Thinking About Politics: American Government in Associational Perspective

By Paul F. deLespinasse, Adrian College

Copyright © 1981 by Paul F. deLespinasse. Details of generous permission to make copies This chapter may not print or copy unless you have clicked here first.

PART 3: Thinking About Rules

Those who would appreciate sausages or laws, it has been observed, should not watch either of them being made.

Phil Cogswell, Portland Oregonian, July 10, 1977.

Footnotes are at the end of the chapter.

To access other chapters, go back
to the Table of Contents

Chapter 7: Congress: The Legislative Procedure

Chapter objectives

After reading this chapter you should understand:

1. The ways in which Congress is and is not like a bureaucratic organization.

2. How the individual interests of congressmen lead them to act in ways that seriously reduce the total power exercised by Congress as a whole.

3. Why it is not possible fully to ascertain the desires of either house of Congress merely by counting heads, and why obstacles to the will of the majority posed by committees and filibusters enhance democracy.

4. Several reasons why some thinkers consider American taxes a national disgrace, and why some of these evaluations may be misguided.

5. How a hypothetical reform of Congress into a one-person legislature illuminates its present strengths and weaknesses.

Key Terms

iron law of oligarchy
collective goods
Legislative Reorganization Act of 1946
single member district
committees on committees
social ordering
paradox of voting
strongly ordered
discharge petition
pork barrel
progressive tax
tax structure

An immense mass, indeed, of the legislation is not, in the proper language of jurisprudence, legislation at all. A law is a general command applicable to many cases. The "special acts" which crowd the statute book and weary parliamentary committees are applicable to one case only. They do not lay down rules according to which railways shall be made, but enact that such and such a railway shall be made from this place to that place, and they have no bearing on any other transaction. *


In their home states, representatives and senators enjoy high status. Their names are well known. Invitations to dine with the rich, or creative, or powerful are certain. Civic groups will give their after dinner remarks serious attentions --a courtesy not always extended to less famous speakers. Representatives and senators have distinguished themselves merely by attaining office. On average, there is only one representative for every half million Americans, only one senator for every two million. From many, the Congressman is indeed one. E pluribus unum.

In Washington, D.C., however, these same Congressmen may well be nobodies. On Capitol Hill, each one can all too easily be submerged in the other 534 members of Congress. An experience related by Oregon Senator Mark Hatfield dramatizes the abrupt reduction in status suffered on arriving in Washington. Hatfield was finishing his second term as state governor when elected to the Senate in 1966. Campaigning for reelection to the governorship in 1962, he had promised to serve out the full term. Unfortunately, Hatfield's term as governor did not expire until a few days after Congress convened and the other freshman senators were sworn in. However Hatfield, whose devotion to principle is unusual, kept his literal word, completed every day in his gubernatorial term, and was sworn in a couple of days late as the junior senator from Oregon. He did not expect that a few days would make much difference.

The Senate Office Buildings have exactly 100 suites of offices, one for each senator. Not all of these suites are equally desirable, however. Offices vacated due to death, loss of election, retirement, or assignment to a better location, become available to the most senior interested senator. But moving the old senator out, redecorating the office, and moving the new senator in takes time. The process is then repeated in the suite of the senator who moved up. The upshot is that the least senior senator, 100th out of 100, has no office at all until everyone else has finished moving. Hatfield, sworn in a few days later than other new senators, was on the bottom of the seniority list and had no office for a number of weeks. Footnote 1

Congressional seniority is important, of course, for more reasons than office space. Sometimes a retiring congressman will resign a day or two early allowing the governor to appoint his elected successor to the vacancy and give him seniority over all the other freshmen. Even with this "leg up" on other newcomers, however, the freshman will not make a big splash when he arrives in Washington: At home, a somebody, within Congress, (almost) a nobody. In Congress, everyone is in much the same boat and as Gilbert and Sullivan noted many decades ago, when everyone is somebody then no one is anybody. Footnote 2

Even so, plenty of people would like to become congressmen. The pay is very good. In 1979, for example, the annual salary was $57,500, plus untaxed expense allowances. The average annual salary in the U.S. as a whole is about $10,000; an average household income, $15,000. Beyond its material rewards, the congressman's job allows some participation in wielding great power, it offers challenges to ingenuity and stamina, and it introduces one to a host of interesting and well-connected people in and out of government.

Election of Congressmen

Because of the great appeal of the job, it is difficult to get into Congress. Openings are limited. There are only 535 seats for a population of 220,000,000. Most congressmen do not retire at the end of a given term, and it is extremely hard to unseat an incumbent. According to figures compiled by the Federal Election Commission, candidates for the House of Representatives spent an average of $70,000 each in the 1976 general elections, ranging from $555,381 for one successful nonincumbent down to $47 for one successful incumbent.[Footnote 3] Incumbents tend to spend less because they have the advantages of their office--staff, travel and correspondence allowances, and name familiarity.

Eligibility to run for Congress is restricted by citizenship, age, and state of residency. Senators must be at least 30; Representatives, 25. However, these requirements are more formal than real obstacles, as few younger people are in a position to campaign successfully. Residency requirements are somewhat more real. One's own state or House district may have well-ensconced incumbents, yet moving to more promising territory is difficult. Newcomers are often dismissed as opportunists and "carpetbaggers,"[Footnote 4] and it usually takes many years to cultivate the local contacts necessary to launch a successful congressional campaign. Indeed, the real residency requirement for the House appears to be harsher than the constitutional one. The Constitution merely requires residency in the state, but for practical purposes one must also reside in the congressional district whose seat one is seeking. This informal requirement offers temptation to gerrymander [Footnote 5] an incumbent of the "other" party right out of his own district during congressional redistricting.

