During the opening days of March, 1933, the U.S. economy resembled a stricken body slowly bleeding out, its organs failing one by one. The Federal Reserve System was hemorrhaging gold, and entire state banking systems were shutting down one after another. If the economy was to survive and recover, FDR had first to staunch the bleeding, and then to arrange for a transfusion. Here I explain how he managed the first of these steps.
Panic at the Fed
Before we consider how FDR managed to end the nationwide banking crisis that broke out on the eve of his inauguration, let's recall just how that crisis came about. It began with the official closing of Michigan's banks in mid-February, 1933. Michigan's decision led other states to start declaring holidays as well. In the meantime, fear of an impending devaluation led to both foreign and domestic runs on the dollar. Together these events set the stage for FDR's declaration of a nationwide bank holiday on March 6th, two days after taking office.
So much for the barest facts. In Drifting Toward Mayhem, his superb but underappreciated history of the depression-era banking crises, historian Robert Lynn Fuller supplies many crucial details. Perhaps the most striking of these is that "hysteria" didn't take hold of bank depositors until after state governments everywhere began shutting-down or limiting withdrawn from banks. Until then it was mainly commercial bankers and (most importantly) Federal Reserve officials who went into panic mode.
During the "wave of fear" that immediately followed Michigan's crisis, Fuller says, "the public remained mostly calm. …[D]epositors did not rush to their banks to withdraw cash." Instead, the wave got started by bankers and state politicians who mostly reacted not to actual depositor runs but to "each other's anxiety about what might happen." The governors of Tennessee and Maryland followed Michigan's example by declaring their own bank holidays, while those of 17 other states allowed their banks to limit depositors' cash withdrawals. Unfortunately such responses only gave bankers elsewhere more reason to panic, by denying them access to correspondent balances that often made up a substantial share of their liquid reserves, or by making them fear they might lose access to those balances at any moment. The ensuing scramble for funds upped the pressure on governors who hadn't yet done so to relieve the pressure on their banks.
As holidays and restrictions multiplied, ordinary bank depositors were bound to join the fray, not because they went berserk, but because they also feared being frozen out of their own accounts. Thus when crowds gathered to withdraw their savings from a bank, as they did in Newark, New Jersey on March 2nd, when they lined-up in front of the Howard Savings Institution, they did so, not because they feared it might be broke, but because they believed that New Jersey's governor was about to declare a holiday that would put their savings out of reach. Fear of local holidays thus joined the growing fear that FDR would suspend gold payments to inspire people to get their money out of banks that were often not only solvent but understood to be so by those who ran on them.
Once one allows for such (perfectly rational) fears of both state bank holidays and devaluation, there isn't much need, after all, for any appeal to depositor hysteria. This doesn't mean that depositors never mistook sound banks for unsound ones during those months, and it certainly doesn't mean that there were no unsound banks left after those around Detroit went belly-up. But it does cast doubt on the view that depositors had come to generally suspect their banks of being broke.
No Holiday for Hoover
As we saw in the last installment, it was the run on the dollar, and especially foreigners' part in it, that ultimately led the federal government to shut down the entire U.S. banking system. Although that run harmed commercial banks, it was less of an immediate threat to most of them than to the Federal Reserve System itself, and the New York Fed in particular, for it was a run out of dollars of all kinds and into gold, where that gold was most kept at the New York Fed. The final, fatal blow to the nation's banks came when the "wave of fear" engulfed New York Fed officials who, Fuller writes (p. 397), "reacted against future expectations of actions by European central bankers, who acted on their own worries about the future of the dollar under President Roosevelt."
It was those officials rather than ordinary bankers who first urged President Hoover to declare a nationwide bank holiday. When that attempt failed, they resolved to get New York State Governor Lehman to declare a state bank holiday. Because state governments lacked the authority to take action on the Fed's behalf, they asked the New York Clearing House Association to join them in making the request. Because the Clearing House banks were themselves in remarkably good condition, and had no desire for a holiday, it did so only with considerable reluctance. Once New York's banks were shut down, it was clear that all the still-open commercial banks in the rest of the country would soon have to close. Fed officials informed other state governors accordingly so that, by the time of FDR's inauguration, state bank holidays had been declared almost everywhere. Most of the banks that were still open stayed so in defiance of state officials' orders, presumably because they felt capable of weathering the storm.
