Federal Reserve Chair Ben Bernanke last week released a statement of how the Fed intends to manage its bloated balance sheet over the next few years. Here I offer my interpretation of what his plan involves.
Bernanke drew a distinction between three different categories of assets that the Federal Reserve has held on its balance sheet. The first involve extension of short-term emergency credit to financial institutions:
our financial system during the past 2-1/2 years has experienced periods of intense panic and dysfunction, during which private short-term funding became difficult or impossible to obtain for many borrowers. The pulling back of private liquidity at times threatened the stability of major financial institutions and markets and severely disrupted normal channels of credit. In its role as liquidity provider of last resort, the Federal Reserve developed a number of programs to provide well-secured, mostly short-term credit to the financial system.
This lending came in the form of a wide variety of new facilities, which summed to almost $1.6 trillion by the end of 2008, but are now almost entirely wound down or phased out, as Bernanke observed:
As was intended, use of many of the Federal Reserve’s lending facilities has declined sharply as financial conditions have improved. Some facilities were closed over the course of 2009, and most other facilities expired at the beginning of this month. As of today, the only facilities still in operation that offer credit to multiple institutions, other than the regular discount window, are the TAF (the auction facility for depository institutions) and the Term Asset-Backed Securities Loan Facility (TALF), which has supported the market for asset-backed securities, such as those that are backed by auto loans, credit card loans, small business loans, and student loans. These two facilities will also be phased out soon.
 |
Bernanke also discussed the emergency measures to support Bear Stearns and AIG in the same terms, while acknowledging that the assets the Federal Reserve acquired through these operations are at best highly illiquid and not something that the Fed is going to be able to unload any time soon.
 |
Bernanke described a third category of the Fed’s assets this way:
after reducing short-term interest rates nearly to zero, the Federal Open Market Committee (FOMC) provided additional monetary policy stimulus through large-scale purchases of Treasury and agency securities. These asset purchases, which had the additional effect of substantially increasing the reserves that depository institutions hold with the Federal Reserve Banks, have helped lower interest rates and spreads in the mortgage market and other key credit markets, thereby promoting economic growth….
With its conventional policy arsenal exhausted and the economy remaining under severe stress, the Federal Reserve decided to produce additional stimulus through large-scale purchases of federal agency debt and mortgage-backed securities (MBS) that are fully guaranteed by federal agencies. In March 2009, the Federal Reserve expanded its purchases of agency securities and began to purchase longer-term Treasury securities as well.
Although Bernanke described these purchases as almost a completely separate set of decisions from the first two categories, it seems not coincidental that, when you look at the total of all the assets the Fed is holding, the expansion of MBS purchases exactly offsets the declines from phasing out the short-term lending facilities. As a result of the MBS and agency purchases, the total assets of the Federal Reserve today exceed the total reached at the peak level of activity for the lending facilities in December 2008.
 |
Where did the Fed obtain the funds to make all these loans and later buy all this MBS? Mechanically, the Fed implements any of these operations simply by crediting new deposits to the account that the receiving party maintains with the Federal Reserve. You can think of those accounts as electronic credits that the bank could ask at any time to redeem in the form of green currency delivered in armored cars by the Fed to the bank. Because the Fed had no desire to deliver this quantity of currency, up until the fall of 2008 it essentially sterilized the new loans by selling off some of its holdings of Treasuries. This sterilization shows up as a $300 billion decline in the bottom blue category in the figure above between August 2007 and September 2008, during which period the height of the graph corresponding to the sum of all Fed assets remained quite stable. The purchasers of these T-bills delivered back to the Fed the same deposits that the Fed had created with its various loans, so that total reserves in the system were unaffected.
You can also keep track of those deposits and where they end up directly in terms of total Federal Reserve liabilities, which are plotted below. The height of this graph for every week is by definition exactly equal to the height of the preceding graph. Whereas the first graph summarizes the assets that the Fed holds, the second tracks where the dollars that the Fed created to purchase those assets ended up. The categories “service” and “reserves” in the figure below correspond to what I described above as electronic credits for cash, whereas “currency” refers to the physical green stuff. As a result of the Fed’s sterilizing sale of its Treasuries, there was essentially no impact of any of the lending programs on the Fed’s total outstanding liabilities until the fall of 2008.
 |
But the great expansion of lending in the fall of 2008 far exceeded anything the Fed had the capacity to sterilize. The Fed therefore asked the U.S. Treasury at this time to do some additional borrowing on the Fed’s behalf (represented by the yellow region in the graph above) and just hold the funds idle in the Treasury’s account with the Fed. Essentially this amounted to the same kind of sterilization action as when the Fed sold T-bills out of its own portfolio– the reserves the Fed created through its new lending facilities ended up in the Treasury’s account with the Fed and then just sat there. The Treasury’s balance with the Fed has subsequently declined, and most of the deposits that the Fed has created are currently simply sitting idle at the end of each day as excess reserves held by banks, represented by the light green area in the graph above.