Individual vs the Organization

As we have already established, the elements of simple action are goals, circumstances, actions, and side effects: A ---> X + Y. Goals, inherently an individual matter, will naturally differ somewhat from congressman to congressman. While organizations may be regarded as acting as if they have goals, this is at best a convenient fiction. At worst, it may be a badly misleading fiction, as Dean Allison shows in his analysis of the Cuban missile crisis of 1962. [Footnote 6] It is probably best to view organizations a having functions rather than goals. That is, their existence and "actions" promote certain types of consequences, but we cannot infer from this fact that the organization intends to bring about these results. Nor can we infer that particular members, even a majority of these members, intended or desired the result. To use a rough analogy, the stock market crash of 1929 obviously resulted from the actions of millions of individual speculators, none of whom desired to become poor.

Personal Goals. We can presume that most congressmen want some or all of the following things, among others:

1. The purchasing power acquired in compensation for their services
2. Power over the actions of government
3. Accomplishments sufficient to make life appear meaningful
4. Reputation
Of course, different congressmen may rank these goals differently. For some, the strongest desire may be to wield power, with little concern about accomplishments. Others may want the power, not as an end in its own right, but as a means to accomplishing things they regard as important. Still others may be concerned primarily about the purchasing power they can acquire directly (salary and perquisites) or indirectly (social and financial connections) through their office. At the other extreme, some congressmen are independently wealthy and do not care about material rewards. Whatever their other goals, however, almost all congressmen have one goal in common: They would like to get reelected, if not, indeed, to attain still higher office. There are few voluntary retirements.

All congressmen must work within complex electoral, legal, and organizational circumstances. Senators face reelection challenges every six years, representatives every two years. Challenges may come from fellow party members, in primaries, or from the opposite party in general elections. They know that rivals will seize on any unpopular or inept things they have done or said, and will try to outpromise them in seeking the favor of the voters. Dealing with fellow congressmen is equally perplexing. As a participant in the congressional organization, the individual member can do nothing without cooperation from many other members. Good ideas for legislation do not implement themselves, and it is not always easy to make one's voice heard above the din. Most congressmen are inundated with communications and must be very selective in their attention or completely lose their bearings. Nor are all good ideas which do get some attention necessarily treated on their merits. Bills must run the gauntlet of a complicated maze of committees and subcommittees in each house and are further held up by disputes between the two houses.

Organizational circumstances are largely the same for all members, although seniority and strategic location on the right committees can help. Electoral circumstances are a major variable, as some members come from "safe" districts in which their party usually receives a comfortable majority, while others must be careful to avoid being here today and gone tomorrow.

The Tensions Between the Individual and the Organization. Otherwise inexplicable behavior of congressmen becomes quite understandable when we see it in the context of their individual goals and organizational circumstances. Enacting wise laws has relatively low priority on the individual's daily agenda, for enlightened legislation does not ease reelection. The ombudsman (trouble-shooting) function takes a lot of time and effort, for people (voters) appreciate favors in the relatively large matters they tend to bring to the representatives. Large organizations such as corporations and unions get a lot of attention, since substantial interests of the people involved in them (also voters) are at stake. Consumers, however, tend to get little attention as such, for everybody is a consumer and small, relatively unappreciated benefits for all consumers may draw the wrath of particular producers who have large amounts at stake.

For example, imagine a regulation that would save every consumer of widgets $10 a year--a total of $2 billion assuming that everyone uses widgets. The $10 is hardly enough to be noticed by these consumers, let alone enough to prompt return favors at the ballot box for those who supported it. However, suppose there are only 10 companies producing widgets in the country, each with an equal share of the business. If the regulation is enacted, it will cost each company $200 million, a substantial sum. The people--owners, suppliers, workers-- connected with these companies have a large enough interest in the matter to retaliate at the next election against those who support the regulation. Of course these people are only a minority of the electorate, but we recall that a coherent minority can often swing a close election and no politician likes to offend such a minority. In an age of specialization, producers are more concentrated than consumers, and legislation tends to proceed accordingly.

For similar reasons, congressmen are happy to participate in whatever vestiges of the patronage or spoils system may still exist. People who think that a politician has helped them land a government job may be able to deliver their own votes and those of their friends on election day. Gratitude, as such, may play little role in politics and has been cynically defined as a "lively sense of future favors." But influence once proven by a successful patronage appointment may be seen as a source of future favors.

One of the remaining patronage posts is the job of local U.S. Attorney, who prosecutes violations of federal law. Congressmen may take a special interest in appointments to and removals from these jobs, for they may thereby be able to avoid legal punishment for their own violations or for those of their important constituents. Candidate Carter, campaigning in 1976, commendably urged that the political patronage system for U.S. Attorneys be abolished. President Carter, governing in 1978, ordered the Attorney General to fire David Marston, the Republican U.S. Attorney of Philadelphia, after receiving an urgent request to this effect from a congressman whose activities were being investigated by Marston. The Administration insisted that the appointment process was going to proceed on merit (though no Republicans need apply). Marston correctly noted that for cover-up purposes the appointment process is less important than the removal process and argued that political corruption among members of the dominant party would only be prosecuted if the U.S. Attorneys-- no matter how they are selected--serve for a fixed term of years like the members of an independent regulatory commission.

Likewise, even participation by staunch fiscal conservatives in the so called pork barrel system is understandable in light of their goals and organizational circumstances. Pork barreling finds a congressman voting for unjustifiable government expenditures in another's district in return for reciprocity regarding equally unjustifiable projects in his own district. "Free money" from Washington is always welcome by the local leaders whose support is so helpful when a congressman seeks reelection.

How the Tensions Are Resolved. Perhaps the most fascinating of all political maneuvers finds a congressman introducing a bill strongly desired by influential constituents but which he personally regards as disastrous. The bill is printed up and copies sent to the constituents, who may be gratified to note that the bill indicates it was introduced "by request"--a gracious but rather abstract acknowledgment of their own public- spirited role in getting it drafted. However "by request" is code indicating to other congressmen that the bill is not really supported by the member who introduced it.

