How the Fed Cornered the Long Bond
Fed Treasury Holdings 5-7-2014
The above link should take you to a PDF showing the Fed’s System Open Market Accounts holdings of Treasury Bonds and Notes which the second link will tell you comprise $2.224 trillion of the total $4.017 trillion of SOMA Holdings, with that total including $1.631 trillion of Fannie and Freddie Mac MBS’s. With the remainder in a variety of other Federal securities. In other words the ‘AFTER’ of three rounds of QE.
The PDF is extracted from a formatted Excel worksheet and shows all Fed SOMA holdings of Notes (1 to 10 years) and Bonds (20 & 30) by Maturity Date, Issue No, Coupon Rate, Par Value, and % (of Issue) Outstanding. As you scroll through the PDF a particular relationship jumps out at you: the higher the Coupon Rate the higer the % of Outstanding actually held by the Fed SOMA with a top limit apparently set at 70%. This isn’t entirely fixed, there is an additional layering of Long Term over Short Term with the Fed holding small percentages of all issues before jumping up to the 44.3% of the 9.25% Feb 15, 2016 and then taking progressively higher chunks of the Long Bonds maturing after that until holding peak as a percentage of issue with the 70% of the 8.13% 8/15/2019. Which percentage holds steady until the 8.00% 11/15/21 but then varies downward with holdings ranging mostly in the 55-70% range for issues between 2021 and 2042.
Feel free to prod at the numbers in this data table as you will. But the first order conclusion is that the higher the coupon rate in the out years of the Long Bond the larger the position of the Fed. With an apparent self-imposed limit of 70% of any given issue. I plan to comment on this at length in Comments but will allow readers the first crack. But would add one little fact nugget for your consideration: the Fed rebates all profits to Treasury. And to the extent that those profits are driven by those 8 and 9% interest payments on Bond issues where the Fed has holdings ranging up to 70% the end result is that a very large percentage of debt service on the Long Bond, a dollar amount that is recorded as an Outlay on the Federal Budget is effectively rebated back to Treasury.
And while this should be obvious the Table shows its most obvious break in 2024 where the percentages held jump up to fairly steady state levels. The reason is obvious, all issues maturing after 2014 HAVE to be either 20 or 30 year bonds. Because 10+ years from now.
A future post will try to put numbers on the total amount of U.S. Public Debt that actually matures after 10 years and the % of that held by the Fed, which in practical terms means held in trust for Treasury. Which makes for a funny kind of ‘debt’.
Webb – the average yield on the Fed’s Treasury holdings is 3.3%. Keep in mind that the Fed has to borrow the money to buy these assets. The borrowing rate today is 0.25%. So the Fed is making a very tidy profit on their boatload of bonds.
But the Fed has said they will be raising rates in the next year. When they do that the spread it earns will decline. When/if the Fed Funds rates rises to 3.3% the whole portfolio will be underwater – no more remittances to Treasury!
So will % rates rise? I think so, but that does not matter. You hang your hat on the forecasts from SSA on things, so use SSA as a forecaster of interest rates. You will see that SSA expects short-term rates to exceed the 3.3% level in a few years.
Webb sees the SOMA holdings as a huge windfall going forward – I see it as a potential loss to the country.
Krasting can you explain to me why the Federal Reserve has to borrow money to buy Treasuries? Isn’t the goal of QE primarily to inject NEW funds into the market? Doesn’t the Fed have its OWN printing presses for Federal Reserve Notes? Isn’t this just new money fresh off those presses?
Since the Fed is easing off QE and maybe won’t be increasing its holdings of Treasuries much going forward wouldn’t its putative borrowing rate be pretty much moot even if it WAS actually borrowing?
More importantly how could the Feds portfolio be underwater if its policy was buy and hold? After all those Bonds are producing an inward flow of funds at the Coupon Rate and if the Fed had no plans to see they could be indffierent to changes in prices on the secondary market. Because these holdings are not the equivalent of a bond fund where all gains depend on the variation in price between new and old issues.
