THE GROWTH OF THE FEDERAL GOVERNMENT IN THE 1920S

The federal government has grown substantially in the 20th century. In 1913, just prior to World War I, federal government expenditures were 2.5 percent of gross national product and by 1990 they had risen to 22.5 percent of GNP. The relatively small size of the federal government before World War I shows that it exhibited minimal growth in the 19th century, in stark contrast with its tremendous growth in the 20th century. [1] Figure 1 shows real per capita federal expenditures from 1800 to 1990, and illustrates the difference between government growth in the 20th century and the 19th. A close look at Figure 1 shows a large percentage increase in federal expenditures during the Civil War, followed by a higher level of expenditures after the war, but minimal growth. Real per capita federal expenditures were $79.76 (all figures in 1990 dollars unless otherwise noted) in 1870, and were $79.56 in 1895. Growth in federal expenditures began in the later half of the 1890s, and Figure 1 clearly shows the wartime peaks of World War I and II. But the most notable characteristic of this graph of government growth is not the peaks that are associated with wars, but the steady growth throughout the 20th century, in stark contrast to the absence of growth in the 19th century. The goal of this paper is to contribute toward an understanding of government growth in the 20th century by examining a subset of that century--the 1920s. [2]

The decade of the 1920s is an interesting subject of study for two reasons. First, as Figure 1 shows, government growth in the 1920s was less than in any subsequent decade, so any light shed on growth of government in the 1920s can help illuminate the entire process of government growth. Second, the 1920s fall between two well-known eras of government growth: the Progressive era prior to World War I, and the New Deal of the 1930s. One is tempted to view the 1920s as an era of government retrenchment that, perhaps, would have produced a repeat of the 19th century had it not been for the Great Depression and the New Deal.

The history of the 1920s is often viewed this way (although not by all historians), [3] perhaps because the government of the 1920s is likely to be closely associated with the presidents of the 1920s. Warren G. Harding and Calvin Coolidge were major personalities of the 1920s by virtue of being presidents, but one must view the government of the 1920s more broadly than the presidents alone. Even there, however, an examination of the policies promoted by Harding and Coolidge (and President Herbert C. Hoover at the end of the decade) show that they were considerably more sympathetic toward growth in government programs and expenditures than general opinion suggests. [4]

The study of government growth in the 1920s is interesting in its own right, but the conclusions of this study have more significant implications for the process of growth in the surrounding decades, and more generally for the entire process of government growth in the 20th century. Popular opinion often credits (or blames) President Franklin Delano Roosevelt's New Deal for the major increase in federal government growth. This paper shows that the foundations of the New Deal go back to the 1920s and before, and that FDR's initiatives were not a crucial turning point in the growth of American government. [5] As an extension of the trends begun during the Progressive era, the 1920s is not a key decade in the growth of American government, but understanding the government growth of the 1920s is a key element in understanding to process of the growth of the 20th century. [6]

At first glance, it might appear that the three Republican administrations of the 1920s sandwiched between the Democratic administrations of President Woodrow Wilson (1913-21) and President Franklin D. Roosevelt (1933-45) would have brought with them a period of conservatism, in much the same way that Ronald Reagan's election in 1980 might be viewed as a reaction against government growth and activism in the 1960s and 1970s. [7] However, before FDR's administration, the Republicans were the party of government activism and the Democrats the party of conservatism. Furthermore, except for President Wilson's election that was the result of a temporary fracture of the Republican party into Republicans and Progressives, the Republicans, along with Republican ideas, dominated the White House. After Abraham Lincoln's presidency, Grover Cleveland was the only Democrat to hold the office until FDR. The ideas of Progressivism, found mostly in the Republican party, provided the intellectual foundation for the substantial growth of 20th century government.

Another factor relevant to the political environment in the 1920s was the relative balance of power between the president and Congress. During World War I, the balance of power tipped considerably toward the presidency, but the 1920s brought a reduced amount of power to the presidency, and increased the power of the Republican-dominated Congress. After the 1920 elections, Republicans held a majority of 303 to 131 in the House and 60 to 36 in the Senate and, particularly when compared with the previous two decades, the political agenda was more controlled by Congress than by the executive branch.

Without strong presidential leadership or an aggressive presidential agenda, political initiatives from the Congress became oriented more toward special interest programs that generated economic benefits for clearly identified groups, but such initiatives involved an expansion of governmental scope and power. [8] While some programs were eliminated after the World War I, many were not, and the orientation of federal government activities changed substantially. The scope of government increased substantially during the progressive era and the war, but the 1920s continued pushing the bounds of government. Ellis Hawley (1979: 60) says the administrations of Harding and Coolidge were able to "lend credibility to claims that permanent peace and prosperity were in train and that ways had been found to realize the dreams of the Progressive era while avoiding the evils feared by conservatives." It was not a period of retrenchment, or even of stabilized governmental activity. [9]

The theme of the Harding administration was a "return to normalcy," which must have sounded especially desirable after World War I. This theme was immediately adopted by Coolidge after Harding's death in 1923. One feature of this return, and an indicator of the conservatism of the Harding and Coolidge administrations, was the slashing of income tax rates, which involved considerable congressional debate. But when the income tax was established in 1913, the highest marginal tax rate was 7 percent; it was increased to 77 percent in 1916 to help finance the war. The top rate was reduced to as low as 25 percent in 1925, but that is substantially higher than the 7 percent rate prior to the war, and the income levels that defined the brackets had also been lowered substantially from their prewar levels. The "normalcy" of the 1920s incorporated considerably higher levels of federal spending and taxes than the Progressive era before World War I.