Congressmen, like all other people, exist and act solely as individuals. Their actions may depend on their organizational circumstances, among other things, but expected consequences will be evaluated from the standpoint of the individual member's own values. In terms of A ---> X + Y, each member will tend to act so as to maximize the net present value [Footnote 7] to him of the future results of his action. The desires of others-- constituents and fellow congressmen included--may affect his decisions to the extent that he values their various anticipated reactions to his own actions. But the side effects that the individual congressman will weigh against benefits and seek to minimize are those affecting him, not those affecting Congress as an institution. One side effect of the self-interested actions of individual congressmen is to reduce the overall power of Congress as a body. Since the individual is a member of the body, this means that he is deliberately acting to reduce the total power of which he has a small share, which means that his share is correspondingly reduced. Yet his actions may be highly rational. Acting otherwise would reduce his chances of remaining in Congress at all because of his electoral circumstances, and most people would rather have a given share in great power than to have no share in huge power.


A Nonbureaucratic Organization

Congress provides interesting contrasts with the bureaucratic forms of organization found in the executive departments.

How Congress Differs from Ordinary Organizations. Congress is an organization in the full sense of the term. It is an association in which individual members occupy offices whose existence is independent of their own brief tenure in them. As with any organization, it can therefore be regarded either as a pattern of interactions among specific individuals, or as a set of relationships among offices. But the Senate and House are clearly not the same kind of organization as the Department of Agriculture, the National Labor Relations Board, or the Federal Deposit Insurance Corporation. There is a basic lack of hierarchy among their members: they all receive equal pay (although expense allowances vary). Each member of House and Senate casts an equal vote in deliberations on the floor of his chamber. Congressmen are not placed in their basic office by superiors within the organization, but all are placed there by their superiors outside the organization--the voters.

Members of Congress are not chosen by virtue of expert training to do specific kinds of work. The educational and professional backgrounds of congressmen are therefore quite diverse, though by no means as much so as the general population they represent. (About half of the representatives and two- thirds of the senators tend to be lawyers, for example, whereas lawyers make up less than I% of the working population. And there are almost no scientists or engineers in either house, a situation that may be unfortunate given the energy, environmental, and public health measures increasingly before Congress.) Most important of all, congressmen are not really full time workers in their organization. Although most of them earn their livelihood from and spend most of their working days doing the work of a congressman, they devote much of this time to dealing as an individual with constituents, administrative agencies, and staff members who are doing likewise. As members of the organization they are essentially part-timers, "honorific" workers, with all of the consequent inefficiencies noted earlier by Weber.

The Hierarchy in Congress

The fact that Congress is not organized bureaucratically is very important. But this fact is only the beginning of our analysis of Congress, not the end of it. The sociologist Robert Michels concluded some decades ago that all large organizations are subject to an "iron law of oligarchy." Oligarchy simply refers to a concentration of decision-making power in the hands of a few individuals. As restated by Chester C. Maxey, the iron law says that:

Society requires organization; organization requires leadership; leadership inevitably becomes oligarchical; oligarchy always becomes self-centered. Footnote 8

Some people have argued that the iron law, it if exists, means that democracy is impossible, but no such extreme conclusions are necessary, as we will see in due course. And this is fortunate for those who believe democracy is desirable, for careful scrutiny of the actual operations of Congress provides considerable evidence that an iron law of oligarchy does in fact exist.

True, each member does have an equal vote on the floor of his house. But the importance of this fact is reduced by another fact: most congressional business is done in committees, not on the floor.

Committees. There are several reasons why Congress depends so heavily on committees to get its work done. It is almost impossible for each member to express his thoughts before the entire membership. Only a limited number of people can speak to an issue in a given number of hours. If the minimum time required to express an opinion meaningfully is five minutes and the house is in session for eight hours a day it would take the House nearly five days and the Senate one day for each member to speak once. [Footnote 9] Under the more reasonable assumption that fifteen minutes is a necessary block of time to say anything and sessions last only six hours the House would require 18 and the Senate four days per round of discussion per issue. This much time is simply unavailable except for occasional crucial issues.

In the early days of the Constitution, when there were only 13 states and a vastly smaller population, Congress was much more agile. The first sessions of the House of Representatives were attended by only 70 members, fewer than today's Senate, while the Senate had only 26 members. As the chambers expanded to today's 435 and 100 members, respectively, reliance on committee work has increased correspondingly. In committees, each person has time to have his say, and this advantage is further multiplied by extensive use of subcommittees for discussing details of proposed legislation.

A second reason for increasing congressional dependence on committees has been the vast growth in legislative business since 1789. In nearly 200 years the national population has multiplied over 50 times, the territory has expanded many-fold, and the federal government has gotten involved in issues that were once ignored by governments or handled exclusively at the state and local levels. No group can talk about more than one thing at once, and if there are too many things that must be discussed the amount of time available for each may drop below that required. Committees alleviate this problem not only by making discussion within each subgroup more manageable by virtue of its smaller size, but also by increasing the number of forums so that many things can be discussed at once.

A third reason for the prominence of committees is that they allow their members to specialize in considering certain kinds of bills and, hence, to develop considerable expertise. Unlike bureaucratic expertise, that of congressmen tends to develop on the job. The job renders them eligible to become expert, whereas in bureaucracies the expertise makes one eligible for the job (in which, to be sure, expertise may be deepened or redirected). Specialization that allows development of expertise, however, does not automatically follow from use of committees. As the British House of Commons showed during its heyday, a legislative body can be divided into committees and bills farmed out randomly to them. Instead of having a committee on foreign affairs, one on atomic energy, and another on civil service, there can be committees A, B, C, and so on, and each committee may in turn consider many different kinds of bills. Even these nonexpert committees, of course, do promote discussion by reducing the size of the group and allow many different bills to be talked about simultaneously. The American committee system inherently results in a great inequality in the importance of the different committees. Some subjects are simply not as important as others, yet each must be the responsibility of a committee when committees have permanent specialties.

Functions of the Committees. There are several important consequences of the power and inequality of congressional committees. First, there tends to be a certain lack of coherence and consistency in legislation, since an overview is not supplied by any one committee and the houses are ill- equipped to provide this service on the floor. Second, there is a tremendous struggle for choice committee assignments at the beginning of each session of Congress.