Although banks may be content at the moment to hold a trillion dollars idle as excess reserves each day, one would suppose and hope that this will change as the economy recovers. But the resulting multiple expansion of credit and withdrawal of the reserves as currency would be impressively inflationary– the $1.1 trillion in credits for cash currently held by banks is a bigger number than the cumulative sum of green currency that the Federal Reserve has delivered to banks week after week since its inception a century ago, and exceeds by a factor of 100 the levels of excess reserves that banks held three years ago. Hence the need for an “exit strategy”, or how the Fed is going to manage its balance sheet when conditions begin to improve. Communicating the Fed’s plans for doing so was a key purpose of
Bernanke’s statement last week:
The FOMC anticipates that economic conditions, including low rates of resource utilization, subdued inflation trends, and stable inflation expectations, are likely to warrant exceptionally low levels of the federal funds rate for an extended period. In due course, however, as the expansion matures the Federal Reserve will need to begin to tighten monetary conditions to prevent the development of inflationary pressures. The Federal Reserve has a number of tools that will enable it to firm the stance of policy at the appropriate time.
Most importantly, in October 2008 the Congress gave the Federal Reserve statutory authority to pay interest on banks’ holdings of reserve balances. By increasing the interest rate on reserves, the Federal Reserve will be able to put significant upward pressure on all short-term interest rates….
Here the Fed Chair lost me. If the purpose of holding the MBS is indeed to produce a monetary stimulus, and if the purpose of raising the interest rate paid on reserves is indeed to produce a monetary contraction, why would the Fed want to do both at the same time? And yet Bernanke made pretty clear that he has no plans to sell off the MBS, which would be the logical way to contract. Instead it sounds like the Fed basically intends to hold those MBS to maturity:
I currently do not anticipate that the Federal Reserve will sell any of its security holdings in the near term, at least until after policy tightening has gotten under way and the economy is clearly in a sustainable recovery. However, to help reduce the size of our balance sheet and the quantity of reserves, we are allowing agency debt and MBS to run off as they mature or are prepaid…. In the long run, the Federal Reserve anticipates that its balance sheet will shrink toward more historically normal levels and that most or all of its security holdings will be Treasury securities. Although passively redeeming agency debt and MBS as they mature or are prepaid will move us in that direction, the Federal Reserve may also choose to sell securities in the future when the economic recovery is sufficiently advanced and the FOMC has determined that the associated financial tightening is warranted.
The puzzle is resolved if the Fed is thinking of raising the interest rate on reserves not so much as a tool for contracting aggregate demand, but instead as a device to persuade banks to continue to sit on a trillion dollars in excess reserves. Indeed, Bernanke went on immediately after discussing the option of raising the interest rate paid on reserves to discuss two other tools available to the Fed for dealing with that bulging light green area in the graph above:
The Federal Reserve has also been developing a number of additional tools it will be able to use to reduce the large quantity of reserves held by the banking system….
One such tool is reverse repurchase agreements (reverse repos), a method that the Federal Reserve has used historically as a means of absorbing reserves from the banking system….
As a second means of draining reserves, the Federal Reserve is also developing plans to offer to depository institutions term deposits, which are roughly analogous to certificates of deposit that the institutions offer to their customers.
I’ve discussed reverse repos
and the Term Deposit Facility at length previously. My position is that these devices, as well as the strategy of paying interest on reserves, all amount to the same essential operation as when the Fed used supplemental borrowing from the Treasury to sterilize its initial lending operations in the fall of 2008. In each case, the financial instruments we are talking about– whether it is the T-bills issued by the Treasury to fund the Fed’s supplemental Treasury account, funds lent to the Fed through the Term Deposit Facility, reverse repos with the Fed, or reserve balances that banks are persuaded to hold in their account with the Fed overnight– all have the same basic structure. The government (either in the form of the Treasury or the Fed) promises to deliver to the holder of that asset (namely to the owner of the T-bill, the lender for the Term Deposit Facility or the reverse repo, or the 5:00 p.m. holder of reserve deposits) the original borrowed sum plus interest. I would describe any such instruments as borrowing from the public. The Fed’s Plan A (raise interest rates on reserves), and Plan B (expand reverse repos), and Plan C (Term Deposit Facility), is that it will continue to borrow, as able and needed, in order to hold on to its MBS until those assets gradually decline as they mature or are prepaid.
I think Benron’s strategy is to flatten the yield curve, and mortgage spreads. The idea being low mortgage rates will help prop up falling housing prices, and mitigate the associated foreclosures, both involuntarily and voluntary.
I can see Benron’s hesitance to state that clearly as being a huge problem for the Fed to solve, after his long time insistence that it wasn’t a problem.
As far as short term interest rate manipulation goes, I guess we can nuance that the Fed isn’t directly trying to influence non real estate borrowing thru the cost of money, but is focused on ways to withdraw liquidity.