The Progressive movement, and the Progressive party, remained vital through the 1920s, the difference being that the Republicans had been able to regain the support of Progressives. In 1924, the Progressive party ran Robert LaFollette, a Republican Senator from Wisconsin, as their presidential candidate. LaFollette gained a respectable 13 percent of the popular vote. Despite the three-way race, Coolidge still won a 54 percent majority, which contrasts sharply with the 1912 election in which the Progressive party split the Republican vote and led to the loss of the Republican incumbent. Normalcy, in the Harding-Coolidge sense, meant peace and prosperity, but it also meant a continuation of the principles of Progressivism, which enabled the Republican party to retain the support of its Progressive element. Despite the popular view of the 1920s as a retreat from Progressivism, by any measure government was more firmly entrenched as a part of the American economy in 1925 than in 1915, and was continuing to grow. Harding and Coolidge were viewed as pro-business, [10] and there may be a tendency to equate this pro-business sentiment as anti-Progressivism. [11] The advance of Progressivism may have been slower than before the war or during the New Deal, but a slower advance is not a retreat. [12]

The Hoover administration, from 1929 to 1933, must be analyzed differently because of the onset of the Great Depression, but compared with his immediate predecessors, it is much easier to make the case that Hoover was an active supporter of increased government involvement in the economy. Hoover served in the Wilson administration as head of the United States Food Administration beginning in 1917 and, as Secretary of Commerce throughout the Harding-Coolidge administrations, was the most active Cabinet member in pursuing increased government involvement in the economy. From 1929 to 1933, under President Hoover's administration, real per capita federal expenditures (graphed in Figure 1), increased by 88 percent. Under President Roosevelt's administration from 1933 to 1940, just before World War II, they increased by only 74 percent. Although Hoover started from a lower base, in percentage terms expenditures under Hoover increased more in four years than during the next seven New Deal years. [13] If a case can be made that federal policies under the Harding and Coolidge administrations were a solidification and extension of Progressive principles, the case is much more easily made for President Hoover's administration.

One might be concerned that looking at the data in real per capita terms, as is done in Table 1 and Figure 1, makes growth look more substantial than it actually was because it ignores the fact that the 1920s were relatively prosperous years, especially when compared to the 1930s. The picture is not much different if federal outlays are examined as a percent of GNP, however, as Figure 2 shows. As already noted, there is no absolute growth in federal outlays as a percentage of GNP during the 1920s, but there is also almost no growth after 1932, when FDR's New Deal began. The substantial growth in Figure 2 occurred during the Hoover administration. One would not want to argue, based on the aggregate expenditure data shown in Figure 2, that the New Deal was not a period of substantial growth in government, although the data in Figure 2 show little growth during the 1930s. This same reasoning applies to the 1920s. These aggregate data can conceal much, and beg for a closer examination of both the 1920s and the 1930s.

Table 1 shows real per capita federal expenditures for the years from 1915 through 1935, which brackets the 1920s by a half decade in each direction. The column next to the year shows total real per capita federal expenditures, which are the numbers for 1915-35 that appear in the graph in Figure 1. The next column shows that total less military, veterans, and interest expenditures, in order to subtract out expenditures directly related to military activity. The relation between military and veterans expenditures is obvious, but in an era where government budgets typically were in surplus, interest on the national debt during the 1920s was also almost entirely an expenditure related to the nation's involvement in World War I. Before the war, in 1915, the national debt was $1.2 billion; by 1920, after the war, it had climbed to $24.3 billion.

Table 1: Real Per Capita Federal Expenditures: 1915-1935 (In Constant 1990 Dollars)
Year	Total	Total Minus Defense and Interest
1915	95.02	27.39
1916	83.60	22.75
1917	193.22	113.81
1918	1,067.31	518.37
1919	1,329.77	477.53
1920	390.98	170.15
1921	338.86	136.16
1922	232.95	78.62
1923	214.57	74.49
1924	194.85	70.36
1925	187.78	71.09
1926	184.80	76.57
1927	180.57	72.75
1928	186.56	79.89
1929	195.41	89.30
1930	211.13	101.81
1931	247.41	130.80
1932	286.39	183.98
1933	367.84	231.20
1934	498.88	361.02
1935	486.81	328.03
Source: Historical Statistics of the United States from Colonial Times to 1970, and author's calculations.

An examination of the data in Table 1 reveals several things about the growth of federal expenditures during the 1920s. First, nonmilitary expenditures before World War I were substantially less than $30 per capita, and even the lowest year after the war (1924) had nonmilitary expenditures of more than $70 per capita--more than double the level of expenditures before the war. This lends some evidence for the ratchets hypothesis, which suggests that during times of crisis, government expenditures are ratcheted up to deal with the crisis, and never fall back down to their prior level after the crisis has passed. [14] However, the table shows more than simply a ratcheting up of government expenditures. Following the ratchets hypothesis, one might expect military expenditures to soar during the war and then decline as nonmilitary expenditures increase. In fact, nonmilitary expenditures soared during the war and declined steeply afterward. Nonmilitary expenditures in 1924 were only 13.6 percent of their peak level in 1918.

The second thing the data in Table 1 suggest is that there are war-related expenditures in the government budget even after subtracting defense, veterans, and interest expenditures. This makes it apparent that one cannot accept nonwar expenditures as unrelated to the war. (These expenditures will be analyzed at more length below.)

A third thing the data show is that after reaching their lowest postwar level in 1924, government expenditures began a relatively rapid growth. From 1924 to 1929, before Depression-related expenditures would have found their way into the budget, nonmilitary expenditures increased by 27 percent, all during the Coolidge administration. If we take the decline in expenditures up through 1924 as a winding down of the war effort, there appears to be a considerable underlying growth in federal expenditures through the 1920s--growth worth examining more closely. What at first appears to be a relatively stable level of federal expenditures in the 1920s actually is substantial underlying growth, masked by a decline in war-related expenditures.

A fourth matter to consider is the appropriate benchmark for looking at government growth during the 1920s. The growth looks substantial if only the years since 1920 are considered, as just argued, but one could argue that it would be more appropriate to use the prewar year of 1916 as the benchmark; if the 1920s really were a no-growth decade, they should have returned federal spending back to its prewar level. The fact that federal spending always remained at more than double its prewar level does say something about growth in the 1920s, especially as the underlying trend was up.

