Chapter 8. Charters in an Emergency: Founding Banks and Rules

8.1Institutions founded in a crisis

Most emergency acts in money and credit stop something. A payment is suspended, a convertibility is closed, a bank's business is halted, a peg is abandoned. A charter or an Act of Parliament does the opposite. It creates something — a bank, a system, an institution — and it is written to outlast the crisis that produced it.

The fourth literature this study maps in Chapter 1 asks what a rule of money does under pressure and what an emergency leaves behind; the readings there judge acts by effects that the records of these cases do not hold. This section adds something narrower and, for this study's question, more direct: what three founding texts actually say about changing themselves, read in the actors' own words where those words survive.

For a founding the questions do not land in the same way as they do for a suspension. A suspension has a term and a reason that runs out; a founding has neither, because ending it would undo the thing it created. The record of a founding still owes a later reader something: not a date of exit, but the provisions that govern how the institution's own rule can be changed, and the identification of who is bound by it. Whether a record holds those provisions is a question this study can ask of every case, and the answer for these three is set out below.

The three cases are the royal charter of 27 July 1694 that founded the Bank of England, the Federal Reserve Act of 23 December 1913, and the IMF's Articles of Agreement adopted at Bretton Woods on 22 July 1944. In Skeptron's grouping, drawn from each record's own domain, the three belong to three different groups: the Bank of England's founding sits in the credit history, the Federal Reserve Act in the money history, and the IMF Articles in the markets history. Reading them together keeps that grouping visible.

Two of the three come with the actor's own words, in the form of public-domain texts this study holds, and in both those words name purposes rather than ends. The title of the Act of 1913 states what the statute is for: "to provide for the establishment of Federal reserve banks, to furnish an elastic currency, to afford means of rediscounting commercial paper, to establish a more effective supervision of banking in the United States, and for other purposes". Safeguards are conditions on the use of resources, not a term on the Agreement; the sections on those conditions are read in full in 8.4. The charter of 1694 has no such text here; what this study can say of it comes from Skeptron's record and from the reading notes on John Clapham and Charles Goodhart. This study reads the 1694 charter as compatible with both accounts: the charter precedes the functions the later Bank acquired.

Two counts from "The ledger's numbers" set the three records against the wider set. Across all eighty-four cases, most records say nothing about whether the act was temporary, only seven give a term, thirty-five name no one who bore the loss or the cost while the act ran, and three name one. Skeptron's recurring sentences — on keeping the event and the document apart, on not merging announcement, decision, implementation and result — are the studio's method, set out once in Chapter 2; they are not findings about any of these cases, and this chapter does not read them as evidence.

8.2The Bank of England, 1694

A founding record in this study is dated, named and purposive, and silent on everything that later made the institution famous. Skeptron's record of 27 July 1694 is the clearest instance: it gives one day, one royal charter and one stated purpose — the financing of the war against France — and it draws, in its own words, a boundary against the later role of a central bank. What this section adds to the debate in Chapter 1.2, on the lender of last resort, is a question those works do not ask: what a record of a founding can carry when the institution it names later becomes the holder of a nation's reserve. And what it adds to the institutional history of Goodhart is an order of things the record can verify: the charter comes first, and the functions the later Bank acquired are not in it.

Skeptron's record for this case opens with one dated act and a stated purpose:

The record answers its own second question with the same facts: the institution was a private bank acting as a government bank, and its main purpose was financing the war against France; the official chronology presents the royal charter of 27 July 1694. The scenario flags that the date is given in the notation used in England at the time, and this study keeps the date as the record gives it, without converting it. The third question asks whether the later record of the Bank entitles a reader to say that the institution possessed all central bank authorities from its launch to the present. The answer is that the ownership and business of the time are distinguished from the later role of a central bank, and that the record is not used to predict the outcome of the war or the performance of government bond investments. That boundary is the most important sentence in the case for this chapter. It forbids a specific error: reading the Bank of England's later lender-of-last-resort functions back into the charter of 1694 through this record. The record holds the founding and keeps the later role apart from it.

Of this chapter's two source texts, the Federal Reserve Act of 1913 and the Articles of Agreement of 1944, neither touches 1694, and no public-domain text of the charter is held here. Skeptron's record is therefore the only dated document of the act this section has, and it is a modern archive's short summary, written long after the charter it names. Skeptron's summaries are dated readings of public sources, not the actors' own words. Bagehot's description of the London money market is that no bank holds any considerable sum in hard cash above what its daily business needs, except the Banking Department of the Bank of England, and that on this structure the reserve of the Bank is the banking reserve of all London and beyond London of England, Ireland and Scotland. That is the doctrine later debates turn on; Chapter 4 sets out the rule and its readings in full. Nothing in the 1694 record anticipates it, and this section does not use the later passage to fill the earlier record.