What they propose are all similar, they just work thru different channels. The least democratic is doing a zillion repos constantly with guaranteed profit for 19 primary dealers. The other two ways spread the wealth around a little more.
But speaking of wealth we can infer an upper limit on how high the Fed can go with interest rates. With $1.4T in MBS paying 5%, $300B in treasuries averaging around 2%, some losses coming, and the Fed needing cover its cost of operations, it’s hard to imagine they can raise rates higher than 4% this way.
At that point they would need to start selling long term assets. Of course that would happen only if the economy did get better, to the point of needing 4% rates to control AD or inflation.
In the meantime looks like a nice welfare program for banks, and I look forward to seeing more news about the big checks our new Welfare Moms write to themselves.
Now if Benron could figure out a way to pay me interest, I would get more excited about it. But it sounds like he wants to get securitation going again, so I’m expected to “take risk” for my tiny interest payments.
No thanks, I have to do it with real money.
How long are we going to endure this nonsense? The US is becoming a straight up Ponzi scheme.
If we leave Bernanke in charge he will destroy the economy forever.
I … don’t theenk so!
I don’t necessarily agree that the Fed’s sterilization was shifted to Treasury accounts after Jan, 2009. First of all the purchase of $1+ trillion of MBS represents a shift of ‘Green Currency’ into the pockets of sellers of otherwise illiquid properties. The 25% increase in stock prices beginning (suspiciously) a few months later also represents a $1 trillion transfer out of Fed reserves toward Wall Street insiders, perhaps not by armored cars but transferred, nevertheless!
Fed has custody of large reserves … reserves of what, exactly? Peanut butter and jelly sandwiches?
The scope of my argument is too great to encompass in a comment to an internet article – but, please catch my drift. There are no real reserves; extend and pretend works both for currently solvent as well as insolvent banks. Banks are ‘lending’ to the Fed and the reserves aren’t ‘real’ Fed liabilities. The Fed is simply custody or bank IOU’s and the public is the sucker.
Excuse me while I go and figure out how I can sell/lend MY illiquid (worthless) assets to the Fed for some long green …
Wow. Feel like I’m obsessively staring at a really nasty automobile accident. These numbers just boggle the imagination I’m capable of….
For perspective, please, what percentage of all US-based mortgaged backed securities (MBS) does the Federal Reserve now own? US$1.4 Trillion of MBS represents what fraction of the overall US-based MBS?
Jim
I enjoy reading your analysis of the Fed’s actions and I completely agree with your assessment that the Fed’s plans amount to borrowing from the public. I find it interesting that Congress has been slow to point out that it, rather than the Fed, is supposed to have the authority to determine how much borrowing from the public the Treasury engages in. I am curious to see how the politics of this strategy plays out.
Best
Andy Atkeson
GNP,
per FT Alphaville
“Firstly, at $9.213 trillion (end Q3) the US mortgage related securities market is still significantly bigger than the US Treasury market, and at $1.4 Trillion, the Feds Agency MBS purchases amount to 15% of total market size.”
http://ftalphaville.ft.com/blog/2010/02/04/141046/filling-the-central-bank-void/
Andy,
You make it sound like all so much nonsense when you put it that way.
“the $1.1 trillion in credits for cash currently held by banks is a bigger number than the cumulative sum of green currency that the Federal Reserve has delivered to banks week after week since its inception a century ago”
Source?
-Carl
You can think of those accounts as electronic credits that the bank could ask at any time to redeem in the form of green currency delivered in armored cars by the Fed to the bank. Because the Fed had no desire to deliver this quantity of currency, up until the fall of 2008 it essentially sterilized the new loans by selling off some of its holdings of Treasuries.
Little green pieces of paper are convenient for small-scale retail transactions and extremely inconvient for anything else. As the volume of such transactions isn’t increasing, there’s no danger of anyone wanting to convert their reserves into extra armored truckloads of paper.
Nor do excess reserves drive additional bank lending. Banks can always obtain the reserves they require to support the loans they make — borrowing at the discount rate if nothing else. So the volume of bank loans is determined by the availability of borrowers who are judged credit-worthy and who are willing to pay the risk-based rate markup above the bank’s borrowing costs — generally the Fed funds rate, but capped at the discount rate in any case.
So why sterilize? Because the Fed was defending a non-zero policy rate, and not yet paying interest on excess reserves. If the banking sector in aggregate has excess reserves, then the overnight market won’t clear — more banks are competing to loan out money than are seeking it, and the rates offered will drop to the “support rate”, which was zero until fall 2008 (briefly set a quarter point below the target Fed funds rate, on Nov 6 the support rate was set equal to the target rate, and the Fed no longer had to worry about sterilizing or whether the overnight market would clear).
Here the Fed Chair lost me. If the purpose of holding the MBS is indeed to produce a monetary stimulus, and if the purpose of raising the interest rate paid on reserves is indeed to produce a monetary contraction, why would the Fed want to do both at the same time?