Finally, Table 1 shows that in percentage terms, the Hoover administration undertook a massive increase in federal spending after the 1929 stock market crash and the onset of the Great Depression. From 1929 to 1933 (the final Hoover budget), nonmilitary federal expenditures increased by 259 percent. Total expenditures increased by a more modest 188 percent during that period because military expenditures were, in 1929, more than half of the federal budget. In percentage terms, growth in federal spending in the Hoover administration far outpaced the growth during FDR's administration.

Although a superficial look at the data suggests that the 1920s represent a lull in government growth between the Progressive era and the New Deal, looking at the data in Table 1 more closely raises several interesting issues and reveals that there was a substantial amount of underlying growth in government expenditures in the 1920s. It would be misleading to try to judge the growth of the federal government in the 1920s only by looking at aggregate expenditures. The next several sections look more closely at the decade to discover how government growth manifested itself.

Table 2 takes a more detailed look at federal expenditures in 1920, 1925, and 1930 to identify differences in budgetary growth among various categories of expenditures. The table shows that when the decade is divided in half, the more substantial growth took place in the second half of the decade. General functions, which include the direct expenses of the legislative, judicial, and executive branches of government, remained stable through the first half of the 1920s, but increased by 35 percent in the second half. Similarly, military expenditures, which fell substantially in the first half of the decade up through 1925, increased by 32 percent from 1925 to 1930. The majority of non-functional expenditures is interest on the federal debt, and that category is relatively stable. The civil functions category is of most interest for current purposes.

Table 2: Expenditures of the Federal Government by Function: 1920, 1925, and 1930 (Thousands of 1930 Dollars)
	1920	1925	1930
General Functions	18,953	18,932	25,642
Military Functions	1,966,358	1,099,711	1,455,093
Civil Functions Less Trans.	426,964	504,891	775,298
Transportation	1,281,878	90,035	87,029
Non-Functional	1,453,747	1,369,446	1,315,742
Total	5,147,900	3,083,016	3,658,804
General Functions are legislative, judicial, and executive expenditures. Military Functions include military pensions. Non-Functional expenditures include interest on the debt, investments, refunds, and deposits into trust funds. Source: Author's calculations from data in Wooddy (1934), pp. 545-546.

Carroll Wooddy (1934) breaks civil functions down into 20 categories, including land transportation and marine transportation. Table 2 lists them separately because changes in these two transportation categories so overwhelm the civil functions category. In 1920 there was a considerable wartime component to the transportation category because of the Merchant Fleet Corporation, which was established to aid in merchant shipping during the war, and because of the nationalization of the railroads during the war. Another major area of marine expenditures was the Coast Guard, which had substantial expenses associated with Prohibition. Because of wartime expenditures and Prohibition, the trends in marine transportation expenditures are not representative of government growth in general.

Civil functions less transportation in Table 2 show relatively modest growth of only 18 percent from 1920 to 1925. From 1925 to 1930, in contrast, civil functions less transportation grew by 54 percent. Part of the reason for the relatively modest growth in the first half of the decade is that increases in some areas were offset by decreases in war-related civil expenditures in other areas. That is easy to document in the transportation category, but is more difficult to see in other categories. [15] Federal relief, for example, declined from $48 million in 1920 to $3.5 million in 1930, undoubtedly because most relief expenditures in 1920 were war-related. [16] But other areas increased substantially over the decade, and deserve a closer look.

Table 3 lists a few areas that saw significant increases in federal expenditures over the 1920s. Those areas are all subsets of the civil functions less transportation category from Table 2; the final row in Table 3, % of Civil less Trans, shows the percentage of that category in Table 2 that is made up of all of the categories in Table 3. Thus, in 1920, the categories of Table 3 are 36.5 percent of civil functions less transportation in Table 2, with the percentage increasing to 66.2 percent in 1925 and declining slightly to 62.7 percent in 1930. The last number in the Total row shows that taken together, these categories were 3.12 times as large in 1930 as they were in 1920. The substantial growth in these categories was partially offset by declines in some other categories, because the Table 2 category was only 1.82 times as large in 1930 as in 1920. That suggests that the categories omitted from Table 3 may reflect declines in war-related expenditures, although other factors may also have been at work.

Table 3: Expenditures of the Federal Government, Selected Functions: 1920, 1925, and 1930
	Thousands of 1930 Dollars
	1920	1925	1930	1930/1920	Annual Growth Rate
Law Enforcement	10,375	24,246	51,537	4.97	17.5
Public Domain	15,402	24,994	37,789	2.45	9.4
Commerce	8,077	14,393	27,354	3.39	13.0
Agriculture	16,644	26,273	48,824	2.93	11.4
Labor Interests	2,065	4,641	6,863	3.32	12.8
Immigration	2,953	5,541	9,512	3.22	12.5
Public Education	5,405	10,483	14,295	2.64	10.2
Public Improvements	73,481	183,411	241,387	3.29	12.7
Territorial & Local Govt	19,964	38,228	46,367	2.32	8.9
TOTAL	155,965	334,044	485,838	3.12
% of Civil less Trans	36.5	66.2	62.7
Source: Author's calculations from data in Wooddy (1934), p. 545.

The second column from the right in Table 3 gives the 1930 expenditure level divided by the 1920 level, so law enforcement expenditures in 1930 were 4.97 times as large in 1930 as in 1920. The rightmost column gives the growth rate over that decade, so law enforcement expenditures grew at a 17.5 percent annual rate from 1920 to 1930. These categories cover nearly two-thirds of the civil functions less transportation from Table 2, so they make up a significant component of nonmilitary federal expenditures. The table shows that their average annual growth rates were very high, so the categories in Table 3 give some indication of where to look for government growth in the 1920s.

Law enforcement expenditures were almost five times as large in 1930 as in 1920, and much of this is related to Prohibition. Agricultural expenditures, which had an average annual growth rate through the 1920s of 11.4 percent, are discussed further below. Immigration and naturalization expenditures more than tripled, partly representing greater selectivity in admitting immigrants. Immigration was sharply down in 1915 due to World War I, and legislation passed in 1917 restricted immigration after the war. In 1924 immigration quotas were established, further adding to immigration and naturalization expenditures (Wooddy 1934: 300-91).

