Saturday, February 26, 2011

Can a central bank make a loss?

Yves Smith quotes from a Gillian Tett article on central bank losses:
Now as some readers have noticed, the Fed has also institutionalized an accounting dodge so that it will not have to admit to balance sheet losses. Commentators appear to have missed that this finesse does nothing to solve the underlying reality. The Fed can “print” its way out of balance sheet losses only to the extent that it is not constrained by inflation. If it does run into inflationary constraints, it would need an explicit bailout, irrespective of the accounting treatment.
This is nonsense.

The US Federal Reserve bank has a negative equity position. It must, as a matter of accounting, given the operational constraints (currently) governing the Treasury. Therefore, if the CB suffers a loss, then this negative equity position simply becomes more... negative? Do we care?

At a normal bank in normal times, negative equity need not stop a bank from functioning so long as its reserve account still works -- either because the Fed is funding it directly via the discount window or because it can find counterparties in the overnight inter-bank market. Regulators can always step in a shut the bank down (indeed, by law they must, although they ignore this law when they choose) but otherwise the lights can stay on, employees get paid, loans get made, etc.

At the central bank this is true, but even moreso. There is no risk that the central bank will be shut out of the reserve system because it is the serve system. It holds all reserves as liabilities. And it lives in a permanent state of negative equity, so the amount of negative equity does not matter, it simply reflects the amount of positive equity that exists in the non-central bank sector. If there is an inflationary constraint, it's because the Fed paid $1 for assets worth 20 cents, and therefore engaged if excessive fiscal policy.

10 Comments:

Blogger Ralph Musgrave said...

I think Yves Smith is right. She is not saying there is a limit to the amount of negative equity at a central bank. She is saying the bigger that negative equity is, the more cash (or other assets) there are in the hands of the private sector. In your own words “…the amount of negative equity does not matter, it simply reflects the amount of positive equity that exists in the non-central bank sector”.

Now if inflation looms, the private sector cannot be given more cash or similar financial assets. That will just exacerbate inflation. Therefor the government / central bank machine has to find another way to curtail inflation. One way is to balance the central bank’s money printing by raising taxes and “unprinting” the money collected, that is, extinguishing the money. But only Congress and the Treasury can do this. So in effect, this amounts to the central bank going cap in hand to the Treasury.

11:09 PM  
Anonymous Anonymous said...

One way is to balance the central bank’s money printing by raising taxes and “unprinting” the money collected, that is, extinguishing the money. But only Congress and the Treasury can do this...

The Fed has all the taxing authority it needs. If it wished, it could control inflation by extinguishing money instead of raising interest rates. Remember this, we kicked George III out of America with the slogan "no taxation without representation". However, our founding fathers were silent on the issue of "user fees without representation".

"The Monetary Control Act of 1980 (MCA) requires the Reserve Banks to recover fully, over the long run, the costs... associated with the provision of most financial services they provide. Currently, Fedwire transaction fees are charged to both the originating institution (debit side) and receiving institution (credit side). Fees are based on the volume of transfers, a fixed monthly participation fee, and a surcharge for originating or receiving offline transfers. In 2009, the volume-based transaction fees range from 8 cents to 26 cents per transfer, per institution. In addition to volume-based transaction fees, a fixed monthly fee of $60 is charged to each participant with activity in that month and a surcharge of $40 is charged to initiate or receive an offline transfer. Electronic access fees are assessed separately.

"In 2008, Fedwire processed an average daily volume of approximately 521,000 payments, with an average daily value of approximately $2.7 trillion. The distribution of these payments is highly skewed, with a median value of approximately $24,000 and an average value of approximately $5.8 million. Roughly 11 percent of Fedwire payments are for more than $1 million..."


Gosh, so if the BOG switched Fedwire from a "fixed fee per volume" to a "percentage of value" fee basis, the BOG could adjust the percentage as needed to siphon off as much bank revenue as they wished in "user fees". Congress would never ever complain because the alternative to Fed-imposed user fees would be Congress-imposed taxes and who wants to vote for new taxes if they don't have to? Now you might say, well banks will just switch from Fedwire to a private competitor like CHIPS... except they can't.

"Because the ultimate CHIPS settlements are provided by Fedwire, CHIPS is a customer, as well as a competitor, of Fedwire. The vast majority of CHIPS members are also Fedwire participants, and the daily value of CHIPS transfers is about 80 percent of Fedwire’s non-securities transfers."
http://www.ny.frb.org/aboutthefed/fedpoint/fed36.html

3:46 AM  
Blogger JKH said...

Winterspeak,

Agree at the highest level of generality, but:

Everything on this earth related to the subject of banking and the financial system can be mapped into existing institutional accounting arrangements, which in turn may be compared to alternative accounting arrangements. Since the financial system is one giant spreadsheet, there is no ultimate "reality", base case or benchmark in this regard. There are just choices about representing the world as we would like it to operate. Accounting arrangements reflect measurement of results under a prescribed operating system. This includes the question of institutional arrangements for governments and their central banks.

