This is a Premium Piece of Notes on the Crises Thank you for being a Paid Subscriber
My last piece for Notes On The Crises was a hit: I wrote about the treasury default memo, which I managed to get from the Federal Reserve through FOIA at the end of June. Besides being widely shared on social media, Toby Nagle of the Financial Times Alphaville blog wrote an entire piece on that memo (along with my commentary). In addition to Nagle’s extensive coverage, Matt Levine at Bloomberg, Oddlots at Bloomberg and Marketwatch all linked to the piece. Given all that, I realized that my followup piece should probably just cover the rest of my current FOIA successes. I have a tendency to want a piece to reflect all my thoughts on a subject, and offer up a grand vision of important macro topics. I need regular reminders to… resist that tendency. Newsletters, after all, are dollops of information.
So let’s keep things straightforward: what else is currently in my FOIA vault?
To my mind all of these different stories fit into a much longer history of accounting gimmicks to avoid the debt ceiling. I have, in fact, had a piece on the “70 year” history of these accounting gimmicks in the works for quite some time. As I researched that piece, I recently discovered in Morgenthau’s diary (Treasury Secretary under president Roosevelt) that there was an extensive and comprehensive plan to use accounting gimmicks to avoid the debt ceiling from 1939 to 1940. That plan was cooked up by Federal Reserve Chairman Marriner Eccles, and then enthusiastically supported by Roosevelt. That means it's more like an 85 year history. (But that’s a story for another time…)Today’s topic is the history of the debt ceiling as we know it, which takes us back to the 1970s. The background context of this batch of FOIA discoveries originates in a little noticed episode at the beginning of the Nixon administration regarding the debt ceiling. The issue of the budget at this time was extremely sensitive because it was so profoundly tied up with the Vietnam War. And the War was already unpopular enough without that. President Johnson didn’t run for reelection because of Vietnam, while Richard Millhouse Nixon had pledged to “end the war” (even if he did not actually intend to…) That made any request to raise the debt ceiling a chance to relitigate the war spending already ongoing. Given that context, a lameduck president (Johnson) was not going to pick an unwinnable fight. And the president that succeeded him was not going to have a legislative solution ready within a week of inauguration.
Enter the accounting gimmicks. The end of the Johnson administration saw a number of federal accounting tricks used, most notably an obscure one regarding the International Monetary Fund (as it’s pretty elaborate, I will cover it in a future piece). So we are decades deep into tricks and illusions like those I have discussed this year. The two memos I’ve secured relate to a more expansive list of accounting gimmicks, some of which require the Federal Reserve’s cooperation. You can expect a lot more from Notes on the Crises laying out how these gimmicks work in the new year.
Continuing the tradition of anodyne memo names, the first memo dated January 30th 1969 was simply entitled "Treasury Cash and Debt Ceiling Dilemma." by Alan Holmes. Holmes was the New York Federal Reserve’s Manager of the Fed’s “System Open Market Account” (SOMA). In other words, he decided what the Fed bought and sold. The second memo “Legal aspects of proposals for assisting Treasury in connection with cash and debt ceiling problems” was authored by head legal counsel Howard Hackley. (Long time readers of Notes on the Crises may recognize Hackley as the man who came up with the legal justification for central bank swap lines — with nothing but the text of the Federal Reserve Act and his creativity.)
Skipping over the Treasury initiated gimmicks for the purposes of this piece, I’m going to focus on the three very interesting Federal Reserve-related proposals. The first proposal is “Immediate credit by Reserve Banks for Government deposits.”But what does that mean?This is more straightforward than it seems on the surface. If you’ve ever sent or received a bank payment (either electronically or by check), you know there is a time difference between when the bank received the payment (or the check) and the point the account was credited. Therefore, this LBJ Treasury proposal is for the Federal Reserve to immediately credit the Treasury’s bank account for checks and other funds deposited by the Treasury rather than having a delay for “processing”. I find this proposal so interesting because, as we’ll see in Hackley’s legal opinion below, it provides insights into larger debates over “what counts” as direct “borrowing” from the Federal Reserve by the Treasury:
The practice by which, under the present time schedules of the Reserve Banks, immediate credit or deferred credit up to two days is given for items received from member banks has itself always been subject to some legal question on the ground that the giving of such credit for items before they are finally collected amounts to an unauthorized extension of credit to member banks. this practice has been defended on the ground that any resulting credit to member banks is incidental to the efficient administration of the check collection functions of the Reserve Banks
However, a departure from the practice that would result in immediate credit for all items deposited by the Government would appear clearly to constitute simply a means of extending credit to the Treasury rather than a proper incident to collection operations and would therefore, in my opinion, be even more subject to legal question.
