Liberty Bonds
April 1917–September 1918
The US Treasury and the Federal Reserve, united under William McAdoo as the leader of both institutions, worked together to finance World War I.
Any great war must necessarily be a popular movement. It is a kind of crusade; and like all crusades, it sweeps along on a powerful stream of romanticism.
William Gibbs McAdoo
World War I began in Europe in 1914, the same year the Federal Reserve System was established. During the three years it took for the United States to enter the conflict, the Fed had completed its organization and was in a position to play a key role in the war effort. Wars are expensive and, like every governmental effort, they have to be financed through some combination of taxation, borrowing, and the expedience of printing money. For this war, the federal government relied on a mix of one-third new taxes and two-thirds borrowing from the general population. Very little new money was created. The borrowing effort was called the “Liberty Loan” and was made operational through the sale of Liberty Bonds. These securities were issued by the Treasury, but the Federal Reserve and its member banks conducted the bond sales.
Generally speaking, the secretary of the Treasury proposes a funding plan for war financing and works with Congress to enact the necessary legislation, while the Federal Reserve operates with considerable independence from both the executive and legislative branches of government. But World War I was different. The Treasury and the Fed, united under one leader, worked together in both the creation of the financial war plan and its execution.
In the congressional debates over the structure of the Federal Reserve, the makeup of the Federal Reserve Board and even its very existence were key issues. The chair of the House Banking and Currency Committee, Rep. Carter Glass, opposed the idea of a central coordinating board. President Woodrow Wilson, however, insisted on a public agency with supervisory powers over the banks. The resulting compromise created a seven-member Federal Reserve Board seated in Washington, DC, with the secretary of Treasury designated ex officio as chair.2 The other members were the comptroller of the currency and five members appointed by the president and confirmed by the Senate. Wilson’s secretary of the Treasury, William Gibbs McAdoo, designed and arranged that compromise, and he emerged from the deal in charge of both the Treasury and the Federal Reserve. Congress cleared the bill in December 1913.3
When the United States entered World War I in 1917, it became immediately evident that an unprecedented effort would be required to divert the nation’s industrial capacity away from meeting consumer demand and toward fulfilling the needs of the military. At the time of the congressional declaration of war, the American economy was operating at full capacity, so the requirements of the war effort could not be met by putting underutilized resources to work. The wartime population would have to sacrifice to pay the bill, and McAdoo understood the point. Shortly after war had been declared, he delivered a speech that he later recorded for posterity:
"We must be willing to give up something of personal convenience, something of personal comfort, something of our treasure – all, if necessary, and our lives in the bargain, to support our noble sons who go out to die for us."
But the question remained: how would the shift in output be arranged? How should the war be paid for? There were three possibilities: taxation, borrowing, and printing money.
For McAdoo, printing money was off the table. The experience with issuing “greenbacks” during the Civil War suggested that fiat money would generate inflation, which he thought would lower morale and damage the reputation of the newly issued paper currency, the Federal Reserve Note. McAdoo also opposed printing money because it would hide the costs of war rather than keeping the public engaged and committed. “Any great war must necessarily be a popular movement,” he thought, “… a kind of crusade.”
McAdoo chose a mix of taxation and the sale of war bonds. The original idea was to finance the war with an equal division between taxation and borrowing. Taxation would work directly and transparently to reduce consumption. Taxes are compulsory, and those who must pay are left with less purchasing power. Their expenditures will fall, freeing productive resources (labor, machines, factories, and raw materials) to be employed in support of the war. Another advantage of taxation was that Congress could set the rate schedule to target those they thought should bear the greatest burden. President Wilson and the Democrats in Congress insisted on a sharply progressive schedule – taxing those with very high incomes at higher rates than the middle class and exempting the poor. The highest marginal rate eventually reached 77 percent on incomes over $1 million.4
Some of the prominent economists of the day suggested that the war should be paid for entirely through such taxes, but McAdoo disagreed on grounds that the eventual cost of war was unknown at the outset. If the tax generated less money than was required, rates would have to be raised again and perhaps repeatedly. Furthermore, changing tax schedules always requires a controversial, complex, and drawn-out political debate. Indeed, as the estimated cost of the war effort escalated, McAdoo came to the conclusion that, despite the high rates, tax revenues would not cover anything like one-half the cost. Given the commitment to the progressive structure of rates, taxation had reached its acceptable limit. The revised goal was one-third from taxes and two-thirds from borrowing.