Yet the crisis still wasn't over. While widespread state bank holidays stopped runs on ordinary banks, so far as the Fed was concerned, they were only a stopgap: although the holidays ended the drain of gold from the Fed to U.S. banks and their customers, they couldn't stop foreign central banks from cashing their dollars in for gold. Only a national bank holiday declared by federal authorities could do that. The point deserves emphasis: the national bank holiday was needed not for ordinary banks' sake but to keep the Fed from being forced off the gold standard. 
Why, in that case, hadn't Hoover declared such a holiday when the New York Fed first urged him to, or even before then? Why did he leave it to FDR, allowing the Fed to lose that much more gold, and also allowing panic to damage more sound banks until state holidays finally shut almost all of them down? It wasn't that Hoover was a cold-blooded Social Darwinist who "wanted to keep banks open and failing from panic until the weak banks had all collapsed," as Eric Rauchway (Winter War, p. 199) claims. Hoover may have been pusillanimous; but he was by no means "inhumane" (ibid.). His problem wasn't a lack of compassion but legal scruples; and those scruples were themselves not at all groundless, for while it was clear that no state governor could legally shut down the Fed, it was far from certain that he himself could do it legally.
It appears that Hoover's advisors first contemplated using the authority of the 1917 Trading with the Enemy Act (as amended in 1918) to suspend or restrict the Fed's gold payments in January 1932. Though nothing came of the idea then, by mid-February, 1933, when Michigan closed its banks, officials at both the Fed and the Treasury were urging Hoover to make use of it. Walter Wyatt, the Fed Board's general counsel, went so far as to draft a resolution for the purpose. Hoover was eventually persuaded enough to consult both Carter Glass and William Mitchell, his Attorney General, about it. But both, and Mitchell in particular, doubted the plan could withstand a legal challenge unless the Democrat-controlled Congress gave it its ex post facto concurrence. Just what the consequences of a finding that the holiday was illegal would have been is far from clear. But Ogden Mills, Hoover's Treasury Secretary, thought it would be "fatal" (Awalt 1969, p. 358).
So Hoover hesitated. Over the coming days, as the monetary system crumbled around him, he made various attempts to persuade FDR to publicly support his resolution, and to otherwise secure Congress's backing. To shorten a long story that flatters neither of its main protagonists, between them Hoover's haughtily ham-handed approaches to FDR, and the fact that FDR had little politically to gain by cooperating with him, ruled-out any hope that Hoover's effort would succeed. Hoover accused FDR of wanting to blacken his reputation at the nation's expense. FDR in turn wondered why Hoover couldn't muster-up the pluck to close the banks without his help. Each had a point; and both therefore deserve some blame for the fact that a national bank holiday wasn't declared until two days after FDR became president.
His Finest Week
Regardless of how it fell to FDR to put it into effect, the national bank holiday that began on March 6th and ultimately ended on the 13th was to be among the greatest achievements, if not the greatest single achievement, of his first term in office. Yet the holiday was much less a New Deal achievement, and much more one for which the Hoover administration deserves credit, than is generally appreciated. Although FDR took the lead and played his part brilliantly, his performance was based on a script much of which had been prepared while Hoover was still president, or by Hoover Treasury officials who stayed on to help FDR's still very green team orchestrate the closing and reopening of the nation's banks.
To realize just how much Hoover's men contributed, it helps to first review the timing of the holiday and public events surrounding it. FDR was sworn in on the afternoon of March 4th, a Saturday. Thirty-six hours later, at 1 A.M. on Monday the 6th, citing the authority granted him by the 1917 Trading with the Enemy Act, he issued his proclamation closing the banks and prohibiting any export or sale of gold until March 9th. On the evening of the 9th, at the end of an emergency joint session, Congress passed the Emergency Banking Act, retroactively amending the Trading with the Enemy Act so it could be used not just during a war but “during any other period of national emergency declared by the President.” Later that same evening FDR signed an executive order, dated Friday, March 10th, extending the bank holiday until March 13th but continuing the moratorium on gold exports indefinitely. Finally, on March 12th, he gave his first and most famous fireside chat, explaining what had taken place and the procedure by which sound banks would begin reopening in stages starting the next day.
