United States: Aldrich-Vreeland Emergency Currency during the Crisis of 1914

As geopolitical tensions rose in Europe during the buildup to the First World War in the summer of 1914, European stock market exchanges closed and the United States experienced significant gold outflows (Jacobson and Tallman 2015; Sablik 2013). Foreign investors began dumping their securities on the New York Stock Exchange (NYSE) as a means to easily and quickly liquidate their holdings (Silber 2007). On August 1, the member banks of the New York Clearing House reported that reserves had fallen by almost $44 million in the previous week, with a deficiency of $17 million (OCC 1914). As a response, Treasury Secretary McAdoo, at the direction of President Woodrow Wilson, spoke directly with the NYSE and requested that it close for an undefined period, which it did. It reopened four months later, on December 12, 1914 (Silber 2007; WSJ 1914).

In previous crises, the banking system largely relied on the emergency liquidity efforts of clearinghouse associations, which served as private lenders of last resort (LOLRs). After the Panic of 1907, Congress formed the National Monetary Commission, which recommended the creation of the Federal Reserve. The Fed would enshrine the LOLR activities of the private clearinghouse associations in a quasi-governmental institution. However, the Fed was not functional at the outbreak of the Crisis of 1914, as Senate nominations dragged on and administrative plans faced delays (Silber 2008, 73).

On July 31, Secretary McAdoo announced that he planned to exercise the Treasury’s powers under the Aldrich-Vreeland Act to allow banks to issue emergency currency, backed by a much wider range of collateral than usually allowed for note issuance.

McAdoo had previously indicated in 1913 a willingness to utilize emergency currency by announcing that $500 million in emergency currency would be available in response to a possible banking panic. The announcement of this power calmed the markets, and emergency currency was never needed or issued at that time (Silber 2007). This event marked a public willingness from the Treasury Department to use its ability to issue emergency currency to arrest a banking panic, which it did in 1914.

In 1908, Congress passed the Aldrich-Vreeland Act, which granted the Treasury Department the power to issue up to $500 million in “additional circulating notes,” or “emergency currency,” backed by collateral other than US bonds (Aldrich-Vreeland Act 1908, vol. 35, sec. 5). National bank notes at the time were backed by US bonds. The Federal Reserve Act, passed on December 23, 1913, reauthorized the Aldrich-Vreeland Act until June 30, 1915; the former act was set to expire on June 30, 1914, under the initial legislation (Aldrich-Vreeland Act 1908, vol. 35, sec. 20; Federal Reserve Act 1913, sec. 27). The Federal Reserve Act also revised the Aldrich-Vreeland Act, lowering the interest rate charged on emergency currency (Aldrich-Vreeland Act 1908, vol. 35, sec. 9; Federal Reserve Act 1913, sec. 27). This would prove to be a key decision that helped prevent the Crisis of 1914 from spiraling into a full panic. It appears that the Aldrich-Vreeland Act, and the powers contained within, was not extended beyond June 1915.

As the crisis unfolded, Secretary McAdoo requested on July 31, 1914, that Congress amend the Federal Reserve Act to address two main issues: the eligibility of large banks to participate and the $500 million limit on total issuance of emergency currency. The Senate debated the amendment for only 10 minutes before passing it (Silber 2008, 77). On August 4, 1914, the amendment officially became law (Silber 2008, 79).^FNRepresentative Victor Murdock, one of only six representatives to vote against the amendment, stated, “Ordinarily [this amendment] would not pass Congress without days of debate, if it could pass at all” (Silber 2008, 78). This amendment is discussed later in the case study as the “August 4 Amendment.”

Media coverage during the early days of issuance of the emergency currency noted that some questions regarding its legality were raised, as this was the first (and would be the only) usage of the Treasury’s authority to issue emergency currency. The New York Times referenced many meetings between government officials to settle certain debates (NYT 1914b).^FNFor example, the deposit location of the collateral backing emergency currency and the exact application form proved contentious (the government decided to keep them with the national currency associations instead of the Subtreasury) (NYT 1914b).