A third consequence of committee importance and inequality is that bills favored by a majority often do not even get reported out of a committee so they can be voted on by the whole house. In the House of Representatives, even if a bill gets out of committee, it may never get considered on the floor because the Rules Committee refuses to issue a "special rule" allowing it to be taken up out of order from the calendar. The crush is so great in the House that few bills have a chance without a special rule; the special has become normal, and the regular has become impossible. Not all bills bottled up in a committee, "pigeon-holed," or amended to death were introduced frivolously. The fact that many bills with predictable majority support on the floor never get there has led many observers to conclude that Congress is undemocratic and that its products do not reflect the will of the majority. Former Senator Joseph Clark of Pennsylvania claimed that

the legislatures of America, local, state, and national, are presently the greatest menace in our country to the successful operation of the democratic process. Footnote 10

Voting: How It Takes Place

A Paradox. Before jumping to rash conclusions, however, let us once again pinpoint our thinking to a small issue for a few minutes. We have seen how a Crusoe society-of-one can help us avoid the complexities around us and achieve basic insights about the human predicament. It is often easier to visualize relationships among people on a small scale, then extrapolate them to larger groups, making necessary changes. In this spirit, let us consider the following scenario in the hypothetical town of Podsville, New State:

Podsville has a three-person city council, the smallest possible legislative body for our purposes of examining the consequences of voting. The councilmen are named Ames, Baker, and Carter. The mayor's limousine has begun to age, and the council must decide what kind of car to purchase as a replacement. Choice has narrowed down to three models, one from General Motors, one from Ford, and one from Chrysler. Councilman Ames's preferences on this issue are as follows:

                                   GM      first choice
                                   Ford    second choice
                                   Chrysler  third choice

Baker's outlook is somewhat different:

                                   Ford    first choice
                                   Chrysler  second choice
                                   GM      third choice

And Councilman Carter has still a different schedule of values on this issue:

                                   Chrysler  first choice
                                   GM     second choice
                                   Ford   third choice

We will assume two things about the value schedules of councilmen A, B, and C: First, each person's values are "strongly ordered." That is, none of the councilmen is indifferent between his first and second choices, or between his second and third choices. Second, each individual councilman's choices are "transitive," or consistent" If Ames is asked to choose between GM and Ford, he will pick GM, if he must choose between Ford and Chrysler, he will opt for Ford, and if the vote is between GM and Chrysler he will support GM.

According to an article, "The Paradox of Voting," if we assume that individual values are strongly ordered and transitive we can also assume that the "social" ordering is also strong provided that we

consider only voting patterns where the number of judges [voters] . . . is odd. This precludes a voting pattern where the social ordering is indifferent between X and Y, i.e., where X is preferred to Y as often as Y is preferred to X. Footnote 11

The three person city council of Podsville satisfies this requirement, of course. The paradox article also assumes that "The social ordering [choice of the group] is determined by simple majority voting between pairs of issues." Footnote 12

For convenience, we can summarize the preferences of the individual councilmen in a table:

                        Ames     Baker    Carter
     First choice       GM       Ford     Chrysler
     Second choice      Ford     Chrysler GM
     Third choice       Chrysler GM       Ford
If the pair of possibilities voted on is GM and Ford, GM will win (i.e., will be the "social" preference of the group): Ames and Carter both prefer it over Ford, though Baker will vote for Ford. If the choice is between Ford and Chrysler, Ford will win, with Ames and Baker prevailing over Carter.

Collectively, then, the city council prefers GM over Ford (as indicated by the first vote) and its prefers Ford over Chrysler (as indicated by the second vote):

      GM             Vote 1
      Ford           Vote 2
Since GM is "socially" preferred over Ford and Ford over Chrysler, we would expect that GM would be preferred over Chrysler. However, a vote between this pair of options produces a surprise or "paradox": Chrysler wins, supported by Baker and Carter. (Check the table yourself to confirm this conclusion.) The council's preferences appear to be intransitive.

What is going on here? There is, of course, no such thing as a true paradox. If two statements which appear to be mutually contradictory are both true then we know that there must be something wrong with our analysis whereby they appear contradictory. There is an old joke in which the difference between an optimist and a pessimist is explained. The optimist looks at a bottle and says it is half full, while the pessimist looking at the same thing perceives it as half empty. The premise of this little witticism is of course that these are merely two different ways of saying the same thing. A bottle one-half full is one-half empty, automatically, or expressed mathematically:

                  1/2  full   =  1/2  empty

Now let us apply standard mathematical logic to this equation, multiplying both sides by 2:

                 2 x 1/2  full = 2 x 1/2  empty

                          full = empty
We have thus "proved" that full is equal to empty! Figure out where we have jumped the rails here yourself, and be duly warned: the thought process producing the "paradox" of voting is equally sleazy.

The paradox results only if we assume that the intensity of each councilman's preference for his first over his second choice is equal to that of his preference for his first over his third choice. This assumption is incompatible with the assumption that the choices of each individual councilman are strongly ordered. Imagine that each councilman could cast votes proportional to the intensity of his enthusiasm for one option compared with the other. Imagine further that each councilman feels three units of enthusiasm for his first choice, two units for his second choice, and one unit for his last choice-- compared with 0 units of enthusiasm for doing nothing and letting the mayor scrape along in his old clunker. We will place these "enthusiasm" levels in parentheses next to each choice in the revised table:

                      Ames        Baker       Carter
      First choice    GM (3)      Ford (3)    Chrysler (3)
      Second choice   Ford (2)    Chrysler(2) GM (2)
      Third choice    Chrysler(1) GM(1)       Ford (1)

Voting on a pair of cars, each councilman will thus cast votes equal to his enthusiasm for the one minus his enthusiasm for the other. Between GM and Ford the vote would be:

          Ames 1 vote for GM (3 - 2 = 1)
          Baker 2 votes for Ford (3 - 1 = 2)
          Carter 1 vote for GM (2 - 1 = 1)
In spite of the odd number of voters, the social ordering is not strong-the group's vote is a tie: 2 to 2. And we will get the same results if the councilmen choose between Ford and Chrysler or between GM and Chrysler. Adding enthusiasm scores for the individual cars, each gets six points. GM = 3 plus 1 plus 2, Ford = 2 plus 3 plus 1, Chrysler = 1 plus 2 plus 3. Again, a tie. Thus one possible explanation of the "paradox" is that the group is unable to arrive at consistent decisions because its members have no collective preference for one car over another. As individuals, each has his preference, but the group is indifferent.