We've already seen how the idea of using the Trading with the Enemy Act to close the nation's banks and suspend gold payments had first been considered by Fed officials in 1932 and proposed to Hoover well before FDR took office. FDR's proclamation of the 9th was also one Hoover's Attorney General had prepared for his signature, with some advice from Fed officials, several days before. Several of the five Titles of the Emergency Banking Act had also been drafted before—in some cases months before—FDR took office, and some outgoing Hoover officials spent inauguration day helping to draft the rest. One of them—Treasury Secretary Ogden Mills—came up with the plan that, with minor modifications, was used to reopen the banks. Finally, apart from astute rhetorical changes made by FDR himself, the text for FDR's famous fireside chat was drafted by Hoover's Treasury Undersecretary, Arthur Ballantine. Nor is it likely that the incoming administration could have managed without all this help. On the contrary: FDR's own team, including William Woodin, his Treasury Secretary, came to Washington with no similar plans of their own, and were accordingly grateful for the help.
Under the circumstances, it seems only fair to conclude that, although chronologically speaking the national bank holiday, the steps taken during it, and the reopening of the banks that followed, all belong to the New Deal, which must therefore be credited for any part they played in the recovery, Hoover's "old Deal" also deserves some of that credit.
Whoever may deserve credit for it, the fact remains that the nationwide banking holiday marked the turning point of the Great Depression. With it the "Great Contraction" that began in 1929 came to an end. And afterwards, as the banks reopened, a recovery began that would continue, with fits and starts, until the summer of 1937. That recovery wouldn't have been possible had the public not been convinced not only to quit withdrawing money from their banks but to start putting it back into them.
Conventional wisdom attributes this amazing revival of trust in the banking system to Roosevelt's confidence-inspiring "chat," and especially to the public's willingness to take him at his word when he promised that banks would reopen only once they were determined to be sound. According to this view, depositors no longer had to judge for themselves whether an open bank might fail, and so had no need to hesitate to place their life savings with it.
It's a pleasant story. But it's hardly plausible. Though FDR's words were certainly encouraging, and may have assuaged some depositors' concerns, they could hardly have won over the many sophisticated bank customers who knew perfectly well that there weren't enough bank examiners around to thoroughly go through thousands of banks' books, with less than a week at their disposal, to determine which ones were solvent. The situation that confronted bank examiners in the New York Fed district was presumably not unique. As Barry Wigmore (1987, p. 752) reports, they complained that it wasn't "humanly possible…to appraise with accuracy" the banks they'd been asked to appraise. More importantly, neither FDR's remarks nor any proof of commercial banks' soundness could quell the fears of devaluation that informed the biggest run of all—namely, that by foreigners, and foreign central banks especially, to convert dollars into gold. It was, once again, that run, rather than any general loss of faith in the country's surviving commercial banks, that had been the final straw that made a federal banking holiday necessary in the first place.
Yet within three days from the end of the national banking holiday, thousands of banks had been licensed to reopen, including more than three-quarters of the 5,916 national banks closed by proclamation. And their customers did not rush to withdraw the money still left in them. What, then, was the real key to the reopenings' success? William Silber claims that it consisted of steps that amounted to a full if only implicit guarantee of deposits in reopened banks. First, among its other provisions, the Emergency Banking Act allowed Federal Reserve Notes to be backed, not only by gold, commercial paper, and U.S. government securities, but by any “notes, drafts, bills of exchange, or bankers’ acceptances”—that is, by just about any Federal Reserve bank assets. Second, in his fireside chat, FDR promised that new currency was “being sent out by the Bureau of Engraving and Printing to every part of the country,” so that “banks that reopen will be able to meet every legitimate call.” He in turn had his Treasury Secretary promise to indemnify the 12 Federal Reserve banks for any losses they incurred in fulfilling his commitment to the public. This arrangement, Silber says, amounted to a 100-percent deposit guarantee—that is, a guarantee far more complete than the one that took effect when the FDIC began insuring deposits the following January.
