A central bank that lends in a panic acts under two kinds of limit. One is the law: what the statute lets it lend, to whom, and against what. The other is policy: what it chooses to do inside the law, and what it says it will and will not do next time. Walter Bagehot's Lombard Street, published in 1873, is the classic statement of the doctrine of the lender of last resort and is cited throughout later debates about rescues; its rule, as it is usually summarised, has a part for each kind of limit: lend freely, at a high rate, on good collateral, and announce beforehand that this is what will be done (Bagehot 1873). The last clause is the one this article follows. It asks what the authorities actually wrote down, at the time and in public, about their power to lend and about where that power stopped.
Two episodes are read. In September 1998 the Federal Reserve Bank of New York brought together the creditors of Long-Term Capital Management, a large and heavily borrowed hedge fund, and on 23 September those creditors agreed to put new capital into it. On 1 October the Chairman of the Board of Governors, Alan Greenspan, explained the episode to a committee of the House of Representatives. On Sunday 14 September 2008, the Board announced that it was widening the collateral it would accept at two of its lending facilities and relaxing, for a time, a limit on what banks could lend to their own affiliates. Skeptron's record of the next day's event dates the bankruptcy of Lehman's holding company to 15 September 2008.
Skeptron's record of Lehman holds two entities, keeping the holding company's bankruptcy of 15 September 2008 apart from its dealer subsidiary's use of the Federal Reserve's existing credit facilities, and this study does not say that Lehman as a whole received or was refused a loan.
The second episode sits at the centre of a live dispute. Laurence Ball argues that the Federal Reserve had the legal authority and the means to lend to Lehman Brothers to prevent its bankruptcy, that the reason it gave, a lack of sufficient collateral, does not fit the evidence, and that the decision not to lend was a matter of policy rather than of legal limits; his book is a critical reading of the official account and is disputed by some of the officials involved (Ball 2018). Ben Bernanke, chairman of the Federal Reserve at the time, gives in his memoir the view that Lehman could not be saved lawfully for lack of collateral, and describes the crisis as a financial panic of the classic kind that the Federal Reserve met by acting as lender of last resort (Bernanke 2015). The Financial Crisis Inquiry Commission, set up by Congress, describes the failure of Lehman and the rescues of others within a broader finding that the crisis was avoidable and was caused by failures of regulation, supervision, governance and risk management; its report was divided, with dissents (FCIC 2011).
These positions argue about what was possible, using evidence gathered afterwards: testimony, internal records, balance sheets. This article does not try to decide between them. Its contribution is to add to the debate the wording of the record itself: what the public texts of 1998 and 2008, read in full and in order, say about the power to lend and its limits, and what the modern summaries of the same events keep and drop. The question is narrower than the one Ball and Bernanke dispute, and earlier in time. It is also one that the disputants' evidence does not answer, because a public text is a commitment of a different kind from an internal memorandum: it is what the market, the Congress and later readers were told.
A second debate runs alongside the first. Writers on the lender of last resort disagree about what the classical doctrine asks for and whether it still fits. Thomas Humphrey presents it as a duty to protect the money stock and the banking system by lending to solvent institutions on good collateral at a penalty rate, rather than to rescue insolvent individual firms (Humphrey 1989). Perry Mehrling argues that modern finance rests on market-based funding, so that the central bank has become a dealer of last resort for markets as well as a lender of last resort to banks (Mehrling 2011). Paul Tucker asks a different question, about legitimacy: he holds that decisions to rescue institutions are political choices that need rules set in advance and a framework agreed with the government (Tucker 2018). The texts of 1998 and 2008 speak to all three positions, and to the old question of moral hazard, which Greenspan's testimony raises in its own words.
The placement of the secondary works in these debates, and their application to the two episodes, are this article's own; what is said of each work stays within the published reading notes of the parent study, and no work is quoted. The primary texts are quoted verbatim and at length.
The argument runs as follows. Section 2 sets out the debates. Section 3 describes the texts, the records and the ledger. Section 4 reads Bagehot, briefly, for the question he put about writing a lender's duty down. Sections 5 and 6 read the two episodes, each beginning with its historical claim. Section 7 sets the two side by side, and Section 8 concludes.