Why is Government So Small in America?⁽¹⁾

The budget is the skeleton of the state, stripped of all misleading ideologies (Joseph Schumpeter).

Anyone who witnesses the current debate on health care reform in the U.S. knows that a (or the) critical constraint on the expansion of public assistance in this area is our inability to raise sufficient funds to finance a comprehensive program -- even when over 80% of the American public believe that "the government should insure that no one goes without health care because they cannot afford it" (Gallup, 1991). The case of health care policy is not unusual. The vast majority of public spending programs in America are widely supported by strong majorities of the public, by the politicians who sit on the committees that oversee the particular programs, and by the civil servants that administer them. It is very clear that Americans do not like the symbol "big government," but when asked about government programs for public roads, public education, the national guard, Aid to Families with Dependent Children, the environment, public health, drugs, crime, space exploration, and virtually every other specific thing that government provides the public, large majorities of Americans say they either like the programs as they are, or want to see them increased.⁽²⁾

A reasonable question, then, is: Why are these programs not larger than they are? Why, indeed, does the United States spend less on social programs than any other western democratic nation? Virtually all of the scholarly explanations for America's exceptionally small welfare state, focus almost exclusively on the spending side of the public budget. I suggest that this spending focus can tell us, at most, half of the story. Given the overwhelming evidence demonstrating both public and elite preference for continued and/or increased spending in most specific areas of public activity, I believe that we need to focus on the other side of the public coin -- taxes.

The evidence about public and elite attitudes towards taxation is even more clear than the evidence we have about spending. People hate taxes.⁽³⁾ I believe that the key to understanding the "exceptional" size of the modern welfare state in America, lies not in differences in spending preferences between America and other democratic nations, but instead in the American state's relative inability to raise sufficient tax revenues to fund these spending preferences. In short, public spending in the U.S. is lower than elsewhere, because taxes are lower.

In this paper I will argue that the key to understanding the low tax revenues in the United States is that our unique political institutions bias the system against raising revenues. Specifically, I suggest that two basic features of the American tax structure explain its low revenue yield. The first is that the U.S. has no national consumption tax. The second is that the U.S. gives away between 5 and 10 percent of GNP in "tax expenditures" each year.⁽⁴⁾

This paper will show that the specific character of the American tax system is best explained through an examination of the unique institutional structure through which tax policy must pass. I have no quibble with those that argue that Americans have unique or exceptionally individualistic values, ideologies or culture: But this paper will demonstrate that these broad characteristics provide us little leverage for understanding why the U.S. has made the particular tax policy choices it has over the past century. I instead suggest that the institutionally defined incentive structures facing American tax policy makers provides powerful analytic leverage over the questions, how and why the U.S. has built its "exceptional" tax system.

Exceptional Taxes

Before proceeding, I would like to highlight some basic features of the modern American tax system. First, as table 1 indicates, it is important to note that all taxes are lower in the U.S. than in our neighboring democracies. Moreover, the U.S. collects near or above average amounts of revenue in some of the most progressive kinds of taxes.

Tax revenue of main categories, 1989

Shown as percentage of GDP

Country

Personal

Income

Social Security

Property

Goods & Services

Profit

All Taxes

Sweden

22.0

14.7

1.9

10.4

2.1

56.1

France

5.2

19.2

2.2

12.6

2.4

43.8

Germany

11.2

13.8

1.2

9.7

2.1

38.1

U.K.

9.7

6.4

4.6

11.3

4.5

36.5

U.S.

10.7

8.8

3.1

4.9

2.6

30.1

EEC avg.

10.4

11.3

1.9

12.6

3.0

39.9

Source: OECD Revenue Statistics 1965-1990, OECD 1991 Table 6 Table 2a

All levels of government included.

Secondly, contrary to what many of us believe, tax rates on the very wealthy have historically been quite high. Indeed the U.S. has had higher top rates on personal income, corporate profits and inheritances than even Social Democratic Sweden for most of the Twentieth century.⁽⁵⁾

Thirdly, consumption taxes in America are much lower than they are in most of the rest of the world. In 1989 the US collected 4.3% of GDP in consumption taxes while the OECD Average was 11.1% and the European average was 12.0%. The key to this discrepancy is the fact that the U.S. is the only advanced industrial democracy which has never instituted a broad based national consumption tax. It is significant to note that if the U.S. had average consumption taxes, total tax revenues would be 36.9% of GDP, which is only 1.5% less than the OECD average.

Finally, the structure of the American tax system is defined by the literally thousands of "tax expenditures" or "tax subsidies," or simply "loopholes" embedded in the system. Tax expenditures undermine the efficiency of the system, and lose huge amounts of tax revenue. There are no reliable international statistical comparisons of the revenue costs of tax expenditures across countries, but both my own studies and interviews with tax policy experts around the world confirm the view that the U.S. tax code has far more tax expenditures than any other tax system in the advanced industrial world.⁽⁶⁾ Though the 1986 tax reform act did reduce the total value of these expenditures, it did so mostly by radically cutting tax rates and thereby reducing the dollar loss of the tax expenditures. But even with this "historic reform" the U.S. Federal government will still offer tax subsidies worth an estimated 361 billion dollars in 1993 (U.S. House Ways and Means Committee, 1990:235).

In sum, the American tax system is distinguished from that of other rich industrial democracies by its relatively heavy reliance on progressive taxes, the very low levels of consumption taxes and finally the extreme use of tax expenditures. Taken together these three features explain the low levels of tax revenue collected by American governments.

The rest of this essay seeks to explain these curious tax policy outcomes. I do so through a historical examination of some major tax policy developments in the twentieth century. I do not attempt a comprehensive history of American tax policy development, but instead focus on several key epochs demonstrating the ways in which America's unique institutional structure shaped particular policy choices. Taken together these choices have made the American tax system one of the most complex and weak tax systems in the world. The consequence of this has been less tax revenue and thus less money to spend on the programs Americans quite clearly desire.

The apportionment of taxes on the various descriptions of property is an act which seems to require the most exact impartiality; yet there is, perhaps, no legislative act, in which greater opportunity and temptation are given to a predominant party to trample on the rules of justice. Every shilling, with which they overburden the inferior number, is a shilling saved in their pocket.⁽⁷⁾

James Madison was an institutionalist. Fearing that one could not depend upon enlightened rulers or a transformation of the basic human character, he argued for political institutions which would fragment authority and thereby make it exceedingly difficult for any faction to wield public power against the interests or rights of other factions. As the quote above demonstrates, the abuse of the power to tax was of paramount concern to the Federalists.

The central argument of this essay is that Madison's institutions worked: The fragmentation of public authority within national political institutions in America has in fact made it extremely difficult for any faction, no matter how well or ill intentioned, to impose taxes on the rest of society. Indeed - except during the emergency of war - it has been nearly impossible for one faction (even the majority faction) to raise taxes in America.

The following pages will demonstrate how the combination of a single member electoral system, the separation of powers, and American federalism have shaped the incentive structure of politicians and interest groups alike to a create a tax system that provides less revenue than the tax system of any other advanced democracy. But as important as the basic constitutional structure is, it does not define the structure of modern decision making institutions. Not only do these political institutions change over time but they do not operate in an historical vacuum.

In the first century of American political development the national state remained loosely structured, decentralized and relatively unspecialized. Decision making tended to be made in informal manner in which the institutions could be adapted to the particular needs of the day. But as Stephen Skowronek convincingly shows, this institutional structure was insufficient for the challenges of the modernizing nation. "Industrialism, in all its dimensions, exposed severe limitations in the mode of governmental operations that had evolved over the nineteenth century and that supported the powers and prerogatives of those in office" (Skowronek, 1982:13). Though political parties had worked to provide some cohesive authority to the national system (along with the courts) federalism worked to undermine their national unity. Moreover, as the tasks of government increased in complexity, the lack of substantive specialization in the legislatures inhibited their abilities to govern.

