The Reconstruction Finance Corporation, the gold standard, and the banking panic of 1933.

[Publication of RFC loans] was the most damnable and vicious thing that was ever done. It counteracted all the good we had been able to do.

Atlee Pomerene, Chairman of the RFC

1. Introduction

By the summer of 1932 the United States banking system had been severely battered by three years of depression, recurring banking panics, and a run on the dollar following Britain's departure from gold. Depression and price deflation depreciated asset values, rendering bank portfolios illiquid. The establishment of the Reconstruction Finance Corporation in February 1932 afforded relief to the beleaguered banking system. The upturn in industrial production in August gave evidence of an incipient recovery.

However, the recovery was short-lived. Industrial production peaked in October, contracting thereafter. The stability provided by RFC lending was undermined by publication of the names of banks receiving new RFC loans, beginning in August, and culminated with publication of the names of all banks receiving RFC loans in January 1933.

Concomitant with the economic downturn was an increase in bank suspensions, leading to the final crisis and bank holiday in March 1933. Every discussion of the Banking Panic of 1933 attributes the crisis to one or more specific causes. Einzig (1933), Hodsen (1938), James (1938),(1) Kennedy (1973), Cleveland and Huertas (1985), and Keehn and Smiley (1988, 1993) blame the run on banks on the publication of the names of banks receiving loans from the Reconstruction Finance Corporation. Sullivan (1936), Friedman and Schwartz (1963), and Yeager (1966) cite the publication of RFC loan lists as well as the ultimately well-found fear that Franklin Roosevelt would devalue the dollar upon assuming the presidency as possible causes of the final panic of the Great Depression. Donaldson (1992) and Wigmore (1987) argue that a run on the Fed's gold reserve through the banking system forced the bank holiday. Wicker (1996) blames the crisis on the declaration of a bank holiday in Michigan in February 1933. However, none of these studies presents tests to distinguish among alternative explanations.

Evidence presented in this paper demonstrates that neither general fear of devaluation nor the publicity about the RFC loans ignited the final crisis. Although publication of the identity of banks receiving RFC loans did increase failures, the increase in failures is small relative to the number of banks identified as recipients of RFC loans.

Similarly, the daily behavior of the major dollar exchange rates shows no evidence of a speculative attack on the dollar until after the declaration of a banking holiday in Michigan. Several short-lived depreciations of the dollar resulting from political events and local banking difficulties are evident in the exchange rate data, indicating that events affecting confidence in the dollar did affect exchange rates. However, the dollar strengthened relative to the franc following Roosevelt's election and remained strong until the Michigan crisis. Banking difficulties in Illinois, Nevada, and Iowa had minor transitory effects on the value of the dollar, whereas a one-day "business holiday" in Louisiana did not weaken the dollar. However, the Michigan holiday, closing the major banks in a major industrial center, and the ensuing series of bank holidays resulted in a sharp depression of the dollar. The declaration of bank holidays in state after state increased uncertainty about the stability of the banking system. These events aroused fears that the situation in the United States compared with the banking crises that spread through Europe during the summer of 1931, driving country after country off gold.(2) Although the Michigan crisis centered around a loan request made to the RFC, the real cause was a clash of egos between old rivals.

Section 2 of this paper examines the effect of publication of RFC loans to banks on bank suspensions during the final months of 1932 and the first months of 1933. Section 3 reviews daily exchange rate data for evidence of a speculative attack on the dollar. Section 4 compares the Michigan crisis with earlier crises? Section 5 summarizes the findings of the paper.

2. The Reconstruction Finance Corporation and the Final Banking Panic

Butkiewicz (1995) has demonstrated that RFC loans to banks reduced bank suspensions prior to the publication of the names of banks receiving loans in late August 1932.(4) The effectiveness of RFC lending decreases at that point, due possibly to loss of depositor confidence resulting from publicity and/or from banks' reluctance to borrow from the RFC due to a fear of adverse effects of publication. The number and dollar value of bank loan applications to the RFC declined substantially in the months following the beginning of publication of the names of banks receiving RFC loans,(5) which indicates banks were reluctant to borrow.

If publication caused bank runs, then bank suspensions should be higher in states where a high percentage of banks are identified as receiving RFC loans. Keehn and Smiley (1988, 1993) compare RFC lending records with the banks that failed between March 1932 and June 1933. Comparing the publication lists with the names of banks which suspended activity, they find that publication may have been a factor in 264 bank suspensions between March 1, 1932, and March 4, 1933. Although they compare the list of banks identified as receiving RFC loans with lists of suspended banks, it is possible to assess the effect of the publication of the loan lists econometrically, which controls for other possible causes of bank suspensions. Also, publication may have resulted in suspensions of banks not borrowing from the RFC, as bank runs spread from borrowing banks to other local banks.(6,7) Finally, public confidence in banks in general may have been eroded when local banks were identified as having borrowed from the RFC.

In modeling bank suspensions during the interwar period, several factors have been established as determinants of suspensions. Previous research by Friedman and Schwartz (1963), Temin (1976), McCallum (1990), and Butkiewicz (1995) found bank suspensions to be related to changes in either aggregate or agricultural income and a continuation or persistence of failures, typically proxied by previous suspensions. The basic model relates suspensions to changes in agricultural income and to previous suspensions. A measure of the extent of publicity regarding RFC loans is added to this basic model.

First, two simple cross-sectional econometric models of bank suspensions by state are used to assess the effect of publication of RFC loan recipients. The models are:

[Mathematical Expression Omitted] (1)

and

[Mathematical Expression Omitted], (2)

where:

[PS.sub.it] = Number of banks suspending operations in a state, i, in a given month, t, as a percentage of the number of banks in operation in a state on December 31, 1931.

[CPS.sub.it] = Cumulative number of banks suspending operations in a state, i, in a given month, t, as a percentage of the number of banks in operation in a state on December 31, 1931. Cumulative totals are from October 1932 (September and October) through February 1933 (September 1932 through February 1933).

%[Delta][FI.sub.i] = Percent change in nominal farm income in a state from 1929 through 1932.

[PVS.sub.i] = Bank suspensions in state i from January through August 1932 as a percentage of banks in operation in state i on December 31, 1931.

[Pub.sub.i, t-1] = Cumulative number of banks listed as receiving RFC loans in state i in all months prior to the current month, t, as a percentage of banks in operation in state i on December 31, 1931.

The numerators of the dependent variables are the number of banks suspending operations in a given state for each month, September 1932 through February 1933, and the cumulative number of failures in a state in each month, beginning in September 1932. The first monthly cumulative total is September and October, and each subsequent month's failures are added to the previous total.(8) The denominator is the number of banks in operation in each state on December 31, 1931.