But there is another possibility equally compatible with the voting behavior of the councilmen. For one last time, let us restructure our assumptions about the relative enthusiasm felt by each councilman towards each car. Instead of each councilman's values having the same relative intensities (3-2-1) for the cars, imagine that they are as follows:

                    Ames         Baker        Carter
    First choice    GM (8)       Ford (9)     Chrysler (7)
    Second choice   Ford (7)     Chrysler (6) GM (6)
    Third choice    Chrysler (2) GM (2)       Ford (4)
The social ordering obtained by adding up enthusiasms for each car separately are:

            Ford      7 plus 9 plus 4 = 20
            GM        8 plus 2 plus 6 = 16
            Chrysler  2 plus 6 plus 7 = 15
This indication that there is a strong ordering of the social preference is confirmed by voting weighted in terms of the relative individual enthusiasms:
               Ford versus GM
               Ames 1 vote for GM (8 - 7 = 1)
               Baker 7 votes for Ford (9 - 2 = 7)
               Carter 2 votes for GM (6 - 4 = 2)
Here, Ford wins, 7 to 3.
               GM versus Chrysler
               Ames 6 votes for GM (8 - 2 = 6)
               Baker 4 votes for Chrysler (6 - 2 = 4)
               Carter 1 vote for Chrysler (7 - 6 = 1)
Thus, GM wins, 6 to 5.

So far, so good. The council's social preference is for Ford over GM and GM over Chrysler. But will there be an intransitivity paradox when the vote is between Ford and Chrysler?

               Ford versus Chrysler
               Ames 5 votes for Ford   (7 - 2 = 5)
               Baker  3 votes for Ford   (9 - 6 = 3)
               Carter 3 votes for Chrysler (7 - 4 = 3)
Here, Ford wins, 8 to 3. Thus there is no paradox. And yet the very same pattern of council preferences produces the paradox when the vote is a simple head count--"one councilman, one vote."

How Preferences Are Communicated. In many respects the paradox of voting is wholly academic in the worst sense of that term. No evidence of any such paradox in actual practice has been put forward. The paradox is found only in certain kinds of formal theory. But many valuable insights can be gained from thinking about this purely hypothetical paradox. It suggests that a group's social ordering of its preferences cannot always be ascertained simply by counting heads on questions related to a single issue. Such head counts make no allowance for the extent to which each individual member feels strongly about the options. But surely democracy requires that the intensity of individual preferences be taken into account in making organizational decisions. Nobody can win all the time, and most people would certainly prefer to have the ones they do win concern the issues they feel strongly about.

In a small group like this hypothetical city council, members can easily communicate the intensity of their feelings. Since the council must make many decisions, not all of which are equally important to each member, other members will tend to defer to the desires of one who feels most strongly about a given issue. In return, he defers to them on issues they are more concerned about. To the extent formal votes are taken at all in such a small group, therefore, they may not reflect actual individual preferences on any particular issue. Rather, they reflect actual preferences discounted by the desire for cooperation on other issues.

Large legislative bodies such as the American House and Senate cannot take preference intensities into account by the informal ways possible in smaller groups. However a great many mechanisms have evolved enabling these houses to consider intensities as well as head counts. Perhaps the most dramatic and controversial of these mechanisms is the filibuster in the Senate. A filibuster is an effort to block a floor vote on a bill that a majority intends to support by debating it to death. It is a tool employed by a numerical minority that feels very strongly against the bill. If the numerical majority is sufficiently large and itself has strong feelings about the matter, it can invoke cloture and choke off the filibuster. Cloture presently requires a 60% head count support, but cannot be gotten if any substantial part of the majority is only lukewarm in its support of the controversial bill. A vote for cloture will not be forgotten when the bill's supporters seek cooperation of its opponents on other matters.

Likewise, in the House, intensities are taken into account through the complicated procedures by which a bill moves through committees and on to the floor for action. When a bill that a majority of the Representatives say they favor fails to make it out of committee or fails to secure a special rule, this may be a case in which an intense minority has checked a feeble majority. It takes only a bare majority o Representatives' signatures on a discharge petition to pry any bill out of a recalcitrant committee. It is probably safe to say that a Representative who says he favors a bill but refuses to sign a discharge petition is only weakly in favor of the bill and is more concerned about other matters that might be jeopardized if he signs the petition.

There is also a good deal of vote-trading, "log-rolling," (see Glossary) and other dealing that goes on among members of both houses of Congress, all of which allows intensities to influence the outcomes of particular votes. The interesting thing is that these practices, along with filibusters and committees that defy the numerical majority, are almost universally condemned as undemocratic.

Congress has plenty of problems, and we will not try to minimize these in this analysis. But the problem with Congress is not that it is undemocratic or insensitive to majority sentiments.

Organizational Rationality and Congress

T. V. Smith once jocularly defined the art of legislation: "Do something, then wait to see who hollers, and then relieve the hollering as best you can to see who also hollers." [Footnote 13] Smith's formula is not a bad description of congressional behavior. But does this not mean that Congress as an organization behaves with a very low degree of rationality? According to our previous analysis, rational action requires a search for the action that will produce the best net ( X minus Y) consequences possible under existing circumstances as measured by a set of values. When we speak of the rationality of individual action, the relevant values are those of that individual. The relevant set of values in determining the rationality of organizational action is the social ordering discussed in connection with the paradox of voting earlier. But how could "do something, then wait to see who hollers.." constitute rational action by any conceivable standard?