European parliamentary systems dealt with the complexities of modern governing through both the centralization of political authority and concentration of responsibility for decision making into the hands of the executive. In the U.S., however, the executive and legislative branches of government were constitutionally separated and intentionally set at odds with one another. Thus the legislature in America was better able to fight off the encroachment of its power by the executive than were its European counterparts. Rather than conceding authority over substantive policy to the Presidency, Congress created an elaborate committee system and devolved political authority for particular policy decisions to individual committees and subcommittees (Polsby, 1968).

This division of institutional responsibility has at least two important consequences relevant to this study. The first is that Committee Government weakens political leadership. Each member of Congress has access to his or her own base of institutional power and therefore are less dependent upon those higher up the institutional hierarchy for support. This means that it becomes exceptionally difficult for those in leadership positions to push or pull Congress into policy positions with which individual legislators are uncomfortable. Politicians are always uncomfortable with taxes.

In the following pages we will see numerous examples in which the President and/or the party leadership in Congress has failed to bring Congress to support their tax policies. In more centralized parliamentary systems, individual legislators are bound to support their party's position on important revenue bills. This institutional fact gives these same legislators the 'political cover' to support measures which they may find political distasteful in the short run. Members of the US Congress, in contrast, are individually accountable for their actions. There are, thus, powerful incentives for these legislators to pay closer attention to the short-term electoral consequences of their votes than to the long-term policy effects of their actions (cf. Mayhew, 1974; Fiorina, 1977).

The second important consequence of Committee Government for tax policy is that authority and responsibility for taxing and spending decisions are separated from one another. This institutional fact has had enormous implications for the development of taxation policy in the U.S. throughout the 20th century. In more centralized parliamentary systems political actors who must take the blame for tax increases can at the same time take the credit for the spending increases that the new revenues enable. Under Committee Government, the incentive structure works quite differently. Those who must take the political responsibility for tax increases get little political credit for the spending increases which new revenues make possible; but they must suffer the political ire of their constituents for the tax increases which they impose. Members of Congress who sit on revenue committees can, however, take credit for introducing tax reductions - particularly when they are focused on specific constituencies.

In sum, though the US Constitution does not precisely define the character and structure of the tax decision making process in the US, it does set the foundation upon which "Committee Government" has been built. Committee government, in turn, undermines both political leadership and fiscal coordination in the modern American decision making process and shapes the incentive structures of those that participate in tax decision making. The inefficient and low revenue tax system I have described above is the direct consequence.

Region, Class and the Income Tax

The limitations imposed be the United States Constitution have, in several cases, an important effect on the scheme of the Federal Income Tax (Harrison Spaulding, 1927).

The politics of taxation in the U.S. at the turn of the 20th century was both a politics of class and a politics of region. For nearly forty years after the Civil War the Republican and Democratic parties were divided more by region than by political philosophy or ideology. This regional division allowed the parties to focus on providing constituent benefits and representing local interests rather than championing divisive political ideologies or explosive policy demands. But by the last decade of the nineteenth century, this system proved increasingly incapable of integrating the new political forces and demands evoked by the rapid industrialization. Republicans had dominated the national politics since the end of the Civil War, and were widely perceived (especially in the South and the West) as representing the monied interests of the industrialist and financier classes of the North and East.

Popular discontent grew slowly at first, but when the Republican's introduced their famous Mckinley Tariff in 1890 frustration with the system turned into a firestorm⁽⁸⁾. The Democrats attempted to capture the moment and moved quickly to steal the platform of the Populists, promising to introduce a series of populist policies including tariff reform and a national progressive income tax. The Democrats were immediately rewarded - if only temporarily. Despite the fact that they received a smaller percentage of the popular vote (46.1%) for members of the House of Representatives, than they had just two years before, they won 71.1% of these seats in Congress. With the democratic/populist landslide and the party's new found radicalism, it seemed certain that the redistributive taxes would soon become an integral part of the Federal tax system.

Within two years Congress passed the nation's first permanent progressive income tax⁽⁹⁾ (which was not signed by President Cleveland, but was allowed to pass into law). However, this law was never put into practice. In the United States, unlike most other democratic nations, those who opposed the Income Tax had an additional institutional layer in which they could block the legislation not available in Parliamentary democracies: The Supreme Court. Opponents to the tax successfully convinced the Court to rule the new law unconstitutional.

In what Waltman has subsequently described as "specious" logic, the Court justified their decision to block the tax on the grounds that it was a "direct tax" and that the Constitution specified that a "direct tax" had to be apportioned according to each state's population (Waltman, 1985: 4). The Court did not specify what it meant by a "direct tax", but was clear about the intent of their decision: Justice Field argued that the Income Tax was in fact class legislation . He put the bottom-line issue most bluntly when he wrote, "The present assault upon capital is but the beginning. It will be but the stepping stone to others, larger and more sweeping, till our political contests will become a war of the poor against the rich; a war constantly growing in intensity and bitterness" (Concurring Opinion, 157, US: 428).

Political Reform

Justice Fields was not right; the war between rich and poor did not escalate to the massive national class confrontations which he feared. The strong class-based parties like those emerging in Europe at the time continued to be very were weak in the United States. I believe that these different outcomes were largely the product of different institutional structures though which these new political forces were channelled. Although there were many similarities in the general political demands of both middle and working class interests in Europe and the United States, the institutions through which these demands had to be expressed differed dramatically.

In Europe as well as America, political reformers at the turn of the century They argued that political power was held by a ruling class of economic elites and that this concentration of political and economic power was both illegitimate and unjust. European reformers concluded that the way to redress this problem was though the extension of suffrage and the construction of mass based political parties which could seize the reins of government and use political authority for the benefit of all citizens (Vallinder, 1962; Fraser, 1973; Beer, 1965). But in the United States, male suffrage had already been extended; as a consequence, the bitter political fights evoked by the economic changes of the era took a peculiar turn.

In the U.S., political parties were perceived to be obstacles to democracy rather than agents of its implementation. The expanding middle class was particularly fearful of the potential for demagoguery in mass based parties. The political agenda for democratic reformers in the United States, then, required the dismantling of these obstacles to democratic accountability rather than strengthening them (Hayes, 1957; Burnham, 1970). Progressive reformers thus demanded, and eventually won, a series of measures specifically designed to reduce the power of political parties and their leaders. Some of these measures included the introduction of the Australian ballot, the direct primary, the direct election of United States Senators, the initiative process, and manipulation of voting qualifications and requirements (Burnham, 1970: 72-90). These reforms reduced the power of party elites and weakened political parties generally. They also had profound consequences for the character of tax policy making for the rest of the twentieth century.

This tide of 'democratic' reformism eventually swept into the halls of Congress. Here too reformers argued that the centralization of political authority in the hands of party elites (particularly under the near autocratic leadership of Speaker Joe Cannon) undermined the democratic principles which the nation was founded upon. Speaker Cannon had tried to hold together his splintering Republican Party through control over appointments to congressional committees. But his control did not last for long. Insurgent Republicans and Progressives joined together and revolted. Not only was Speaker Cannon deposed from the Speakership in 1911, but the reformists also instituted a series of reforms specifically designed to protect themselves from the will of their party leaders. The most important of these reforms for our purposes was the institutionalization of seniority selection for committee chairmanships (Polsby, 1969). From this point forward, individual Congressmen would dramatically expand their individual sphere of influence and thus become substantially freer to each steer his or her own policy course - even when opposed by party leaders and/or the President.

The Right to a Class Tax -- The Sixteenth Amendment

Somewhat surprisingly, it was the Republican President William Taft who sponsored the idea for a constitutional amendment permitting a federal income tax. He hoped this strategy would undercut the demand for an income tax as well as demands for political ("democratic") reforms which were wedges which threatened to split his party. Knowing how difficult it would be to pass a Constitutional Amendment, Taft felt certain the amendment would fail. It had, after all, been nearly forty years since any amendments to the Constitution had passed.