Financing a war by borrowing need not be inflationary if the public diverts income away from consumption to purchase bonds. Higher saving as a share of income would necessarily mean lower consumption. Such a change in saving behavior, however, would be difficult to engineer and far from certain. A high rate of return on the war bonds would be unlikely to work. High rates might tempt some to take momentary advantage and save more. But there is also an opposite effect. With high interest rates, a household’s wealth would accumulate more rapidly. With that mechanism working on behalf of the saver, less saving from current income would be required to ultimately reach a target level of wealth. The two opposite tendencies would tend to cancel each other. Another problem with offering a high interest rate on the war bonds is that it might divert funding away from investments in physical capital when the war effort warranted an increase in productive capacity.
It is unclear if McAdoo understood that offering high rates of interest would not work. In any case, he was opposed to high rates because that would be a sign of weakness and would reward the rich – the very group the income tax was designed to target. He chose to keep the interest rates competitive with the current return on comparable assets. To many observers, a massive bond sale on these terms seemed to be an imprudent gamble. The worry expressed by bankers and bond dealers at the time was unanimous: the bonds might not sell without the promise of an extra-attractive return. Moreover, the critics pointed out, only a few Americans had any direct knowledge about bonds, and fewer still actually owned any.
It was at this point that McAdoo conceived of the Liberty Loan plan. It had three elements. First, the public would be educated about bonds, the causes and objectives of the war, and the financial power of the country. McAdoo chose to call the securities “Liberty Bonds” as part of this educational effort. Second, the government would appeal to patriotism and ask everyone – from schoolchildren to millionaires -- to do their part by reducing consumption and purchasing bonds. Third, the entire effort would rely upon volunteer labor, thereby avoiding the money market, brokerage commissions, or a paid sales force. The Federal Reserve Banks would coordinate and manage sales, while the bonds could be purchased at any bank that was a member of the Federal Reserve System.
To the war planners, the appeal of borrowing funds from the public was that it would be good for morale. Individuals could demonstrate their support for the war by purchasing bonds. Indeed, during the bond campaigns, purchasers were given buttons to wear and window stickers to display, thus advertising their patriotism. If bond sales were strong, if the offering was oversubscribed, that would demonstrate American resolve.
Yet there was a risk. Poor sales would be a sign of weak support and insufficient patriotism. To avoid a failure to sell the entire bond issue, the government arranged to sell them in a series of brief but intense campaigns by subscription. The first campaign was announced on April 28, 1917, twenty-two days after the declaration of war. The first offering of bonds was to be for $2 billion and promising a 3.5 percent rate of return. That was slightly below the rate paid by savings banks on customers’ deposits (which ranged between 3.5 and 4 percent) or the yield on high-grade municipal bonds (3.9 to 4.2 percent). The fear was that individuals with preexisting savings accounts or municipal bond holdings would use those funds to purchase Liberty Bonds if the bonds’ promised return was greater than what a savings account was earning. Such a rearrangement of portfolios would not have increased saving or reduced consumption. McAdoo also knew that financial institutions would resist mightily any competition for their deposits from the government.
Endnotes
-
1
Richard Sutch is the Edward A. Dickson Distinguished Professor of Economics (Emeritus) at the University of California, Riverside and Berkeley, and a research associate at the National Bureau of Economic Research.
-
2
The provision that the secretary of Treasury chair the Federal Reserve’s Board of Governors was eliminated in 1936.
-
3
For details on this maneuvering, see the autobiographies of McAdoo [1931: 242-245] and Glass [1927: 101].