You are making the same kind of category mistake that you do when looking at the Social Security Trust Funds through the green eyeshades of an ex bond trader. The Trust Funds simply don’t function in the same way that a bond fund would or even in the way that a public pension plan like CalPers does. Because the holdings are simply indifferent to theoretical price (because SocSec Special Issues are non-marketable anyway). The Trustees of Social Security ALWAYS collect interest on Specials at EXACTLY the Coupon Rate and so too would the Fed if it adopted a policy of buy and hold.
Plus the key here is not the ‘profits’ but the apparent fact that it costs Treasury nothing to service bonds held by the Fed, not in real terms.
Or maybe you can explain this to me. Starting with demonstrating your contention that the Fed has to borrow money at any rate at all before decidinng to buy a Treasury. And then how if their policy was to buy and hold they could EVER endure a ‘loss’? Because I think you are both confused and confusing (the issue) here.
Why speculate when you have Google? Turns out the Fed publishes a handy FAQ called “Purchases of Longer Term Treasuries” which spells out the policy and procedures and in passing confirms that 70% limit.
http://www.newyorkfed.org/markets/longertermtreas_faq.html
I am still digesting it, so everyone feel free to snack on it in the meantime.
A first note. The Fed has weighted targets for puchases based on maturity. For example it directs that 29% of such purchases have maturities of 7-10 years compared to 27% with maturities from 20-30 years. But the spread is not as even as this might suggest because the overall pool of 1 to 10 years is nearly 6X as large as the pool of 20 and 30 year meaning that the 27% of allocated purchases directed to the Long Bond builds the % of Outstanding significantly faster than the 29% allocation directed at shorter term Notes.
Krasting
i want to avoid commenting here because i really don’t know anything about what Bruce is talking about. But it does seem to me you are confusing the issue.
I at least don’t “hang my hat” on SS forecasts. I just try to tell people what those projections (not forecasts) mean in terms that mean something to average workers.
Meanwhile, I suspect SS Trustees (the actuaries actually) know what the forecasts are and incorporate them into their projections… which would mean that when I break down those projections into terms that mean something to the workers who pay the taxes and expect the benefits, those “forecasts” are included in the projections.
This doesn’t mean the projections won’t change. My position will remain, however, based mostly on the projections of the last ten years or so, that the new projections will not amount to anything nearly as dire as it may seem and that the best way for workers to insure that they will be able to retire when they need to is to raise their own “Federal Insurance Contribution” rate (you call it the “payroll tax”) a few tenths of a percent… pennies per week…. Because however bad things get, that SS savings is what will change “not enough” to “enough” when they need it most.
I don’t think the SSA is that great at predicting interest rates in the long term. They seem to assume in the Trustees’ Reports that the future will eventually be like the past, so the intermediate forecasts for most quantities go to the long-term average, usually over a span of a few years. So if you look at interest rates for 2014 in the 2009 – 2011 reports, they predict values of 5% or more (for the Special Treasuries). In 2013, they don’t predict that until 2016 – but it’s always just a few years away. I think they only depart from using the past for quantities that have a definite trend, like life expectancy.
By the way, does anyone have any idea of when the 2014 Trustees’ Report is coming out? Is that kept secret until the actual release?
Mike B
that’s what i was trying to say by distinguishing “forecasts” from “projections.”
Mike B taking your last question first.
I posed that question to a ListServ that includes everyone on the pro-SocSec side that matters and the word that came back from a knowledgeable source was “early to mid June”. This despite the fact (which I pointed out in my query) that current law mandates a Report date “no later than April 1”. Something the Obama Trustees have not met even once in six Report years. What this says about their commitment to Soc Sec generally I don’t know exactly. But nothing good.
As to interest rates and ‘ultimate assumptions’ there is an operating principle that future variations of al relevant variables with be around some sort of ‘natural’ mean and that for projection purposes you can just set all future years to that average. As far as interest rates specifically I don’t see 5% nominal and 2% real on the 10 year note which OACT uses as their baseline is that out of whack. On the other hand nothing has been normal on this front since 2008 and maybe we need to knock a point off both metrics.