We learn two interesting things from Hackley’s legal analysis here. First, the Federal Reserve’s check collection practices had long been subject to extensive internal legal debate that I was not previously aware of. This debate is based on the idea that by crediting the accounts of banks before checks had been fully processed, they were providing interest free loans that they weren’t demanding collateral for. The justification for this practice, that they aren’t “intending” to extend credit but instead its “incidential” to “efficient administration of the check collection functions”, has big implications for other practices that provide “incidential” benefits but are justified for other reasons.
Most interesting however, is the discussion of the Treasury “borrowing” aspect. Recall that, at this time, the Federal Reserve could buy up to 5 billion dollars of treasury securities directly from the Treasury. The issue from the debt ceiling point of view is that if such “credit extensions” counted as a “direct purchase”, they would count against the debt ceiling. The trick is getting the Treasury’s account credited without any corresponding “debt subject to limit”. Alan Holmes memo informs us that “counsel at the Federal Reserve Bank of New York has expressed the opinion that the Federal Reserve is not authorized to grant immediate credit to the Treasury while it denies such credit to other depositors”. The interesting implication of this statement is that if the Federal Reserve switched to immediate crediting of all checks, this would not be subject to the same legal objections. This leads to the obvious question: what other things could the Federal Reserve do that “incidentially” filled up the Treasury’s bank account without creating more debt “subject to limit”?
The next option discussed by the memos is labeled by Holmes as “Speed-up of payments to Treasury of interest on uncovered Federal Reserve notes”. I’m not going to get into the ins and outs of this antiquarian phrase of art except to say that it is referring to remittances of profits from the Federal Reserve to the Treasury. Today these payments are legally required by the 2015 FAST act. The idea with this option is simply that the faster these remittances fill up the Treasury’s bank account, the less treasuries they have to issue. Hackley finds this proposal so uncontroversial that he spends all of two sentences affirming its legality:
Payments to the Treasury of interest on Federal Reserve notes at a rate fixed by the Board pursuant to section 16 of the Federal Reserve Act are now made on a monthly basis. I see no legal objection to a procedure under which such payments would be made on a weekly or even a daily basis
The wide discretion the Federal Reserve has over its own accounting rules creates many opportunities to fill up the Treasury’s account by reporting more profits (or less losses) at the Federal Reserve Bank level.
However, the most legally questionable approach would be to talk in terms of “prepaying” remittances. This is a proposal economist JW Mason made in Barron’s this year. As we’ve seen above, “prepaying” remittances would most likely be interpreted similarly to providing immediate credit for government deposits. In other words, an extension of credit to the Treasury without statutory authorization and, even if authorized, producing a debt obligation that would count against the debt ceiling. That being said, even this provocative approach would have more legal grounding if some other purpose was proffered that made the higher flow of remittances an “incidental” outcome. This brings us to the final Federal Reserve related proposal. This suggestion is the most familiar gimmick to a wider audience of scholars. The “Warehousing of foreign currency assets” did actually happen in 1969 and again in 1994, as I briefly wrote about in the Financial Times blog. Thus, what is interesting about the discussion of this proposal in these memos is understanding the legal justifications Hackley provided in 1969 and the wider implication of his legal arguments.
Let’s back up; what is “warehousing”? The Treasury has historically had primary responsibility for “intervening” in foreign exchange markets to affect the dollar’s exchange rate with foreign currencies. To do this there is an “Exchange Stabilization Fund” where the Treasury “holds” this foreign currency (The role of the ESF has greatly expanded beyond foreign currency operations since its inception). The issue is that the ESF is just a legal and accounting device (dare I say “gimmick”?) When the “ESF” buys foreign currencies, what operationally happens is that the Treasury sends out dollar payments from its bank account at the Federal Reserve. The Treasury fills this account by issuing treasuries- there’s that debt ceiling again.
The solution the Treasury suggested was to, as long as the treasury was near the debt ceiling, sell the foreign currency to the Federal Reserve with an implicit “promise” to buy the foreign currency back when the debt ceiling no longer binds. There are a few notable things about this. First, it is operationally identical to depositing a platinum coin with the Federal Reserve. So, to the extent you fear what would happen if the Treasury deposits a trillion dollar platinum coin with the Fed, “warehousing” tells you what would happen: not much of anything. This proposal is as much, arguably more, of an “accounting gimmick” than the platinum coin. The Federal Reserve has gone along with “warehousing” multiple times.