If the purpose was only to add liquidity to the system, there are plenty of long-dated Treasuries out there that the Fed could have bought instead. The way I see it, the Fed chose to support the market for a particular asset class, MBSs and Agencies, because they believed that that market was not functioning properly — deep-pocketed investors had grown skittish, no one knew how to price the risk. Quoting Friday’s IMF staff paper:
“Markets are segmented, with specialized investors operating in specific markets. Most of the
time, they are well linked through arbitrage. However, when, for some reason, some of the
investors withdraw from that market (be it because of losses in some of their other activities,
loss of access to some of their funds, or internal agency issues), the effect on prices can be
very large…. When this happens, rates
are no longer linked through arbitrage, and the policy rate is no longer a sufficient instrument
for policy. Interventions, either through the acceptance of assets as collateral, or through their straight purchase by the central bank, can affect the rates on different classes of assets, for a
given policy rate.”
So the Fed is anticipating that they might want to tighten monetary policy broadly without withdrawing their support for MBSs as an asset class. Paying a higher support rate on excess reserves is a way of raising the policy rate. That rate rise will be passed through in all bank loans (the risk-based markup that they charge each potential borrower isn’t affected), so the pool of people desiring to borrow will be reduced. And, as you say, that money will continue to sit in excess reserves.
Carl Lumma: Total currency outstanding is the number reported weekly on the H41 and is the number plotted in dark green in the fourth graph above.
lilnev: Supply must equal demand for each asset, whether we are talking about green pieces of paper or deposits with the Fed. If there is an excess supply of an asset– and it is not hard to imagine a situation in which $1.1 trillion represents an excess supply of reserve deposits– then other factors must adjust until equilibrium is restored. And this particular asset must either be held as reserves, withdrawn as cash, or absorbed back by the Fed through a separate operation.
I am suggesting that adjustment of the price level is one possible mechanism by which the demand for nominal cash could rise to meet the available supply.
I don’t get this. Isn’t the reason the banks are keeping all these excess reserves as cash, precisely because they are offsets against expected losses the banks expect to have to swallow in the next year or two from Option ARMS and commercial mortgages and credit card defaults?
I thought the banks were keeping the cash aside because they know they are currently in the eye of the storm and the second wave of defaults will begin shortly.
I’ve been assuming, once all the losses are written off, there won’t be any surplus cash left to lend out.
Professor, Don’t the reverse repos and the term deposit facility sound like temporary fixes to reduce the reserves?
So the fed is hoping for a best case scenario of inflationary pressures not picking up before the MBS mature. Does that make sense?
GNP: The Fed’s Flow of Funds Table L.125 lists $5.3 trillion in agency- and GSE-backed mortgage pools outstanding for 2009:Q3, of which the $977 B currently held by the Fed would represent 18% of the total.
Keid A: The term “reserves” is used in various capacities in banking. One is “loan loss reserves,” which are designated for purposes such as you mention. Here we are talking about Federal Reserve deposits, which simply represent accounts held at the Fed. There is no connection between Federal Reserve deposits and loan loss provisions.
Spry:That this is a temporary fix is exactly my point. By its nature, borrowing means putting off the repayment. The Fed plans to put off absorbing these reserves back in until the MBS mature.
Keid A,
Official bank “loan loss reserves” (or sometimes called “provisions”) are quite small so far relative to what actual losses may turn out to be over the next couple years. Banks don’t like scaring their stockholders, and it has been fashionable lately to shore up capital thru secondary equity offerings.
My bet is that part of the reason excess reserves are so high is that banks know this and are keeping a large unofficial cash cushion.
Plus mark-to-market accounting rules have been relaxed, so once these go back to normal, and higher capital requirements are adopted as financial stability proponents suggest (shouldn’t they talk to Benron and solve his problem for him???), and some Tier 1 or 2 capital is written off, much of the reserves won’t be excess anymore.
The inflation bug’s mind works on the basis of fear; specifically the fear that rising prices deter and ultimately destroy savings and investment.
And the inflation bug is 100 percent correct.
To see this, all we need to do is to compare the price level for consumer goods and services today with level prevailing in 1933. Yes, that’s right; prices today are much higher than they were in 1933.
And what has been the effect? Yes, just as the inflation bugs predicted, investment has plunged dramatically in the last 75 years, rendering the U.S. a byword for staggering economic decline.
And those living on fixed incomes or from savings have, en masse, been forced to beg whatever crumbs they can garner from illegal immigrants working as crop pickers or car washers. The creation of Social Security merely exacerbated the pauperism that our decades-long monetary inflation has wrought.
The only exceptions are those wise and fortunate souls who have hoarded gold.
If only we as a nation had not inflated our currency starting in 1933, our children and grandchildren would have all the marvellous pre-1933 era blessings afforded by Sound Money policies.
But the inflation bug at least can be consoled by the KNOWLEDGE that he was RIGHT, and that the inflationists were wrong.