In earlier empirical work (Butkiewicz 1995), previous suspensions and a measure of farm income were found to significantly affect bank failures. The farm income variable used here is the percent change in nominal farm income in each state from 1929 through 1932. Nominal income is used since farmers were adversely affected by both price declines and loss of real income. The previous suspension variable is the total number of bank suspensions in a state from January 1932 through August 1932 as a percentage of the number of banks in the state on December 31, 1931.

The publicity variable is the cumulated number of banks in a state listed as receiving RFC loans in a given month, as a percentage of banks in the state on December 31, 1931. Because some banks borrowed from the RFC in more than one month, cumulation double-counts these banks. However, repeated borrowing could have been interpreted as indicative of continuing problems for a bank, increasing the likelihood of a run. For September, the variable is the number of banks reported in the August 23 newspaper report. For October, the number of banks reported in the October 8 list is added to the previous total. For November, the October 23 list is added to the previous total. The December variable adds the November 29 list, the January variable includes the number of banks in the December 23 report, and the February variable adds the January 27 lists to the previous totals.

The list of RFC loans published in August was short; only a 10-day time period was covered, so any effect in September might not be evident in the data.(9) The lists published in January were the most extensive, including all loans made in December and in the period February 2 through July 21, 1932.(10) However, the effect in February is not expected to be large because most states declared bank holidays or restricted withdrawals to forestall panics.

The results of the estimated bank failures models are reported in Table 1. Because preliminary estimates indicated evidence of heteroskedasticity, all equations are estimated using the White correction to obtain heteroskedasticity-consistent standard errors and covariances. Panel A contains the results for the monthly percentage of failures by state, and panel B contains results for the cumulated percentage of failures.

[TABULAR DATA FOR TABLE 1 OMITTED]

In panel A, the farm income variable is significant at the 1% level in two of six months. The other variables are insignificant at conventional levels in all months. In panel B, the farm income variable is again significant in two months, while the other variables are insignificant in all months.

The insignificance of the publicity variable in all months fails to support the hypothesis that publication caused a large number of suspensions. Clearly, there are documented instances of silent runs. However, in the period August 1932 through January 1933, more than 3300 banks were identified as borrowing from the RFC, many on multiple occasions. Yet in the period September 1932 through February 1933, only 795 banks suspended operation, compared with 662 suspensions from February through August 1932, the first seven months of RFC operations.(11) Olson (1977) and Keehn and Smiley (1988, 1993) reported that approximately 10% of banks receiving RFC loans ultimately failed.

The results of cross-section estimates presented above do not provide much insight into the cause of bank failures. Data limitations make state-level estimation difficult. Two of the independent variables, farm income and previous suspensions, are the same in all months. Monthly state-level income data does not exist for this period.

An alternative is to estimate a time-series model of failures. A reason to investigate the [TABULAR DATA FOR TABLE 2 OMITTED] effect of publication at the aggregate level is that contagion effects are not limited by state boundaries. If publication of RFC loan lists aroused fear about the safety of banks in one state, fear likely spread to neighboring states.

A time-series model of bank failures used by Butkiewicz (1995) to investigate the effect of RFC lending is

[Mathematical Expression Omitted] (3)

where

[F.sub.t] = bank suspensions in a month as a percentage of banks in operation during the quarter.

[LIFP.sub.t] = the natural logarithm of a farm price index.

[RFC.sub.t] = net RFC lending to banks and trust companies as a percentage of total commercial bank demand and time deposits in a given month.

Two different publicity variables are added to the failures model:

PUBM = number of commercial banks listed in the monthly report as a percentage of the number of commercial banks in the quarter.

PUBC = cumulative number of commercial banks listed in all monthly reports to date as a percentage of the number of commercial banks in the quarter.

The time-series model is estimated for the period 1921:02-1933:02. The estimates are reported in Table 2. As expected, failures are inversely related to the farm price index and RFC lending. Failures are positively related to previous suspensions. The estimated coefficients for all of these variables are statistically significant.

The estimated effects of the two publicity variables are statistically insignificant. The estimated coefficient for the monthly publicity variable has an incorrect sign. Thus, time-series data also fail to provide evidence that publication of RFC loan lists increased bank failures.(12)

Why was the effect of publication so small? Why wasn't the number of silent runs greater, resulting in a much larger number of suspensions? The results presented here fail to support the belief that publication caused a significant increase in bank suspensions. Thus, the most likely harm resulting from publication was that it increased banks' reluctance to borrow from the RFC, even though these banks might have benefitted from the loans.

Table 3. U.S. Monetary Gold Stock (Millions of $)

1931        $      1932      $      1932        $      1933      $

Sept. 2    4712   Jan. 6    4171   July 6      3635   Jan. 4    4237
Sept. 9    4714   Jan. 13   4170   July 13     3645   Jan. 11   4262
Sept. 16   4729   Jan. 20   4164   July 20     3665   Jan. 18   4279
Sept. 23   4610   Jan. 27   4139   July 27     3674   Jan. 25   4269
Sept. 30   4454
                  Feb. 3    4119    Aug. 3     3700   Feb. 1    4261
Oct. 7     4355   Feb. 10   4116    Aug. 10    3718   Feb. 8    4248
Oct. 14    4137   Feb. 17   4082    Aug. 17    3759   Feb. 15   4224
Oct. 21    4050   Feb. 24   4063    Aug. 24    3777   Feb. 22   4173
Oct. 28    4002                     Aug. 31    3801
                  Mar. 2    4063    Mar. 1     4056
Nov. 4     4025   Mar. 9    4075    Sept. 7    3818   Mar. 8    3956
Nov. 11    4060   Mar. 16   4087    Sept. 14   3841   Mar. 15   3964
Nov. 18    4084   Mar. 23   4094    Sept. 21   3878   Mar. 22   3977
Nov. 25    4121   Mar. 30   4101    Sept. 28   3897   Mar. 29   3985
Dec. 2     4133   Apr. 6    4109    Oct. 5     3914   Apr. 5    3996
Dec. 9     4153   Apr. 13   4093    Oct. 12    3921   Apr. 12   4006
Dec. 16    4171   Apr. 20   4090    Oct. 19    3940   Apr. 19   4026
Dec. 23    4178   Apr. 27   4081    Oct. 26    3970   Apr. 26   4023
Dec. 30    4171
                  May 4     4057    Nov. 2     3979
                  May 11    4027    Nov. 9     3983
                  May 18    3988    Nov. 16    3996
                  May 25    3920    Nov. 23    4032
                                    Nov. 30    4053
                  June 1    3820
                  June 8    3692    Dec. 7     4065
                  June 15   3622    Dec. 14    4081
                  June 22   3630    Dec. 21    4200
                  June 29   3633    Dec. 28    4218

Source: Banking and Monetary Statistics, 1914-1941 (Board of
Governors of Federal Reserve System 1943, pp. 368-9).

