Paging Bill Woolsey on the Fed’s Accounting Skullduggery
Like the Spaniard in The Princess Bride knew he needed the Man in Black, when I try to parse the Fed’s notorious January 6 accounting change, I know I need the economist who wrote this article (“Who Owns the Fed?”).
So here’s the Fed’s announcement, with my placeholders and bolding added:
The Board’s H.4.1 statistical release, “Factors Affecting Reserve Balances of Depository Institutions and Condition Statement of Federal Reserve Banks,” has been modified to reflect an accounting policy change that will result in a more transparent presentation of each Federal Reserve Bank’s capital accounts and distribution of residual earnings to the U.S. Treasury. [#1] Although the accounting policy change does not affect the amount of residual earnings that the Federal Reserve Banks distribute to the U.S. Treasury, it may affect the timing of the distributions. Consistent with long-standing policy of the Board of Governors, the residual earnings of each Federal Reserve Bank, after providing for the costs of operations, payment of dividends, and the amount necessary to equate surplus with capital paid-in, are distributed weekly to the U.S. Treasury. The distribution of residual earnings to the U.S. Treasury is made in accordance with the [#2] Board of Governor’s authority to levy an interest charge on the Federal Reserve Banks based on the amount of each Federal Reserve Bank’s outstanding Federal Reserve notes.
Effective January 1, 2011, as a result of the accounting policy change, [#3] on a daily basis each Federal Reserve Bank will adjust the balance in its surplus account to equate surplus with capital paid-in and, in addition, will adjust its liability for the distribution of residual earnings to the U.S. Treasury. [#4] Previously these adjustments were made only at year-end. Adjusting the surplus account balance and the liability for the distribution of residual earnings to the U.S. Treasury is consistent with the existing requirement for daily accrual of many other items that appear in the Board’s H.4.1 statistical release. The liability for the distribution of residual earnings to the U.S. Treasury will be reported as “Interest on Federal Reserve notes due to U.S. Treasury” on table 10. [#5] Previously, the amount necessary to equate surplus with capital paid-in and the amount of the liability for the distribution of residual earnings to the U.S. Treasury were included in “Other capital accounts” in table 9 and in “Other capital” in table 10.
For those who don’t see what the big deal is in the above–and you’re not alone, it took the financial sector a good two weeks to realize it–see Robert Wenzel’s analysis.
So my questions for Woolsey, who (by my back-of-the-envelope calculations) has written a novella in his arguments with Captain Freedom on this blog, and so should jump at the chance to share his wisdom with a wider audience:
#1: Here, are they referring to the fact that if the Fed suffers a capital loss, it will halt the flow of remitted earnings to the Treasury, until that loss is whittled away? If so, are they being honest when they say it won’t affect the total flow of earnings, but rather just the timing? Before this rule change, how would the Fed ever have recapitalized itself after taking a hit? Would it have happened more gradually, so that the flow of remitted earnings to the Treasury would have been reduced more modestly, but for a longer period, than will be the case now?
#2: What the heck are they talking about? I don’t think I ever knew that the Fed had to pay interest on Federal Reserve Notes. Is that just some internal costing mechanism to allocate the burden of remitted earnings among the various Fed banks? And what is the interest rate?
#3: Can you say this in English?
#4: They are making it sound as if they are simply doing something daily, which previously would have been done yearly. Is that correct? And if so, then why is everybody acting as if the Fed is doing something crooked? Was it standard practice to skirt insolvency on an annual basis before, and now it’s done weekly?
#5: This is the money line, right? The Fed is making it sound like it’s just putting its “surplus” somewhere besides its “capital account,” when this also opens the door to them putting a negative surplus–aka a loss–somewhere besides its capital account. If I’ve read that correctly, then can you relate it to a standard balance sheet? A lot of people are saying stuff like, “The Fed is moving its losses off the left-hand side of its balance sheet.” But that’s not how I’m reading that. I think they will still reduce the value of their assets, but then on the Liabilities and Capital side, they won’t reduce Capital correspondingly. In order to make the right hand side fall as much as the Assets fell, they will instead put in a negative number in the Liabilities. This pulls down the right hand side of the balance sheet, while still maintaining Assets > Liabilities.
Is this about right? And of course people other than Bill Woolsey can chime in.
Paging Jeffrey Rogers Hummell! Hummell is very good on figuing out the Fed balance sheet.
The “surplus” is required retained earnings. The Fed has to have retained earnings equal to paid in capital, which is the amount of stock that banks are required to hold. The paid in capital is the $100 shares banks must buy equal to 3% of their net worth.
It looks like the Fed will be calculating that requirement for surplus (retained earnings) every day instead of at year’s end.