Unfortunately for Taft, his strategy backfired. As he predicted, the current income tax legislation quickly died. Little did Taft know that by forwarding the Constitutional Amendment he would undermine the last bulwark against the "socialist scheme" his supporters so bitterly opposed.⁽¹⁰⁾ By June, 1913, 42 states, six more than the two thirds necessary, ratified the 16th amendment. The Federal Income Tax Act followed ratification very quickly. This tax imposed a "normal" tax of 1% on all individual incomes over $3,000 or family incomes over $4,000 and a "surtax" which rose from one to six percent was levied on incomes over $20,000. Only a tiny fraction of citizens were affected by the new law.

The process of putting the income tax proposal up before each state legislature, however, had its costs. In their efforts to build support for the amendment, proponents promised a variety of tax exemptions to hesitant supporters in exchange for their acquiescence. Cordell Hull, the author of the 1913 law, subsequently disparaged in the following way: "A possible danger to the successful and permanent operation of an income-tax law, as is true of all tax laws, is the disposition of its friends to insert additional exemptions here and there and to add liberal qualifications, thereby opening many doors to those who would evade or avoid their full share of taxes" (Waltman, 1985: 79). For example, as the Amendment went from statehouse to statehouse, the fear that this new tax might undermine the revenue generating capacity of state governments began to grow. In an attempt to assuage this fear, and ensure its passage through the various state houses, proponents of the tax promised the introduction of the first explicit tax "loophole." It was therefore agreed that interest from state and local bonds - which was becoming an increasingly important source of revenue for these governments -- should be exempted from the new federal tax. It was clearly understood that not only would this provide an incentive for wealthy individuals to buy these bonds, but also that these same individuals would be able to use the bonds to protect their income from the federal taxman.

In sum, even in the very first stages of the construction of the "modern" tax system, the fragmentation of political authority in the United States had begun to shape the development and structure of our national tax system. At the very outset the Federal Income Tax was forced to open itself to specific loopholes in order to wind its way through the institutional labyrinth created by the United States Constitution. This was an inauspicious beginning: It would not take long before the doors swung wide open.

The Institutionalization of Progressive Taxation - Finance and War

By the second decade of the 20th century the US had finally passed both national personal and corporate income taxes. It would not take long for these taxes to be transformed into precisely the kind of political weapons and financial honey pots which their opponents had feared. Within five years tax rates would shoot up to 77% on the very wealthy, and in less than ten years these two taxes would generate more revenue for the federal government than all other taxes combined.

The primary explanation for this immediate transformation in these taxes was WWI. While there were a great deal of isolationist sentiment in Congress at the time, there was little support for being unprepared. The new income tax, it was quickly discovered could be a remarkably efficient revenue source. The tax came at a particularly convenient time because of a precipitous decline in tariff revenues. Moreover, it made great political sense to tax those who were making huge riches from the war instead of increasing taxes paid by the very same families who spilt blood in this rich man's war.

Still, the income tax was becoming a significant burden to politically powerful interests. And it did not take long for these interests to begin their campaigns for tax reductions. But in the context of both the strongly populist/progressive climate of the decade, the increasing lack of coherence in Congress and the reassertion of Congress' power vis-a-vis the President, it was politically infeasible to reduce tax rates significantly. But it was not impossible to get narrow tax breaks for specific interests. At first it was the "worthy causes" whose political appeal was difficult to resist: but it would be too long until others followed suit.⁽¹¹⁾

In response to the erosion of the tax base, President Wilson introduced what was probably the first Presidential effort at "tax reform" on May 27, 1918. But his proposals did not go far in Congress. Both the House and Senate tax committees were "lobbied heavily" and instead decided that they should both increase marginal rates on the wealthy and open new loopholes for many of these same interests (Waltman, 1985:60).

Return to "Normalcy"

When the war in Europe was finally over and the soldiers began to return home it became obvious that both personal income and excess profits taxes would have to fall. But who was to benefit? One might expect that with the resounding Republican victory of Harding over Cox in 1920 -- and particularly given President Harding's decision to appoint Financier Andrew Mellon to be his Treasury Secretary -- that marginal tax rates on the wealthy would soon fall. But again, the system of 'checks and balances' confounds predictions which would, in a parliamentary system, be certainties.

By the 1920s, the committee system became firmly established in Congress and the system of seniority appointment to committee chairmanships was equally well ensconced. "Party cohesion and majority rule through the caucus gave way to majority rule in the House based on a powerful cross party alliance which supported and maintained the seniority system.... [Therefore] decentralization took a major leap forward" (Polsby, 1969:802) This institutional reality would shape tax policy in the United States at least until the next Great War.

Immediately after Harding's election, business organizations mobilized a campaign to replace the Corporate Profits Tax with a national sales tax. Mellon agreed to their suggestions and proposed the replacement of these taxes along with cutting personal income tax rates to a maximum of 20%. But neither of these suggestions would come to pass. Cordell Hull described the politics of this "tax reform" as follows:

It was most unfortunate that the attempted revision legislation of 1921 degenerated measurably into a wrangle between champions of large income taxpayers and those of smaller taxpayers each striving to see which could unload the largest amount of taxes first. The legislative situation thus became so confused and demoralized that but scant opportunity for consideration and comprehensive scientific tax revision was afforded (quoted in Witte, 1986:91-2).

Even though Republicans controlled both the House and the Senate, as well as their tax writing committees, Mellon could not get them to go along with his tax proposals. Not only was the sales tax rejected,⁽¹²⁾ but even his income tax rate reductions were cut back substantially. Though rates, particularly the "normal" tax rates paid at the bottom of the tax scales, were cut in the 1921 through 1925 tax bills, most of the political action turned around what have subsequently been called "tax expenditures." The Senate alone added 833 amendments to the 1921 tax bill.⁽¹³⁾ Mellon was furious at his Republican colleagues.

Eventually Mellon was able to get the rate cuts he wished for and by 1925 the top tax bracket was reduced to 25%. What he could not do, however, was prevent Congress from undermining the fiscal integrity of the tax code. As the Income Tax became a significant economic issue for powerful interests, it also became a significant political issue for the politicians who wanted to please their constituents. Jerold Waltman's superb study of the politics of the early income tax describes the political dynamics already in force.

We have already alluded to how the question of rates was sidestepped by affected interests.... What they could press for, however, was special treatment of some items of receipt or expenditure. Lowering income by either of theses expedients could mean substantial savings on one's tax bill. Congress was to be faced from this point on with pressure from various economic interests to alter this or that section of the tax code. Of course, once such an exception was granted, others presented themselves pressing analogous arguments, frequently with some justice.....The result is not only a panalopy of interest groups pleading that their business is different from and deserving of some special treatment; it is also that the code itself becomes more complex. This very complexity creates its own anomalies, which means more attempts must be made at rectification. This in turn begets more complexity, leading to an unbreakable circle (Waltman, 1985:78).

Many Presidents, Treasury Secretaries, party leaders and Chairmen of the Ways and Means or Finance Committees would subsequently attempt to break the "unbreakable circle." Occasionally their efforts would meet with some modest success and some of the more outrageous loopholes in the tax code would be removed or reduced. It was far more common, however, that tax reforms would be turned into tax giveaways.

Taxation under Roosevelt

One might have expected that FDR, who is widely recognized to have been one of the most successful Presidents of the twentieth century, to have been able to break Waltman's "unbreakable circle." But the reality was somewhat different.

At first, Roosevelt was not particularly interested in taxation policy. Only after he and his administration became committed to a large series of public programs did he begin to change his attitudes about taxation. Roosevelt understood quite well that the Southern Conservative Democrats who now controlled the revenue committees in Congress would not go along with big tax increases on corporations or the wealthy.⁽¹⁴⁾ Much as with proposals for National Health Insurance, Roosevelt chose not to introduce tax policy proposals which would almost certainly make it more difficult to pass other (hopefully more acceptable) New Deal programs. In short, faced with the institutional realities of Committee Government, Roosevelt apparently decided not to fight a two-front war.