Many banking institutions were not able to issue emergency currency because they did not meet the eligibility standards. As in previous crises, local clearinghouses acted as private lenders of last resort, with the NYCH taking a leading role (Fulmer 2022). The NYCH issued clearinghouse loan certificates to member banks, which used the CLCs as temporary substitutes for legal tender when settling clearing balances. This allowed member banks to avoid contracting lending and liquidating their assets in order to make payment. The Aldrich-Vreeland Act modeled emergency currency’s structure and function on CLCs, albeit with several significant changes discussed later in this case study (Jacobson and Tallman 2015).

CLCs could not be used outside the NYCH or for any other purposes. However, other clearinghouses (particularly those in the South) allowed CLCs to circulate outside their memberships (Andrew 1908). At its peak, 12 clearinghouses across the United States issued almost $200 million in CLCs to member banks, with the NYCH accounting for $109 million of that total (Fulmer 2022). Denomination was a significant difference between CLCs and emergency currency in New York. The NYCH issued CLCs only in denominations larger than $5,000, while emergency currency largely came in denominations less than $100. This illuminates the different use of the emergency measures in New York, as CLCs circulated in interbank transactions at larger denominations while emergency currency could be used to handle depositor withdrawals at lower denominations (Jacobson and Tallman 2015).

A study by Jacobson and Tallman (2015) focusing on New York concludes that emergency currency and CLCs were not substitutes for New York banks but rather complements that together constituted a replacement for legal tender. Emergency currency issuance grew the money supply available to banks, while CLCs granted banks greater flexibility in handling payment for settling clearing balances and could be issued to the state banks and trust companies that were barred from utilizing emergency currency (discussed in Key Design Decision No. 6, Eligible Participants). Many New York banks took out both forms of liquidity, implying that there were unique benefits to both (Jacobson and Tallman 2015).

Outside of New York, Sprague (1915) notes, emergency currency was preferred to CLCs, as banks could use emergency currency across regions, outside set clearinghouse associations. Additionally, the law as revised on August 4 set the interest rate on emergency currency much lower initially than the rate on CLCs (3% versus 6%), leading borrowing banks to prefer emergency currency while creditor banks preferred CLCs. There was a significant decline in CLC usage in 1914 compared to the crisis of 1907. The number of clearinghouse associations that issued CLCs dropped from 51 in 1907 to 12 in 1914, while the peak outstanding amount of CLCs went from $227 million to $196 million (OCC 1915).

Both the Treasury Department and national currency associations operated as managers of the emergency currency program. The Treasury Department had to approve the formation of national currency associations, which consisted of a board filled with one representative of each member bank. Additionally, each association had a president, vice president, treasurer, secretary, and five-person executive committee. The Treasury Department required associations to file bylaws for its approval. Furthermore, the Treasury secretary had the authority to make “any such rules and regulations and exercise such control over the organization and management of national currency associations” (Aldrich-Vreeland Act 1908).

At the end of 1913, there were 21 national currency associations in operation, representing 352 national banks (OCC 1914). This would increase to 45 associations, representing 2,197 banks, by the end of 1914 (OCC 1915). It does not appear that national currency associations had any significant role to play after the opening of the Federal Reserve System. In 1913–1915, the annual reports of the OCC contain entire sections on national currency associations, while the 1916 annual report does not mention them once. The Aldrich-Vreeland Act expired in 1915, and there does not appear to be any indication that national currency associations continued to exist beyond that expiration.

If a bank wished to issue emergency currency during the crisis, it submitted an application to its national currency association along with the pledged collateral, which was held in trust by the association. If the association approved this application, the application as well as a schedule of pledged collateral, were forwarded to the comptroller of the currency and the secretary of the Treasury. Once the secretary of the Treasury approved the issuance of emergency currency, the bank notes were shipped to the borrowing bank (OCC 1914).