On the other hand for the purposes of this post that doesn’t matter much. Because Coupon Rates on existing 20 and 30 year Bonds are what they are and will be until maturity. Just as a look at the linked PDF shows Treasury paying off Long Bonds with Coupon Rates over 11% this year! presumedly 30-Years from 1984.
My argument is that between Treasury retiring high coupon rate Bonds from the 80s and 90s and the Fed gobbling up overlapping 20-Year and 30-Years from the 90s and 00s we have a collapsing effective yield curve. Already the average interest rate on outstanding Notes and Bonds is a combined 2.007%, even before you calculate in the effects of the Feds ‘corner’ of the higher rate ones. We are already at the point where average interest rates on all Treasuries are at the Feds inflation target and are on the cusp of having that average cross with actual inflation leaving us with negative interest rates on our debt. At some point this has to correct but as it is it makes a mockery of the claim that we are saddling future generations with unsustainable debt service. I mean even if we concede that a lot of infrastructure is fraying we are still passing forward an impressive amount of dams, water projects, energy projects and grids and functioning highways, bridges, seaports and airports for a pretty small amount of needed debt service. And have the opportunity of funding a bunch of upgrades with fresh rounds of long term debt with coupon rates in the 4-5% range. Precisely because the Fed gobbled up 70% of existing issues carrying higher rates. I just want to have a discussion of the implications of all this for politics and policy.
Bkrasting,
The Fed does not borrow to buy the bonds They simply buy in the open market with the push of a button swapping reserves for the bonds. Not much more than an accounting entry. The interest remitted to the treasury on the bond portfolio is simply money destroyed (it does not go in any savings account), versus being paid into the economy. The Fed has done this for a long time it is basically open market operations, and the method used to mange the FFR. Now at the zero bound they use interest on reserves to mange the FFR floor.
If the Fed wants to raise rates they can simply sell the bonds back, and drain reserves from the system.
Treasuries really serve a couple of purposes – interest rate maintenance mechanism, and a risk free savings asset for pensions and other savers. But they are not required to execute on those items.
Remember, if the Federal government wanted they could continue to spend and never issue any treasuries, they would just push one button instead of two, and the accounting entry ends up as excess reserves at the Fed.
Thanks, Bruce, for the information on the Trustees’ Report. I’ve been eagerly awaiting its release, but it looks like I’ll have to wait a while longer.
I have no argument with you about interest rates (or about your post). I agree that debt service is not a problem and I think it is unlikely to be a problem in the near future (and I don’t think anyone can predict the far future). One reason I don’t expect interest rates to rise is because the government is not spending enough, and it doesn’t look like that will change anytime soon. I think the way SSA does its projections is understandable, but I wouldn’t be at all surprised if the economy does worse than projected for some time to come. Personally, I’m particularly interested in projections of the Average Wage Index (AWI), since this is important in determining benefits. This is projected to increase by at least 5% per year from 2015-2018, and that seems optimistic to me.
Mike I fully agree on spending. And in fact this post is a small effort to debunk the crowding out argument that Austerians use to tell us “That is why you can’t have nice things” (like critical infrastructure and the wage jobs that would go with them).
I suspect that few people know that the ratio of Notes (1-10 Years) to Bonds (20 & 30 years) is so high as it is (>5:1) and that even less have thought about the effect of QE on Long Bonds effectively outstanding. Mainly because I have been asking for some time now before concluding I needed to have a little conversation with Mr. Google.
I would also note that even a tapering of QE still means tens of billions of dollars of U.S. Securities or guaranteed loans will be bought before it is totally wound down. And on a different point see no reason why the Fed would ever need to sell down their portfolio instead of effectively retiring those portions of the issues. Then again I am asking questions here in hopes that people like you and McOsker will chime in (BK not so much).
And Matt McO. An upcoming post will be on the actual tracking of debt service. Although, as you point out, interest ‘paid’ to the Fed and then ‘rebated’ is really just destroyed it does seem to count as an Outlay for Budget calculations. I am going to try to track down some hard numbers but it would seem that even our rapidly shrinking Deficit is somewhat overstated because of this. And we are not talking chump change, accepting BKs numbers for the minute a 3% average interest earnings on a $2.2 trillion portfolio we are talking some $66 billion in scored Outlays hat wouldn’t in real world terms be that at all.