But Silber's argument isn't the whole story. Like arguments that credit FDR's soothing words or people's naive confidence in bank examiners for restoring depositors confidence, it overlooks the fact that many depositors had never distrusted their banks in the first place: if they took their money out, it was because they feared either that the dollar would soon be devalued or that state authorities were about to make their accounts inaccessible. For banks that had not been suspected of being unsound before the national holiday, no new guarantees were needed.
The nine large banks of the New York Clearing House Association were a case in point. With one notable exception, they were, despite everything, in remarkably solid condition when Governor Lehman closed them; and it's for that reason that their examiners, for all their carping about lacking time for proper audits, approved all but the one for reopening on March 13th—the day the nationwide holiday ended. Nor were New York's banks unique: more than half of the 15,000-odd national, state and private banks closed en masse by state authorities, representing 90 percent of closed banks' deposits, were judged perfectly safe and ready to open that week on the basis of perfunctory assessments that mainly reflected the general understanding, shared by savvy depositors and regulators alike, that the banks in question were never in danger of being insolvent. The remaining, closed banks were another matter; but these were opened, if at all, only after steps were taken to assure their solvency, whether by bolstering their capital or otherwise.
But what about gold? So long as devaluation fears persisted, how could people be induced to hold either bank deposits or paper dollars instead of gold itself? They had, first of all, to be prevented from withdrawing any more gold from their banks. This much had been largely accomplished by FDR's March 6th bank holiday proclamation, which among other things provided that no "banking institution or branch shall pay out, export, earmark, or permit the withdrawal or transfer in any manner or by any device whatsoever, of any gold or silver coin or bullion or currency or take any other action which might facilitate the hoarding thereof." Although FDR's subsequent proclamation of March 10th announced that the banks would begin reopening on March 13th, the same proclamation continued the suspension of gold dealings indefinitely. The domestic run on gold that played so large a part in the banking troubles of the last several weeks would never trouble U.S. banks, or the Fed, again.
This leaves but a single conundrum. For people didn't just quit exchanging paper dollars for gold. Many also hastened to return gold they'd been hoarding, and did so well before the notorious April 5th, 1933 Executive Order (No. 6102) prohibiting gold ownership outright. By May the Federal Reserve System had recovered more gold than it had lost since the start of February. Yet devaluation, far from having become unlikely, was daily becoming more likely than ever.
Why, then, were Americans seen literally lining up to hand their gold back to the Fed on March 9th, before their banks had even reopened? Eric Rauchway (The Money Makers, p. 52) claims that they were glad to take advantage of the governments' offer to take "a great burden from them…by accepting their gold for safe paper money," as if they hadn't always had the option of holding paper dollars instead of gold, and as if gold were not "safer" than nominal paper equivalents that everyone expected to see devalued!
Evidently Professor Rauchway was either unaware of, or chose not to draw attention to, a more obvious reason why gold started flowing back into the Fed, to wit: that on Wednesday, March 8th, just before all those long lines formed, the Federal Reserve Board announced that the Fed banks had compiled lists of those "unpatriotic" persons who had withdrawn gold from the system during the preceding weeks, and that it planned to have the press publish them if they didn't bring the gold back at once (Fuller, p. 465). At least one Fed bank—Philadelphia's—also threatened to refuse to supply currency to member banks that didn't help it identify such unpatriotic citizens. According to Fuller "This tack served its purpose." That Friday alone $200 million in gold coin and bullion came back (details here). Soon Fed officials were able to report that they now had all the gold backing they needed for their notes, so that the requirement no longer had to be suspended. Whatever FDR's fireside chat accomplished, it can't be credited with this gold inflow, most obviously because the gold started coming back before the chat, but also because FDR offered no assurance against devaluation, fear of which had prompted gold hoarding in the first place.