In the opening days of emergency currency issuance, newspapers reported concerns from bankers about the administration and efficiency of the program. In the New York Times on August 3, one banker remarked that the Treasury lagged in its printing, claiming that less than half of the more than $100 million available to New York City banks had been printed and brought to the city (NYT 1914a). The Aldrich-Vreeland Act directed the Treasury Secretary to print the emergency currency after its passage in 1908, which it did and stored in a DC vault (Silber 2008, 80). When Secretary McAdoo authorized the issuance on August 3, the Treasury immediately shipped $46 million of notes to New York City, which was the amount eligible for the initial set of New York City banks (NYT 1914a). However, extra last-minute printing and shipping was required for the remaining $100 million. Therefore, while the banker is correct that the Treasury needed to print and ship more currency, it was not a predictable change that could have been anticipated.

The NYCH swiftly began issuance of CLCs on August 3 after the emergency currency authorization, likely due to its institutional experience in crisis fighting. The New York Times explicitly made the comparison between the NYCH’s prompt action and the Treasury Department’s slower rollout of relief (NYT 1914b).

According to the Aldrich-Vreeland Act, a group of at least 10 national banks could form a national currency association, which would serve as the only association in its city. Membership to these associations, and consequently access to emergency currency, was limited to only national banks at the time of passage of the act in 1908. This excluded state banks and trust companies from participating in the issuance of emergency currency. The Federal Reserve Act required every national bank to join the Federal Reserve System upon its creation (Federal Reserve Act 1913).

The Aldrich-Vreeland Act and the Federal Reserve Act established that a national bank could borrow emergency currency only if it had outstanding national bank notes equal to at least 40% of its capital stock (Aldrich-Vreeland Act 1908, vol. 35, sec. 3; Federal Reserve Act 1913, sec. 27). This requirement restricted many national banks from borrowing emergency currency as they had not issued enough national bank notes (Sablik 2013). However, as war in Europe erupted, Congress amended the Federal Reserve Act on August 4 to allow the Treasury secretary to suspend the 40% condition at the secretary’s discretion (August 4 Amendment 1914).^FNCongressional deliberations revealed that some in Congress viewed this amendment as unfairly supporting and prioritizing the needs of New York City banks (Silber 2008, 78).

Additionally, in the August 4 Amendment, Congress gave the Treasury secretary the authority to grant access to state banks and trust companies, on the condition that they join the Federal Reserve System by August 19 (August 4 Amendment 1914). Despite the set of eligible participants technically widening, state banks and trust companies largely declined to join the Federal Reserve System because membership imposed high administrative and cost burdens (Jacobson and Tallman 2015; Sablik 2013). By May 1915, only 17 trust companies and state banks had joined the Federal Reserve System, compared to 7,605 national banks in the system (Fed 1915b). Therefore, the vast majority of these institutions were not eligible for emergency currency and had to rely on clearinghouse loan certificates if eligible in their cities.

In total, 1,366 banks in 41 national currency associations issued emergency currency. There were 45 national currency associations at the time, with 2,197 members in total (OCC 1915). Therefore, about 62% of eligible national banks issued emergency currency.

Alternatively, individual national banks could apply directly to the Treasury Department for emergency currency, without the need to join a national currency association. A bank applying for emergency currency in this manner could utilize only state, county, and municipal bonds as collateral, subject to a 10% discount (Aldrich-Vreeland Act 1908, vol. 35, sec. 3). In total, only eight national banks took this option, for a small total of $910,500 (OCC 1915). However, more than $70 million of state, county, and municipal bonds were pledged as collateral for emergency currency (OCC 1915). Therefore, national banks pledged the vast majority of this type of collateral through national currency associations, despite the ability to directly request emergency currency from the Treasury Department. This implies that national banks likely preferred the risk-pooling dynamic of the national currency associations and the ability to pledge commercial assets as collateral for emergency currency. Additionally, it cannot be ruled out that dealing directly with the Treasury Department was administratively and bureaucratically slow, since this was the first issuance of emergency currency.