But that will take some extended conversations with Google and the Treasury website to nail down.
Mike B
i think the projections about AWI are nominal. about 3% is inflation and about 1.1% is real growth in wages.
Webb – so many questions!
The Fed electronically creates money and transmits it to the primary dealers (banks). The banks immediately transfer it back to the Fed in the form of a deposit. The banks earn a risk free 0.25% return for this. This is called Interest On Excess Reserves. If your interested, Google “IOER” – there are hundreds of references to this.
The following is a link to a recent Fed Balance Sheet. Got to page 5. You will see assets and liabilities. Under liabilities is Deposits of $2.6T. This is how QE is funded. A liability is the same as a loan for accounting purposes. So the Fed borrows (or accepts deposits – same thing).
http://www.federalreserve.gov/monetarypolicy/files/quarterly_balance_sheet_developments_report_201403.pdf
Now, why is it done like this? If there were no IOER (a floor rate of 0.25%) then interest rates would fall below zero. That would ‘break the buck’ of most money market funds. When QE started in 2009 the Fed did not want to create another source of panic. So it set a minimum interest rate.
The round trip through the banks, and the IOER have been controversial. Many economists have said the Fed should have set IOER at a negative rate. This would have forced the banks to find more productive uses for the money deposited by the Fed. If, for example, IOER was set at -2.0%, the banks would be lending, but Money Funds would be trading at 97% of par. That would have scared a bunch of folks.
Now, “Can the Fed lose money?” Yes, of course it can. Just google that question and you will find many articles on this. The following is from (2013) The Economist (note that the assumptions back then are pretty much where we are today):
In a recent paper five Fed economists calculated that if the Fed buys $1 trillion of bonds this year and starts tightening in 2014, then the Fed’s profit will turn to loss by 2017. Cumulative losses could eventually reach $40 billion, from higher interest expenses and realised losses on MBS sales (the economists assume the Fed will hold its Treasuries to maturity). If interest rates rise more sharply than expected, losses could peak at $125 billion, and the Fed would pay no profit for six years.
Link to full story:
http://www.economist.com/news/finance-and-economics/21570753-what-happens-when-fed-starts-losing-money-other-side-qe
What could cause the Fed to lose money in any given year? This happens when Expenses > than income. More specifically if IOER is greater than (1) Coupon interest income LESS (2) Premium Amortization.
Say the Fed has average Coupon interest of 4% and the amortized premium is -1%. Net Income is 3%. If IOER is 4% then, in that year, the Fed has a loss of 1%. That would be on $2.6T (minimum) So an implied 26B loss.
Is this a possible outcome? That the Fed suffers two/three/more years of annual losses of $100B? I would say no, that would be an unlikely outcome. But it is not out of the question.
IOER = Federal Funds. (Just trust me that this is true). The following link is to FRED, it charts the long-term rate of Federal Funds. If IOER were to rise to the average rate of FF, then the Fed would be underwater:
http://research.stlouisfed.org/fred2/series/FEDFUNDS
Again, I don’t see this happening, but even if if it did, it would not matter. An accounting trick prevents an annual loss from impairing the Fed’s balance sheet. If there were to be an annual loss, a Differed Asset is created equal to the loss. So the loss is realized, but it does not flow through the balance sheet (BS).
The accounting logic is that there is ample evidence that the Fed has been profitable in the past and will be be profitable again in future years. A simplified example:
Year one income from operations = -$10b
Year one payment to Treasury = 0
Year one BS = -$10 (loss),+ $10B New Deferred Asset – 0 Treasury = No Change
Year two income from operations =+20B
Year two payment to Treasury = $10B
Year two BS= +20B (income) -10B Reversal of Deferred Asset -10B Treasury = No Change
So in the end, Treasury (AKA the tax payers) suffer the loss as the deferred asset is unwound. In theory, the Fed could go on with annual losses for a few years, the result would be that Treasury gets nothing until the loss is worked off.
