using the red herring we want to get them lending again as cover for recapitalization. Why would a bankrupt (close to bankrupt) bank lend money in an asset deflationary environment to consumers who are already overleveraged? And, how does shuffling bad collateral from one place to another solve anything expect to extend the horizon. Fed is all but pathetic at this point. Would someone please put them out of their misery already. Read Caroline Baums Bloomberg piece for a suggestion.
BTW wasn’t Bosky convicted of securities parking? Oh how times have changed
Brad,
The issue I have with Krugman’s analsysis is that the Fed isn’t actually buying anything- yet- merely lending against it. And it isn’t clear to me that these are effectively the same thing.
Directly buying these securities- even if the excess monetary base were to be sterilized- would do two things: give a market that has lacked one a bid to go with the offers and, to the extent that the sellers are the financial institutions wanting to get out from under, improve their problematic capital ratios. By contrast, lending against these securities on such favorable terms as the Fed is promoting- while clearly all else equal incentivizing market making in these securities and holding inventory- doesn’t address capital issues and doesn’t change the fact that central banks, SIVs, conduits and hedge funds are not buying (and in the latter cases, incapable of buying, or even actively selling by strategy or by margin call). These markets calcified around certain logistics that appear to have broken down- what, if any, buyers are there to step in?
In short, if a lack of buyers is the underlying problem, and if that is the cause and consequence of the dramatic continuting capital losses on these securities, and if the pace of those losses combined with the dubious capital ratios of financial institutions are responsible for spiraling margin requirements, how does this Fed move help all that? I don’t see it. What am I missing?
“If you believe that large-scale sterilized intervention doesn’t matter — and the allocation of the emerging world’s massive reserves across different assets doesn’t matter — then none of this matters.”
What matters with the emerging world is that they are massively distorting the Treasury market as they reduce repo-able supply and the Treasury market is tighter than during the short period when the US had a budget surplus. The treasury market is dysfunctional and the Fed would be right to address that issue with something like the TAF/TSLF. Eventually that would solve the bond conundrum and result in higher treasuries yields. What the TAF/TSLF has to do with solving the current credit crisis, I’ll like to know, as far as I can tell absolutely nothing, correcting a false market in Treasuries is not going to create a market for junk. In fact…it could well do the reverse…
“I’ve updated my thinking on the effect of CB intervention being pronounced because maintenance of FX stocks matters more than flows represented by FX turnover. It’s those who buy for keeps, not those who merely buy and sell, that we must watch.”
That is also my view. It seems a bit unreal to me that the question of whether intervention can matter even comes up. Intervention to support a currency is hard - speculators will break the bank. But determined intervention to suppress a currency seems to work just fine. Would anyone seriously argue that the yuan is at the free market value? Or is the argument over the time period - short run versus long run? But the yuan has been undervalued for some time, and the yen was undervalued for a long time at something like 250 per dollar.
About the recent Fed machinations - what if the basic problem is not a liquidity crisis, but a shortfall in loan demand? Changing the method of propulsion won’t make pushing on a string any more effective.
jck — good point re the impact of EM central banks soaking up so many treasuries. Presumably the fed has dramatically increased supply to help meet a surge in private demand, as banks and broker-dealers that post agency collateral for treasuries can then sell them in the market.
majorajam — good points. my sense is that the fed is hoping to help create a bid for agencies at least from the broker dealers who can use them to get treasuries, and thus help out those institutions now forced to sell agencies b/c of margin requirements. the goal isn’t so much to help out specific institutions as to break a spiral that was pushing prices out of line with what the fed views as the agencies fundamentals, and perhaps more importantly, a process that risked feeding on itself.
but i am still thinking it all through. the high quality of the comments on this thread has been helpful to me, and is most appreciated.
As Waldmann’s commenters pointed out, it’s not nationalization because the Fed is not part of the government.
But the nation *is* paying for these follies.
RebelEconomist: Given that a US government commitment to let the GSEs go bust would probably not be credible, it might actually be best to give them an explicit guarantee.
Hard to do. In order to give GSE’s a “full faith and credit” guarantee would require Congressional legislation, and even introducing that sort of legislation would cause a drop in confidence in the GSE’s which would make the problem worse.
It’s actually much, much harder to give a credit guarantee than to actually just write a check to pay for a default. If the GSE’s do default, then you just write a check, and then that’s the end of it. Giving a credit guarantee is harder because once you give a guarantee you can’t withdraw it later.