According to the Aldrich-Vreeland Act, the national currency association member banks were jointly liable for the emergency currency issuance. If member banks failed and were unable to pay off their emergency currency, the federal government had a lien on the assets of “all banks belonging to the association” (Aldrich-Vreeland Act 1908, vol. 35, sec. 1). This resembled the structure of CLCs, which were jointly guaranteed by the clearinghouse membership.

Under the Aldrich-Vreeland Act, if any currency association had outstanding emergency currency, it had to pay into a redemption fund 5% of the outstanding amount.^FNUnrelated to the issuance of emergency currency, banks at the time already maintained a redemption fund at the Treasury equal to 5% of circulation of national bank notes backed by US bonds (Laughlin 1908). If banks did not pay into or failed to maintain the redemption fund, the Treasury could remove money from the redemption funds of other banks in the same currency association to account for this issue (Aldrich-Vreeland Act 1908).

The Aldrich-Vreeland Act set a hard cap on outstanding emergency currency issuance at $500 million (Aldrich-Vreeland Act 1908, vol. 35, sec. 5). However, the August 4 Amendment allowed the Treasury secretary to discretionarily suspend this ceiling (August 4 Amendment 1914, 13:683). Although the amendment did not establish a new limit and the Treasury secretary did not announce a new limit, news reports estimated that eligible banks could issue more than $1 billion in emergency currency, based on the collateral available to them (NYT 1914b). After the Crisis of 1914, the comptroller of the currency estimated that the theoretical limit on emergency currency issuance in 1914 was almost $1.5 billion (OCC 1915).

In total, participants issued $386 million in emergency currency, with a peak outstanding of about $364 million on October 24, 1914. Therefore, banks issued less than one-quarter of the theoretical limit on emergency currency. Emergency currency functioned as a temporary relief measure before the Federal Reserve System was operational. Eighty percent of the issuance during the Crisis of 1914 occurred in cities that hosted Federal Reserve banks. Figure 2 shows the state-level data on emergency currency issuance as a share of the theoretical limit. This can serve as a proxy for financial stress, as states with a larger utilization of their limits (which is a function of banking sector size) indicate a greater need for additional liquidity. Issuance appears most concentrated in New York and the Southern states. Banks in nine states did not issue any emergency currency (OCC 1915).

Source: Author’s creation; OCC 1915.

The August 4 Amendment limited individual participating institution issuance to 125% of their capital and surplus (August 4 Amendment 1914). Additionally, the Aldrich-Vreeland Act restricted issuance of emergency currency within each state to its proportion of the nation’s total capital and surplus. However, the Treasury secretary could reallocate the emergency currency issuance cap of states that had not used their full issuance amount to states within the same region that wished to exceed their limits (Aldrich-Vreeland Act 1908, vol. 35, sec. 7). Research could not confirm whether individual currency associations had internal participation limits.

According to the Federal Reserve Act, emergency currency carried an interest rate (“tax”) of 3% for the first three months, based on the average amount of emergency currency in circulation held by the institution. After that point, the interest rate increased 50 bps until a ceiling of 6% was reached, in the ninth month (August 4 Amendment 1914). The Aldrich-Vreeland Act initially established an interest rate of 5% for the first month, which then increased by 1% every month until a ceiling of 10% was reached (Aldrich-Vreeland Act 1908, vol. 35, sec. 9). The comptroller of the currency referred to this initial rate as “onerous” (OCC 1914). He then claimed the lower rate charged in the Federal Reserve Act was “practicable and efficient” (OCC 1914).

The interest charged on emergency currency went to the Division of Redemption of the Treasury (Aldrich-Vreeland Act 1908, vol. 35, sec. 9). This amounted to almost $3 million for the entirety of the program (OCC 1915).