Federal expenditures on public education also increased substantially during the 1920s, funding programs such as land grant colleges and federal subsidies for vocational education. These expenditures were undertaken by grants to the states, but the federal government also contributed significant sums directly to some educational institutions, including Howard University and Columbia Institution for the Deaf. The Library of Congress also falls within the education category, and in 1930 accounted for $2.3 million of the $14 million education category. Two features of educational expenditures are relevant to the consideration of federal growth in the 1920s. First, education, which even in the late 20th century would normally be considered to be an activity undertaken by the states, had considerable and rapidly growing federal involvement in the 1920s. Second, much of the federal government's expenditures involved grants to the states, leading to more federal financing and more federal control in areas traditionally administered by state governments.

Public domain expenditures include the Department of the Interior expenses to manage public lands and national parks, the Federal Power Commission, whose expenditures were largely related to hydroelectric projects, and the Forest Service, which was a part of the Department of Agriculture. About 70 percent of the public domain expenditures in Table 3 are Forest Service expenditures, which went primarily for the construction of roads and trails, and for grants to state governments to aid in their managing of public lands, including road building and fire protection.

The Secretary of Commerce through the Harding and Coolidge administrations was Herbert Hoover, whose department shows remarkable 13 percent growth rate through the 1920s, more than tripling in size. By 1930 about one-third of commerce expenditures were due to aviation, but that still leaves a substantial amount of the budget for non-aviation related expenditures. The expenditures of the Patent Office increased substantially, as did the Bureau of Foreign and Domestic Commerce, which had as one of its main goals the promotion of American goods to foreign purchasers. Expenditures more than doubled in the Radio Commission and the Tariff Commission also. Increased expenditure and activity in the Department of Commerce through the 1920s may reflect the growth of commerce during that prosperous decade, but those expenditures grew under the guidance of Secretary Hoover, lending some insight into Hoover's vision of government that would be applicable to evaluating his presidential administration.

The substantial increase in expenditures on labor interests cannot be assigned to a dominant area, because those expenditures primarily represent the Department of Labor budget, which included a diverse set of activities. The Bureau of Labor Statistics and the United States Employment Service primarily provided information. The United States Board of Mediation helped to settle labor disputes. The United States Housing Corporation was created in 1918 to help facilitate the construction of housing in areas of critical wartime production, but, despite its wartime motivation, lasted into the 1930s. The Harding and Coolidge administrations had a pro-business reputation, but expenditures on labor-related activities soared during the 1920s.

All these areas might be explored in more detail, but this examination illustrates the primary point that after omitting expenditures that are clearly related to World War I, there was substantial growth in underlying civilian federal expenditures in the 1920s. The next several sections select a few specific areas of government activity that have special relevance to the nature of government growth in the 1920s.

One of the methods by which the government has expanded beyond its traditional bounds in the 20th century is the establishment of government-owned corporations. The first wholly owned federal corporation was the Bank of the United States, chartered by an act of Congress in 1791. But after the charter of the Second Bank of the United States expired in 1836, another publicly owned corporation was not formed until 1904, to build the Panama Canal.

The next federally owned corporation was the Merchant Fleet Corporation, established in 1917 as a part of the war effort. After the war, the Merchant Fleet Corporation remained in business through the Great Depression, losing phenomenal sums of money over the decade of the 1920s. [17] Throughout the 1920s there were protests from members of Congress that the Corporation was being run inefficiently, and that it was unnecessary and a government-subsidized competitor of privately owned U.S. shippers. The corporation remained in business, however.

Also in 1917 the Food Administration Grain Corporation was established to help finance the sale of American wheat to Europe, and pushed the frontiers of public corporations because its purpose was to aid private businesses (in this case, farms) in selling their products, although admittedly in a manner related to the war effort. In 1918 the War Finance Corporation was established to help key industries borrow funds and pushed the frontiers even further. The War Finance Corporation aided businesses more generally, and in 1921, well after the war, its charter was extended so that it could undertake more general types of lending activity, including the making of agricultural loans. The War Finance Corporation continued in business through 1931.

Federally owned corporations, a rarity only a few years prior, began to proliferate in the late teens. The Federal Land Bank was established in 1917, and the Sugar Equalization Board, the United States Housing Corporation, and the Spruce Production Corporation were created in 1918. The war effort provided some impetus to these corporations, although some had a tenuous connection with the war, and in any event the war ended in 1918. If the corporations were intended to help win World War I, they could have been dismantled after the war, but the corporations persisted and provided a model for government growth that continued through the 1920s.

In 1923 the Agricultural Credits Act created 12 federally owned banks to make agricultural loans. The Inland Waterways Corporation was created in 1924 to operate boats and barges on the Mississippi River. In 1929 the Federal Farm Board was created to support the prices of agricultural products. The use of public corporations to further the federal government's goals did not begin during World War I, but the late teens were transition years during which the government created more federally owned corporations than had ever been created previously. For the most part, they were not dismantled after the war, but rather remained and became models for additional federally owned corporations that were established in the 1920s. Paul Johnson (1991: 17) argues, "The war corporation of 1917 began one of the great continuities of modern American history, sometimes underground, sometimes at the surface, which culminated in the vast welfare state which Lyndon Johnson brought into being in the late 1960s." Johnson's observation implies that the 1920s were a part of the continuing process of 20th century growth, rather than a "return to normalcy" falling between the Progressive era and the New Deal.

The goals of those corporations are significant. They were not formed to enhance national security, or to protect the rights of individuals, but to provide economic assistance to a subset of the U.S. population. The programs were small in scale when compared with the New Deal, but they were large in scale when compared with federal programs before World War I, and they were built on the philosophy that the federal government has a direct role to play in enhancing the economic well-being of Americans. The New Deal changed the scale of federal programs when compared to the 1920s, but did not change the underlying philosophy.

Federal aid to states was relatively small through the 1920s, but grew relatively rapidly. In 1915 federal aid to states was $6 million, and had increased to $109 million by 1930. Though small in amount (with 48 states, the average state would have received $2.3 million), the percentage increase was substantial, and federal aid rose to $198 million in 1931 and $234 million in 1932. In examining the growth of the federal government from 1915 to 1932 Wooddy (1934: 552) concludes, "There seems little doubt that the development of this device for permitting federal participation of the activities of state and local governments contributed substantially to the growth of the period." For present purposes, the magnitude is not of as much interest as the precedent.