Now, there are some who will say the right-wing inflation-haters are merely absurd cranks, or absolutely whacko crackpots, or indescribably moronic simpletons, who are fed these stupendously erroneous ideas about monetary economics by the propaganda machine that serves the interests of the ruling plutocracy. But as the simple, clear, and dramatic analysis of the facts which I have provided above PROVES BEYOND ALL DOUBT, these idiotic douchebags actually have the facts on their side.
Not, to be sure, the liberal-socialist-collectivist so-called facts, but the facts as derived from the ridiculous, off-the-wall, fruitloop worldview of the inflation bug.
Professor,
The interest rate paid on excess reserves CAN NOT depart significantly from the Fed fund rate.Paying interest on excess reserves that is much higher than the Fed fund rate would automatically drive the Fed Fund rate higher. Why? Simply because Banks would have no incentive to lend to each other at the Fed fund rate since it could park their excess liquidity at a higher rate at the Fed. Therefore, the Fed fund rate would increase in such scenario to match approximately the rate paid on excess reserves.
So when Bernanke is saying that “By increasing the interest rate on reserves, the Federal Reserve will be able to put significant upward pressure on all short-term interest rates”, he is de facto stating that he would be increasing the Fed fund rate as well.
As you pointed out: Reserve balances with the District Federal Reserve Banks are the difference between “total factors supplying reserve funds” and “total factors, other than reserve balances, absorbing reserve funds.
The member banks do not loan out this remainder, or their excess reserves.
Prior to August 2008 (with the exception of bank capital adequacy rules), the member banks were already unencumbered and unimpaired in their lending operations. Bank lending and investing were only constrained by their clearing activities, or the debits & credits to their customers accounts. Vault cash & ATM networks, plus retail deposit sweep programs, had outstripped the need to hold the former larger volume of legal reserves. So, only as it is profitable to borrow, and profitable to lend, would bank credit expand.
However now, IOR’s are member bank earning assets. These earning assets compete with other yields and money market instruments. IOR’s have caused the member bank portfolios to be shifted (and the supply of loan funds to be fall), from low yielding T-Bills (& vault cash), into the higher yielding, risk-free, IORs @.25%. E.g., on November 19th, the yield on a new 3-month U.S. T-bill fell to 0.005%.
Also, IOR’s cannot be accurately juggled to balance the factors absorbing & supplying reserve funds (to manage the money supply), when the system’s expansion coefficient varies widely.
Never the less, it seems likely that the FED will to some degree; utilize a combination of these policy tools (reverse repos, term-deposits, & IORs).
Why? taking the largest factor absorbing reserves, the question is: how does the trading desk forecast the change in excess reserves relative to a change in the remuneration rate on IORs? I think the demand for IORs will initially vacillate, and even after it stabilizes, will in the end, be unpredictable. Weve already seen the FED guess at the proper rate.
This provides a need for the last two policy tools mentioned (reverse repos & term-deposits). These have much more advanced monetary management characteristics as compared to IORs.
For example, term Deposits will enable the FOMC to calculate and sell whatever volumes, durations, rates of interest, at whatever auction times, or to individual banks (small, medium, & large), & whatever other explicit covenants, it wants to target. I.e., the FED can use a systematic, measured (staggered), technique, to unwind IORs (if necessary).
Reverse repurchase agreements are somewhat more limiting, incorporating the price, volume, and duration.
IORs, term-deposits, and reverse repurchase agreements, now act exactly like raising reserve ratios (a credit control device, by raising, or lowering, the volume of outstanding legal reserves).
As of Feb 10, IORs stood @ $1,160,806 trillion. Due to the size of these IORs, excess reserves can be used as a measure to vary the banking systems legal lending capacity (i.e., by increasing, or decreasing, idle, unused, bank deposits). In this environment, the term excess reserves is no longer an accurate label.
With the debt markets saturated, IORs would be less disturbing, and the preferred policy tool, because term-deposits, & reverse repos, compete, and conflict, with the issuance of governments, corporates, municipals, mortgages, refinancing, rolling over, funding facilities, and everything else along the yield curve.
The order of preference will be IORs, reverse repos, and then term-deposits. Fixed volumes, for fixed maturities (no early withdrawal privileges), will make term-deposits and reverse repurchase agreements, less volatile, and more predictable. It makes it easier for the FEDs other tools to supplement, substitute, and balance: public and private debt, their floatation, rollover, & runoff.
There seems to be a great deal of misunderstanding surrounding Excess Reserves.
In a rate-targeting system, the Fed stands ready to supply any funds to banks needed to keep the Fed Funds rate at target — 0%.
So banks should be completely indifferent as to the level of Excess Reserves. If ER’s are zero, and the banking system wishes to increase loans by $1tr, the Fed will simply immediately provide any resulting reserve funds demanded to keep the Fed Funds rate from rising above 0%. Therefore, banks should not care whether they have existing Excess Reserves or whether the Fed provides reserves to the system as the increase in lending occurs.