On the study's own axes, the record gives a date, an act and a purpose. It gives no term, no condition under which the charter or the bank's obligation would end, and no figure — no capital subscription, no loan to the state, no rate. It names no individual as the decider; the charter is named as royal, and nothing further. It names no bearer of any cost: not the subscribers, not the state, not the holders of notes, not the holders of the government's bonds. The record's list of what is unverified says that month, day, time of occurrence, coordinates and personal information are not entered as actual input, and that only the acts, objects and scope stated in the record are confirmed.

Humphrey's survey and Bordo's article describe an institution's later duty to lend in a panic, and neither asks what a record of a founding can carry: no sentence in the 1694 charter, as the record holds it, makes the Bank the holder of the nation's ultimate reserve or names a duty to lend in a panic.

John Clapham's history supplies the kind of material this record holds none of. His two-volume history of the Bank of England, from its foundation in 1694 to 1914, describes the Bank's origins as a lender to the state and the growth of its role in the banking system and the money market, and covers the crises it faced, including the restriction of cash payments from 1797. That growth is a development over decades; the record of 1694 holds one day. Clapham's work is therefore not a correction of the record but a different source on a different span, and this chapter cites it only for what its note says about origins and development. Where the Bank's own suspension appears in this study, it appears as a separate case, dated 1797 and read in Chapter 3.

In The Evolution of Central Banks, Goodhart's account of why central banks emerged points the other way. The 1694 record shows the opposite order from the one Goodhart describes for the mature institution: a private bank chartered to finance a war, with no central-bank function named in the record at all. This study reads that as compatible with Goodhart's account of emergence rather than as a contradiction of it, because the charter precedes the functions the later Bank acquired, and Goodhart's argument concerns how those functions came to be held by such an institution. What the record cannot do is tell a later reader when, by what decision, or on what terms the founding became a central bank. The record names that role only to keep it apart.

In this study's ledger, the case is read as an act of law, charter or institution; the record does not present the act as temporary; it names no condition for its end; it says nothing of what became of it; it names no two dated moments among decision, taking effect and result; it gives no figure; and it names no bearer of cost. Nothing in that description depends on the later history of the Bank; it says only what the record holds. The debate the section serves — over what a founding owes a later reader — therefore rests on a record that is precise about one day and silent about everything that came after it.

8.3The Federal Reserve Act, 1913

The Act's own words state what it is for. The title, as the study's text gives it, reads "to provide for the establishment of Federal reserve banks, to furnish an elastic currency, to afford means of rediscounting commercial paper, to establish a more effective supervision of banking in the United States, and for other purposes". Section 1 gives the statute its name, the Federal Reserve Act.

The sections of the Act that this study holds go further into procedure than any title can. Section 2 charges the Secretary of the Treasury, the Secretary of Agriculture and the Comptroller of the Currency, acting as an organisation committee, with designating not less than eight nor more than twelve Federal reserve cities and dividing the continental United States, excluding Alaska, into districts each containing one such city; the committee's determination is not subject to review except by the Federal Reserve Board when that board is organised. Section 4 carries the same sequence into the founding of an individual bank: once the organisation committee has established the districts, an application blank is sent to each national bank in the district and to such other banks as the committee declares eligible; once the minimum capital has been subscribed and allotted, the committee designates five banks whose applications have been received to execute a certificate of organisation, which shall state the name of the Federal reserve bank, the territorial extent of the district over which its operations are to be carried on, the city and State in which it is to be located, the amount of capital stock and the number of shares into which it is divided; and the Act provides that, on the filing of that certificate with the Comptroller of the Currency, the Federal reserve bank shall become a body corporate and, in the name designated in the certificate, shall have power. The Act, in other words, schedules its own institution. Enactment is followed by organisation, subscription, certificate and corporate existence, each step dated by what happens rather than by the calendar.

The powers the Act confers are given with limits attached. Section 13 provides that, on the indorsement of any of its member banks with a waiver of demand, notice and protest, any Federal reserve bank may discount notes, drafts and bills of exchange arising out of actual commercial transactions — issued or drawn for agricultural, industrial or commercial purposes — and that the Federal Reserve Board has the right to determine or define the character of the paper eligible for discount. The eligible paper is confined by maturity: at the time of discount it may run no more than ninety days, though paper drawn or issued for agricultural purposes, or based on live stock, may run up to six months and be discounted within a percentage of the reserve bank's capital fixed by the Federal Reserve Board. Concentration is limited too: paper bearing the signature or indorsement of any one person, company, firm or corporation, rediscounted for any one bank, may not at any time exceed ten per centum of the unimpaired capital and surplus of that bank, and the restriction does not apply to bills of exchange drawn in good faith against actually existing values. Section 14 turns to open-market operations and, in paragraph (d), gives each reserve bank power to establish, from time to time and subject to review and determination by the Federal Reserve Board, the rates of discount it charges for each class of paper, fixed with a view of accommodating commerce and business.