Incrementalism. In political science literature this issue is posed in the incrementalism (or "muddling through") versus comprehensive rationality debate. [Footnote 14] Rationality is said to represent an effort to maximize attainment of various goals, whereas incrementalism rests content with "satisficing." Rational action seeks the best possible outcome, incrementalism looks for an outcome that is good enough. Now there can be no doubt that Congress engages in a great deal of satisficing. Smith's formula is obviously based on much observation of actual congressional behavior. If we could sharply distinguish between satisficing and maximizing, and if rationality requires maximizing, then we would have to conclude that Congress exhibits a very low degree of organizational rationality.

However the dichotomy between satisficing and maximizing goal attainment is a false one. Like most decision makers, Congress must make hundreds of decisions every year. No doubt every single one of these decisions could be improved on if Congress devoted unlimited amounts of time to it. But the side effect of devoting more time to one decision is that others must be made in less time, with a consequent decline in the acceptability of their consequences. The classical expression for this uneven approach to decision-making is "penny wise and pound foolish." At the individual level we can plainly see the folly of disproportion in the following example:

A one-hour college class test consists of two essay questions, each worth a maximum of 50 points. On question number one, you can write an answer worth 38 points in half an hour, but if you spend the full available hour on it you can produce an answer worth 45 points. (Points do not increase proportionately to time because of decreasing marginal returns.) But it would probably not be very rational for you to spend the extra half hour on question one, because a side effect of doing this is to leave no time to work on question two:

    Option one Spend half an hour each on questions 1 and 2:

                     1  38 points
                     2  36 points (or so)
                 Total  74 points

      Option two Spend the whole hour on question 1:

                       1  45 points
                       2   0 points
                   Total  45 points

When many decisions must be made, to satisfice on individual issues is to maximize from the overall point of view. As financial analysts say, it is the "bottom line" that counts. To the extent that the bottom line from Congress, its legislative output, is inadequate, it is therefore probably not due to the incremental ways in which it makes its decisions. Satisficing is a highly rational way for Congress to act under its complex circumstances. The fundamental problem with Congress may rather be that it tries to make too many decisions.

Options of Congress. Devoting too much time to one decision leaves too little time for others. But a certain minimum time is necessary to decide anything at all, let alone to decide wisely. Only a limited number of decisions can therefore be made by anybody. And this is especially the case when the decision-maker is a nonbureaucratic organization, like Congress, rather than a bureaucracy or an individual. Given the broad responsibilities of Congress, we therefore find that it has few options: (1) It can function ineffectively by making many poor decisions, or by abdicating some of its responsibilities completely. (2) It can delegate some of its legislative responsibilities--to its own committees, to the bureaucracy, or to the courts--and exercise a mainly supervisory or appellate role, looking into things only when someone "hollers" very loudly. This technique makes more sense than option one, but has serious dangers for an actor organized like Congress. Delegating legislative power opens the floodgates so that enactments can easily proliferate to the point that Congress cannot even effectively supervise them. Like all decision- makers, even more than most, Congress suffers from the span of control problem, and once it delegates legislative powers it is very likely the tail will wag the dog. (3) Congress can make its decisions at a sufficiently wholesale level to bring their quantity down within manageable limits.

To maximize its constructive power, Congress must make rules. Abandoning concentration on decisions about particular government actions, Congress must decide on whole classes of actions. It must avoid decisions in the form expressed by A ---> X + Y, and emphasize those we have referred to as R ---> X + Y. Only thus can Congress wield power effectively over the actions taken by government.

Another Look at the Individual vs. the Organization. Here, though, we encounter a sharp conflict between the requirements of rational action by Congress as an organization and by, congressmen as individuals. Congressional clout as an organization can be wielded only by enacting general rules. But individual congressmen can best advance their own fortunes by doing particular favors for constituents and for other congressmen. Individual mileage is gained by acting as ombudsman for disgruntled constituents. It is gained by adding amendments--complexities reducing the generality of proposed rules and making their adoption more like a retail action--to protect one interest or another. It is gained by "log-rolling" and mutual backscratching arrangements with other congressmen. And, perhaps most of all, it is gained by participation in the details of the congressional budgeting process, a process that inherently smacks more of retail than of wholesale decisions.


Money for the Government

The government of the United States wields a tremendous power of the purse in its capacities as contractor and trustee. According to the U.S. Department of Commerce:

During fiscal 1975-76 it is estimated that the Federal Government, the 50 State governments, and the more than 78,000 local governments . . . collected revenues of $572.6 billion, spent $626.1 billion and were $871.4 billion in debt. Compared to fiscal 1974-75 these data mark a $55.5 billion increase in revenues, a $68.7 billion increase in expenditures and a $106.0 billion increase in indebtedness. Footnote 15
Since it is difficult for the average person to visualize a billion dollars, one writer tried to clarify the federal budget:

A billion seconds ago Pearl Harbor was bombed.
A billion minutes ago was 40 years after the death of Christ.
A billion hours ago man had not yet set foot on the earth.
A billion dollars ago in federal spending was yesterday. Footnote 16

Another way to think of billions is to imagine one dollar bills taped end to end to make a long chain stretching from the earth to the moon and back to earth again. This round trip chain will contain slightly more than five billion dollars.

In recent years per capita taxes have been the fastest rising component in the cost of living, and now actually amount to more than the average family spends for food, shelter, or clothing. An increase in Social Security taxes voted in 1978 was so stiff that the staff of the congressional Joint Economic Committee predicted it would increase unemployment by 1.3 million people. Footnote 17

Who Pays and How

Basically, all governments raise their money at the point of a sword. Rather than threatening to cut off government services (inducements) to people who do not pay their taxes, one threatens them with jail or with seizure of their property. Perhaps one reason that threatening withdrawn inducements is insufficient to get most people [Footnote 18] to pay their taxes is that many government services are "collective goods." [Footnote 19] They must be provided for everybody in order to be provided for anybody. National defense, for example, against aggressors cannot be supplied to some people in a country but not others. Nor can highways (except in the special case of toll roads), or public health.