3. Speculative Run on the Gold-Backed Dollar

The other oft-cited reason for the final banking crisis is that rumors that president-elect Roosevelt would devalue the dollar caused a speculative run on the dollar through the banking system. However, exchange rate behavior, especially behavior of the dollar-franc exchange rate, shows no evidence of a run on the dollar until after the Michigan crisis. Before examining exchange rate behavior, the fluctuations in U.S. monetary gold holdings during this period are discussed.

A run on the dollar began following Britain's suspension of the gold standard in September 1931 and continued through June 1932. After June, the U.S. gold position strengthened. Table 3 reports the U.S. monetary gold stock(13) for weekly reporting dates (Wednesdays) from September 1931 through April 1933. The gold drain following Britain's suspension of the gold standard is readily apparent. The dollar recovered beginning in June 1932, and gold holdings peaked on January 18, 1933. During the following four weeks, total gold losses were $55 million. During the next three weeks, following the Michigan crisis, total gold losses were $268 million. The gold drain accelerated during the week of the Michigan bank holiday, which began on Tuesday, February 14.(14)

The gold drain beginning in the third week of January can be attributed to the strengthening of the British pound. At this time, the pound was floating, and Britain enjoyed a balance of payments surplus. The Bank of England chose to accumulate gold rather than let the pound appreciate.(15) In December 1932, Britain made a war debt payment to the United States, significantly reducing its gold holdings. During the next two months, Britain increased its gold holdings to a level above the amount held prior to its debt payment in December.

Monthly gold holdings of the central banks of England and France, and end of the month holdings for the United States are listed in Table 4 for 1932 and 1933. During the run on the dollar in 1932, both England and France gained gold. France began losing gold in December 1932, while the U.S. gold position continued to strengthen. In an external speculative attack on the dollar, France, the heart of the European Gold Block, should have gained gold, as was the case during the previous run on the dollar. The international gold flows, especially the gold drain from France in the last half of 1932, is not consistent with the hypothesis of a run on the dollar during the presidential interregnum.

It is possible, but unlikely, that a gold drain was strictly internal. A significant internal drain would certainly have precipitated a simultaneous external drain. This is evident by the sensitivity of the foreign exchange markets to rumors that the dollar might be devalued.(16) Figures 1, 2, and 3 graph the daily dollar price of the pound, franc, and reichsmark, respectively, from June 1, 1932, through March 3, 1933. Several events did unsettle the foreign exchange market in the months prior to the crisis. In a campaign speech in Des Moines, Iowa, on October 4, 1932, President Hoover claimed that his leadership had saved the gold standard, as the U.S. was within two weeks of leaving gold (New York Times, October 5, 1932, pp. 1, 18). Although subsequent reports indicated the gold standard was never in jeopardy and could not even identify the month to which Hoover referred,17 the dollar-franc exchange rate depreciated markedly at this time, moving from the gold import point to the gold export point.(18) No move is apparent in either the dollar-pound or dollar-reichsmark exchange rate, as neither of these currencies was tied to gold at this time. This date, October 5, 1932, is indicated by the second solid vertical line labeled "Hoover Speech" in all three figures.

On January 3, 1933, Senator Borah of Idaho revealed plans to introduce legislation to devalue the dollar. The pound and franc strengthened slightly during the next two days. The [TABULAR DATA FOR TABLE 4 OMITTED] press reported a one-day "run" on the dollar.(19) After the Borah incident subsided, the dollar-franc exchange rate hovered close to the U.S. gold import point of 3.9024 cents per franc until the Michigan bank crisis began. The Borah incident, occurring on January 3, 1933, is marked by the "Borah" solid vertical line in each figure.

The behavior of the dollar-franc exchange rate provides a good indication of fears of devaluation, especially as evidenced by the Hoover speech in October. That the dollar remained strong relative to the franc after the November election is inconsistent with the argument that fears of devaluation by the new president caused a run on the dollar.(20) The strength of the pound can be attributed to the improvement of Britain's international position following the suspension of the gold standard in 1931, rather than a run on the dollar.

The dollar depreciated markedly relative to the franc and reichsmark immediately after the bank holiday was declared in Michigan.(21) This date, February 14, 1933, is marked by the "Michigan" vertical line in each figure. The fall of the dollar relative to the reichsmark is especially noteworthy, as this exchange rate had shown little movement in either direction for months. Just as banking crises had ended in suspensions of the gold standard in Austria, Hungary, Germany, and Britain in 1931, the Michigan crisis fueled fears that the dollar now faced a similar fate.

The failure of the dollar to depreciate relative to the pound following the Michigan crisis is the result of intervention by the British. The British established the Exchange Equalisation Account in April 1932. Its goals were to smooth exchange rate fluctuations resulting from (i) seasonal variations, (ii) speculation, and (iii) capital flight (Waight 1939). Stabilization was at first achieved through the purchase of gold currencies, but in early 1933, the British bought gold directly to avoid exchange rate risk (Einzig 1935, p. 241). Einzig (1933, p. 66) argues that whereas the pound was held at $3.45 at the time of the Michigan crisis, it would have appreciated to between $3.80 and $4.00 if the Exchange Equalisation Account had not been used to stabilize the currency. Thus, the behavior of the dollar-pound exchange rate following the Michigan crisis reflects a policy of a managed float, and not market forces.

4. The Banking Crisis of 1933

Friedman and Schwartz (1963, p. 324) argue that the final banking crisis of the depression began in Western states in late 1932 and spread East. However, the evidence presented here indicates that there was no developing crisis until the Michigan holiday in February 1933. The evidence supporting this alternative view includes the behavior of the dollar-franc exchange rate, the currency/demand deposit ratio, and the Federal Reserve's measures of money in circulation(22) and member bank excess reserves. The events leading to the Michigan crisis are discussed, followed by an examination of the evidence.

A banking crisis spread throughout the United States following Britain's suspension of the gold standard. As the panic subsided and the RFC began operations in early 1932, the banking situation stabilized. Relative calm in the banking sector characterized 1932. An exception was the banking crisis in Illinois. Bank runs, which began in the Chicago suburbs, hit the "loop" banks on June 22, 1932. The dollar did depreciate moderately relative to the franc at this tame, as indicated by the "Chicago" line in Figure 2. However, the appreciation of the dollar resumed within a week.

The next crisis was the declaration of a bank holiday in Nevada on October 31, 1932 (Nadler and Bogen 1933). The holiday was sparked by the financial problems of the Wingfield group of banks comprising 12 of the 25 banks in the state. This problem was followed by a brief depreciation of the dollar relative to the franc, which was reversed almost immediately.

On January 20, 1933, the Iowa General Assembly passed a law authorizing the state superintendent of banking to operate any bank without placing it in receivership (New York Times, January 21, 1933, p. 23). Iowa had 26 bank suspensions in December 1932, 21 in January 1933, but only three in February 1933. This legislative action forestalled an incipient panic. Again, the dollar experienced a one-day depreciation.