Still Roosevelt's strategy did not work. And in 1935, increasingly frustrated by both business and Congressional opposition to his New Deal programs, Roosevelt finally embraced a new tax strategy. Roosevelt proposed a dramatic "Share the Wealth" Budget in 1935 ⁽¹⁵⁾ and introduced the Social Security Plan. Roosevelt's new tax strategy did not work. Instead, the administration's relationship with Congress began to deteriorate even though many of them had come to office on his electoral coattails. For example, the Treasury decided to use the 1937 budget to take aim at many of the increasingly obvious tax loopholes used by the economic royalists in legally avoiding their fair share of the tax burden. A modified version of the administration's bill did pass through Congress but, as Ratner notes:

The 1937 Revenue Act was a step forward in closing some of the loopholes in the income tax laws, but it failed to deal with such important tax avoidance devices as tax-exempt securities, undistributed profits, capital gains, single premium life insurance policies issued by fictitious companies, pension trusts, community property laws, percentage depletion, and multiple trusts for accumulating income... tax attorneys still saw plenty of ways for their clients to escape paying the price for civilized society. (Ratner, 1942: 478)

After 1937, Roosevelt's influence in tax matters declined even further. Soon it became clear that he could not realistically hope to gain any of his radical tax policy goals. Rather than raise taxes on the wealthy to finance more public spending and to redistribute wealth as FDR proposed, the Congress cut effective tax rates for most individuals and introduced (or sometimes re-introduced) tax loopholes for powerful constituents. Congress was increasingly coming under fire from the President, the Treasury and from academic analysts for ignoring the continuing deficit and for using the tax code as a political plum to pay off constituents. But their incentive structure led them to listen to their constituents and not academics or Presidents. In their 1940 text, The Federal Income Tax, Roy and Gladys Blakey describe the passage of the 1938 tax bill in the Senate with more than a small amount of frustration: "As the clerk speedily read the bill for amendments, the chairman punctuated his steady flow of words with the formula, 'Without objection amendment agreed to,' spoken as though it were one long word. At the end of twenty minutes Vice President Garner turned the gavel over to Minton with the remark that they had already passed 224 pages of the measure" (Witte, 1985: 107-8).

In sum, Even Franklin D. Roosevelt, probably the most powerful and successful president in the 20th century, was unsuccessful in changing the basic dynamics of tax policy making in the U.S. "The realignment of 1928 to 1932 gave the New Dealers the power to make many changes in federal tax policy, but they did not fully control Congress, the Supreme Court, or state governments. It is therefore not surprising," Susan B. Hansen tells us, "that the impact of the New Deal was not as great as its supporters had hoped or as its conservative opponents had feared" (Hansen, 1983:86).

The policy result was that this meant that though representatives of both the Administration and the Congress could not agree on broad tax policies, specific members within Congress could agree to a large number of particular measures designed to benefit particular constituents back home. The Treasury as well as presidents from both parties fought the Congress over these amendments, but in the United States, Congress, not the President, writes tax law.

A Return to Wartime Finance

It did not take long after the outbreak of hostilities in Europe for national political elites in America to realize that massive increases of revenues were necessary. Almost all elites understood that war preparation would have to be financed through mutual sacrifice.

Still, the Federal government had significant problems to surmount in translating this general agreement into specific tax policies. The institutional fragmentation of American politics did its work even during the wartime crisis. Whereas the European parliamentary systems allowed for coalition governments, in the American case the administration could neither formally invite its antagonists into the government - and thereby share the glories and defeats of government - nor could it suspend popular elections, as had been done in Britain between 1935 and 1945. These institutional facts left both business interests and Republicans free to sit outside and criticize the government. At the same time, not formally including these interests into a governing coalition had the effect of preventing them from gaining a stake in the policies developed during the war. Rather than building support for the administration and its reformist ambitions, the war in many ways had the effect of undermining the New Deal coalition.

Though America's wartime tax policies were in many ways quite similar to those of most European nation's tax policies - there were also important differences. As in Europe, the massive revenue needs of WWII demanded that the modern income tax be expanded dramatically. Policy makers by now well understood that this tax was both revenue rich and relatively efficient to administer. In addition to dramatically lowering the income tax threshold, a new Pay-As-You-Earn (PAYE) system was introduced. Under this system the tax would be calculated and paid with each income earner's paycheck, rather than quarterly or annually as had been the case before. This innovation both brought in much needed revenues more quickly and efficiently and made it easier to collect the tax from virtually anyone who earned a paycheck. Thus, with WWII the rich man's income tax became "mass tax." In addition, in the U.S., as in Europe, "Excess Profits" taxes were passed and marginal income tax rates paid by the very wealthy shot up to rates nearing 100%.⁽¹⁶⁾

The most significant difference between European and American wartime tax policy was the American government's inability to impose a national consumption tax. The administration fought quite hard for a national "Spendings Tax" early in the war, but the Congress could not be convinced even though it was widely agreed that additional revenues were needed⁽¹⁷⁾ and it was commonly assumed that some form of consumption taxation would in fact be passed. Once again, however, because of fragmentation of authority within Congress, no specific proposal could wind its way through the legislative labyrinth. Each was shot down by a different objection; some were too regressive, others were too complicated. In the end, none could generate broad enough support to move through both Houses. Treasury official Randolf Paul described the fate of the administration's spending tax proposal in the following way.

The spendings tax foundered on the rocks of administrative intricacy. There was the perennial dilemma. When a revenue proposal is presented in general terms, Congressmen want to know how it will be implemented. When the particulars of a proposal are immediately offered, the mass of details makes it seem too complicated. The Treasury was impaled on the later horn. (Paul, 1947: 103)

A second important difference between American and our allies' wartime tax policy had to do with the number of tax exemptions offered to particular industries and interests. Despite the broad agreement that taxes needed to be raised, there were powerful forces demanding special treatment. These forces often had powerful friends on the revenue committees. As a result, Congress agreed to very steep income and excess profits taxes, but also opened a plethora of loopholes to some of the very industries whose profits increased due to the war.

As a result of the large numbers of tax expenditures offered during the war, the American tax system was becoming dramatically more complex than its European counterparts. Randolf Paul described the 1942 law with distaste: "The rate structure had reached the point where loopholes resulted in drastic loss of revenue, and where inequities and discrepancies threatened to be not only troublesome, but even disastrous to taxpayers" (Paul, 1947: 318). John Witte bluntly states, "The complexity of the bill was overwhelming" (Witte, 1985: 118).

As the war progressed, resistance to high taxes and congressional catering to special interests only became worse. In 1943, for example the year after the 1942 election undermined progressive forces within the Democratic party, Congress essentially ignored the tax bill sent to them by the administration. The law finally passed by Congress was so utterly littered with special provisions that Roosevelt labeled it "not a tax bill, but a tax relief bill" as he vetoed the measure. This was the first revenue bill ever to be vetoed by an American president -- it was also the first revenue bill to have the president's veto overridden by Congress. Yet Congress was still controlled by the Democratic Party when it overrode Roosevelt's veto.

In any parliamentary system, a conflict over revenue policy of this magnitude would not only bring down the government, but might also evoke a constitutional crisis. In America, the conflict between the President and the Congress over the tax policy had by now become routine. Rather than bring down the government, it was becoming commonplace that the president/Treasury's tax measures -- designed to champion the broad general interest and take away tax privileges from the special interests -- be fundamentally rewritten by Congress. Even during wartime, Congress' constituency is not "the nation." It is the particular constituencies which congressmen (and some women) represent.

Post-War Tax Policy, Revenue Gain without Political Pain

The essential outlines of the modern American tax system were thus well established during WWII. After this period, no new taxes are passed and in fact the general story line is of repeated tax cuts. During the wartime emergency political elites were able to push through the kinds of tax policies which would have been impossible absent the crisis. But once the modern tax system was set in place the politics of revenue was forever changed.