This accounting treatment is identical to a Deferred Tax Asset for a public company. GE has used boodles of them. They suffered a big loss in 2010, the loss hit the income statement but did not hit the balance sheet. Since then GE has worked off all of those Deferred Assets by paying Uncle Sam less money.
Me? I think the annual remittance to the Treasury will fall starting in 2015. They might go as low at +$20B in 2017/18. They will not hit zero.
bk
PS
IOER is an interesting tool of monetary policy. This is from the NY Fed Economics folks, a discussion and a few more links.
http://libertystreeteconomics.newyorkfed.org/2012/08/interest-on-excess-reserves-and-cash-parked-at-the-fed.html#.U3U3Ei-d62w
-Janet Yellen has said she anticipates raising interest rates in 2015.
correction
mike b
your 5% appears to be the projection for AWI for about the next five years as wages rise from a low base during the “recovery.”
in the long term the Trustees project about a 3.7% increase: 1.1% real and 2.6% inflation.
Bruce agreed. The interest remitted to the treasury does offset out lays thus reducing the deficit, which I term as “money” destroyed. And you are very right that interest is not chump change, and exactly why some feel QE is deflationary pulling interest income out of the economy.
coberly – Right, the AWI numbers assume that as the economy recovers, wages will rise faster than their long-term average to make up for the slow growth of the last few years (even a drop in one year). I hope they’re right, but I don’t see it happening with demand so low and so many people still unemployed or underemployed. Their projection is for wages to be almost 30% higher in 2018 than in 2012 (the latest year available). If it’s only 20% higher (as it would be with current growth rates), then that’s 10% lower benefits for those who turn 60 in 2018.
Krasting I just finished a year long sequence of College Accounting and am scratching my head at your assertion that “a liability is the same as a loan for accounting purposes”. That is backwards, while a loan might well be credited to Notes Payable, indeed a liability account, it would equally be debited to Cash which is an Asset Account. But that doesn’t mean anything that ends up in a liability account is the same as a Note Payable, you are confusing a class member with a class.
Similarly I never said the Fed couldn’t lose money, only that they couldn’t lose money on Treasuries held to maturity. Which you counter by citing an article that suggests that a trillion dollars of bond purchases which INCLUDE MBS’s could lead to losses on MBS’s. With an immediate parenthetical acknowledging that the assumption is that the Fed will hold Treasuries to maturity.
Which was my point. Your introduction of possible losses from MBS holdings reducing (but as you admit probably not eliminating) Fed rebates to Treasury has as far as I can see jack shit to do with this post that focuses on Treasuries. And certainly doesn’t excuse your egregious error that the Fed ‘borrows’ to purchase Treasuries.
But maybe our Commenters can make mor sense of this. Which is why I released it from Moderation where for whatever reason it got stuck (in case you were wondering about its delayed display)
Webb- If you know about accounting then surely you know what a Liability is on a a Balance Sheet. It is some form of IOU.
The Fed PAYS 25BP on the 2.6T. The Fed pays the banks for these deposits % of ~$6B a year. When something is a liability on the BS and interest is paid on that liability then it is a liability, and a liability is an IOU.
Write to the Fed, they are very friendly. They will help you out. The Fed borrows the reserves. That’s just the way it works.
Krasting Income Tax Payable is a standard liability account. As is Wages Payable and of course for most companies maybe the biggest liability account, that of Accounts Payable. And at some level all those are from the point of the firm ‘IOUs’ but you seem to be claiming that they are ‘loans’ from Uncle Sam, employees, and vendors respectively. This is absurd: that A is a C and B is a C doesn’t mean that A is a B. You are confusing transitivity for identity.
Bill Clinton was ridiculed for saying “It all depends on what the meaning of “Is” is”. But grammarians and logicians were not laughing. Because it really does.
And what the hell does interest on excess reserves have to do with holdings of the Feds SOMA to start with? Doesn’t the former accrue to banks that are not involved in selling Treasuries to the Fed? That the payment for such sales comes in the form of crediting reserves of the sellers doesn’t seem to establish an identity.
More explanations and less advice about ‘helpful links’ please.