Another reason why the threat of withdrawn inducements is not enough may be that people do not desire many of the services and regard even those they do want as overpriced. H. L. Mencken was expressing this point of view when he said:

The intelligent man, when he pays taxes, certainly does not believe that he is making a prudent . . . investment of his money; on the contrary, he feels that he is being mulcted in an excessive amount for services that, in the main, are useless to him. . . . In . . . the exploiters constituting the government . . . he has no confidence whatever. . . . They constitute a power that stands over him constantly, ever alert for new chances to squeeze him. . . . If they leave him anything at all, it is simply prudentially, as a farmer leaves a hen some of her eggs. Footnote 20

It is often said that American income taxes are collected by a self-assessment method on a "voluntary" basis. We should not be deceived by such statements into thinking that the association between taxpayers and government is voluntary. Of course, the taxpayer's actions are voluntary, as are those of the victim handing over wallet to robber. But all taxes are based on sanctions. Taxes are collected by government-as-legislator or government-as-bandit, or they are not collected at all.

A Look at the Present Tax System

During the 1976 campaign Jimmy Carter described the American tax system as "a national disgrace." It is hard to disagree. Carter apparently thought that American tax rates are not "progressive" enough. But even if we reject the idea that the rich should pay a higher percentage of their income than the poor, we can still agree that the system is disgraceful. Sanctions, we have argued, should be applied only to people who have violated general rules of action. Progressive income taxes, seen in isolation, conflict with this requirement. Under a progressive tax there is no general rule that declares "anybody who does not pay 22% of his income to the government shall suffer specified sanctions." Instead, people are classified in terms of yearly income, and different rules are applied to people with different incomes: One income bracket pays 22% in taxes, a higher bracket pays 26%, still higher pays 47%, and so forth. However, the premise that sanctions should be imposed only for violating general rules provides no real basis for concluding that the present American tax system is a disgrace. Conclusions that elements of that system are unacceptable do not automatically apply to the system as a whole. The federal personal income tax is only one element in that overall system, and produces only about 36% of total federal government tax revenues and only about 26% of the total tax revenues of federal, state, and local governments combined. [Footnote 21] Other elements in that tax system are flat rate. Still others are regressive, with a higher percentage of income being paid by the poor and a lower by the wealthy. When these different taxes are combined, a careful study at the Brookings Institution indicates that the overall rate is "flat." In other words, if all federal, state, and local taxes were repealed and replaced with a single flat-rate personal income tax producing the same total revenue as the present system, it would have little effect on the total taxes paid by anybody. Since a flat- rate income tax can be expressed as a general rule--all shall pay X%--this means that the present system is not incompatible with the generality requirement. The present system is the equivalent of a general rule, even though its elements are not.

What undoubtedly is disgraceful about the American tax system is its monumental complexity. Consider just the federal income tax. An annual Federal Tax Course, a basic textbook published by the Commerce Clearing House, runs approximately 1400 pages, most of which deal with income taxes. The Internal Revenue Code extends for more than 2000 pages. "Interpreting and implementing the statutory provisions of the Code are the Regulations issued by the Treasury Department. . . . In recent years the Regulations have grown extensively and now number in the thousands of pages in the federal income tax area alone. " [Footnote 22] On top of all this the Internal Revenue Service issues 40,000-50,000 formal "Ruling Letters" each year, as well as an uncountable number of informal advisements to taxpayers.

A person who wants to understand what the federal income tax law requires may also have to consult the many decisions of the U.S. Tax Court and the Court of Claims, as well as those of the District Courts, the Courts of Appeals, and the Supreme Court.

The complexity of the federal tax code derives in part from the following six factors:

Its original purposes lost, each loophole breeds new ones. For example, the deductibility of interest paid on home mortgages, originally intended to encourage people to buy their own house rather than rent, is currently being used as an equity argument in favor of a similar deduction for a portion of rents paid. The complexities come basically from the purported efforts to redistribute income in favor of the poor through the tax system, and then to remove the inequities caused by the actions (like progressive taxes) directed towards this goal. Objectively, the beneficiaries of the present tax structure are the professional people who work to administer these complexities, on the one hand, and those who help people figure out how to avoid or minimize their taxes, on the other hand: the tax lawyers and the accountants. From the overall point of view the work of these people therefore largely cancels out in the wash and is wasted effort.

What Is Wrong with the Present System?

Responsibility for federal tax laws rests squarely with Congress. Congressional performance here may be strongly influenced by internal House and Senate procedures and organization. Imbalances between revenue and expenditures, for example, have produced a vast and rapidly increasing national debt, or accumulated deficit. Of course, deficits could merely reflect the popularity of spending and the unpopularity of taxing. And they do not disprove the economic axiom sometimes known as TANSTAAFL--"there ain't no such thing as a free lunch." When government spends money to purchase goods and services, it reduces individual purchasing power correspondingly, and if this is not done in the form of taxes it will be taken care of by inflation. Individual purchasing power can just as well be reduced by decreasing the purchasing power of a given amount of money as it can by taking away some of that money. However unbalanced budgets are undoubtedly encouraged by the way taxing and spending measures are separated for committee treatment in Congress. Both kinds of taxation issues--those regarding tax structure and those of tax levels--are handled in one committee in each house. Expenditures are processed in entirely different committees, nor can one member sit on both committees. This is expressly prohibited by the rules of each house. The individual interests of congressmen in low taxes, high spending, and extensive reliance on inflation to make up the difference encourage normal tendencies for the full house to defer to these uncoordinated committee decisions. Inflation, of course, can be demagogically blamed on greedy corporations, labor unions, the sunspot cycle, anybody except Congress!

Coupling of tax levels and structure decisions in a single committee further confuses the issues in Congress. Proposals to increase or reduce taxes get bogged down in arguments over which taxes should be changed. Proposals to reform the tax structure are opposed or supported because of the impact the reforms will have on revenue levels, not on the structural merits.

A Radical Prescription for Reform

Let us forget about political feasibility for a moment and ask what an ideal tax system for the federal government might look like. Based on our earlier discussion, let us assume that an ideal system would be:

1. simple, rather than complex;
2. equitable--flat rate or the equivalent of flat rate;
3. one producing few harmful side effects in the general economy.