Not only had the income tax now been expanded to include the vast majority of income earners; but, given the progressive nature of this tax and the PAYE revenues from this tax would now expand almost effortlessly. The mass based income tax thus introduced a new fiscal logic to the financing of the state. Because income taxes were taken out of each paycheck, taxpayers would scarcely notice increases in their tax burdens. Additionally, because the tax was progressive income tax revenues would grow faster than national income.⁽¹⁸⁾ Thus a new political logic was born: Traditionally increases in spending had to funded specifically through tax increases, now increases in spending could be financed while politicians cut taxes. It was a politician's dream come true. They could take credit for cutting tax rates and/or providing special tax breaks to constituents while at the same time they could take credit for spending more of the taxpayer's money.

This would be the pattern followed for most of the next four decades. When new tax legislation was desired, whether by the president or the congressional leadership, the Republicans or the Democrats, tax benefits had to be yielded throughout the system if the tax legislation was to have a chance getting through the system of checks and balances. Once again, given the fragmentation of power on America, opposing sides could stymie each other's first preferences (whether tax cuts for the poor or rate cuts for the wealthy), and force them into choosing compromise policies. This general rule has prevailed even when the President and the majority in Congress were of the same party, as was the case from 1949-1951, 1953-1955 and 1961-1969. No one has been able to pass tax legislation without literally buying off opposition from representatives interested in protecting their constituents.⁽¹⁹⁾

The difficulty was not, of course, that policy makers were unaware of the problems tax expenditure policies were creating for the tax system. Rather, the electoral incentive encouraged Members to cater to the short-term interests of their constituencies even when this meant overlooking the longer term interests of the nation (Witte, 1985: 142; Manley, 1970: 338).

Even after President Eisenhower was elected and his party controlled both the House and the Senate, Republicans were unable to pass broad tax rate cuts as they promised voters. They were instead forced to choose tax cuts by way of exception, special rule, and particular allowance.⁽²⁰⁾ The Republicans, too, fundamentally accepted the role of the federal government in the modern economy and understood that the modern tax system must reflect this reality. John Witte sums up the politics of taxation in the 1950's in the following way:

[T]he U.S. economy had become much more complex by the 1950s, and the rudimentary laws of the early income tax were no longer sufficient. To match this complexity, the code needed to distinguish between corporations with income earned abroad, partnerships, holding companies, closely held corporations, and a wide variety of tax-exempt organizations. Complex organizations lead to complex sources and flows of income and costs, which in turn lead to demands for different treatment. Although one interpretation of this situation is that each type of group and organization uses arguments about differential status solely to gain exclusive benefits, another is that such laws are simply based on honest efforts to treat different cases fairly relative to others. Either way, the result is bound to be a long and complex tax code. (Witte, 1986: 149)

What Kind of Tax Cuts?

The collection of all those interests in this country represent what the political system is -- and the political system creates the tax system. And that's why the tax system is sorta' like an inner tube that has been patched about 150 times (Rep. Byron Dorgon, D-ND).

All modern capitalist democracies introduced a variety of tax expenditures intended to effect certain social and/or economic outcomes in the post-war period. There were, however, important differences between the kinds of special amendments implemented in the American and our European counterparts. The differences were largely a result of the respective processes by which these rules were passed. Throughout Europe a variety of special tax rules favoring (and in some ways regulating) industrial investment were introduced in the 1940s and 1950s. In general, these rules attempted either to regulate the timing of investment decisions, or to encourage investment in capital intensive industry and machinery. In most cases, they were not attempts to micro manage the economy via special measures to particular industries or companies.

The American tax rules were quite different. Rather than set general rules regulating the timing or general character of investment in the economy, Congress preferred to write tax laws benefiting particular industries, corporations, electoral districts and even individuals. "Tax incentives" were political plums.

From Committee to Subcommittee Government

Don't expect leadership from us. Congress isn't equipped to lead. It's a representative institution, with all that means for a stalemate (unidentified member of the House Ways and Means Committee, quoted in Light, 1985).

Once again, institutional structures are not static. After the war, Congress began a concerted effort to regain policy authority and autonomy from the executive branch which it felt it had lost during the wartime crisis. However, as Dodd and Schott point out, the 1946 reforms "contributed to the isolation of most members from congressional power. In streamlining the committee system, the 1946 Act left a relatively small number of positions that carried with them real power and status" (Dodd and Schott, 1979: 92). The revenue committees were the prime examples. Given the importance of revenue to the state, it made good sense to insulate these committees from normal politics. But this organizational solution created its own problems. These Committees - particularly the Ways and Means Committee under Wilbur Mills - became too powerful. Not only were they largely impervious to the influence, leadership, or power of the President, they also developed autonomy from (and ultimately power over) Congress itself (Fenno, 1966; Manley, 1970; Strahan, 1986; Rudder, 1977).

By the mid-1970s, the problems with this organizational structure became increasingly obvious. On the one hand, committee chairmen, all too secure in their positions, often used their positions of authority to bottle up legislation that was favored by the majority in Congress and often the President as well. The revenue committees were notorious sinners in this regard. On the other hand, as the power of the committee chairman grew, the ability of junior members to effectively develop policy making authority with which they could provide benefits to constituents shrunk. Again, the revenue committees were seen as the greatest problems.⁽²¹⁾ The Ways and Means Committee, in particular, was seen as "a 'bastille' that symbolized the inequities of the old order" (quoted in Strahan, 1986: 5).

As is well known, these strains combined to build momentum for another major reform era in the United States Congress. It is perhaps less well known that the congressional reforms of the 1970s were aimed particularly at Chairman Mills of the Ways and Means Committee (Strahan, 1986). Three important changes in Congressional rules were directed at his committee. First, all committee mark-ups were henceforth to be held in public, unless a majority of the committee agreed to a closed session. Second, the Democratic Steering and Policy Committee was given authority to individually appoint chairman to committees, rather than simply follow the seniority rule. Finally, the closed rule was altered so that it would be easier to offer amendments to Ways and Means bills on the floor of the House.

The explicit intent of these measures was to both unseat Wilbur Mills, and to democratize the revenue process (Rudder, 1977: 118-130). The effect of these rule changes, however, was to break open the tax policy making system and undermine the already weak forces of restraint. One reason Mills had been accused of being autocratic and unresponsive was precisely because he did not give in to every request from House members for special tax amendments to favored constituents. To be sure, over the years thousands of special amendments had been introduced, but many more were rejected. This made the inequities of the system even more difficult for junior members to accept. Only those who had Mill's confidence were able to get tax breaks for their supporters, and this cronyism was clearly unfair.

Unfortunately for the coherence of the American tax code, the legislature's answer to the cronyism of revenue politics was to distribute responsibility for taxation ever more widely. This made an already overly open process even more open, and made an already porous system even more loophole-ridden. In the context of the continuing institutional conflict between the legislative and executive branches of government, Congress had no incentive to yield tax writing power to the executive branch. Thus, the option of shifting responsibility for tax policy management to the executive, as had been done in Europe, was scarcely considered in the United States.

But the strategy of centralizing power within Congress into the hands of the Congressional leadership - as was clearly favored by the leaders themselves - was also foreclosed. Absent strong control in Congress, the leadership were forced to yield powers to quite junior members if their reform efforts were to be successful. Congressional reforms thus went in two opposing directions. On the one hand, the Democratic Party Steering Committee was given new powers (most importantly, the power to select and in effect fire committee chairmen.) On the other hand, junior members of Congress were given more powers as well. In exchange for their support of the Bolling Committee's attempts to centralize power in Congress and make it a more effective instrument in its dealings with the President, Congress also passed the "Sub-Committee Bill of Rights." This set of reforms decentralized policy making authority into the literally hundreds of subcommittees that had proliferated since the 1946 Reorganization. The reform thus allowed each member of Congress to be a subcommittee chairman, and gave these chairmen substantive legislative powers. With this reform, virtually every member of Congress would have the power to bring home specific benefits to constituents.

Loophole Madness

The President blames the Congress, the Congress blames the President, and the public remains confused and disgusted with government in Washington (James Sundquist).