The one-day Louisiana holiday was a bizarre aberration, sparked by controversy over RFC loans.(23) Representative Hamilton Fish of New York, in testimony before the House Banking Committee, called for an investigation of bad loans made by the RFC. On February 3, 1933, newspapers carried reports of his criticism of Rudolph Hecht, president of Hibernia Bank and Trust Company of New Orleans. Hecht also served as chairman of the RFC's regional advisory committee. The RFC made $4 million of loans to the Union Indemnity Company of New Orleans and the Union Title Company. Hecht also served as a director of Union Indemnity. The loan proceeds were used to repay loans from the Hibernia Bank, and Union Indemnity subsequently failed. Fish charged Hecht with conflict of interest. Hecht responded that he had abstained from voting on the loans to the indemnity company, and Fish ultimately dropped his call for an investigation.

The news reports started a "wire-run" on Hibernia on Friday, February 3, by out-of-town depositors. The story was kept out of the New Orleans newspapers, so there was no local run. Senator Huey Long and Governor Allen came to Hibernia's rescue. An RFC loan for $20 million was arranged but could not be transferred until Monday, February 6. Lacking any other justification, Governor Allen declared a "business holiday" on Saturday, February 4, in commemoration of the sixteenth anniversary of the severance of diplomatic relations with Germany.

By Sunday, the RFC loan was announced,(24) subject to an agreement that all customers with deposits greater than $10,000 freeze one-half of their balances. The bank also raised $4 million from other sources. On Monday, February 6, New Orleans banks opened early and closed late, restoring public confidence.(25)

In the Chicago crisis, the major loop banks survived the runs. The Nevada holiday involved an economically unimportant western state, with no consequence for the rest of the country. The legislative action in Iowa was designed to keep banks open. The Louisiana holiday occurred on a weekend, and was resolved by Monday. These four events did not affect banks in other localities. In all cases, the dollar depreciated briefly, but recovered. Each time, the depreciation did not move the dollar even close to the gold export point, and, in the case of the New Orleans crisis, the quarter-cent decline of the exchange rate is inconsequential.

Conversely, the Michigan crisis resulted in the declaration of a week-long holiday and quickly spread to other states. The details of the Michigan crisis have been recounted by Kennedy (1973), Wigmore (1985), and in Stock Exchange Practices (U.S. Senate Committee on Banking and Currency 1934). A brief summary of the crisis based on these accounts and the New York Times and Detroit Free Press reports 26 is presented here.

Two major bank groups, the Guardian Detroit Union Group and the Detroit Bankers Company, dominated the Detroit banking business. One of the Guardian banks, Union Guardian Trust, had sustained substantial losses on its security and real estate loans. The Guardian Group was attempting to pay all depositors of the Union Guardian Trust, terminate its deposit business, and limit the bank's activities to the trust business. The bank had sustained deposit losses since the first of the year, but there were no reports in the local press of a run, and the stock price of the parent group traded in a narrow range for several months through the last business day before the holiday (Table 5).

Prior to the crisis, officials of the Guardian Group approached the RFC for $60 million to pay all depositors. The RFC was apparently willing to make the loan even though all good collateral had been used to secure previous loans. The RFC hoped to avoid a crisis, as loans to Chicago banks had done previously in June 1932.

Michigan Senator James Couzens chaired a committee investigating bad RFC loans. Couzens was once Henry Ford's partner in the auto business, but the men were now rivals. The Guardian banks were known to be the Fords' banks, and Henry Ford had $7 million of deposits in Union Guardian Trust. Couzens demanded that Ford subordinate his deposits as a condition of the RFC loan. Ford not only refused, but threatened to withdraw his $20 million in deposits from First National Bank, the lead bank in the Detroit Bankers Company group. Fearful of a collapse of both major Detroit banking groups and unable to find a workable alternative solution, Governor Comstock proclaimed a one-week, statewide bank holiday on February 14, 1933, the first business day of the week.

The lead article in the Wednesday, February 15, 1933, issue of the Detroit Free Press blames the conflict between Couzens and Ford for the failure to resolve the crisis. The front page also carries a prominent picture titled "Dramatis Personae" showing Couzens and Ford in an obvious face off. Subsequent accounts of the crisis by Sullivan (1936), Kennedy (1973), Wigmore (1985), and Wicker (1996) all emphasize the role this personal conflict played in precluding a resolution to the problems in Detroit.

What was different about the Detroit crisis that led to a nationwide crisis when previous problems in Chicago, Nevada, Iowa, and New Orleans were contained? Nevada and Iowa were rural or agricultural states, with no industrial centers. The entire interwar period was plagued by banking problems in agricultural and rural areas. Banking problems in these areas were common and not a cause for concern in urban and industrial centers. Chicago and New Orleans were major urban/industrial centers. However, the RFC was able to forestall a general crisis in both of these episodes.

Detroit was a major urban and industrial center and the home of the automobile industry. The failure of the RFC to resolve the crisis was a major blow and increased the perceived likelihood that similar problems could occur in other industrial centers.(27) In the cases of Chicago and New Orleans, the crises were resolved with no interruption of banking operations. The Detroit crisis resulted in the declaration of a state-wide bank holiday. The crisis spread to adjacent states, which in turn also declared bank holidays. The resulting deterioration of both domestic and international confidence in the U.S. banking system is evident in the behavior of a number of economic series.

The international loss of confidence is evident in the behavior of exchange rates and central bank gold holdings. The dollar depreciated rapidly against the franc and reichsmark during the week following the Michigan holiday and experienced a slight weakening relative to the pound. The New York Federal Reserve Bank experienced a significant gold drain late in the week. Table 6 lists the daily change in the New York Fed's gold holdings as reported in the New York Times. The New York Fed lost $25 million of gold reserves in the week ending February 11. This gold drain was attributed to British efforts to accumulate gold in the Exchange Equalisation Account, rather than allow the pound to appreciate. During the holiday-shortened week of the Michigan bank crisis, the New York Fed lost $38 million in gold,(28) and U.S. monetary gold holdings fell by $51 million.