At any given target policy rate, the level of Excess Reserves is irrelevant to lending and the money supply. Why do Excess Reserves matter? Because they are the funding mechanism for the Fed’s quasi-fiscal subsidies of mortgages and long term Treasuries. The issue is what will happen when those subsidies are stopped (in March) or reversed.
I *think* I understand.
1. “$977 B currently held by the Fed,” and this too is ultimately backed by the taxpayer?
2. This is the amount, at par?, that the banking system had itself relieved of, that is now “excess reserves” that the Fed is now paying interest on (or in JDH’s parlance, the Fed is now borrowing from the banks)?
If these statements are true,
3. what are the estimated gains & losses these MBS are likely to deliver? Is there a proxy? and
4. What are the expected proceeds to the banks for these excess reserves at some average interest rate of x%–which gets us to what are the maturities of these MBS?
5. Could the Fed achieve keeping these reserves from turning into Benjamins by jacking up the reserve requirements on THESE reserves, keeping this interest rate (paid on excess reserves) low and at the same time hiking the rate which affects main st (I do get the terminology confused on the two rates, or are there more?
Professor,
Question going back to the Aug 2007 to Sept 2008 timeframe.
In Aug 2007, the Fed initiated two types of actions: 1) it began lowering the Fed Funds rate and 2) (slowly at first) it began selling treasuries and creating various loan facilities (Term Auction, Repos, Swaps — H41 T10).
It seems to me that these actions were in opposite directions? Lowering the Fed Funds Rate makes borrowing to cover reserves cheaper and signals a loose monetary policy, but selling $300B in treasuries pulled reserves out of the system forcing banks to access the newly created lending facilities.
It seems to me that this combination of actions weakend the banking system as a whole? The end result was to replace bank reserves with Fed Loans. That is non-interest bearing reserves got replaced by Fed loans requiring interest payments. For a banking system that was facing uncertain loses on mortgages and MBSs, forcing a cash outflow doesn’t seem smart to me.
Additionally, I would think the loss of reserves would create significant pressure to reduce lending?
What am I missing?
Cedric and JDH: Thank you. I’m happy with rough orders of magnitude.
But in the interests of better understanding this data, please bear with. I read from the fed fund flow table L.108 Monetary Authority (1) US$824 B on Line 13 for Agency- and GSE-backed securities held by the fed. That number is seasonally-unadjusted. The seasonally adjusted number is US$1 Trillion from table F.108.
I do not understand where the figure $977 B comes from. What am I missing?
Keith Eubanks: Selling T-bills by itself is indeed contractionary (drawing in reserves), but the Fed at the same time extended all these loans (putting out reserves), so that the net effect was sterilization (no change in reserves).
GNP: I’m reading $977 B off the Feb. 10 value from the latest H41 release.
Yes, pegging the FFR ever since 1965, has assured the bankers that no matter what lines of credit they extend, they can always honor them, since the Fed assures the banks access to costless legal reserves, whenever the banks need to cover their expanding loans deposits.
The effect of tying our monetary policy to a pegged rate is to supply additional (and excessive legal reserves) to the banking system when loan demand increases.
The FED can directly control the FFR, IORs, & the Primary Credit rate in the short-run, but the effect of Fed operations on all other interest rates in the long-run, is still INDIRECT, and varies WIDELY over time, and in MAGNITUDE.
Our money supply can never be managed by any attempt to control the cost of credit (i.e., thru pegging governments, or thru “floors”, “ceilings”, “corridors”, “brackets”, etc).
The FED can hold the remuneration rate on IORs for longer periods as compared with previous methods. Thus, IORs will become an engine of inflation.
The banks are indifferent, but as Dr. Hamilton points out, the FED continues to sterilize, (H.4.1) or offset, the role excess reserves plays in the banking system. It’s the MATCHING PRINCIPLE:
In the Week Ending 2/10/2010:
Total factors SUPPLYING reserve funds: 2,292,193
Total factors ABSORBING reserve funds
(other than reserve balances): 1,138,143
RESERVE BALANCES at the FED: 1,154,050
Total factors ABSORBING reserve balances 2,292,193
Professor,
Yes, I see the sterilization in the data. But why implemment the policy at all: selling treasuries and creating loan facilities? The bottom line for the Fed’s balance sheet didn’t change; total reserves didn’t change much; but banks now had interest payments to make.
If boosting reserves was a goal, why not simply buy Treasuries on the open market? The fed went to a great deal of trouble to shift things around. Why?
Keith Eubanks: Yes, that is the question. If the Fed had wanted to conduct true quantitative easing, they would not have sterilized. I think the explanation, both for the behavior then, as well as for Bernanke’s plans for the future, is that the Fed is thinking in terms of two distinct objectives, trying to control both the level of short-term interest rates (the traditional focus of monetary policy) and the spread between the yields on risky and safe assets (the objective of the various new measures such as the MBS purchases).