The establishment of post offices and post roads is one of the activities of the federal government explicitly given in the Constitution. By charging for its services, the Post Office has been able to operate in a relatively businesslike manner, even though it was not, until recently, a publicly owned corporation. Through 1851 the Post Office operated under an implied balanced budget constraint, but legislation in 1851 specifically prevented the Post Office from diminishing any of its existing services while encouraging continued expansion. In 1857 and 1858 the Post Office established new routes to the Pacific costing considerably more than the revenues they generated (Kelly 1932: 63-65). The 1850s marked the beginning of a postal policy designed to keep rates low to expand its service, which it was able to do thanks to the backing of the federal Treasury.

While postal deficits were not uncommon, in the years before World War I the Post Office was able to roughly balance its budget; it showed a substantial surplus during the war years. The apparent financial success of the Post Office during the war years was bolstered by two factors. First, much mail was carried by the military during the war, and second, postal rates were increased during the war. But military assistance in carrying the mail never reduced the expenditures of the Post Office. In 1915 the Post Office spent $299 million, and with increases every year, expenditures were $454 million in 1920. [18] Despite some cost advantages due to the war, the Post Office delivered all mail of military personnel without postage, and began selling war-savings certificates in 1917, adding to its responsibilities. More than 80 percent of the war-savings certificates were sold through post offices.

In the Wilson administration, Postmaster General Albert Burleson, appointed in 1913, ruled the Post Office with a heavy hand, trying to do away with unions and increase the efficiency of the Post Office by controlling costs. [19] Burleson's explicit goal was to have the Post Office operate at a surplus (Kelly 1932: 83), and he succeeded. In 1917 postage for first class letters was increased to 3 cents, and for post cards to 2 cents, increasing Post Office revenues, but in 1919 the increases were reversed, which created substantial losses for the Post Office.

Postmaster General Burleson was not popular with postal employees and, after Harding's election, several Postmaster Generals were appointed that were considerably more sympathetic to postal employees than Burleson. But under President Harding, postal expenditures increased tremendously. Expenditures of $454 million in 1920 were followed by expenditures of $621 million in 1921 and a huge postal deficit. Although expenditures fell to $546 million in 1922, they were still 20 percent higher in 1922 than in 1920, and the Post Office showed a deficit every year after 1919, when the rates were lowered to their prewar levels.

Despite the lowering of rates, in 1920 the salaries of letter carriers were raised from $1,400 a year to $1,800, and were raised again in 1924 to $2,100. Harding's first Postmaster General, Will H. Hayes, explicitly stated that the Post Office did not exist to make a profit, but rather to provide service. Further, Hayes wanted to "humanize" the postal industry, arguing that labor is not a commodity. The result was not only higher wages, but an expansion in the role of the Post Office. It expanded rural free delivery and established more city delivery offices. The Post Office also issued many more money orders and increased its sales of savings bonds (an activity that began in 1917 and continued through 1931). As Wooddy (1934: 270) notes, postal deficits in the 1920s were caused by the expansion of postal services and the provision of many services without charge or considerably below cost.

The deficits in the 1920s might be looked at as a return to the idea that, because of the Post Office's service to the public, postal deficits should not be considered objectionable. However, that approach is in stark contrast to Postmaster Burleson's ideas under the Democratic Wilson administration. Burleson's ideas were more in line with the contemporary vision of limited government and bureaucratic accountability and were the status quo for eight years under President Wilson. If the Harding-Coolidge administration was a return to limited government, it could have continued President Wilson's businesslike approach. The Post Office under the Harding-Coolidge administration shows a turn away from the conservative principles in the Wilson administration and a turn back toward the Progressivism that preceded it.

Prohibition corresponds almost exactly to the Republican administrations of Harding, Coolidge, and Hoover. After the passage of the 18th Amendment to the Constitution, Prohibition became effective in 1920 and continued until it was repealed in 1933 by the 21st Amendment. While Prohibition itself slightly predates the Harding administration, the way in which it was enforced was a function of the executive branch of government. Enforcement really goes back to the Harrison Act of 1914, which regulated the use of narcotic drugs. Because taxes were supposed to be paid on legal narcotics, the enforcement of the Harrison Act was undertaken by the Bureau of Internal Revenue, within the Treasury Department. Because there were also limited legal nonbeverage uses for liquor, the Bureau of Internal Revenue also was charged with enforcing Prohibition.

Expenditures to enforce Prohibition extended beyond the Bureau of Internal Revenue, because the Customs Service, Coast Guard, and Department of Justice were also involved. As Table 3 shows, federal law enforcement expenditures increased by nearly 400 percent in the 1920s, more than doubling from 1920 to 1925, and then more than doubling again from 1925 to 1930. However, while there may be some debate about how costs of agencies like the Coast Guard and the Department of Justice should be associated with Prohibition, it is interesting to note how rapidly expenditures of the Bureau of Internal Revenue were affected. [20] In 1920 a Prohibition Unit was established in the Bureau of Internal Revenue, and in 1927 a separate Bureau of Prohibition was created within the Treasury Department. The Treasury's Prohibition-related activities were obviously not intended to generate revenue, as Table 4 shows. In no year did Treasury collect more revenues from enforcement of Prohibition than it expended in enforcement activities. This does not include expenditures undertaken by the Justice Department, Customs, or the Coast Guard.

Table 4: Treasury Revenues and Expenditures From Enforcement of Prohibition (Thousands of Current Dollars)
Year	Revenues	Expenditures
1920	1,149	2,060
1921	4,571	6,301
1922	4,356	6,544
1923	5,095	8,135
1924	6,538	7,509
1925	5,873	9,203
1926	5,647	9,573
1927	5,162	11,685
1928	6,184	11,610
1929	5,475	12,328
1930	5,357	13,507
1931	4,138	9,624
1932	3,954	11,058
Source: Wooddy (1934), p. 101.