Skeptron's record holds one date and one act — the signing on 23 December 1913 — and refuses to merge it with what followed. The record names Woodrow Wilson as the United States president who signed the Act. Its third question asks whether the signing may be regarded as the same-day opening of operations of each regional Federal Reserve Bank. The answer is a refusal:

That answer is the record's own. The Act's text supports the same separation in a different register: the statute itself holds within it the steps by which reserve banks come into existence, none of which is the signing. A later reader who wants the date of opening has to find it elsewhere, and this study's record does not hold it.

The historical link between this statute and the panic that preceded it is in the reading notes on Friedman and Schwartz. Chapter 4 reads the case itself — the clearing house of 22 October 1907, keeping its resources for its members, refusing the Knickerbocker Trust, and the trust's suspension of business. This chapter reads the statute that came six years later.

Whether the Act names its own end cannot be answered from what this study holds. The sections received here — 1, 2, 4, 13 and 14 — contain no clause on repeal or expiry, but they are not the whole Act, and this study does not hold the rest. What the received sections do show is a permanent design: powers granted, limits set, rates subject to review by a board, districts that the Federal Reserve Board may readjust, and new districts it may create from time to time. Revision, where the Act names it, is internal — by board, by committee, by capital subscription — not by a date.

In "The ledger's numbers", this case shows an act of law, charter or institution; it does not say whether the act was temporary; it names no condition for its end; it says nothing of what became of it; it names no second dated moment among decision, taking effect and result; it gives no figure; and it names no bearer of cost. The record insists that the date of the legislation and the start of operations must be checked separately, but it supplies only one of the two dates; the separation is named as a duty on the reader, and the second date is not given. That gap — whether a record names two dated moments among decision, taking effect and result — is the third of this study's registered hypotheses, and its counts are reported in Chapter 12.

8.4The IMF's Articles, 1944

The record's reference date is 22 July 1944, and the act is the adoption of the Articles of Agreement at the United Nations Monetary and Financial Conference in Bretton Woods, New Hampshire. The record's third question asks whether that date can be treated as the date on which all member countries completed their accession and ratification. The answer refuses:

The text of the Articles that this study holds is a United States Government printing of 1946, and its printed title note gives the same separation from the other side. The note records the agreement as formulated at the United Nations Monetary and Financial Conference at Bretton Woods, New Hampshire, between 1 and 22 July 1944, and signed at Washington on 27 December 1945, then records an instrument of acceptance by the United States deposited on 20 December 1945. The printed title note records formulation at the conference from July 1 to July 22, 1944, signature at Washington on December 27, 1945 and the deposit of the instrument of acceptance on December 20, 1945, and the note is cut off after "effective December". It says "Formulated", not adoption, and so carries no separate moment for adoption on 22 July; the record gives none of these as the date of entry into force.

The Articles' own words set out what the institution is for, and Article I closes by stating that the Fund is to be guided in all its decisions by the purposes set forth in it. The phrase that carries the most weight for this chapter's question is the fifth item's: resources are made available under adequate safeguards. Safeguards, not a term. The purposes are permanent; the conditions attached to the use of the resources are the operative rule.

Those conditions are set out in Article V. Section 4 provides that the Fund shall consider the member's willingness to pledge as collateral gold, silver, securities or other acceptable assets, and may require such a pledge as a condition of waiver.

Article IV, section 5 governs changes in a member's par value — the exchange-rate rule the Agreement set. a member is not to propose a change in the par value of its currency except to correct a fundamental disequilibrium, and a change may be made only on the member's proposal and only after consultation with the Fund. The section then sets bands and, inside one of them, a clock: where the proposed change exceeds the first ten per cent of the initial par value but not a further ten, the Fund may concur or object but is to declare its attitude within seventy-two hours if the member so requests. Among the three texts read in this chapter, this is the only express time limit this study's excerpts hold. It governs the speed of a consultation, not the life of the Agreement.

Benn Steil's The Battle of Bretton Woods (2013) is the account of how the text came to be. Steil describes the negotiations of 1944, centred on John Maynard Keynes for Britain and Harry Dexter White for the United States, the competing plans — Keynes's international clearing union and White's stabilisation fund — and the American position that prevailed in the final design of the Fund and the World Bank; his argument is that the outcome reflected American power and the weakness of Britain. That argument is about the drafting room, and it bears on this chapter's question only indirectly: who held the pen determines which of the texts' provisions were made permanent and which were left open. The record of 22 July 1944 holds no draft, no negotiation and no vote.