Thanks JDH.
Cedric: I just realized that the Financial Times article you point to Filling the central bank void states that £1.4 Trillion of fed-owned MBS is measured in British pounds, not American dollars. I did not check to see if that affects the calculation of the percentage.
OFF-TOPIC: I dunno about this GSE acronym. GSE = government sponsored enterprise. I believe something like Crown Corporation would be much more elegant. Sovereign Corporation might work better in the Republic of the USSA (United Socialist States of America). Or how about POEs? People- or public-owned enterprises.
GNP,
Yes, I noticed that too, but assumed it’s a typo.
I think the original Fed plan was to end up with 1.4T of agency MBS and 300B of agency corporate bonds. The latest numbers that are projected when the Fed is complete in March are 1.25T of agency MBS and somewhere north of 200B in agency debt.
The 9T+ total market size of all US MBS includes private issued MBS as well. Also known as CDOs backed by residential mortgages.
So the 15% estimate is close enough for government work.
GNP,
I think the acronym GSE needs re-work too.
How ’bout PHA, Public Housing Authority?
Plus we can sell it to the Chinese someday. (along with the Pentagon so the Chinese have plenty of rent and tax collectors)
JDH: clear and factual, so thank you. It also stimulated some good discussion, so thanks again.
Regarding the Fed – their car has a loud public address system, but they drive it by looking at the rear view mirror. It works when the road is smooth, but we’ll see . . .
Keith Eubanks, JDH
Not sterilizing would drive the Fed fund rate to an absolute zero. This would de facto means that the Fed abandoned any form of interest rate targeting. No Central Banks have done that during the financial crisis since a zero rate policy would mean that Banks have no incentive to lend to each other (for a private bank, there is no difference between sitting on liquidity at 0% or lending to another bank at 0%). Bank of England started paying interest on reserves (similar to the Fed), while all interventions by the Bank of Canada were sterilized by a corresponding emissions of Canadian T-Bills.
To my knowledge, quantitative easing without sterilization has not been tried by any Central Banks precisely because they want to maintain the target rate above zero.
So is the FED sterilization and acquisition of MBS’s and longer dated Treasuries an example of not letting a good crisis go to waste?
As it stands now, the FED should be able to pop any future housing bubble or treasury bubble by selling from its stock of these securities. The resulting increase in mortgage and/or treasury rates would effectively pop any equity bubble as well.
as you Dr. Hamilton stated: the shifts in the composition & asset substitutions nets out.
but how is the liability side of the balance sheet today, different from the liability side of the balance sheet, say, when percentage of required reserves to note & deposit liabilities was 91.1 in 1941?
or when our monetary gold stocks increased from $3.6 to $23 billion (between 1932 to 1941), due to an influx in gold and the rise in the price of gold from 20.67 to 35 per ounce? (and there was no offsetting movement of Reserve bank credit)?
the difference between factors supplying reserves, & factors absorbing reserves (other than reserve balances) accumulates in the bucket entitled “reserve balances”.
I guess that the balance sheets in these two periods are similar (they reflect a large absolute volume of reserves)
but does it make a difference what we call the remainder? total, required, excess, IORs ?
if the liabilities on the balance sheets are comparable, then the FED is following a “TIGHT” money policy. we will see in the next couple of months. gdp falls off in May. too bad GDP isn’t reported on a monthly basis.
MF: I just want to correct a misinterpretation of what JDH wrote. Overnight rates are the traditional tools that the US central bank “focuses” on.
The US has not targeted interest rates per se for a long time. It continues to focus on two mandates: price stability and economic growth. The US central bank has not yet officially adopted an inflation targeting policy.
If the US federal reserve had adopted inflation targeting prior to the 21st century–like so many other rich, developed economies–I believe it is an open question whether the Greenspan federal reserve would have kept overnight rates so low for so long earlier this decade. My naive belief is the following: the US federal reserve operating under a clear inflation-targeting mandate would have avoided creating the conditions that generated the financial collapse of late 2008.
Correction, that should have been EXCESS reserves as percentage of note & deposit liabilities. It was 91.1% in 1941.
So where are we today? EXCESS reserves are 13% of M2, or $1,119,423,000,000 (Feb. 10, 2010)divided by $8,415,000,000,000 (Feb. 1, 2010).
Concern about the level of excess reserves has been exaggerated.
GNP:
The FED is a short term interest rate setter, not a short term interest taker. The FED will do what is necessary for the target overnight rate it has itself established to be achieved. I agree that this is the policy tool, and that the policy objectives are price stability and economic growth.
Whether we like it or not, the FED by setting the target for the overnight rate, set the rate on its short term debt (treasury bills). It would makes no sense for the yields on short term treasury bills to depart significantly from the Fed fund rate. Arbitrage possibility for traders guarantee this. It is weird that economists still don’t get it.