The interest rate did not vary by jurisdiction. At least initially, emergency currency had a much lower interest rate than clearinghouse loan certificates at the New York Clearing House Association, which had a uniform 6% rate (Fulmer 2022). Figure 3 presents money market rates in New York City during 1914, as there is no nationwide interest rate data to compare emergency currency against. In New York City alone, it appears that emergency currency had a below-market price, at least before the escalating structure kicked in. However, this rising rate stopped at 6%, which could still be considered a market rate, per Figure 3.

Note: Shaded area represents the period in which the New York Clearing House Association issued CLCs generally reflecting the crisis period. The dashed horizontal line shows the 3% initial rate charged on emergency currency, although this rate began increasing after three months to a maximum of 6%. The dotted horizontal line shows the 6% rate charged on CLCs. Data was provided monthly as a range.

Source: Author’s creation; OCC 1914; OCC 1915.

According to the Aldrich-Vreeland Act, eligible collateral for emergency currency included “any securities, including commercial paper, held by a national banking association” (Aldrich-Vreeland Act 1908, vol. 35, sec. 1). Seemingly any securities held by national banking associations were eligible as collateral.^FNLaughlin (1908) claims that the Aldrich-Vreeland granted “sweeping” powers, by simply referring to “any securities.” Laughlin states that some members of Congress assumed that they were limiting the securities to “legitimate commercial paper,” but the wording clearly indicated that “any securities” were allowed (Laughlin 1908).

For state and municipal bonds, the state or locality had to exist for at least 10 years and not have defaulted in the same period (Aldrich-Vreeland Act 1908, vol. 35, sec. 4). There was no limitation on the issuance size of state and municipal bonds as a proportion of the state’s or municipality’s capital and surplus.

As for commercial paper, the act defined it as “notes representing actual commercial transactions . . . [that] bear the names of at least two responsible parties and have not exceeding four months to run” (Aldrich-Vreeland Act 1908, vol. 35, sec. 1). Consistent with banking in the early 20th century, the act did not indicate what qualified a party as responsible. Borrowing banks could not use commercial paper to issue emergency currency in excess of 30% of their capital and surplus (OCC 1914).

In the 1915 annual report by the OCC, the Treasury divided pledged collateral for emergency currency into four categories:

• State, municipal, and county bonds;
• Miscellaneous securities, including industrial bonds, and other securities, mainly city and town notes and warrants;
• Commercial paper; and
• Notes secured by warehouse receipts for cotton, tobacco, and naval stores (OCC 1915).^FNThis category appears to have been a concession to Southern states, especially the inclusion of cotton receipts (Silber 2008, 87).

The national currency associations had the authority to request either additional collateral or a collateral substitute if the securities deposited failed to meet its standards. Ten days after a borrowing bank failed, the association could sell the deposited collateral on the market and return the proceeds to the Treasury. If this sale was not enough to cover the emergency currency, the association had a lien on the assets of the failed bank (Aldrich-Vreeland Act 1908, vol. 35, sec. 1).

The Aldrich-Vreeland Act established that the state and municipal bonds received at least a 15% haircut, while commercial paper and all other securities received at least a 25% haircut (Aldrich-Vreeland Act 1908, vol. 35, sec. 1). However, as can be seen in Figure 4, stricter haircuts were applied to every type of collateral pledged.

Source: OCC 1915.

Emergency currency functioned similarly to notes secured on US bonds in that they could be retired at any time without a set maturity (OCC 1914).

Research did not determine other conditions attached to emergency currency.

Due to the period of history that this case focuses on, this Key Design Decision assesses the effect of emergency currency on the money supply, rather than monetary policy, as the gold standard was in effect. The issuance of emergency currency had a significant effect on the money supply. According to Silber (2008, 86, citing Friedman and Schwartz 1963), the money supply grew by an annual rate of almost 10% in the first three months of the crisis. At the time, news reports stoked fears of excessive issuance of currency. Silber details an opinion piece from the New York Times that starts by supporting the emergency currency issuance but ends with a strong warning of potential negative consequences of excessive issuance. Additionally, he documents a Wall Street Journal article that attacks the relatively low interest rate charged on emergency currency, referring to the “craze for easy money” (Silber 2008, 87).