If the Fed suffered a loss, then it would recapitalize itself by building up retained earnings. it would pay out less to the Treasury. (The Fed can also require banks to buy newly issued stock up to another 3% of their net worth.)
This didn’t say that the Treasury will get the same amount regardless of what the Fed earns. It won’t. The Treasury is the true residual claimant. But they are saying this accounting change doesn’t effect what the Treasury gets. If the Fed makes less, or losses money, then the Treasury will get less from the Fed.
Losses will come out of these retained earnings first. And then paid in capital. I am not really sure about the banks being compelled to buy more stock, but that is legally the next place to cover losses. After that, the Fed is bankrupt.
It has always been true that if the Federal Reserve goes bankrupt, the Treasury is liable for Federal Reserve notes. The Federal Reserve has to post collateral for them and pays interest on them to the Treasury. Before the end of the gold standard, the Treasury would need to come up with gold if necessary.
After the end of the gold standard, this obligation from the Treasury could be understood as an obligation to issue Treasury currency in place of Federal Reserve notes, I guess. I don’t think the Treasury is obligated for reserve balances.
The “interest” that the Fed pays on these Federal Reserve notes is a residual–the profits of the Federal Reserve operations. Dividends, on the other hand, are really like interest payments on paid in capital.
I don’t really understand Wenzel’s post. It seems little related to the accounting changes above.
For example, suppose banks have reduced net worth, and so cash in stock at the Fed. The Fed reduces the amount it must have in surplus (retained earnigns) and increases what it must pay to the Treasury. They start paying the Treasury more now rather that waiting to the end of the year to make this adjustment.
Suppose the banks have additional net worth and so buy new stock in the Fed. The Fed must increase its surplus (retained earnings) and so reduces what it owes to the Treasury and presumably reduces its payments right now rather than at the end of the year.
Suppose the Fed loses money and has less than the required retained earnings. It reduces payments to the Treasury right away until it builds back up to what it is supposed to have. Under the old system, if the required amount had increased during this period, then the Fed would have started paying the Treasury again, and then noticed that it need to increase its retained earnings according to the requirement at year end. And so it would have to stop (or reduce) the payments to the Treasury again.
If, on the other hand, the amount the Fed needed had fallen, then it wouldn’t resume Treasury payments soon enough. It would have to wait until the end of the year.
Maybe it really means that the Fed will always keep the required amount of retained earnings (surplus) and if the amount owed to the treasury turns negative, then the Treasury owes the Fed money. Maybe. I doubt it;
My guess is that the Fed will recognize capital losses or credit losses on some assets only when it has sufficient interest income so that the result is lower earnings rather than losses. I really doubt that these changes mean that if the Fed takes a big loss this leaves its net worth unchanged but generates a loan from the Treasury in the form of negative payments owed to the Treasury.
But maybe.
Actually, losses come out of payments to the treasury first, and then dividends to the member banks. But really, that is just lower profits and not losses. After that, is is actual losses, and it comes out of the retained earnings, then the paid in capital, and then in theory, all the banks having to chip in another 3% of their net worth.
If this change really means that the Fed pays the dividends and keeps its retained earnings, if there isn’t enough income to do it, the Treasury lends to the Fed through negative payments due the Fed, then it is accounting skuldugery. While I don’t think there is anything wrong with the Treasury earning less from the Fed if the Fed makes less money (much less takes a loss,) dividends to the banks, retained earnings, and paid in capital are there to cover losses.
Losses will come out of these retained earnings first. And then paid in capital. I am not really sure about the banks being compelled to buy more stock, but that is legally the next place to cover losses. After that, the Fed is bankrupt.
Bill, could you explain what you mean by “the Fed is bankrupt”? How can an institution that has the legal authority to create new money out of thin air ever go “bankrupt”? If their liabilities are ever greater than their assets, than can’t they just issue additional checks drawn on itself, much like they do whenever they buy government bonds from the primary dealers? Every check they issue, every dollar contained therein, becomes legal tender.
Or is it the case that by “bankrupt”, you only mean that the Fed’s accounting figures can show liabilities greater than assets? If that is what you mean, then all banks in the entire country would be bankrupt, since every bank engages in fractional reserve banking, which generates accounting figures of instantaneous liabilities being greater than assets (which is why bank runs can manifest a bank’s bankrupt status).
Why would an institution that can create money out of thin air perceive a need to change the way it accounts for accounting losses? Why would they perceive a need to reduce the accounting loss they incur on toxic waste, and transfer that loss over as an increase in accounting liability they owe the Treasury?