Table 5. Daily Closing Stock Prices of Major Detroit Bank Groups

Date            Guardian Group           Detroit Bankers

1932

12/19               7.00                      13.50
12/20               6.75                      12.75
12/21               6.875                     13.00
12/22               6.375                     13.00
12/23               6.50                      13.00
12/24               6.625                     13.00
12/25              Christmas/Sunday - Markets closed
12/26                 Holiday - Markets closed
12/27               6.50                      13.00
12/28               6.625                     12.875
12/29               6.875                     12.625
12/30               7.875                     13.50
12/31               8.00                      13.75

1933

1/1                New Year's/Sunday - Markets closed
1/2                     Holiday - Markets closed
1/3                 7.75                      13.75
1/4                 8.50                      14.75
1/5                 8.50                      14.00
1/6                 8.375                     14.00
1/7            In memory of Calvin Coolidge - Markets closed
1/8                       Sunday - Markets closed
1/9                 8.50                      14.00
1/10                8.25                      14.00
1/11                8.125                     14.125
1/12                7.875                     13.75
1/13                7.625                     14.00
1/14                7.625                     14.00
1/15                    Sunday - Markets closed
1/16                7.625                     13.625
1/17                7.375                     13.50
1/18                7.50                      13.50
1/19                7.50                      13.25
1/20                7.50                      13.00
1/21                7.375                     13.25
1/22                    Sunday - Markets closed
1/23                7.375                     13.625
1/24                7.375                     13.50
1/25                7.50                      13.875
1/26                7.50                      13.875
1/27                7.375                     13.50
1/28                7.375                     13.375
1/29                    Sunday - Markets closed
1/30                7.25                      13.375
1/31                7.25                      13.50
2/1                 7.375                     13.25
2/2                 7.00                      13.125
2/3                 7.125                     12.875
2/4                 7.25                      12.625
2/5                     Sunday - Markets closed
2/6                 7.125                     12.125
2/7                 6.875                     11.50
2/8                 6.375                     11.625
2/9                 7.00                      12.00
2/10                6.875                     11.875
2/11                7.00                      11.875
2/12                    Sunday - Markets closed
2/13                    Holiday - Markets closed

In the days following the Michigan holiday, some banks in the state remained open for business. However, it is evident the problems quickly spread from Michigan to adjacent states. By February 17, the governor of Ohio was denying that a holiday was needed in his state (New York Times, February 18, 1933, p. 5). By the week of February 28, the New York Times devoted a page each day to summarize the banking situation. States either declared holidays or restricted withdrawals to 5% of deposits. By Saturday, March 4, the day Roosevelt was inaugurated, all states, including New York and Illinois, had holidays or restrictions, and London suspended its dollar foreign exchange market.(29)

Table 6. Change in Federal Reserve Bank of New York's Gold Holdings
Before and After Michigan Crisis

Week ending February 11, 1933                -$25,000,000
February 13, 1933                                 Holiday
February 14, 1933                            -$10,482,500
February 15, 1933                             +$2,176,700
February 16, 1933                             -$7,607,000
February 17, 1933                            -$10,099,700
February 18, 1933                            -$12,413,800
Net loss for week, February 14-18            -$38,426,300

Source: Daily issues of the New York Times.
Table 7. Ratio of Currency to Demand Deposits

Month        1929     1930     1931     1932     1933

January      0.172    0.171    0.184    0.295    0.318
February     0.172    0.169    0.183    0.293    0.388
March        0.174    0.164    0.185    0.290
April        0.172    0.165    0.191    0.295
May          0.175    0.171    0.195    0.301
June         0.176    0.170    0.201    0.320
July         0.171    0.169    0.206    0.334
August       0.174    0.173    0.217    0.328
September    0.169    0.170    0.225    0.324
October      0.157    0.168    0.250    0.316
November     0.178    0.172    0.252    0.308
December     0.168    0.180    0.266    0.311

Source: A Monetary History of the United States, 1867-1960
(Friedman and Schwartz 1963).

Wicker (1996), who also identifies the Michigan crisis as the cause of the final crisis, argues that the declaration of bank holidays was the propagation mechanism. That is, the declaration of a holiday in one state resulted in a deterioration of public confidence and a perceived increase in the likelihood of a holiday in surrounding states.

Wigmore (1987) attributes the crisis to a run on the dollar and specifically on the drain of gold reserves from the Federal Reserve Bank of New York. He states that the run on the dollar began in the last half of February (p. 754), but acknowledges that "foreign deposit withdrawals and demand for gold were partly related to the surge in state bank holidays" (p. 744).

It is likely that all of these factors are interrelated. At this time, three years of depression had severely weakened the banking system. Financial crises had already driven several nations off the gold standard. Crises in Chicago, Nevada, Iowa, and New Orleans did not create a crises in public confidence. The declaration of a state-wide bank holiday in Michigan did spark a crisis of confidence. As successive holidays were declared in adjacent states, the likelihood that the United States would be forced from the gold standard increased. Internal and external gold drains were a rational response to expectations of devaluation. The runs on banks and the run on the dollar were both manifestations of a lack of confidence in the battered financial system.

The deterioration of public confidence following the Michigan holiday is evident in the behavior of several data series. The currency/demand deposit ratio is listed in Table 7 for the period January 1929 through February 1933. This ratio shows a marked increase at the time of the stock market crash and at each of the first three crises noted by Friedman and Schwartz (1963). At the time of the stock market crash, from September to November 1929, the currency/demand deposit ratio increased 5.3%. The currency ratio increased 9.5% from October 1930 through January 1931, the time of the first banking crisis. The ratio increased 9.8% from February through June 1931, the time of the second banking crisis. The increase from August 1931 through January 1932, following Britain's departure from gold, is 36%. The currency ratio even increased 11% from May through July 1932, the time of the Chicago crisis.

Although the currency ratio increased during each crisis noted above, it fell from October to November 1932 after the Nevada holiday was announced on October 31, 1932. From the November low through January 1933, the currency ratio did increase by 3.2%. This increase is smaller than occurred during any previous crisis. If measured from October 1932 through January 1933, there is an increase of less than 1%, and the ratio is lower in January 1933 than in September 1932. This hardly seems consistent with a mounting banking crisis. However, from January through February 1933, the currency ratio increased 22%. This is consistent with the hypothesis that the Michigan crisis was the spark that ignited the final panic.

A similar pattern is evident in the Federal Reserve data on money in circulation and estimated member bank excess reserves. Money in circulation includes currency in the hands of the public and vault cash in banks. Member bank excess reserves are estimated by the Fed. These weekly data are reported in Table 8 for the period beginning January 1932 through the end of March 1933.

Prior to June 1932, member bank excess reserves were growing, and money in circulation was stable. The Chicago crisis began in June 1932. The crisis is evident in both the increase of money in circulation and the drop of excess reserves. Following the Chicago crisis, member bank excess reserves resumed, increasing steadily.

Money in circulation actually fell from mid-December 1932 through mid-January 1933. Similarly, excess reserves increased during the same span. This is not indicative of a growing crisis.(30) Money in circulation increased by 1.7%, and excess reserves were stable on the two reporting dates prior to the Michigan holiday. Money in circulation increased by 2.75% and excess reserves fell by 32% on the first reporting date following the Michigan holiday,(31) with similar dramatic changes reported in each subsequent week.