By the end of the decade the scrambling for special tax breaks for ever smaller groups had reached absurd proportions. Reese provides an example of 'tax politics as usual' of this era, reciting an exchange between Senator Nelson (D. Wisc.) and Finance Chairman Long: "I understand," said Nelson "that in my absence we passed the tax breaks for railroads, but omitted railroad-overwater ferries --such as the one in Wisconsin." "I'll be happy to give you that one without need for further discussion," replied Long, "but I expect you in exchange to vote for this next tax credit we're about to discuss." (Reese, 1980:163)⁽²²⁾ Table 5.3 shows the total revenue loss from tax expenditures between 1973 and 1986. Keep in mind that the vast majority of tax expenditures in the American tax code are focused quite narrowly and therefore do not individually cost great sums of money.

Table 3

Revenue Losses from Tax Expenditures, 1973-86

Source: calculated from CBO, Tax Expenditures: Current Issues and Five Year Budget Projections for Fiscal Years 1983-1987, Nov. 1982, Table 3; CBO, The Economic and Budget Outlook: an Update, Nov. 1986, Tables, I-3, II-3 and II-8; and Witte, 1985, Table 13.4

There were, of course, attempts at reform. Tax expenditures were by now a major explanation for the continuing and growing budget deficits, and they also inspired numerous attacks on Congress over its ability to govern. But until the mid 1980s, these efforts at reform failed miserably. In 1978, for example, President Carter introduced a major tax reform bill based on Blue Prints for Tax Reform, the comprehensive review of the system conducted under the Ford administration. Even with the strong support of the chairman of the Senate Budget Committee, (Muskie) and a substantial media campaign, Carter's tax reform was a dead letter. As the bill worked its way from Ways and Means to House floor to the Finance Committee to the Senate floor and finally to the Conference Committee, it "abandon[ed] any pretense of tax reform." Indeed, "The final Conference bill was a complete renunciation of the Carter tax proposals and any notion of tax reform" (Witte, 1985: 209, 213).

Faced with a growing fiscal crisis, the new chairman of the Ways and Means Committee, Al Ulman, took a different tack and introduced a proposal for a new Value Added Tax (VAT). Ulman and Finance Chair, Russell Long, agreed about the need for revenue and also agreed that, given the increasing fragmentation of authority in Congress, neither income nor profits were likely to generate enough revenues in the foreseeable future

The revenue committee chairmen, moreover, were acutely aware of the need for revenue leadership and of Carter's inability to provide that leadership.

I see my role as altogether different than chairmen used to see theirs, (said Ulman in an interview in 1978.) They were worried about image and not losing any bills and not bringing a bill to the floor unless they had all the votes in their pockets. You can't operate that way anymore. I see my role as one of leadership and trying to expand the thinking of Congress in new directions in order to meet the long-term needs of the country (Strahan, 1986:11).

In October, 1979, Ulman introduced the "Tax Restructuring Act of 1979," which would institute a 10% Federal VAT to raise $130 billion a year and cut income and social security taxes by an equal amount. Trying to distinguish the VAT from the increasingly discredited Income Tax, he introduced his bill by saying "The VAT in my bill is virtually a tax without loopholes." I will not recite here the history of this bill, it follows precisely the path of all other consumption tax bills introduced in previous decades. It failed.

Ullman reintroduced the VAT the next year, but the 1980 bill also went nowhere. It did, however, teach Ullman and his congressional colleagues a lesson about attempting to provide "leadership" on tax matters. Though Al Ullman was noted has having one of the safest seats in Congress when he took the chairmanship of Ways in Means, the VAT proposal became a decisive factor in his failed 1980 re-election campaign. Ullman stuck his neck out on the tax issue; as a result, he had his head chopped off. Some years later, a Republican member of the Ways and Means committee confided to the author that he genuinely felt that a VAT is the only possible alternative to solving the United States's fiscal deficit. When asked why he wouldn't say this publicly, he responded, "Are you kidding? Nobody wants to get the Al Ullman syndrome."

The "Historic" Tax Reform

By the early 1980's it was widely agreed that the tax system had become too complex, raised too little revenue and was becoming an impediment to economic growth. Aaron and Galper neatly summarize the elite consensus attitude towards the tax system in the 1980's.

The US tax system has become a swamp of unfairness, complexity and inefficiency. The accumulation of credits, deductions and exclusions designed to help particular groups or advance special purposes conflict with one another, are poorly designed and represent no consistent policy. (Aaron and Galper, 1985, p.1).

Not only was the tax system perceived to be a mess, the political system which had created these tax laws was also suffering from a crisis of legitimacy. "The house of horrors" as Wilbur Mills called the Federal Revenue Code, became the penultimate example of Congress' inability to "just say no" to the special interests. "What are we" one Washington columnist rhetorically suggested that Packwood asked Rostenkowski, "a bunch of prostitutes, who take the PAC's money and give them any tax loophole they want?" This question was no joke. And it is precisely the columnists implication that motivated Congress to reform a policy arena that they had never reformed before.

The 1986 tax reform was indeed "historic." But as impressive as the 1986 tax reform surely was, it is important to remember that despite virtually unanimous consensus in Washington that the tax code was in need of simplification, this act did not simplify the code. Whatever the original intentions of the administration and Congress with respect to this reform, clearly the most important political and fiscal dynamic which made the bill possible was (in this act as in every act since WWII) Congress' desire to bring home tax benefits to their constituents. In 1986 this meant steep cuts in taxes for the "average" American.⁽²³⁾ The product of these two acts has been to take an already immensely complicated tax system and to make it substantially more complicated.

Though there were clearly a large number of tax loopholes which were cut or reduced in the 1986 law, there were also a great many major tax expenditures which were broadened.⁽²⁴⁾ There were also a huge number of highly particularistic measures introduced or retained for specific groups, industries, cities, regions of the country, and even specific individuals. Thus, despite the huge fanfare and extraordinary attention these two reform acts have received in the media, they are in fact excellent examples of the ways in which the American tax policy making process shapes tax policy outcomes. They, like the tax systems which they were built upon, are products of the fragmented, decentralized system of "committee government" which America has adapted over the last century. These tax laws are more complicated and inefficient than the tax laws of twenty years ago, because the institutions through which tax laws must pass have become more decentralized and more fragmented in this period.

Social Security Taxation

The reader may have noticed the conspicuous absence of virtually any attention to Social Security Taxation since its inception in 1935. The reason for this is that Social Security Taxation provides an important exception to the general rules of tax policy in America. It is, however, the exception which proves the rule.

The politics and history of Social Security taxation stands apart from the general pattern of tax politics in America on several grounds: The Social Security tax is explicitly regressive.⁽²⁵⁾ It is the only new major tax to be approved by Congress since the Income Tax of 1913. Social Security taxation marks the only non-wartime example in American tax history in which Congress has explicitly legislated tax increase on its constituents. Finally, instead of being a social security "laggard", America has developed a very generous public pension plan by international standards. (See Wilensky, 1975, pp 59-61, Tommasson, 1984.) Social Security taxation is no less substantial in the US (8.1% of GNP in 1983) than it is in the average for OECD nation (8.3% of GNP) and substantially higher than those found in Britain (6.4% If we combine tax expenditure subsidies, the U.S. spends 9.1% of *** on public pensions-compared to 8% in the UK and 5.5% in Canada). Indeed if we add tax expenditures to direct expenditures, we see that the U.S. is well out ahead of others in this intergenerational welfare program.

What explains this exceptional level of public commitment? Is it that American political culture has historically had a soft place in its heart for the aged? I suggest not. In fact, the U.S. was very slow to start its Social Security program (cf. Weir, Orloff and Skocpol, 1988; Flora and Heidenheimer, 1977). A more potent explanation for the uniqueness of Social Security Tax policies is that the institutional structure in which Social Security policies are made differs quite substantially from other tax policies:⁽²⁶⁾ The key difference between Social Security taxes and general fund taxes is that decisions for Social Security taxes and spending are made by the same committees. As a result Social Security tax and spending policy is considered simultaneously.