My suspicion for why the Fed changed its accounting practices is to stave off Treasury bankruptcy. A Treasury that defaults on its debt will result in a rapid rise in interest rates in the US, and that will collapse most of the financial institutions, and a good portion of the real economy that was malinvested due to the record low interest rates set by the Fed for the last few years.
The Fed has been very clear that they are going to seek to hold interest rates at record lows for the indefinite future. An economy that “needs” almost zero interest rates just to stay out of depression is an economy that contains so much malinvestment, that continued monetary stimulus can only affect the economy primarily by in a rise in prices, and eventual monetary breakdown.
From 2008 to QE2, the Fed has kept interest rates at record lows, and since QE2, interest rates have risen somewhat, which was not what the Fed wanted or intended (according to their public statements). Could this accounting change be a way for the Fed to “subsidize” the Treasury’s increasing interest liabilities in particular, and “non-discretionary” liabilities in general?
Is it just a coincidence that the Fed would suddenly declare a new liability to the Treasury of billions if not trillions of dollars, while the Treasury is cash strapped and is in danger of going bankrupt, e.g. if the debt ceiling is not continually raised?
Here is an article by Dr. Murphy on the fed becoming insolvent.
http://mises.org/daily/4869/Can-the-Fed-Become-Insolvent
Though I agree that all the banks are technically insolvent their balance sheets show assets equal to liabilities.
Sorry, I didn’t read everything you wrote.
The Fed can be insolvent because its liabilities are less than its assets.
It can’t really default in the usual sense becaues it isn’t obligated to pay
off its liabilities with anything.
If the Fed is subject to some kind of rule other than one controlling the quantity of its
liabilitiies, then the closest thing it can come to default is an inability to follow the rule
because it has too few assets. In particular, if it needed to contract the quantity of money
to follow the rule, and it sold off all of its assets, and had no more to sell, then it couldn’t
keep to the rule. The real value of money would be lower (the price level higher) than it
would be if the rule had been kept.
Fractional reserve banks can be insolvent, but they aren’t necessarily. (You need to review
your basic concepts.) They can default if they fail to pay off their liabilities. They are obligated to
pay off in currency, or in practice, meet clearing obligations with reserve balances.
Forget banks. Suppose you own a million dollar home, and have borrowed $100 from a friend and
promised to pay him back at your front door in a week. You have no other debts. You forget about
him coming by and don’t have the $100. You default. But you are solvent because your net worth
is $999,900.
If you had that same house, had no other assets, and owed $1,100,000 you would be insolvent. Your assets
are less than your liabilities. If the next payment on your debt is due in a year, you haven’t defaulted (yet.)
The Fed can be insolvent because its liabilities are less than its assets.
How can the Fed become insolvent if its liabilities are less than its assets?
Did you mean to say “greater than its assets”? Or am I missing something?
Fractional reserve banks can be insolvent, but they aren’t necessarily. (You need to review your basic concepts.) They can default if they fail to pay off their liabilities. They are obligated to pay off in currency, or in practice, meet clearing obligations with reserve balances.
In accounting terms, all fractional reserve banks are technically insolvent. Their instantaneous liabilities are always greater than their assets because they grant checking account money out of thin air, which are instant liabilities, through the loans they offer. A fractional reserve bank always has more instantaneous liabilities than they do in assets.
The only reason they don’t go bankrupt in practice is because their instantaneous liabilities in accounting terms rarely become actual liabilities that have to be paid out in cash (since they only keep on hand enough money to satisfy the daily withdrawals of demand deposit holders, who typically do not all withdraw their money at the same time).
They can default in practice if all the owners of those instantaneous liabilities (checking account owners) ask to be paid in cash at the same time.
Forget banks. Suppose you own a million dollar home, and have borrowed $100 from a friend and promised to pay him back at your front door in a week. You have no other debts. You forget about him coming by and don’t have the $100. You default. But you are solvent because your net worth is $999,900.
If you had that same house, had no other assets, and owed $1,100,000 you would be insolvent. Your assets are less than your liabilities. If the next payment on your debt is due in a year, you haven’t defaulted (yet.)
Yes, that argument is used quite often to justify fractional reserve banking, and in denying frb bank insolvency. The problem with that argument is that it fails to take into account the fact that the market prices of the real assets, that supposedly “back” the loans, are themselves dependent on the practice of fractional reserve banking. The reason why the market price of the home is $1,000,000 is precisely because of fractional reserve banking generating an increased aggregate nominal demand, which increases the market prices of all goods, including homes.
Thus the assets “backing” the backed-by-nothing money given out in loans are in fact still backed by nothing. As fractional reserve inflation later turns into bankruptcy and deflation, the market prices of the real assets will fall.