Central banks do not run negative equity under existing arrangements. They could do so if their governments allowed it, but they generally don’t. They do not live in “a permanent state of negative equity”.

(This is separate from the tendency of left wing bloggers to argue and declare that most banks, central and commercial, are insolvent due to accounting lies. But this is driven by ideological obsession rather than persuasive financial analysis.)

Governments do run negative equity under existing accounting arrangements. They don’t show a balance sheet as such, but a cumulative operating deficit equates to negative equity.

An alternative combined entity would obviously run negative equity. But that’s a different institutional accounting world, reflecting a different operating system. It’s the world that MMT really desires.

A government with its own sovereign central bank ultimately can do what it wants – including allowing the distribution of negative equity to slip into the accounts of its central bank (instead of recapitalizing it), or simply combine the two entities institutionally.

The latest Fed accounting tweak that Tett refers to remains within the general category of existing, separate, institutional accounting arrangements. It allows short term losses to be funded from cumulative future profits. Anybody who understands the monetary system shouldn’t get too riled about it. Fed profits have been massive over the past two years, and given that big handover of positive equity to the government, there’s no sensible reason why it shouldn't get a proportionate accounting pass in terms of its ability to absorb any future short term losses that might result from some of their past risk taking – risk taking that will end up being proven by history, beyond the shadow of a doubt, to have been very modest indeed in proportion to what it has accomplished. The Fed has the balance sheet to handle any such scenario very easily, although the blogosphere generally speaking seems immune to the idea of examining such balance sheet simulations at an analytical level.

So I agree generally with your conclusion, but would define things and present it a little differently - as usual.

I've argued for a long time that MMT would be well served just to recognize the government’s negative equity position in a direct fashion, whether considered separately or combined with its central bank, instead of avoiding it as an analytical reality, or being fearful that this dark little secret may leak out to the public. Dealing with it directly as it does with deficits would make the overall MMT accounting paradigm story much more powerful in my view. There’s no need to fear misinterpretation if your story is the right one.

4:56 AM  
Blogger winterspeak said...

JKH: Blogger seemed to have eaten this, so I'm re-posting.

JKH has left a new comment on your post "Can a central bank make a loss?":

Winterspeak,

Agree at the highest level of generality, but:

Everything on this earth related to the subject of banking and the financial system can be mapped into existing institutional accounting arrangements, which in turn may be compared to alternative accounting arrangements. Since the financial system is one giant spreadsheet, there is no ultimate "reality", base case or benchmark in this regard. There are just choices about representing the world as we would like it to operate. Accounting arrangements reflect measurement of results under a prescribed operating system. This includes the question of institutional arrangements for governments and their central banks.

Central banks do not run negative equity under existing arrangements. They could do so if their governments allowed it, but they generally don’t. They do not live in “a permanent state of negative equity”.

(This is separate from the tendency of left wing bloggers to argue and declare that most banks, central and commercial, are insolvent due to accounting lies. But this is driven by ideological obsession rather than persuasive financial analysis.)

Governments do run negative equity under existing accounting arrangements. They don’t show a balance sheet as such, but a cumulative operating deficit equates to negative equity.

An alternative combined entity would obviously run negative equity. But that’s a different institutional accounting world, reflecting a different operating system. It’s the world that MMT really desires.

A government with its own sovereign central bank ultimately can do what it wants – including allowing the distribution of negative equity to slip into the accounts of its central bank (instead of recapitalizing it), or simply combine the two entities institutionally.

The latest Fed accounting tweak that Tett refers to remains within the general category of existing, separate, institutional accounting arrangements. It allows short term losses to be funded from cumulative future profits. Anybody who understands the monetary system shouldn’t get too riled about it. Fed profits have been massive over the past two years, and given that big handover of positive equity to the government, there’s no sensible reason why it shouldn't get a proportionate accounting pass in terms of its ability to absorb any future short term losses that might result from some of their past risk taking – risk taking that will end up being proven by history, beyond the shadow of a doubt, to have been very modest indeed in proportion to what it has accomplished. The Fed has the balance sheet to handle any such scenario very easily, although the blogosphere generally speaking seems immune to the idea of examining such balance sheet simulations at an analytical level.

So I agree generally with your conclusion, but would define things and present it a little differently - as usual.

I've argued for a long time that MMT would be well served just to recognize the government’s negative equity position in a direct fashion, whether considered separately or combined with its central bank, instead of avoiding it as an analytical reality, or being fearful that this dark little secret may leak out to the public. Dealing with it directly as it does with deficits would make the overall MMT accounting paradigm story much more powerful in my view. There’s no need to fear misinterpretation if your story is the right one.

2:08 PM  
Blogger winterspeak said...

Ralph: Yves must have come to MMT too late in life. She knows about it, but can never quite get "in paradigm". When the Fed buys an asset, the money printing is done. Whether it paid 100 cents to the dollar to 20 cents to the dollar depends on whether there was a fiscal transfer or not.