The revenues in Table 4 include fines, penalties, and taxes collected as a result of enforcing Prohibition, and the expenditures include only Department of Treasury expenditures. Note that revenues are never higher than in 1924, when they still fall short of expenditures by about $1 million, but expenditures continue to increase substantially up through 1930. Those expenditures probably represent less than one-third of total Prohibition-related federal expenditures. The government's powers related to Prohibition expanded as rapidly as its expenditures, and the creation of a separate Bureau of Prohibition was an initiative of the Coolidge administration to better enforce Prohibition. That is yet another area in which government expanded its powers during the 1920s, and did so with the approval of the Coolidge administration.

In the 1920s American agriculture did not fare as well as American industry, creating the impression of an agriculture versus industry antagonism, and the impression that the Harding and Coolidge administrations favored industry over agriculture. [21] Much of the blame for agricultural problems must be placed on increased worldwide production, creating a glut on world markets. The industry could have been allowed to adjust to market conditions, creating more transitionary hardship among farmers, but it was not. With regard to government policy in the 1920s, tariffs on farm products were passed by Congress in 1921, the Capper-Volstead Act was passed in 1922, exempting agricultural cooperatives from antitrust laws, and in 1926 the Division of Co-operative Marketing was set up in the Department of Agriculture with the warm endorsement of President Coolidge (Hicks 1960: chap. 9). The Agricultural Credits Act of 1923 made it easier for farmers to get intermediate credit from the Federal Farm Loan Board.

Looking at the facts, one has to conclude that government intervention for the assistance of the agricultural industry was significant in the 1920s, and in no case was any government protection or assistance taken away from farmers. But farmers wanted more than they got from President Coolidge. After much debate and substantial controversy, the McNary-Haugen bill passed Congress in 1927, which would provide government support prices for agricultural commodities. Coolidge vetoed the bill, but it was redesigned and passed again the next year, and again vetoed by Coolidge. After Hoover's election, the Agricultural Marketing Act of 1929 was passed, creating a Federal Farm Board with the power to buy and store "any quantity" of agricultural commodities for the purpose of supporting prices. [22]

The government did not treat farmers as generously as they wanted to be treated in the 1920s but, despite the "industry versus agriculture" impression that some historians have of the period, the 1920s saw no reversals of government policy to aid agriculture, and a substantial growth in new agricultural policies. Benjamin Anderson (1979: 125-27) has argued that the original introduction of the McNary-Haugen bill in 1924 marks the true beginning of the New Deal. From 1924 on, legislation was increasingly designed to help control the economy and to support the economic interests of well-defined interest groups, and farmers were major beneficiaries. As Table 3 showed, in 1920, federal expenditures on agriculture were $17 million (in 1930 prices), and had increased by 193 percent to $49 million by 1930. Whether evaluated financially or with regard to programs, the 1920s saw considerable government growth in the agricultural industry, and laid the foundation for more federal involvement that was to follow in the New Deal.

Expenditures are the easiest measure of the size of government, but tell only a part of the story of government growth. Government regulation also has a substantial impact, but is harder to measure. [23] Starting with the Sherman Act in 1890, the federal government began its antitrust activity to try to limit the economic power of businesses. Only 22 cases were brought before 1905, but the pace started picking up later in that decade, which saw 39 cases brought between 1905 and 1909. From 1910 to 1919, a total of 134 cases were brought, showing increasing antitrust enforcement. But there was little slowdown in the 1920s, which saw a total of 125 cases. [24]

As Thomas McCraw (1984: 145) notes, "By the 1920s antitrust had become a permanent part of American economic and political life." One might anticipate, after an increase in cases, that firms would be more cautious in their activities to avoid antitrust cases being brought against them. But McCraw (1984: 146) further notes that in the 1920s a large proportion of antitrust cases were brought against firms that were not normally regarded as being highly concentrated. Antitrust enforcement in the 1920s was vigorous and increasingly broad in scope. Antitrust enforcement is another dimension in which the federal government was exercising increased power in the 1920s, despite the conventional wisdom of the 1920s as a decade friendly to business.

In a famous passage, J. M. Keynes (1936: 383) remarked, "The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed, the world is ruled by little else." Reinforcing the ideas of Progressivism, the academic concept of scientific management was gaining currency early in the 20th century and enhanced the interface between academic institutions and public policy. The creation of the National Bureau of Economic Research, which owes its origin to the Progressive technocratic reform movement that wanted to use government to improve the economy's ability to produce and distribute income, was one manifestation of an increasing influence of academic institutions on government.

With private funding, and support from both the public and private sector, the NBER was established in 1920 with Wesley Clair Mitchell as its first director. Mitchell's goal was "to organize the bureau's research program around questions that were, first, of primary social importance, and second, capable of the statistical resolution needed to surmount social science's internal crisis of scientific legitimacy" (Alchon 1985: 59). The NBER worked closely with the Department of Commerce, and Secretary Hoover was very much interested in making the government more actively involved in economic policy. Indeed, many of the new Keynesian ideas of the 1930s were a part of the accepted wisdom of American economists in the 1920s (Davis 1971). The 1920s saw an alliance of academic institutions, private foundations, and government to create organizations like the NBER that furthered the scientific aspects of social science and provided data and ideas for government management of the economy.

The 1920s, falling between the Progressive era and the New Deal, and directly following World War I, may appear as a decade of stable government sandwiched between major episodes of government growth. In some sense, this is true. Government expenditures were declining dramatically from their peak levels during the war, and Calvin Coolidge, president during most of the decade, was not acting aggressively to expand the scope of government. But in other ways, the 1920s was a decade of increasing government activity, in expenditures, in regulation, and in attitude, as the federal government was increasingly willing to expand its role in the economic lives of its citizens. Indeed, the foundation for the New Deal was established in the 1920s, and it certainly would be wrong to conclude that were it not for the New Deal, government would have remained more confined later in the 20th century. The New Deal was really an extension of the type of government growth that occurred in the 1920s.