With a great enough inflation (2001-2007), the greater the threat of deflation, and thus the tendency of inflation to accelerate. Each deflationary threat requires ever more inflation to counteract, since the inflation adds to the house of cards.
Back to your example. To make it more accurate and more reflective of the banking system, each bank could have total liabilities of $100 in checking accounts, and $1,000,000 in collateral assets, of which say $10 is in cash. If this one bank represents all banks, and the demand deposit clients all want to withdraw their cash, then the bank cannot just sell its collateral and be sure to fetch enough money to pay off their liabilities. For in a credit induced depression, if many banks scrambled for cash, they would primarily seek to sell their assets. The prices for those assets would plummet. Banks could then go bankrupt in the apparent, obvious sense.
Only if the Fed keeps inflating, to counteract this deflation, can the banks stay afloat. The collateral by itself cannot do this, and so it cannot be the case that the banking system is solvent as long as they write down $1,000,000 as the market value of their collateral for example.
I think Murphy is on the right track here. The bottom line appears to be that the Fed is attempting to use an accounting gimmick to turn a loss into a “negative liability”. This is not as much about banks paying capital into the Fed, as Woolsey states, but about Fed branches recognizing losses. It is about operating losses that simply accumulate as a negative liability/deferred asset ad infinitum. This means a dividend payment made when there is a neg liability de facto simply adds to the existing neg liability in the amount of the dividend because surplus cannot be drawn down, under the new rules. Same goes for any other expense.
This is the key. The accounting change from annual to daly is a bit of a smoke scereen, that won’t change anyting< Actual payments to the Treasury, or the lack thereof. will change things and that is what needs to be focused on in addition to losses above interest payments to the Treasury for any year.
Part of the key to understanding what is going on is to look at the 1922 Opinion that is an attachment to current Fed manual (Note:Bob English dug this up for me). The reason the Fed attaches the opiiniom appears to be a round about way of establishing the ability of the FR Branches to pay any and all expense obligations from surplus (which can never be drawn down) when they operate at a loss.
Anyway, this was clearly written by someone with an advanced degree in obfuscation, amd not desigend to be understood by anyone who does not have handy the original copy of the Asereth ha-D'bharîm.
Bill, just to make sure you are aware of this, it’s not Wenzel out in the wilderness spouting off a crank theory. The standard interpretation–among friends and foes alike–is that this rule change makes it impossible (or at least much more unlikely) for the Fed to go bankrupt. See the CNBC story for example.
So are you saying those people are all wrong? It wouldn’t be the first time, I just want to make sure you realize this is the conventional wisdom.
Just yesterday I told a friend that he ought to check out your bog, at least on Sunday and defend my position against Gene. Go figure, you’re obviously too busy. This is no knock to you. It’s a lament to my timing.
If this negative entries get larger than the Fed’s capital, I’m going to call them bankrupt.
There are negative entries now for some Federal Reserve banks. They are much smaller than the bank’s net worth. And the entries are still positve for the system as a whole.
Personally, I think it is too obvious for this to be some accounting trick.
While it makes sense to call the positive amounts fund due the Treasury.
The negative amounts are shortfalls from required surplus (retained earnings.)
The Fed cannot go bankrupt in any meaningful sense unless US citizens refuse to accept FRNs in transactions.
Can’t happen as long as the government taxes the people in US dollars, which coerces US taxpayers to accept US dollars in their exchanges. Legal tender laws cannot be easily overruled.
Yes, but if enough people refuse to accept them, taxes won’t matter- people will just stop paying them.
They still have to pay taxes in US dollars if they make any gains from trade.
For example, suppose I get rid of my US dollars and buy gold. I just hold onto it for say 1 month. Suppose the paper price of gold doubles. I would then owe the government US dollar taxes on that gain. That means I have to go out and find enough US dollars to pay taxes, which means I once again have to accept them in trade somehow.
Then there are property taxes. Even if I tried to live “off the grid”, I would still owe property taxes on any land I come to own.
you all need to learn the difference between bankrupt and default.
I read the CNBC article.
It is hard to take the quoted financial analysts seriously.
The accounting change doesn’t effect the Fed’s underlying financial condition.
If interest on Federal Reserve notes due to the Treasury is more negative than
capital, it is insolvency.
In practice, if these numbers stay negative and become more negative, it is a matter of
concern. If they are negative some weeks and rapidly return to zero (or turn positive,) then
it really means nothing.
I think it has more to do with changes in the required amount of retained earnings than anything
else.
My guess is that the Fed will account for any losses in a way that doesn’t turn these negative.
Unless, of course, the Fed has to sell off assets at a loss to meet monetary policy objectives.
And then, yes, the negative amounts in this figure will show how much of a bailout the Fed needs from
the Treasury.