We have seen above that the separation of taxing and spending authority in the U.S. Congress creates an incentive system in which members on revenue committees can get little credit for the increases in public spending which their tax increases enable. They will instead be held accountable for tax increases. They can, however, get credit for tax expenditures. In the case of Social Security the situation is quite different. Members of the committees have responsibility for taxation but also receive the political credit of raising benefits. Thus, Social Security is the only national tax in which those who must take the blame for tax increases can be compensated politically with the credit for the benefit increases. The incentive structure faced by members of the revenue committees, then, is fundamentally different for Social Security taxes than for other taxes because of the different institutional structure through which social security policies must pass.

Corporate Taxation

The second, and in some senses the most surprising exception to the general trends in American tax policy is that of corporation taxes. While the US collects less than the average democracy from every other revenue source, has traditionally collected more than the average western democracy from corporate taxes. We collect more from companies not because our tax rates are higher because corporations do not have access to as broad and generous tax expenditures through which they can protect their profits from the taxman. In this sense, America appears to have had a comparatively revenue efficient system of corporate taxation. Closer examination reveals, however, that the American corporate tax system is more complex, particularistic and, indeed, regulatory, than the corporate tax system in virtually any other advanced capitalist nation. The remarkable irony is that America, the land of "free market opportunity" and anti-state ideology, has in many ways had one of the most micro- interventionist corporate tax systems in the world.

Again it is America's unique institutional fragmentation of the tax decision making structure which explains this exceptional tax policy outcome. Just as political fragmentation has allowed congressmen to veto tax increases, fragmentation has allowed congressmen to veto broad corporate tax reductions. During both world wars, taxes on corporate profits increased dramatically everywhere. There appeared to be an international consensus that workers should not have to spill their blood while the corporate managers got rich. Since the last war, however, all countries have begun to use their corporate tax systems as instruments through which they could influence corporate decision making - they have introduced corporate tax expenditures. US corporate tax policy has differed because Americans have not introduced the broad and generous tax expenditures favoring corporate investment typical in European tax systems. The American Corporate Income Tax is riddled with more specific tax expenditures than the Europeans, but collect more revenue because the tax expenditures which are available are substantially less generous and substantially more specific.

With the exception of the Investment Tax Credit and the accelerated depreciation schedules in effect between 1982 and 1986, America has not relied heavily on broad, and open ended corporate tax expenditures like those found elsewhere. Instead of promoting business interests generally, American tax policy makers have preferred to offer specific tax breaks to specific industries, professions, and individual companies. There are literally thousands of particularistic measures affecting corporate interests in the Federal Revenue Code, specifying particular rates, depreciation schedules, allowances, accounting procedures, types of employees, types of investments, regions of the country and even weather conditions. As complex as these measures make the tax code, however, their specificity means that they have smaller revenue effects than the general measures available in the rest of the OECD.

Again, the American tax policy making process is distinctive because the administration does not control tax policy making. In the US, as elsewhere, there has always been an important faction within Congress (parliament) who opposed broad corporate tax reductions. But in the US, 'backbenchers' have real legislative power. Reducing taxes, whether by way of tax expenditures or through general rate decreases, requires legislation. And, as scholars of American politics have pointed out ad infinitum, our political system presents numerous opportunities for those who are opposed to legislation, to veto it. "It is very easy to defeat a bill in Congress", said President Kennedy. "It is much more difficult to pass one."

Nothing said here implies, however, that the same legislators who favor heavy corporate taxation in general, want the employers, industries and interests in their districts to pay heavy taxes. The electoral incentive structure facing Congress leads them to introduce tax legislation in favor of their district, no matter what their political party or general philosophical orientation towards taxation (Einstein, 1961; Reese, 1980; Manley,1970; Witte, 1985). In addition, legislators prefer to introduce legislation which they can claim credit for, particularly if it helps their district in particular. Broad incentives are also less attractive to individual Congressmen because their benefits are less concentrated. When benefits are concentrated, credit can be more concentrated as well.

In sum, American corporate tax policy differs from European not because American's are more anti-capitalist than Europeans - nor because Americans are more inclined towards government intervention in private business decisions: Rather, the U.S. taxes corporations heavily and writes corporate tax law in such specific and detailed ways because the incentives facing locally elected politicians bias their tax decisions in these ways.

In this essay I have emphasized the extent to which the basic institutional features of American politics have shaped the incentives of tax policy makers. I have argued that the exceptional institutional structure of divided government has created and incentive system in which legislative elites have been able to retain control over the tax policy making process by decentralizing power and building the committee system for making public policy choices. This devolution of public authority both prevents the authoritative use of public power by central political elites (eg. the President and/or Congressional elites) and reinforces incentives for legislators to cater to the most narrow and short-term interests of their particular (often locally defined) constituencies. In short, it is this institutional fragmentation of political authority in the U.S. which best explain how and why the U.S. has been unable to impose a national consumption tax (which every other modern capitalist democracy has imposed) and the heavily reliance on tax expenditures which characterizes the American tax system.

I have not argued that Americans hold the same philosophical beliefs, ideological prejudices or "national values" as others. But I have not found much evidence suggesting that these general American values have been particularly important for defining tax policy choices. Instead, tax policy makers operate within a peculiar set of institutions which both critically shapes their political behavior and yields disproportionate amounts of power to particular constituencies and interests.⁽²⁷⁾

I believe that this analysis has important implications for understanding "American Exceptionalism" in two ways. First, if my argument is correct, the critical explanation for America's meager social welfare policies is that state officials have been consistently unable to find the funds to introduce or increase the programs to the extent that both they and the citizenry desire. Once again, it is clear that the majority of American's would like to see more government spending on virtually every category of public expenditure (cf. McClosky and Zaller, 1984:272; Coughlin, 1980; Light, 1985;). The paucity of revenue dollars thus explains the extent to which many (though not all) public programs receive comparatively lower levels of funding than do their European counterparts.

Secondly, this analysis contributes evidence in support of the "institutionalist" approach within political science. The implication is that the institutional fragmentation of American politics explains other "exceptional" features of the American public policy. I would argue, for example, that precisely this kind of analysis would provide fertile ground for explaining a broad range of public policy outcomes in the U.S.. To take but one example: Why is the U.S. the only advanced democracy which does not have a comprehensive universal health care system? Listening to both public and scholarly debate on this question would lead one to believe that the answer is simply that such a program violates our basic ideas and values. I submit, instead, that we do not have national health insurance for precisely the same reasons that we do not have a national sales tax: Committee Government yields disproportionate power to entrenched locally elected politicians (historically from the south) and/or well financed and organized narrow interest groups: Madison's institutional fragmentation protects small factions against large ones.

Rather than comprehensive or universal programs - which would be administratively more sensible as well as politically more invulnerable - American political institutions encourage reformers to compromise in particular ways. First, reform ambitions have to be scaled back in order to be successful even when there appears to be huge public consensus favoring reform (eg. medicare/medicaid not national health insurance).⁽²⁸⁾ Secondly, the specific character of the programs is altered because so many specific constituencies must literally be 'bought off' in order to push the reforms through Congress.

I submit that these institutionally dictated compromises have a profound impact on attitudes and values in society. My argument is based on the premise that citizens and elites are broadly rational.⁽²⁹⁾ Thus, given the inefficiencies and persistent incoherence of our national political institutions and the public policies which finally trickle through the labyrinth, it is perfectly reasonable (indeed predictable) that Americans would distrust government. My proposition here is that not only have the fragmented political institutions shaped specific public policies in America; they have also, over time, shaped our national values and political culture. In the end then, Americans may well be the most anti-state citizens in the modern world -- but this has less to do with the character of the citizens, than it has to do with the character of the state.

Endnotes:

1. The author would like to thank Chris Howard, Margaret Levi, Charles Lockhart, Seymore Martin Lipset, Joseph White and Aaron Wildavsky for helpful comments on earlier versions of this essay.