-
4
Accompanying the personal income tax was an increase in the corporate income tax, an entirely new “excess-profits tax,” and excise taxes on such “luxuries” as automobiles, motorcycles, pleasure boats, musical instruments, talking machines, picture frames, jewelry, cameras, riding habits, playing cards, perfumes, cosmetics, silk stockings, proprietary medicines, candy, and chewing gum. These ad valorem taxes ranged from 3 percent on chewing gum and toilet soap to 100 percent on brass knuckles and double-edged dirk knives. A graduated estate tax on the transfer of wealth at death exempted the first $50,000 and rose progressively thereafter from 1 percent to 25 percent on amounts over $10,050,000. There were also a variety of miscellaneous war taxes introduced. These included taxes on transportation services; admissions to places of entertainment; social, athletic, and sporting club dues; a stamp tax on legal documents; a tax on the value of outstanding corporate stock; and a tax on the use of yachts.
-
5
A Liberty Bond with its coupons. Twice each year the owner would clip out one of the coupons and cash it in at the local bank. This 4 percent bond sold for $50, and earned $1 every six months. Reproduced with the kind permission of the Joe I. Herbstman Memorial Collection of American Finance. Additional images of Liberty Bonds can be viewed at: http://www.theherbstmancollection.com/#!liberty-loans/c1cwm
-
6
A War Thrift Stamp. McAdoo felt that it was important to enroll children and women in the home front’s war bond drives. Children were encourage to purchase these stamps and collect them in a booklet which might be exchanged for a Savings Certificate. Women were asked to take their change in thrift stamps when shopping. Note McAdoo’s signature was reproduced on every stamp. Reproduced with the kind permission of the Joe I. Herbstman Memorial Collection of American Finance.
-
7
Celebrities were enlisted to promote bond sales. Some of the most famous personalities in America toured the country holding bond rallies attended by thousands. This photograph captures Douglas Fairbanks boosting Charlie Chaplin to “Boost the Liberty Loan” with all Wall Street looking on with approval, April 8, 1918.
This article is drawn from:
Sutch, Richard, “Financing the Great War: A Class Tax for the Wealthy, Liberty Bonds for All,” Berkeley Economic History Laboratory Working Paper WP2015-09, September 2015. http://behl.berkeley.edu/files/2015/09/WP2015-09_Sutch.pdf.
For firsthand accounts of the Liberty Bond Campaign and the collaboration between the Treasury Department and the Federal Reserve see:
Glass, Carter. An Adventure in Constructive Finance: An Account of the Federal Reserve System. New York: Doubleday, Page and Co., 1927.
Leffingwell, R. C. “Treasury Methods of Financing the War in Relation to Inflation.” Proceedings of the Academy of Political Science in the City of New York vol. 9, no. 1 (June 1920): pp. 16-41.
McAdoo, William G. “American Rights” [transcription of a sound recording], American Leaders Speak: Recordings from World War I and the 1920 Election, American Memory Project, Library of Congress, no publication date.
McAdoo, William G. Crowded Years: The Reminiscences of William G. McAdoo [with W. E. Woodward]. Boston: Houghton Mifflin, 1931.
United States Congress Joint Commission on Agricultural Inquiry. “Statement of Hon. Benjamin Strong, Governor of the Federal Reserve Bank of New York,” in Agricultural Inquiry: Hearings, Before the Joint Commission of Agricultural Inquiry, Sixty-Seventh Congress, First Session Under Senate Concurrent Resolution 4 (August 2-5, 8-9 and 11, 1921).
Bibliography
Craig, Douglas B. Progressives at War: William G. McAdoo and Newton D. Baker, 1863-1941. Baltimore: Johns Hopkins University Press, 2013.
Federal Reserve Bank of Kansas City Staff. “Federal Reserve Act Signed by President Wilson: December 23, 1913.” Federal Reserve History, last updated November 22, 2013.
Ghizoni, Sandra Kollen. “Reserve Banks Open for Business: November 1914.” Federal Reserve History, last updated November 22, 2013.
Herbstman, Joshua T. “Joe I. Herbstman Memorial Collection of American Finance.” http://www.theherbstmancollection.com/#!liberty-loans/c1cwm
Mock, James R., and Cedric Larson. Words That Won the War: The Story of the Committee on Public Information, 1917-1919. Princeton: Princeton University Press, 1939.
Schuffman, Lawrence D. “The Liberty Loan Bond.” Financial History vol. 113 (Spring 2007): pp. 18-19.
Wheelock, David C. “The Fed's Formative Years: 1913-1929.” Federal Reserve History, last updated November 22, 2013.
Written as of December 4, 2015. See disclaimer.