The other notable element of this proposal is the idea that the Treasury is informally promising to “buy back” the foreign currency at an unspecified later date. That’s kind of like a “credit extension” is it not? Indeed, Hackley comments on this issue at length in his memo:
It must be recognized that adoption of the proposed arrangement could subject the System to criticism. It might be charged, for example, that the proposed warehousing transactions would constitute a direct extension of credit to the Treasury by the Federal Reserve and would be contrary to the spirit if not the letter of the law, particularly in view of the express provisions contained in section 14(b) of the Federal Reserve Act for direct borrowing by the Treasury from the Federal Reserve within prescribed statutory limits. However, for the reasons here indicated, I believe that the transactions would be legally defensible as not being designed primarily to aid the Treasury but as intended to avoid developments that would have an adverse impact upon the "credit situation of the country."
This conclusion is, of course, foreshadowed by Hackley’s legal reasoning regarding the first proposal I discussed. It, in fact, goes further. Now the implicit “credit extension” is permissible not just as a byproduct of another policy goal, they can go further. They can specifically aim to fill up the Treasury’s account with some other statutorily authorized action if that is merely a means to the end of avoiding the negative impact of treasury default on the “credit situation of the country”. This is truly extraordinary and in some ways is stronger confirmation of my arguments this year than the 2011 memos.
Hackley reconfirms this in multiple places in the memo. He states that as long as the Federal Reserve has statutory authority to undertake an action (in this case buying foreign currency on the “open market), “the fact that their purpose may appear to be solely to provide the Treasury with additional cash does not affect their legality”. That is an extremely strong statement. He expounds on this point at length such that his view is unmistakable:
Section 12A of the Act provides that open market purchases and sales "shall be governed with a view to accommodating commerce and business and with regard to their bearing upon the general credit situation of the country." The proposed warehousing transactions might not be regarded, in the usual sense, as conforming with this requirement. However, Mr. Holmes' memorandum expresses the view that either a breach of the debt ceiling or failure of the Treasury to pay its bills "could have financial repercussions, both at home and abroad, that might seriously impair the dollar," Moreover, circumstances might arise under which, in order to avoid these alternatives, the Treasury might be obliged to sell in the market large amounts of foreign currency holdings of the Stabilization Fund; and any such action could, of course, result in "disorderly conditions in exchange markets" within the meaning of the Foreign Currency Directive. Consequently, if the proposed warehousing transactions are designed to avoid any of these possibilities, they would clearly have a very vital "bearing upon the general credit situation of the country."
Since Section 12A of the Federal Reserve Act is still good law, it is clear that this interpretation is still valid. Furthermore, the 2011 memo (and FOMC transcript discussion of it) indicates that this legal interpretation has not changed, even if the Federal Reserve does not like the implications of its internal policy to become fully understood by the public- or congress for that matter. Holmes suggestion that “In the meantime the Treasury could be assured of the desire and willingness of the Federal Reserve to cooperate in meeting a national emergency if one, in fact, developed.” defines the Federal Reserve’s posture. It will do everything in its power to avoid treasury default, it defines its legal powers quite widely and it simply wants to keep these facts quiet.
As far as I can tell, this Hackley memo has only been previously cited by secondary literature authored or coauthored by Owen Humpage. Humpage is a former senior economist at the Cleveland Federal Reserve Bank. Presumably, he got access to the memo thanks to being an employee of the Federal Reserve System. Additionally, he only focuses on one element of the memo: the discussion of the “warehousing” proposal and its supposed threat to “central bank independence”. Unlike me, he is not particularly interested in the wider question of the debt ceiling that the memo is actually focused on. As such, I believe this is the first published discussion of the entire contents of the Hackley memo and, as far as I can tell, the first public mention of Alan Holmes’ memo from the previous day.
These memos are far older than the 2011 memo I unveiled at the beginning of the month. Superficially, their contents are less explosive. However, I think my discussion here, and their text themselves, illustrate that they are similarly important in subtler ways. They show that the “financial stability” logic motivating preventing treasury default has a long history within the Federal Reserve’s internal legal reasoning. These memos also exemplify that given the Federal Reserve’s deep commitment to preventing treasury default, they prefer whenever possible to be seen as not making a choice. This is why, as I wrote about for much of the first half of the year, options which have the Treasury solely rely on the Federal Reserve as a fiscal agent are not just acceptable from the Federal Reserve’s point of view, they are preferred.