2. Only "Defense" and "Welfare" tend to get majority negative responses to questions about increased spending. The American public is very similar to the publics of other nations in this regard. See, for example, Coughlin, 1982; Lewis, 1983; Public Opinion, 1985; Hansen, 1983; Survey Research Consultants, 1987.

3. Americans are by no means exceptional in this regard. The vast majority of the public in every democratic country - even in Sweden, the country which has increased the tax burden on citizens the most in the past 50 years - believes that the tax burden is "too high" and ought to be reduced. Here too the evidence is voluminous. See Coughlin, 1980; Lewis, 1982; Taylor-Gooby, 1987; Inglehart, 1989; Hadenius, 1986; Steinmo, 1993.

4. . If we add tax write-offs for dependents, small business and a number of other tax deductions generally considered "normal" parts of the American tax code (but often not of other advanced country's codes) then revenue losses from tax preferences may rise to 20 percent of GNP.

5. . See Steinmo 1993. High tax rates does not, of course, necessarily mean that these tax papers actually paid more taxes than in other countries...see point four below.

6. . Truly satisfactory comparative statistical data which could be used to neatly summarize this point unfortunately does not exist. Though an increasing number of countries have begun to take more accurate fiscal accounts of their tax expenditure budgets., efforts to implement international standards of accounting have so far been unsuccessful. Though international comparisons have been made, the exact values lost via particular tax expenditures remains illusive for most countries. See OECD, 1986. Still it is possible to examine the use of tax expenditures in particular countries, to make comparisons and to draw general conclusions from these somewhat more idiosyncratic data. Cf. Steinmo, 1989, 1993; McDaniel and Surrey, 1985; Rose, 1984; Heclo, Heidenheimer and Adams, 1986; Peters, 1991.

7. . See Thelen and Steinmo, 1992.

8. . For a general discussion of the politics of this era see Burnham (1970) Ronald King (1983) discusses the economic background and populist revolt which specifically lead to the demand for the income tax. See also Stephen Skowronek, (1982) for a discussion of political and administrative changes in this era.

9. . The US had a "temporary" income tax during the Civil War, see Hansen, 1983, p. 79-81.

10. . Taft's administration also conceded to the demand to tax corporations directly in an attempt to take the wind out of the sails of the personal income tax. The Corporate Income Tax was passed in 1909. For an excellent discussion of the summary of the politics surrounding these reforms see Susan Hansen, 1983-81-84.

11. . For example, as the 1917 tax bill was being considered charitable organization put on a full court press for tax deductions for charitable contributions. The nation's college and universities joined the battle. The tax committees resisted these interest group pressures, on the grounds that the benefits for these tax breaks would go to the same wealthy individuals these taxes were attempting to squeeze. But in what became a familiar story in the politics of taxation the US, even the powerful Finance Committee they could not fight these 'worthy causes' once amendments had been put forward in their behalf on the floor of the Senate.

12. . The sales tax was defeated by a scout one vote in the Senate. See White, 1986, p. 90-91. Anti sales tax forces quickly mobilized a lobbying campaign of their own. Some of the group who denounced the proposals included: The National Grange; The Farmer's Federal Tax League; American Federation of Labor; The Railroad Brotherhood; and a collection of business groups who wanted the Excess Profits Tax repealed, but did not want it replaced with the sales tax. Some of these groups included the National Association of Credit Men; The Tax Committee of the National Industrial Conference Board; the United States Chamber of Commerce; and the National Association of Retail Grocers. See Ratner, 1967, p. 402-405.

13. . Waltman, 1985, p. 88. The amendments covered a wide array of interests and the final legislation provided special measures for interests as diverse as oil and gas and "the American Legion or the women's auxiliary units thereof" Waltman, p. 94.

14. . In what must of seemed a bitter irony to the New Dealers, Roosevelt's landslide elections had the effect of further entrenching Southern Democrats into key committee chairmanships.

15. . This package of measures included a) the introduction of an inheritance tax to go along with the existing estate tax; b) a graduated corporate profits tax; c) a tax on intercorporate dividends; and d) an increase in income surtax rates on individuals with incomes over one million dollars. See Witte, 1985: 100.

16. For more detailed discussion of wartime tax policies see Hansen, 1983; Witte, 1987; Steinmo, 1993.

17. . In 1942, despite the massive increases in income and profits taxes already legislated, the budget deficit skyrocketed 13% of GNP (and rose to 29% of GNP by 1933). Exactly like their European counterparts, American officials argued that national consumption taxes were appropriate because they also served the purpose of helping manage the macro economy. Thus the administration claimed that, in addition to the need for new revenues, consumption taxes would also help cool inflation. See "Morgenthau's Statement" New York Times Sept. 4, 1942, p. 16.

18. This is called "bracket creep." It works like this: Because the income tax is progressive its revenues grow "automatically." As incomes increase - through inflation and/or real income growth - individuals taxpayers are pushed up into higher tax brackets. This means that revenues from a progressive income tax will grow faster than inflation and national personal income even if tax rates are not increased.

19. . President Truman at first refused to accept these institutional rules. As a consequence he and 3 separate vetoes of tax bills overridden by Congress. See his annual report. Annual Report of the Secretary of the Treasury on the State of Finances for the Fiscal Year Ending 6/30/1951. Page 491, exhibit 32.

20. . Democrats were far from innocent of this behavior, and in many ways were able to have it both ways. They could, on the one hand, decry the Republicans for their attempts to gouge the working people and offer huge tax breaks to the "fat cats"; while, on the other hand, support special provisions for the powerful and wealthy who lived in their districts. For a powerful indictment of both parties' tax policies during this era, see Einstein, 1961: 218-219.

21. . One should note the conflict between these complaints, and reform objectives. On the one hand, reformers wanted the committees to be more responsive to party elites. On the other hand, individual members wanted the committee policy outputs to be responsive to individual members constituency needs. The Ways and Means Committee under Mills was considered a sinner on both accounts. Mills neither bent to the wishes of the party leadership, nor was he willing to willy nilly pass out tax benefits for individual member's constituencies. The Finance committee and the Senate generally was considered much more amenable to individual politicians needs. Strahan, 1986: 4.

22. . (Congressional Quarterly Weekly Report, 35, Sept. 10, 1977, p.1906) quoted in Reese, 1980: 163.

23. . The 1981 tax act yielded even greater benefits to the wealthy, but the 1986 act took back some of the tax benefits the rich had received in 1981 and was in fact a marginally progressive tax reform. See Henry Aaron, 1988. See Birnbaum and Murray, 1987; Conlin et. al. 1989 for more complete discussions of the politics surrounding 1986 tax reform. Witte, 1986 discusses the 1981 tax reform in detail.

24. . Some examples include: the personal exemptions (for the taxpayer and each of his/her dependents), standard deductions, the earned income credit, the deductibility of state and local income, real estate and property taxes, charitable contributions of itemizers, interest payments on first and second homes, medical expenses, the exclusion of a wide variety of employer provided fringe benefits, as well as government transfer payments.

25. . There are in fact several separate taxes which go into one broad payroll tax category which I will refer to in the singular.

26. . One might think that a key difference between Social Security Taxes and other taxes has been sold to the public (falsely) as an insurance scheme and therefore politicians believe that it is less politically costly to raise tax rates. But the U.S. is not exceptional on this account. Social Security Taxes are generally said to be part of a national insurance system. This has been a popular fiction in almost all countries.

27. . Clearly the policy preferences of well ensconced Southern Democrats have had an enormous impact on tax policy development in the U.S. for much of the 20th century. Equally clearly, their tax policy preferences have consistently been at odds with the preferences of the only nationally elected official in the United States. Perhaps one could argue that the values and preferences of these Southern Democrats have been determinative, but I doubt whether this would lend much support for the "Liberal Traditions" argument.

28. . Such a condition almost never exists for taxation policy.

29. . By this I do not take an narrow homus economicus view of rationality. Rather I mean rational in a more limited "bounded rationality" sense as argued by Simon.

Back to Steinmo's Home Page