How else would you explain the Japanese case?? This country has the highest debt-to-GDP ratio in the world and is able to borrow for three months currently at 0.13%, while Australia with one of the lowest debt-to-GDP ratio in the world is currently borrowing for 3 moths at 3.65%! Ask any economists to explain this phenomenon, and the majority would venture an opinion regarding the “different culture” in Japan. When are economists going to understand that a Government with its own floating currency sets, through its central bank, the short term interest on its own debt?
MF: You are correct. We generally assume the central bank sets interest rates as if they were exogenous. However, if you pay careful attention to Bernanke defending policy earlier in this decade, he makes the fed appear an awful lot like an interest rate follower. Granted, he could be simply finger-pointing at the global capital glut. Who in the their right mind would take credit for contributing to the worst economic disaster in the post-war period?
As for economists “not getting it”, do you have any idea how transparently “Strawman” your argument is? Just in passing, I thought Japan was a net exporter of capital but I could be wrong.
GNP:
My “strawman” argument was not directed to you but rather at the economic profession… I have been reading a lot of articles lately talking about the fact that yields would have to increase on T-bills otherwise there would be no taker… There will ALWAYS be taker for T-Bills at approx the price (ie. interest rate) dictated by the FED. Period.
Can we now move to another topic like discussing the unemployment rate? 🙂
A question: what would happen if the Fed decided to mark down its MBS portfolio by, say, 25%? Would the decline come out of reserves?
MF, you wrote: There will ALWAYS be taker for T-Bills at approx the price (ie. interest rate) dictated by the FED. Period.
Are you sure? That’s news to me and contradicts everything I know about financial markets.
Walt,
It’s my understanding that the Fed bought the MBS (don’t know at what price though). Thus the reserves are the property of whatever bank/institution the Fed bought them from and the reserves are on deposit at the Fed, not the property of the Fed, and thus would not be affected by a writedown of asset value held by the Fed. Now if the Fed had done a reverse repo upfront…
GNP:
By definition, when the U.S. government creates a deficit, it de facto creates the liquidity (you can call it cash or money) that will be used to buy its treasury bills. This is not an economic theory, this is an accounting identity: when the U.S. generate a fiscal deficit, someone somewhere get U.S. dollars (ex. through higher transfer payments, or a subsidy for example), and someone somewhere exchange its U.S. dollars for Treasury bills.
Then the argument by mainstream economists is as follow: the U.S. fellow that got cash through the deficit spending will turn around and buy a plasma TV made in China and a Japanese car so this cash will leave the U.S. for good.
Well… no it does not leave the U.S. for good!!! What are the Japanese and Chinese going to do with their U.S. dollars?? They won’t stack it under their mattress, since they want some form of return on it. So they will try to invest it back in the U.S. (where else could they invest it??). So they could decide to invest in real estate, buy Treasuries or build plants… whatever…. in all cases you note that this money flow back to the U.S..
Of course, some would argue, China and Japan could always exchange their U.S. dollars holdings for Euros… and let say the German government is willing to accommodate that. But then… it is the Germans that are caught with U.S. dollars… so they will seek a return on it. This money will flow back to the U.S. in the form of purchase of treasuries for example.
Here’s my straightforward advice to China: if you are fed up with the continuous inflow of U.S. dollars you receive… just stop exporting to the U.S.. Same message to Japan: you don’t want U.S. dollars? So keep your cars! And now let see what this will do to the Chinese and Japanese economy… It is the U.S. that has leverage over China, not the other way around.
Economists are really good statisticians and econometricians… but so clueless when it is time to examine accounting identity and money flow. My friendly advice: don’t listen to the deficit terrorists in DC that tells you that at one point China and Japan will stop buying our stuff. They have no other realistic options. In fact, other than stopping their export to the U.S., their other options is to stack their US dollars under their mattress or burn it.
Lets assume for the sake of it that they do stack their US dollars under the mattress of the People’s Bank of China (China Central Bank)… This money would therefore leave the U.S. for good, and this will create a liquidity shortage in the Fed fund rate market that will tend to push rates up, but then the Fed would intervene through open market operations by purchasing treasuries in order to inject liquidity back in the private economy and maintain the fed fund rate unchanged. So China sleeping on their US dollars holding would result in the Fed buying back the U.S. government debt. This is real great news for the U.S… it now pays interest to itself (Fed’s profit flow to the U.S. Government every year).
Cheers,
MF
Walt French:
To add to Rob’s comment, when purchasing MBS, the Fed did take an hair cut (I am not sure how much however). So in practice, there is room for a decline in MBS value without the Fed’s profit/capital being affected. Assuming that the decline is so important that it would decrease the asset side of the Fed’s balance sheet, then the Fed’s profit would be affected, and the Fed would send less money to the Treasury at year-end.
BTW- The Fed is hugely profitable these days (see:http://articles.latimes.com/2010/jan/13/business/la-fi-fed-profits13-2010jan13), so it could be able to absorb a decline in MBS value without its capital being affected.