However, these concerns about excessive currency issuance proved unnecessary, as banks quickly retired their emergency currency and money supply growth from August-December actually declined when compared to previous years (Silber 2008, 87).

Research did not determine other options considered by the government to respond to the Crisis of 1914, as the Federal Reserve was not operational at the time. Dodge (1922) says that many businesspeople believed that the emergency currency issued were actually Federal Reserve notes, due to the timing of the issuance of the currency amid significant promotion of the new central bank, which began to issue notes only in November.

The emergency currency program did not involve any international coordination.

In one of the first media mentions of the possibility of emergency currency issuance, Secretary McAdoo told the Wall Street Journal on August 1, 1914:

This indicated a proactive willingness to use the Aldrich-Vreeland Act to position the Treasury Department as a lender of last resort rather than a midcrisis decision. Secretary McAdoo officially announced the authorization of emergency currency on August 3 in the newspapers (Silber 2008, 79; NYT 1914a). Newspapers reported on further updates to the emergency currency operation. Additionally, the annual report of the OCC provided detailed information and data on the emergency currency issuance (OCC 1914; OCC 1915).

Section 9 of the Aldrich-Vreeland Act required national currency associations to provide monthly reports to the Treasury Department of outstanding emergency currency in order to calculate the interest charges accumulated (Aldrich-Vreeland Act 1908).

Research could not determine any individual bank-level disclosure of usage during the Crisis of 1914 by the Treasury Department. The annual reports of the OCC  seem to have provided the sole source of data on emergency currency issuance, albeit at a highly aggregated level.

However, some disclosure of individual emergency currency issuance appeared in a contemporary newspaper article. The New York Tribune published on August 6 the names of four banks that requested emergency currency and the amount granted to them. It is unclear from the language used in the article (“it was learned”) if this was found through investigative reporting or as a public announcement from the banks or Treasury Department (New York Tribune 1914). Research could not determine if this was a single instance of disclosure or a common practice.

The Aldrich-Vreeland Act made individual emergency currency notes appear almost identical to national bank notes (Aldrich-Vreeland Act 1908, vol. 35, sec. 11). Banners along the top of national bank notes said, “Secured by United States Bonds Deposited with the Treasurer of the United States”; similar banners along the top of emergency notes said, “Secured by United States Bonds or Other Securities” (Jacobson and Tallman 2015). This would imply that the law intended to block disclosure of individual usage of emergency currency, since holders did not have information on whether the notes were secured by US bonds or other securities (Silber 2008, 76). As a result, depositors asking for withdrawals from their banks likely could not have easily determined if their banks issued emergency currency.

The escalating interest rate structure on emergency currency encouraged borrowers to retire notes quickly. Additionally, it seems that Congress did not extend the Aldrich-Vreeland Act beyond mid-1915, which removed the ability to issue emergency currency.

Although the Federal Reserve became operational in November 2014, Federal Reserve notes did not immediately replace emergency currency. Banks continued to request emergency currency despite the availability of the Federal Reserve notes, indicating that they were not perfect substitutes. Therefore, according to Dodge (1922), emergency currency was simply a necessity during a panic, while institutions used Federal Reserve notes for more regular business activities. However, the Federal Reserve Bank of Cleveland indicated in 1915 that the quick retirement of emergency currency required the reserve bank to issue Federal Reserve notes in response (Fed 1916).

The comptroller of the currency announced that borrowing banks had retired all of the emergency currency issued by July 1, 1915, except for a small sum issued to a failed bank. This nominal amount ($200,000) was recovered by December 1915. As a result, the emergency currency program did not face any losses (OCC 1915). See Figure 1 for a depiction of outstanding emergency currency along with its subsequent aggregate retirement.