Intimidation was all well and good for getting Americans to bring their gold back to the Fed. But it couldn't do a thing about any of the gold that had escaped abroad. Getting that gold, or at least some of it, to come back, would take different measures. Just what those measures were, and how they worked, will be the subject of our next installment.
Continue Reading The New Deal and Recovery:
- Part 1: The Record
- Part 2: Inventing the New Deal
- Part 3: The Fiscal Stimulus Myth
- Part 4: FDR's Fed
- Part 5: The Banking Crises
- Part 6: The National Banking Holiday
- Part 7: FDR and Gold
- Part 8: The NRA
- Part 8 (Supplement): The Brookings Report
- Part 9: The AAA
- Part 10: The Roosevelt Recession
- Part 11: The Roosevelt Recession, Continued
- Part 12: Fear Itself
- Part 13: Fear Itself (Continued)
- Part 14: Fear Itself (Conclusion)
- Part 15: The Keynesian Myth
- Part 16: The Keynesian Myth, Continued
- Part 17: The Keynesian Myth, Concluded
 According to the Commercial and Financial Chronicle for April 8th, 1933, while foreigners' fear that the U.S. would be forced off the gold standard was the proximate cause of heavy gold losses the Federal Reserve Bank of New York suffered during the first days of March, and on Friday, March 3rd especially, those losses were in turn "largely responsible for the domestic distrust which came concurrently and so greatly added to the home propensity to hoard." Based on a painstaking econometric study of the role of fundamentals versus panic-based withdrawals in depression-era banking crises, Calomiris and Mason (2003) find the crisis that began in January 1933 was the only one in which a nationwide "panic" appears to have played an important role. However, they note that what their model identifies as a "panic" effect may actually be fear of devaluation that the Chronicle refers to. That Calomiris and Mason also find some evidence of a nationwide "panic" in the fall of 1931, when Britain's suspension of the gold standard also inspired a run on the dollar, lends further credence to this possibility. (The month of October 1931 witnessed 522 bank suspension—a record that held until March 1933.) Calomiris and Mason don't mention the possibility that fear of state bank holidays may also have contributed to the 1933 crisis.
 Fuller (Drifting, chapter 12) refers to numerous examples of such banks.
 Although state governors technically lacked the authority to close national (as opposed to state-chartered or private) banks, that problem had been separately dealt by a February, 25th Joint Resolution allowing the Comptroller of the Currency to "exercise to such extent as he deems advisable with respect to any national banking association any powers which the State officials having supervision of State banks, savings banks and/or trust companies."
 Concerning the Hoover-FDR stalemate, see Fuller chapter 13, Susan Kennedy, The Banking Crisis of 1933, pp. 135-151; and Jonathan Alter, The Defining Moment, chapter 26. Because Alter's book is generally very sympathetic to FDR, his conclusion (p. 181) that "it is hard to avoid the conclusion that he intentionally allowed the economy to sink lower so that he could enter the presidency in a more dramatic fashion" is especially telling.
 In a very interesting working paper of which I became aware only after this essay was nearly complete, Peter Conti-Brown and Sean H. Vanatta argue that, despite their limited numbers and the short period of time available to them, the bank examiners' efforts were actually crucial to the bank holiday's success, because in identifying and refusing to open unsound banks, they gave credibility to FDR's claim that reopened ones were indeed "sound." I agree that the bank examiners played a more important role than is often recognized; and Conti-Brown and Vanatta argue this much compellingly. I suspect nonetheless that they exaggerate the extent to which the public had lost confidence in banks closed by state holidays rather than because examiners concluded that they were financially unsound. There were, no doubt, depositors who needed to be reassured before returning to such banks. But there were also many who never doubted that their banks were sound. That many proved to be sound in fact made examiners' task less fraught than it might have been otherwise. Their task was nonetheless delicate, for it fell on them to limit the Treasury's exposure to losses it might otherwise have incurred by having promised to indemnify the Fed for any loss it incurred in guaranteeing the deposits of reopened banks.
 The lone exception, Harriman National Bank, went into receivership in October 1933. Joseph Harriman, its chairman and former president, had by then been arraigned for fraud, for which he was found guilty and sentenced to several years in prison in 1934.