Harold James's International Monetary Cooperation Since Bretton Woods (1996) covers what came after. His history spans the decades in which Article IV's rule was tested and Article V's conditions were invoked, which is a way of saying that the Agreement named its own instruments of change but not its own end.

The record that shows those instruments in use later is Korea's stand-by arrangement of 4 December 1997, which the IMF approved at SDR 15.5 billion over three years, of which SDR 4.1 billion was available at once and the rest tied to reviews and conditions. That case is one of the cases Chapter 7 reads in full; the link to this chapter is structural rather than numerical. The Articles' Article V conditions are the standing rule under which a stand-by arrangement is approved in tranches and subject to review, and the 1997 case shows that rule working on a single dated day. Nothing in the 1944 record anticipated it, and nothing in the 1997 record can be read as an amendment of the Agreement.

In "The ledger's numbers", this case shows an act of law, charter or institution; it does not say whether the act was temporary; it names no condition for its end; it says nothing of what became of it; it names no second dated moment; it gives no figure; and it names no bearer of cost. That last silence is worth stating plainly for this case, because the Articles' own fifth purpose names a class of people in general terms — members, correcting maladjustments in their balance of payments — without naming anyone who bore a cost in any particular emergency. The record holds the same silence.

8.5What a founding names of its own revision

The three cases read in this chapter agree on all eight descriptions this study asks of a record, and the table below gives them in the ledger's own words:

On the ledger's own reading, all three are acts; none is presented as temporary; none names a condition of its end; none says what became of it; none gives two dated moments among decision, taking effect and result; none gives a figure; none names a bearer of cost. The ledger gives each of these three cases the same value on all eight descriptions; kind of act is a classification rather than a finding: these are laws, charters and agreements.

What that uniform row supports is the point this chapter makes: a founding's record owes not a term but the provisions that govern the institution's own revision. A suspension's record that says nothing about its term is silent about something the act was expected to have. A founding's record that says nothing about its term is silent about something the act was not designed to have. The two silences are not the same, and putting them on one axis in the ledger makes that difference visible only by reading each one back against the kind of act it sits under. This study reads the "not-said" of a founding as expected rather than as an omission, and reads the "not-named" of its revision provisions as a genuine gap, because revision is the one thing a permanent act must eventually face.

What each of the three texts names about its own revision can be checked against what this study holds of them. The record of 1694 names a royal charter and nothing about how the charter or the bank's obligations could be changed, cancelled or renewed. The Federal Reserve Act, in the sections received here — 1, 2, 4, 13 and 14 — names organisation, subscription, corporate existence, powers, maturities, limits and rates subject to board review, but no clause on repeal or expiry; those sections are not the whole Act, and this study does not hold the rest of it. The Articles, in the articles received here — Article I, Article IV section 5, and Article V sections 3 and 4 — name purposes, conditions on the use of the Fund's resources, a power of waiver, and a procedure for changing a par value, but no clause on amending or terminating the Agreement itself. None of the three received texts contains a rule about ending the institution.

Two counts from "The ledger's numbers" set these three records against the wider set. Across all 84 cases, 27 do not say whether the act was temporary, 7 give a term, 2 call an act temporary or a suspension while giving no term, 8 are unsettled and 40 do not apply.

The difference this chapter has been building towards can now be stated in one sentence. A permanent act and a temporary one differ in what a record of them owes a later reader: a suspension's record owes the term and, where it can be found, the bearer of the cost; a founding's record owes the provisions that govern the institution's own revision and the identification of those bound by it. On the ledger's own counts, the records in this set hold almost none of either. The one record that protects a later reader in a specific way among these three is the Bank of England's, and it does so not by naming a term but by drawing a boundary — keeping the later role of a central bank apart from the charter of 1694 — against the error of reading later powers into an earlier founding. The Federal Reserve Act's record protects a different boundary, between the date of legislation and the start of operations, while holding only one of the two dates. The IMF Articles' record draws the line between adoption and the accession, ratification and entry into force of each country, and the printed text of the Agreement adds a line between formulation and signing and then stops before any effective date.

None of that tells a later reader what any of these institutions should do. What it tells is what the record holds: three founding acts — a royal charter, an Act of Congress and an international agreement — each silent on revision and on cost, and each protected by a boundary rather than by a term. Chapter 11 reads the conditions for ending across all of the study's cases; Chapter 9 takes up the emergency acts that did end on a date — rates and pegs abandoned — where the question of a term is not abstract but immediate.