Mike B
it seems to me that the benefits would not be 10% lower, but 20% higher.
that might be 10% lower than you “expected” or wanted. but it is still higher …. that it is represents an ability to pay out more than was taken in. this “more” is effectively “interest on your payroll tax payment.” This interest could be less, even, than inflation. But in the economy you are envisioning it is going to be hard for any investment to return better than inflation.
in any case, the SS beneficiary is going to be better off than if he didn’t have SS.
I can’t promise you a rose garden, but i think i can almost guarantee you will be better off with Social Security than without it.
I’m not trying to be cure about the rose garden. Too many people today think that god in his heaven and milton friedman… wherever he is… guarantees that life is always going to be better. it might not be. but that government insured savings account (SS) is the still going to be the best way ordinary workers will have to cope with whatever comes.
http://www.newyorkfed.org/markets/longertermtreas_faq.html
http://www.newyorkfed.org/markets/primarydealers.html
“The Fed electronically creates money and transmits it to the primary dealers (banks). The banks immediately transfer it back to the Fed in the form of a deposit. The banks earn a risk free 0.25% return for this. This is called Interest On Excess Reserves. If your interested, Google “IOER” – there are hundreds of references to this.”
20 second of “writing” to the “helpful” people at the Fed revealed that you identification of “primary dealers” with “banks” was bogus. Not all of the former are the latter and somehow I doubt that all the securities brokers who qualify are simply funded by the Fed and paid 25 basis pts on top of that. That is once again you seem to have fallen into an ABC identify/transitivity trap. Or maybe think I am too simple to figure out your ABC Shell Game as you shift the Pea back and forth.
“If you can’t dazzle them with brilliance then baffle them with bullshit”. The operating principle of higher level trolls. While it may be true in the bigger picture that I Fell Off the Turnip Truck on my entry to Big City Finance that fall didn’t happen yesterday.. And I can spot the Hustlers and Grifters.
Mike B
consider this (some simplifying assumptions i made you are free to challenge if you think it makes a material difference):
suppose the average wage is 1000 dollars per week today. and the part of the tax that goes to retirement benefits is 10%. if there are 500 retirees for every thousand workers, that would lead to a benefit of 200 dollars per week per beneficiary. and the worker would take home 900 dollars per week.
now if you expect wages to go up 30% this would change to 1300 dollars per week for the worker and 260 dollars per week for the retiree, with the worker taking home $1170. (remember 2 workers for each beneficiary).
but if wages only go up 20% then the worker makes 1200 per week and the retiree gets 240 per week, leaving the worker with 1080 to take home.
but because of increasing prices the 240 per week for the retiree represents a serious hardship.
since the worker knows that he will one day be a retiree he can tax himself an extra 1%… that is a total of 11%. this will provide the retiree with $264 per week, and leave the worker with $1068.
The question for workers at that time is can they live on “only” $1068 per week while working with less hardship than they would have trying to live on $240 per week when they retire.
I think the answer is likely to be yes. Obviously at some point the answer might become… “errrr maybe not.” But we don’t need to make that decision in advance, for all time, based on projections that no one can reasonably project. and i hope when the time comes that we face such a decision we have all outgrown the stage of life where a shiny new pickup means more to us now than groceries fifty years from now.
please note your “10% less” turns out to be 19% more, even if you raise your tax so granny doesn’t starve. if for no better reason than to protect yourself for when you become the “granny.”
coberly – I basically agree with you. I was not criticizing Social Security, which I think it a great program which should be strengthened (or at least kept the same), not cut. My criticism is of the federal government, for not doing nearly enough to end the current slump. This means that millions of people are needlessly unemployed, and wages are stagnant. I only mentioned the effect on Social Security benefits because we were discussing the Trustees’ Report projections, and the AWI series is one I’m very interested in.
Webb – You’re impossible! When I write something without a reference, you accuse me of lying. When I go out of my way to document what I say with many links you complain there are too many links.