Were it not for the Great Depression, government growth would have been slower in the 1930s than it was, because there would have been no call to respond to a national economic crisis. But government was expanding its programs, its powers, and its budget during the 1920s in the relatively passive presidential administration of Calvin Coolidge, as a part of the trend of government growth that had begun with the Progressive era around the turn of the century. If the trends of the 1920s had continued, federal government growth would have been substantial with or without the Great Depression and with or without the New Deal. For Coolidge, being pro-business did not mean being anti-government (or anti-labor, or anti-agriculture), and Coolidge supported the expansion of government in almost every area, although not to the degree desired by some of his critics.

The point of this paper is to try to shed some light on the remarkable phenomenon of government growth in the 20th century that is illustrated in Figure 1. By looking at the decade before the New Deal, this paper shows that it was not the New Deal or the nation's response to the Great Depression that triggered the growth of the federal government. The seeds were sown in the Progressive era prior to World War I, and the 1920s served to reinforce those principles of government established during the Progressive era by continuing to expand the government's reach. The 1920s did not represent a plateau in government activity that was reversed by the New Deal; rather, the foundations for the New Deal were established by the increasing scope of federal government activity during the 1920s.

Alchon, G. (1985) The Invisible Hand of Planning: Capitalism, Social Science, and the State in the 1920s. Princeton, N.J.: Princeton University Press.

Anderson, B.M. (1979) Economics and the Public Welfare. Indianapolis, Ind.: Liberty Press.

Anderson, G.M., and Tollison, R.D. (1991) "Congressional Influence and Patterns of New Deal Spending, 1933-1939" Journal of Law & Economics 34(1)(April): 161-75.

Anderson, T.L., and Hill, P. J. (1980) The Birth of a Transfer Society. Stanford, Calif.: Hoover Institution Press.

Beard, C.A., and Beard, W. (1930) The American Leviathan: The Republic in the Machine Age. New York: Macmillan.

Borcherding, T.E., ed. (1977) Budgets and Bureaucrats: The Sources of Government Growth. Durham, N.C.: Duke University Press.

Davis, J.R. (1971) The New Economics and the Old Economists. Ames: Iowa State University Press.

Fabricant, S. (1952) The Trend of Government Activity in the U.S. Since 1900. New York: National Bureau of Economic Research.

Hawley, E.W. (1979) The Great War and the Search for a Modern Order, A History of the American People and Their Institutions, 1917-1933. New York: St. Martin's Press.

Hicks, J.D. (1960) Republican Ascendancy: 1921-1933. New York: Harper & Brothers.

Holcombe, R.G. (1992) "The Distributive Model of Government: Evidence from the Confederate Constitution." Southern Economic Journal 58(3)(January): 762-69.

Holcombe, R.G. (1993) "Are There Ratchets in the Growth of Federal Government Spending?" Public Finance Quarterly 21 (1) (January): 33-47.

Horowitz, D.A. (1984) "The Perils of Western Farm Politics: Herbert Hoover, Gerald P. Nye, and Agricultural Reform." In C.E. Krog and W.R. Tanner (eds.) Herbert Hoover and the Republican Era: 220-43. Lanham, Md.: University Press of America.

Hughes, J.R.T. (1977) The Government Habit: Economic Controls from Colonial Times to the Present. New York: Basic Books.

Johnson, P. (1991) Modern Times: The World from the Twenties to the Nineties. Rev. ed. New York: HarperCollins.

Keller, R.R. (1982) "Supply-Side Economic Policies during the Coolidge-Mellon Era." Journal of Economic Issues 16 (3) (September): 773-90.

Kelly, C. (1932) United States Postal Policy. New York: D. Appleton and Company.

Kendrick, M.S. (1995) A Century and a Half of Federal Expenditures. New York: National Bureau of Economic Research.

Keynes, J.M. (1936) The General Theory of Employment, Interest, and Money. New York: Harcourt, Brace and Company.

Metzer, J. (1985) "How New Was the New Era? The Public Sector in the 1920s." Journal of Economic History 45(1)(March): 119- 26.

Murray, R.K. (1973) The Politics of Normalcy: Governmental Theory and Practice in the Harding-Coolidge Era. New York: W.W. Norton & Company.

Musgrave, R., and Culbertson, J.J. (1953) "The Growth of Public Expenditures in the U.S. 1890-1948." National Tax Journal 6(2)(June): 97-115.

North, D.C. (1985) "The Growth of Government in the United States: An Economic Historian's Perspective." Journal of Public Economics 28 (3) (December): 359-82.

Peacock, A.T., and Wiseman, J. (1961) The Growth of Public Expenditures in the United Kingdom. Princeton, N.J.: Princeton University Press.

Posner, R.A. (1970) "A Statistical Study of Antitrust Enforcement." Journal of Law & Economics 13 (2) (October): 365-419.

Rassler, K.A., and Thompson, W.R. (1985) "War Making and State Making: Governmental Expenditures, Tax Revenues, and Global Wars." American Political Science Review 79 (2) (June): 491-507.

Schmeckebier, L.F. (1929) The Bureau of Prohibition: Its History, Activities, and Organization. Washington: Brookings Institution.

Shughart, W.F., II, and Tollison, R.D. (1986) "On the Growth of Government and the Political Economy of Legislation." Research in Law and Economics 9: 111-27.

Wallis, J.J. (1984) "The Birth of the Old Federalism: Financing the New Deal, 1932-1940." Journal of Economic History 44 (1) (March): 139-59.

Wooddy, C.H. (1934) The Growth of the Federal Government: 1915- 1932. New York: McGraw-Hill.

Wynn, N.A. (1986) From Progressivism to Prosperity: World War I and American Society. New York: Holmes & Meir.

Randall G. Holcombe is the Devoe Moore Professor of Economics at Florida State University. The author gratefully acknowledges research assistance from Dual South, and helpful comments from Gary Anderson, Bruce Benson, Donald Boudreaux, Gary Founier, Robert Higgs, Russell Sobel, Gordon Tullock, and seminar participants at Florida State University, George Mason University, and West Virginia University.