Yves belief that a loss would need to be made up by money printing (which might create inflation through some mystery mechanism) is the usual muddled monetarism writ small.

JKH: Am I still getting the accounting wrong here? If the private sector has positive net financial assets (equity) then, somewhere there must be an entity with negative net financial assets (equity). If that isn't the Fed, where is it?

2:11 PM  
Blogger JKH said...

Winterspeak,

The Fed as an institution normally has positive equity equal to its capital position. However, it has a zero NFA position as an institution because it holds financial assets offset by liabilities and capital that are held as financial assets by others.

A government running a cumulative deficit has negative equity equal to its outstanding debt. As an institution, it has no financial assets (to speak of) offsetting that debt.

NORMALLY, the Fed’s balance sheet consists MOSTLY of assets in the form of treasuries, liabilities in the form of currency and a small amount of reserves, and capital. The Fed's Treasury bond assets = currency + reserves + capital, for the most part. The non government sector holds the Fed’s currency and reserve liabilities and capital position as part of its direct NFA position. But the Fed’s own position from an institutional perspective is neutral in terms of NFA – it has financial assets offset by financial liabilities and capital. From an institutional perspective, it is the government liability in the form of treasuries held by the Fed that is the ultimate offset to this particular component (currency, reserves, and capital) of non government NFA. Putting it all together, the Fed’s treasury holdings plus treasuries held directly by the public constitute total non government NFA as well as the total negative equity position of the government.

If the Fed retained its normal annual profits, its positive equity would grow, like that of a commercial bank. It doesn’t. It normally dividends some of its profits and remits the rest to Treasury. That annual remittance is in effect a reduction in the government’s budget deficit. It is a transfer of positive equity from the Fed to the government. However, this is not really an issue for NFA accounting - if the Fed didn’t transfer those profits, it would hold more treasuries as assets, and non government would hold more Fed capital as its direct NFA instead of the treasuries that it effectively would have sold to the Fed.

If the Fed runs unusual losses, its normally positive equity will become negative unless it is recapitalized. Again, this would have no significant NFA accounting effect, since the decline in the value of Fed capital held by non government would be offset by an increase in Treasury issuance to finance the resulting difference in the budget deficit.

However, the Fed has recently been permitted to implement an accounting adjustment (the one that Tett refers to) that will change the impact of potential Fed losses on its capital position. It is somewhat technical, but the effect will be for the Fed to be allowed to offset the negative equity impact of losses with anticipated future profits. It is in effect a bet that any losses will be temporary, and that future profits will eventually offset those losses, provided those profits are retained instead of being remitted to Treasury. This is a reasonable bet, given the definitively finite nature of such losses, and the reliability of future seigniorage to eventually offset those losses. The Fed doesn’t have to “print money” currently to do this; it merely has to operate long enough to return to normal institutional profitability, which it most certainly will do at some point as the result of the attrition/maturity of the assets that are the source of the losses and the persistence of seigniorage.

Given the already complex nature of the above description, I’m not going to get into the current ABNORMAL balance sheet of the Fed. It’s not a critical complication for the point you are making in your post.

4:16 PM  
Blogger Ramanan said...

JKH,

"dark little secret" ... :-).

Btw PKEist of the non-MMT variety openly write about the government's negative financial net worth position. And keep the central bank and the government's account separate but sometimes combine them.

8:07 AM  
Blogger NKlein1553 said...

JKH,

Perhaps this is a stupid question, but can you give an example of what "Fed capital," consists of?

6:14 PM  
Blogger JKH said...

NKlein1553,

First reference I found:

http://www.federalreserve.gov/releases/h41/current/h41.htm

scroll down to # 8 (the second # 8 under "continued") at the very bottom of the liability list

you'll seen about $ 53 billion in Fed capital

various rules govern the Fed's capital position, which you can probably discover if you search around the Fed's website enough

10:45 AM  
Blogger Matt Franko said...

Winterspeak,

"Yves must have come to MMT too late in life. She knows about it, but can never quite get "in paradigm".

Please consider that she is another that does not possess the same Mathematical Maturity as you do, or JKH does, or Ramanan, and many others 'in paradigm'.

I do not mean this in a derogatory way. We all have our gifts, and I look at her's (and also for instance Barry Ritholtz at Big Picture blog) as those of a 'lawyer' rather than those of a person who has high quantitative skills, she seems to excel at breaking down the legal aspects, but math is not her strongest skillset. She is 'rules based', the concept of 'negative equity' is 'breaking the rules' for her, to you it is just a negative number on a giant spreadsheet, no big deal, to her it is 'insolvency'.

Folks like her just cannot 'see' MMT. At least when it is explained to them verbally in emails and blogs. You would have to get her in front of a chalkboard with diagrams and ledgers, etc... help her visualize the flows, and then she would perhaps catch on, probably would. There were reasons for having chalkboards in math class ;)

Resp,

6:28 PM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home