Anyway, more links
This is from FRB. On this page you can look at the Balance Sheet of the Fed for years 2009 -2012.
http://www.federalreserve.gov/monetarypolicy/clbs_2009.htm
This link is for the years after 2012
http://www.federalreserve.gov/monetarypolicy/quarterly-balance-sheet-developments-report.htm
If you look at a few of these reports you will see the growth of the Fed’s BS over time. This is QE. Now look at the liabilities side of the BS. You will see that Deposits from banks increases lock step with QE. In May of 2007 the Deposit number was ~$25B, It has grown to $2.6T today.
Now if you are still skeptical about how this works consider looking at one of the big banks 10Qs. You will see the growth of the assets with the Fed in the form of deposits. At the end of the day Assets = Liabilities.
For a different discussion on how the Fed might lose money, the use of IOER, the creation of Deferred Assets and the consequences of rising interest rates, have a look at this FRB paper:
http://www.federalreserve.gov/pubs/feds/2013/201301/revision/201301pap.pdf
You said:
“I never said the Fed couldn’t lose money, only that they couldn’t lose money on Treasuries held to maturity.”
While I consider it highly unlikely, it is possible for the Fed to lose money on Treasury securities even if they are held to maturity. The risk is that that the cost of borrowing the money to fund the QE purchases exceeds the interest income on the bonds.
A link to the most recent Fed forecasts on ST interest rates. Look at Figure 2 of the report. This is the Fed’s members ‘guess’ on where ST % rates will be. The Fed believes (as does SS and CBO) that % rates will be normalized over the next few years. When this happens the IOER (or the Federal Funds rate) will rise. When that happens the deposit rate on those $2,6T of deposits will increase. This will cause the Fed to have higher expense and lower net income (and conceivable a loss in a given year).
http://www.federalreserve.gov/monetarypolicy/files/fomcprojtabl20140319.pdf
Finally, a link to a discussion re a ‘famous’ 60 Minutes story with Bernanke. In this Bernanke explains that the Fed does not print money. QE is a swap of short term liabilities for long term assets. The Fed is borrowing short and lending long with QE. In the process it is assuming what is called Duration Risk.
http://dailybail.com/home/bernanke-on-60-minutes-were-not-printing-money.html
Bkrasting,
1) The Fed has unlimited capacity to create reserves, they just push a button. There is no technical limit – Fed losses on their balance sheet are irrelevant. Don’t make it more complicated than it is.
2) When the Fed buys a bond they swap reserves for that bond, on net nothing in the system changes as the bond goes away to a parking lot – out of circulation. The composition of the banking system’s portfolio changes.
3) Banks don’t lend reserves. The total quantity of reserves in the system won’t change, and the only way to drain reserves is:
a) The Fed sells bonds from its balance sheet
b) The federal government runs a surplus
c) Bank lending causes excess reserves to shift to required (but banks don’t lend the reserves this is mechanical accounting)
d) People withdraw lots of cash at the ATM
Good piece on banks and resevres:
http://www.standardandpoors.com/spf/upload/Ratings_US/Repeat_After_Me_8_14_13.pdf
From the piece:
“. A key distinction to bear in mind (hinted at in the last previous paragraph) is between individual
banks and banks in aggregate. Neither individual banks nor banks as a whole can “lend out” reserves, but individual
banks can and do offload their reserves (particularly excess reserves) by lending them to other banks or by buying
assets; but the banks in aggregate cannot do this–in such cases, the reserves that leave one bank’s balance sheet just
pop up on another, remaining on the central bank’s balance sheet all the while.”
Mike B
thanks. i was pretty sure you knew, but not at all sure other readers would know.
i agree that we should be doing more, or different, about the unemployment and poor wage growth in this country. just trying to remind people that expected bad times is not the smartest time to cut your bad-times insurance.
“Webb – You’re impossible! When I write something without a reference, you accuse me of lying. When I go out of my way to document what I say with many links you complain there are too many links.”
Krasting I never accused you of using “too many links”. Mispresenting the content of the links you did supply or using links as a substitute for actual argumentation perhaps but I defy you to cite me complaining about volume as such.
Burning strawmen much?
In any event I plan to continue this discussion via a New Post so don’t count on me responding to this now dead thread.