That the loss of public confidence is due to the Detroit crisis is evident in the distribution of the changes in Federal Reserve notes in circulation in each Federal Reserve district. The changes in weekly reported Federal Reserve notes in circulation for the system total and for the Chicago District, which contains the state of Michigan, are reported in Table 9 for the weeks of July 6, 1932, through March 29, 1933.

Increases in Federal Reserve notes in circulation become apparent in the weeks preceding the Detroit crisis. In the three-week period from January 26 through February 15, the increase in notes in circulation is $185 million. Of this increase, 29% occurred in the Chicago district, and 42% of the total increase was in the Chicago and Cleveland districts. In the first full week following the Michigan holiday, total notes increased $109 million. Of this increase, 52.9% came in the Chicago district. In the next two reporting weeks, Chicago accounted for 20.2% and 30.4% of the total change. During these weeks the crisis was spreading throughout the nation, so a decline in the percent increase in the Chicago district is expected. The absolute change in notes in circulation in the Chicago district in each of these three weeks increases. The changes are $57.7 million, $116.8 million, and $193 million, respectively.(32)

The widespread and rapid deterioration of the U.S. financial system following the Michigan crisis is evident in a number of data series. The dollar depreciated relative to both the franc and the reichsmark. The New York Fed experienced sizable gold losses, and total Federal Reserve System gold reserves declined by an even larger amount. Currency in circulation increased, and bank excess reserves declined. The increase in Federal Reserve notes in circulation originated in the Chicago Federal Reserve district, which contained the state of Michigan. That the behavior of all these series changed at the time of the Michigan crisis supports the hypothesis that the collapse of the banking system in the city of Detroit sparked the final banking panic of the Great Depression.

Table 8. Money in Circulation and Member Bank Excess Reserves
Weekly Reporting Dates, January 1932 through March 1932 (Millions
of $)

                                              Excess Reserves
Date                Money in Circulation        (Estimated)

1932

January 6                  5374                     62
January 13                 5332                     47
January 20                 5326                     48
January 27                 5302                     34
February 3                 5344                     56
February 10                5338                     35
February 17                5322                     43
February 24                5306                     27
March 2                    5296                     52
March 9                    5258                     69
March 16                   5235                     58
March 23                   5193                     84
March 30                   5152                     81
April 6                    5171                    110
April 13                   5136                    175
April 20                   5138                    128
April 27                   5111                    259
May 4                      5162                    289
May 11                     5144                    281
May 18                     5162                    331
May 25                     5124                    354
June 1                     5180                    270
June 8                     5165                    279
June 15                    5180                    274
June 22                    5218                    249
June 29                    5362                    220
July 6                     5488                    162
July 13                    5427                    216
July 20                    5448                    247
July 27                    5403                    282
August 3                   5441                    213
August 10                  5420                    267
August 17                  5419                    277
August 24                  5397                    337
August 31                  5405                    330
September 7                5438                    323
September 14               5381                    403
September 21               5355                    374
September 28               5318                    428
October 5                  5362                    437
October 12                 5364                    399
October 19                 5334                    461
October 26                 5297                    537
November 2                 5329                    494
November 9                 5364                    455
November 16                5342                    502
November 23                5348                    493
November 30                5361                    498
December 7                 5382                    485
December 14                5377                    517
December 21                5443                    525
December 28                5400                    554

1933

January 4                  5385                    582
January 11                 5302                    627
January 18                 5315                    609
January 25                 5324                    573
February 1                 5365                    499
February 8                 5418                    501
February 15                5567                    340
February 22                5701                    401
March 1                    6432                    272
March 8                    7251                    129
March 15                   6982                    309
March 22                   6321                    238
March 29                   6066                    293

Source: Banking and Monetary Statistics, 1914-1941 (Board of
Governors of the Federal Reserve System 1943, pp. 386-7).

5. Conclusion

The Michigan bank crisis clearly sparked the final banking crisis of the Great Depression. The publicity regarding RFC loans caused some bank suspensions, but the number was small relative to the number of banks identified as receiving RFC loans. The main harm inflicted by publication was that banks became reluctant to borrow from the RFC, even when the loans would have provided needed liquidity.

News reports abounded with calls to inflate prices or devalue the currency, including speculation that president-elect Roosevelt may support inflationary policies. During this period, the dollar depreciated relative to the franc when confidence in the dollar was shaken. However, for several months prior to the Michigan crisis, the dollar-franc exchange rate was at or very close to the dollar gold import point. By February 17, three days after declaration of a banking holiday in Michigan, the dollar depreciated beyond the gold export point for the United States. The depreciation of the dollar and increased gold drain following the Michigan holiday are indicative of a dramatic change in perception of the strength of the dollar.

The behavior of other key indicators of banking crises also indicates a dramatic shift in response to the Michigan holiday. The currency/demand deposit ratio and member bank excess reserves both indicate a banking crisis at the time of the holiday in Michigan, but not before February 14.

[TABULAR DATA FOR TABLE 9 OMITTED]

Banking difficulties in Illinois, Nevada, and Iowa caused concern, as evidenced by brief and relatively minor devaluations of the dollar relative to the franc. The problems in Chicago could potentially have caused widespread panic, had any of the major "loop" banks failed. Fortunately, all survived. Nevada and Iowa were rural or agricultural states, and since the problems were contained to those states, no general panic followed.

The Michigan holiday, a bank crisis in a major industrial state, aroused fears that the situation in the United States was comparable to the situation in central Europe in the spring and summer of 1931, when banking crises forced country after country to abandon the gold standard. In the Michigan crisis, the banks in Detroit and throughout the state were ultimately closed. This appears to be the factor differentiating the Michigan crisis from earlier problems, especially the Chicago crisis, where RFC loans avoided the need for widespread bank closures or a bank holiday.

While president-elect Roosevelt closely guarded his views about the gold standard, which naturally aroused suspicions, the dollar appreciated relative to the franc following his election in November. There is little evidence of a speculative run on the dollar prior to the Michigan crisis, other than moderate gold losses, which can be attributed to a strengthening British pound rather than a weakening dollar. Once the crisis began, a run on the dollar did develop. This dollar run and the spread of bank holidays from state to state together forced newly inaugurated President Roosevelt to declare a nation-wide bank holiday.

The usual reasons cited for the final banking crisis of the Great Depression, publication of the identities of banks receiving RFC loans and speculation against the dollar prior to the Michigan crisis based on fears of adoption of inflationary policies by President Roosevelt, are not strongly supported by the data.

The dramatic response of exchange rates, the currency/demand deposit ratio, member bank excess reserves, Federal Reserve notes in circulation in the Chicago district, and the U.S. monetary gold position following the Michigan crisis are consistent with the hypothesis that the Michigan holiday sparked the final panic. It is impossible to determine how events would have evolved had the problems of the Union Guardian Trust been resolved without recourse to a bank holiday. It is at least plausible that, had the Michigan holiday been avoided, there may have been no banking crisis in 1933.