Notes

[1] The idea that there was minimal government growth in the 19th century is subject to question, especially if government regulation is included in the definition of government. See Hughes (1977), Anderson and Hill (1980), and Holcombe (1992) for discussions of various aspects of 19th century government growth in the United States.

[2] Borcherding (1977: 45-70) examines the growth of the federal government from 1870 to 1970 and concludes that it has grown about twice as rapidly as one might have expected based solely on changes in underlying economic variables. By the same methodology he would surely have found government growing less rapidly than expected during the 19th century.

[3] Wynn (1986: 217) remarks, "Even the view of the Harding administration as 'conservative' is an oversimplification, and one not now generally accepted by historians. The new president had adopted a fairly liberal and sympathetic attitude to groups who had been dealt with quite harshly by the previous administration." A more extreme view comes from Anderson (1979: 125-27) who calls 1924 the beginning of the New Deal. Anderson calls the 1924 introduction of the McNary-Haugen bill designed to protect farmers from foreign competition the first piece of the New Deal legislation that continued through Franklin Delano Roosevelt's presidency. (The bill was not actually passed until 1927.) See also Higgs (1987), Johnson (1991), and North (1985), who see changes around the turn of the century as pivotal events in the history of government growth.

[4] Keller (1982) argues the similarities in the policies of Coolidge and Reagan--a similarity Reagan himself tried to draw on. In contrast, Metzer (1985) agrees with the thesis of this paper, arguing that public policy in the 1920s was a continuation of the prewar expansion of the federal government. For a more general treatment of the history of the time as being a part of a transformation to government growth, see Polanyi (1944). Also see Beard and Beard (1930), who approvingly refer to American government as a Leviathan, and even dedicate their book "to the thousands of men and women who loyally serve the public on land and sea under the auspices of the Government of the United States."

[5] In response to the Depression, Roosevelt started many new programs, of which Social Security is the largest survivor. But Roosevelt also abolished many of the programs he started in response to the economic crisis. The Civilian Conservation Corps, the Works Progress Administration, and the National Youth Administration were among the programs that were terminated by FDR.

[6] Wallis (1984) sees a major change in the 1930s as federal government replaced local government as the dominant spending unit, but Metzer (1985), in agreement with the current study, argues that the underlying government growth was present in the 1920s. Other writers, such as Fabricant (1952) and Musgrave and Culbertson (1953) have noted the rising trend of government expenditures prior to the New Deal. Kendrick (1955) anticipates the ratchets hypothesis of Peacock and Wiseman (1961) by arguing that wars become bridgeheads for maintaining larger postwar expenditures.

[8] Murray (1973: 41) says, "Almost by default, the Republican senatorial leadership attempted to speak for the party before Harding's inauguration in 1921." Murray goes on to describe the move toward government programs favoring special interests. As Holcombe (1992) notes, the problem of special interest legislation was well recognized even before the Civil War, and Anderson and Tollison (1991) document the influences of special interests on New Deal expenditures.

[9] As Wynn (1986: 219) notes, many of the goals that the Progressives had early in the century had been accomplished by the 1920s, which would naturally lead the movement to lose some of its force.

[10] One political issue in the 1920s that is often cited as an example of Coolidge's pro-business leanings was his attempt to sell a hydroelectric plant and nitrate plant in Muscle Shoals, Alabama, to Henry Ford. Coolidge was blocked by Congress (Ables 1969: 70-71), and the facility later became the beginning of the Tennessee Valley Authority.

[11] This impression may be furthered by the Progressive elements in the Republican party that continued to oppose Coolidge. See, for example, Hicks (1960: chap. 4).

[12] Harding's Cabinet is viewed by some as the strongest Cabinet in American history. Of the four most prominent members, three--Charles Evans Hughes, Secretary of State; Henry C. Wallace, Secretary of Agriculture; and Herbert Hoover, Secretary of Commerce--were associated with Republican Progressivism. Andrew Mellon, Secretary of the Treasury, was the only major Cabinet officer without Progressive leanings (Hawley 1979: 58-59).

[13] In 1929 real per capita federal expenditures in 1990 dollars were $195.41. They increased to $367.84 by 1933, an 82 percent increase. By 1937, an equal number of years into the Roosevelt administration, per capita federal expenditures were $545.23, only a 48 percent increase, and they had gone up to $638.78 by 1940.

[14] This ratchets hypothesis was originally put forward by Peacock and Wiseman (1961), and plays a significant role in discussions of government growth by Rassler and Thompson (1985) and Higgs (1987). While these ratchets undoubtedly exist, Holcombe (1993) casts some doubt on this hypothesis as an explanation of government growth.

[15] One category, listed by Wooddy (1934: 545) under "general and financial administration" as a subset called "other," declined from $8,672,000 in 1920 to $601,000 in 1930. One must suspect that there are significant war-related expenses in this "other" category.

[17] The Merchant Fleet Corporation was originally established under the name United States Shipping Board. Additional details on its operation are given in Beck (1932).

[18] Figures on postal expenditures are from Wooddy (1934) unless otherwise noted.

[19] One effect of Burleson's tenure as Postmaster General was to reduce morale among postal employees. See Cullinan (1968: 124-40) for a discussion of Burleson's tenure as well as those who succeeded him during the Harding administration.

[20] The Coast Guard's own figures on the subject show that more than half its budget during the 1920s, amounting to about $15 million a year, went toward enforcing Prohibition (Schmeckebier 1929: 14). The Justice Department estimated that about one-third of its budget, or about $10 million, would be spent enforcing Prohibition in 1930 (ibid.: 308).

[21] This impression was accentuated by Andrew Mellon's relative prominence as Secretary of the Treasury, a position he held through the Harding and Coolidge administrations. Ables (1969) calls the Harding-Coolidge administrations the rule of Mellon.

[22] See Hicks (1960: 196-218), and for a discussion of Hoover's role in agricultural policy both during his and Coolidge's administration, see Horowitz (1984).

[23] Shughart and Tollison (1986) show a tremendous increase in bills introduced into Congress (and a less dramatic increase in laws enacted) before the 1920s, providing another indicator of the growth of government.