Data Appendix

Aggregate monthly bank suspensions. Federal Reserve Bulletin, September 1937.

Banks identified as RFC borrowers. New York Times reports: September, August 23; October, October 8; November, October 23; December, November 29; January, December 23; February, January 27.

Central bank gold holdings. Board of Governors of the Federal Reserve System (1943, pp. 537, 639, 642).

Changes in gold position of Federal Reserve Bank of New York. New York Times. Reports of changes in New York Fed's gold position were published daily.

Daily exchange rate data. New York Times. Summaries of previous week's high, low, and closing prices for major currencies were published each Monday.

Demand and time deposits. Friedman and Schwartz (1963).

Farm income in 1929 and 1932. Personal income by state, supplement to the survey of current business (1956). Farm income is the sum of wage and salary disbursements for farms and farm proprietors' income.

Federal Reserve notes in circulation. Weekly reports were published in the Commercial and Financial Chronicle.

Forward exchange rates. Data for Wednesdays were published in the Economist and for Fridays in The Commercial and Financial Chronicle. The two reports use different sources, so the data are not strictly comparable.

Index of farm prices. Yearbook of Agriculture, 1935.

Member bank excess reserves. Board of Governors of the Federal Reserve System (1943, pp. 386-7).

Money in circulation. Board of Governors of the Federal Reserve System (1943, pp. 386-7).

Monthly bank suspensions by state. Federal Reserve Bulletins, January 1932 through March 1933. Publication of this data was terminated as a result of the banking holiday in March 1933. The totals for each month differ slightly from the monthly data reported in the September 1937 Federal Reserve Bulletin, which is the standard source for suspensions data during this period. However, the 1937 data do not provide monthly suspensions data by state, so the earlier, unrevised data are used.

Number of banks by state in December 31, 1931. Federal Reserve Bulletin, October 1932.

Number of commercial banks. Board of Governors of the Federal Reserve System (1933) and Federal Reserve Bulletin, October 1937. This is the number of commercial banks at call dates in each quarter. Data for quarters with no call dates are interpolated by the author.

Prices of shares of the Detroit Bankers Group and the Guardian Detroit Union Group were reported in the Detroit Free Press.

Ratio of currency to demand deposits. Friedman and Schwartz (1963, pp. 712-13).

RFC loans. Quarterly Report of Reconstruction Finance Corporation, July 1 to September 30, 1933. Data are disbursements minus repayments.

U.S. monetary gold stock. Board of Governors of the Federal Reserve System (1943, pp. 381-7).

The author acknowledges the helpful comments of Burton Abrams, Pawan G. Bareja, Stacie Beck, Anne Dushel, Mary Eshelbach Gregson, Evangelos Falaris, Farley Grubb, Richard H. Keehn, Joseph R. Mason, Jeffrey Miller, the participants in the weekly University of Delaware economics seminar, an anonymous referee, and the research assistance of Shuang Wong and Xiang Xu. Responsibility for any errors rests solely with the author.

1 James (1938) does recognize that the Michigan crisis was the spark that caused the final panic.

2 Donaldson (1992) and Wigmore (1987) blame the crisis on an attack on the Fed's gold reserves in the last weeks of February and early March 1933 as the cause of the nationwide bank holiday. However, the crisis in Michigan preceeded the general gold drain. Wigmore does acknowledge that the gold run was "partly related to the surge in state bank holidays." (p. 744)

3 The Michigan crisis has been well detailed by the U.S. Senate Committee on Banking and Currency (1934), Kennedy (1973), and Wigmore (1985).

4 See Mason (1995) for an alternative viewpoint.

5 For data, see Butkiewicz (1995, table 6, p. 208).

6 Keehn and Smiley (1988, 1993) recognize the possibility that publication of individual RFC loans may have affected banks not identified as RFC borrowers but do not address this issue empirically.

7 A case in point is the suspension of the Pleasantville National Bank, which was reported as closing on February 4, 1933, due to excessive withdrawals following the closure of two Atlantic City Banks (New York Times, February 5, 1933, p. 9). Pleasantville is six miles from Atlantic City. The Atlantic City banks suspended activity on January 27, 1933, and January 30, 1933. The news reports indicated that both Atlantic City banks were RFC borrowers, and noted that the Pleasantville bank had never borrowed from the RFC (New York Times, January 28, 1933, p. 3; January 31, 1933, p. 14).

8 Data sources are listed in the data appendix.

9 In fact, suspensions declined in September. Upham and Lamke (1934) and Olson (1977) argue that the decline in the number of suspensions in September is evidence that publication of the identities of banks receiving loans was not harmful.

10 The original legislation requiring the RFC to report the identity of all loan recipients to Congress covered only new loans. Subsequent legislation required retroactive reporting of all loans made since the inception of RFC operations. The legislation required only that the RFC report to Congress. House Speaker John Nance Garner's decision to make the reports public aroused considerable controversy (Butkiewicz 1995).

11 The number of suspensions reported in the September 1937 Federal Reserve Bulletin is 688 for February through August 1932, and 809 for September through February 1933. See the data appendix for a discussion of these differences.

12 The models were also estimated for sample periods ending in September 1932 through January 1933 (not reported). In no case was either publication variable statistically significant.

13 This data, from the Board of Governors of the Federal Reserve System (1943) includes Treasury gold, gold held by Federal Reserve Banks exclusive of gold under earmark, and an estimate of gold coins in circulation.

14 All banks were closed on Monday, February 13, in observation of Lincoln's birthday.

15 See the New York Times, January 21, 1933.

16 Although no formal tests are conducted, foreign exchange markets are assumed to reflect the rational expectations of market participants. Information regarding the gold positions of the United States and major foreign central banks appeared daily in the New York Times, so relevant information was readily available to the public.

17 See New York Times, October 8, 1932, p. 1; October 10, 1932, p. 27; November 11, 1932, pp. 20, 31; and November 17, 1932, p. 23.

18 Morganstern (1959) computes gold points for various dates during the interwar period. He calculates that the gold export and import points on March 3, 1933, were $0.039441 and $0.039024 for the franc and $0.239654 and $0.237261 for the reichsmark.

19 For details, see the New York Times, January 4, 1933, p. 1; January 5, 1933, p. 29; January 6, 1933, p. 27; and January 7, 1933, p. 21.

20 Expectations of devaluation may also be reflected in forward exchange rates. Data on forward exchange rates are sparse during this period. However, published data are available for Wednesdays in the Economist, and for Fridays in the Commercial and Financial Chronicle. The spot and forward rates move synchronously. Although no formal tests of interest parity are conducted, as data on transactions costs are not readily available, the forward pound was typically at a discount because U.S. short-term interest rates were below comparable rates in London. In sum, the limited forward exchange rate data exhibits behavior similar to the spot rates.

21 Another unsettling event that week was the attempted assassination of president-elect Roosevelet in Miami on February 15. Roosevelt was unharmed, but Chicago mayor Anton Cerniak died on March 6 from wounds incurred in the assassination attempt (Schlesinger 1983).

22 The Federal Reserve measure of money in circulation includes currency and bank vault cash.

23 Details of the New Orleans crisis are from the New York Times, February 3, 1933, p. 6; February 5, 1933, p. 25; February 6, 1933, p. 10; February 7, 1933, p. 27; and February 12, 1933, sec. 4, p. 1; and from Kennedy (1973).

24 It is noteworthy that an RFC loan was involved in the resolution of this crisis.

25 The closing dollar price of a franc rose from $0.039025 on Friday to $0.03905 and fell to $0.0390375 on Monday.

26 See New York Times articles in February 1933: 14, p. 1; 15, pp. 1-2; 16, pp. 6, 18; 17, p. 6; 18, pp. 5, 21; 21, p. 23; 22, p. 31; 24, p. 25; 26, sec. 2, p. 7; and March 2, 1933, p. 8; and Detroit Free Press articles in February 1933: 15, p. 1; 16, p. 1; and 17, p. 1.

27 This is similar to the Friedman-Schwartz (1963) argument that the failure of the Fed to resolve the problems of the Bank of United States dealt a major blow to public confidence. The difference is that the Bank of United States was a single bank, whereas the Detroit crisis resulted in a state-wide bank holiday, a far more frightening outcome.

28 Wigmore (1987) and Donaldson (1992) argue that the crisis resulted from a run on gold, but neither links the gold drain to the Michigan holiday. Here, the Michigan crisis is identified as the spark that ignited the financial panic and resulting gold drain.

29 Wicker (1996) argues that the closure of all state banking systems, and not the loss of gold by the New York Fed (Wigmore 1987), forced the new president to declare a nationwide bank holiday.

30 Excess reserves fell during each crisis period for which these data are available (September 1931, June 1932, and February 1933).

31 Although these percent changes are of different magnitude, the dollar value of the changes are of similar magnitude.

32 The increase in notes in circulation was not uniform outside of the Chicago and Cleveland districts. The only other district experiencing a substantial increase in notes in circulation during the weeks ending February 15 and February 21 was New York, with the largest increase coming during the crisis week, ending February 15. This increase in notes in circulation in the third district likely reflects reserve drains from New York to the Chicago district.

References

Board of Governors of the Federal Reserve System. 1933. Nineteenth annual report of the Federal Reserve Board. Washington, DC: United States Government Printing Office.

Board of Governors of the Federal Reserve System. 1943. Banking and monetary statistics, 1914-1941. Washington, DC: Board of Governors of the Federal Reserve System.

Butkiewicz, James L. 1995. The impact of a lender of last resort during the Great Depression: The case of the Reconstruction Finance Corporation. Explorations in Economic History 32(2):197-216.

Cleveland, Harold van B., and Thomas F. Huertas. 1985. Citibank, 1812-1970. Cambridge, MA: Harvard University Press.

Commercial and Financial Chronicle. Various dates.

Detroit Free Press. Various dates.

Donaldson, R. Glen. 1992. Sources of panics: Evidence from the weekly data. Journal of Monetary Economics 30(2): 277-305.

The Economist. Various dates.

Einzig, Paul. 1933. The sterling-dollar-franc tangle. New York: The Macmillan Company.

Einzig, Paul. 1935. World finance, 1914-1935. New York: The Macmillan Company.

Federal Reserve Bulletin. January 1932-March 1933.

Federal Reserve Bulletin. September 1937.

Federal Reserve Bulletin. October 1937.

Friedman, Milton, and Anna Jacobson Schwartz. 1963. A monetary history of the United States, 1867-1960. Princeton: Princeton University Press.

Hodson, H. V. 1938. Slump and recovery. London: Oxford University Press.

James, F. Cyril. 1938. The growth of Chicago banks. New York: Harper and Brothers Publishers.

Keehn, Richard H., and Gene Smiley. 1988. U.S. bank failures, 1932-1933: A provisional analysis. Essays in Economic and Business History 6:136-56.

Keehn, Richard H., and Gene Smiley. 1993. U.S. bank failures, 1932-1933: Additional evidence on regional patterns, timing, and the role of the Reconstruction Finance Corporation. Essays in Economic and Business History 11: 131-45.

Kennedy, Susan Estabrook. 1973. The banking crisis of 1933. Lexington, KY: The University of Kentucky Press.

Mason, Joseph R. 1995. The determinants and effects of Reconstruction Finance Corporation assistance to banks during the Great Depression. Working Paper, Office of the Comptroller of the Currency.

McCallum, Bennett T 1990. Could a monetary base rule have prevented the Great Depression? Journal of Monetary Economics 26(1):3-24.

Morganstern, Oskar. 1959. International financial transactions and business cycles. Princeton: Princeton University Press.

Nadler, Marcus, and Jules L. Bogen. 1933. The banking crisis: The end of an epoch. New York: Dodd, Read, and Company.

New York Times. Various dates.

Olson, James S. 1977. Herbert Hoover and the Reconstruction Finance Corporation, 1931-1933. Ames: Iowa State University Press.

Reconstruction Finance Corporation. 1933. Quarterly report of the Reconstruction Finance Corporation to Congress, July 1 to September 30, 1933. Washington, DC: United States Government Printing Office.

Schlesinger, Arthur M., Jr., general editor. 1983. The almanac of American history. New York: G.P. Putnams's Sons.

Sullivan, Laurence. 1936. Prelude to panic. Washington, DC: Statesman Press.

Temin, Peter. 1976. Did monetary forces cause the Great Depression? New York: Norton.

United States Department of Agriculture. 1935. Yearbook of agriculture. Washington, DC: United States Government Printing Office.

United States Department of Commerce. 1956. Personal income by state, supplement to the survey of current business. Washington, DC: United States Government Printing Office.

United States Senate Committee on Banking and Currency. 1934. Stock exchange practices. Washington, DC: United States Government Printing Office.

Upham, Cyril B., and Edwin A. Lamke. 1934. Closed and distressed banks. Washington, DC: The Brookings Institution.

Waight, Leonard. 1939. The history and mechanism of the Exchange Equalization Account. London: Cambridge University Press.

Wicker, Elmus. 1996. The banking panics of the Great Depression. Cambridge: Cambridge University Press.

Wigmore, Barrie A. 1985. The crash and its aftermath. Westport, CT: Greenwood Press.

Wigmore, Barrie A. 1987. Was the bank holiday of 1933 caused by a run on the dollar? Journal of Economic History 47(3):739-55.

Yeager, Leland B. 1966. International monetary relations. New York: Harper and Row.