One possible package might include the following four changes: (1) The social security tax and other payroll taxes would be repealed. These taxes are regressive, since they apply only on earnings up to a certain point. They encourage employers to hire a few people for full-time work or even overtime work rather than spreading available work more evenly. [Footnote 23] By purporting to collect part of the tax "from the employer" they confuse and mislead workers into thinking that the money is not all coming out of their pockets. (From an employer's point of view these taxes are just as much a labor expense as are wages. Without the tax, he would be equally willing to pass the money along in raises or in reduced prices charged to consumers, and all workers are also consumers.) (2) All tariffs would be repealed. Such tax laws are no longer revenue measures, but are manifestations of the differential mobilization problem. There is no reason why consumers in general should be forced to pay higher prices for some goods just to benefit some workers in favored industries. Legitimate concerns for preserving certain industries in case war cuts off foreign supplies can be accommodated by paying subsidies to the relevant firms, thus putting the expense of national defense insurance on all the taxpayers, where it belongs. (3) Inheritance taxes would not be used as a source of government revenue. (4) The present graduated or "progressive" income tax on individuals would be replaced with a flat-rate tax on individual and, perhaps, corporate income. This tax would have a number of virtues. It would be simple, and therefore understandable. As we have seen, much of the complexity in the present income tax stems from the effort to make it fair in spite of being progressive. With a flat rate--all to pay X%--there would be no need for different rates for the married, the married filing singly, the single, or heads of household. There would be no need for income averaging, and there would be less reason for taxing capital gains at lower rates than other income. [Footnote 24] Finally, and perhaps most important, the tax would be honest. With a flat-rate tax it is impossible to create the illusion that increased government expenditures can be paid for with other people's money, by "soaking the rich." The fact is that there are not enough rich individuals to soak. The Tax Foundation estimates that if all taxable income over $32,000 a year were confiscated-taxed at a rate of 100%-the extra revenue generated would amount to less than 5% of the federal budget. Footnote 25

Taxing Corporate Profits. The advisability of having any tax on corporate profits at all is debatable. Economically, such a tax may help to promote stability by providing a "cushion" of costs that disappear whenever a corporation starts losing money. This cushion may enable the corporation to avoid bankruptcy and putting its employees into unemployment, giving it time to pull itself back together. [Footnote 26] Politically, it seems unprincipled to tax individual persons but not to tax (at the same rate) corporations which claim the rights of persons under the Constitution. On the other hand it can be argued that all of the people who make up the associations we call corporations will be paying income tax on their salaries, dividends, and interest, and that taxing the corporation too represents a double burden. Alone, this objection is not insurmountable, but there is also a serious psychological problem with taxing corporate profits: many corporations are big and generate huge volumes of business, and it is tempting to think of them as something rich that can be "soaked."

In fact, of course, corporations cannot really be taxed at all. Taxes represent the difference between what an individual produces and what he can consume, and corporations as such neither produce nor consume anything. Rather, they are associations in which individual efforts are coordinated and organized for maximum productive efficiency. All production and all consumption are done by individuals, and hence all taxes are paid by individuals. For one reason or another some of these taxes may be collected wholesale rather than retail. Taxes collected from corporations reduce the money available to pay salaries and dividends, and thus are taxes on the individuals whose receipts from the corporation are correspondingly reduced. To the extent that such payments are not reduced by the corporate tax, the tax will be passed along to consumers as higher prices--a hidden sales tax. And since the poor must spend more of their income than the rich, sales taxes are regressive. Perhaps George McGovern did not appreciate the economic humor of his 1972 campaign proposals to reduce personal tax rates and increase the rate to corporations.

Perhaps on balance there should be a corporate income tax, in order to get the economic stabilization it can provide. But the tax rate on corporations should be at the same flat rate that it is on individuals. Using the same rate would eliminate temptation to hornswoggle voters by decreasing their direct (retail) taxes while increasing their wholesale (indirect) payments. Using the same rate for all taxpayers, individual and corporate, would also be a more general rule than having separate rates for each, and thus would comply with the requirements of the rule of law.

Why this Reform is not Possible. The chances of any reform along these lines are very poor. From the congressman's point of view, it has all sorts of tactical disadvantages in addition to any problems it might have as policy: (1) The proposed tax system is simple. But congressmen like to load bills up with complexities and loopholes, compromises designed to reconcile as many interests as possible to a given law. (2) It is honest. But congressmen appear to prefer tax schemes that confuse taxpayers so they are unable fully to comprehend what is being done to them. To this effect we have "soak the rich" rhetoric cancelled out in practice by economic realities and by compensating loopholes. We have inflation as a hard to understand substitute for higher taxation. And, perhaps best of all, we have the "withholding" system under which most workers never have to "pay" any income tax at all. Imagine the howls that would arise if federal income taxes came due in a lump sum at the end of a year. Economically, there is little difference (except for implicit interest) between subtracting $50 a week from someone's paycheck and handing him a bill for $2500 at the end of the year. Psychologically and tactically, there is a world of difference which most congressmen are presumably well aware of. (3) The proposed reform is a radical departure from current practice, and Congress is strongly inclined to make incremental rather than major changes in any program.

Perhaps if the electorate were more sophisticated Congress would be willing to adopt a simplified tax system along the lines suggested here. But if the electorate were economically sharp, the reform would not be as necessary as it is under actual conditions. As things stand now, the only way we are likely to get basic tax reform would be if the Supreme Court struck down the progressive income tax as a violation of the Equal Protection and Due Process Clauses of the Constitution. Such a decision would leave Congress with no choice but to pass a flat rate--i.e., simpler-tax. But do not hold your breath waiting for the Supreme Court to act.

A More Reasonable Reform

A somewhat less sweeping reform might be more feasible for Congress to adopt. This reform would do two things to the way Congress organizes itself to consider money issues: (1) Separate the issues of tax levels and tax structure by assigning them to different committees. (2) Combine the issues of tax levels and appropriations, so that the onus of raising taxes is borne by the same people receiving the credit for increasing spending.

Under this system there would still be two money committees in each house, but their functions would be re-allocated: The Committee on Appropriations and Tax Levels would operate much as the Appropriations committees do now. But a new rule would provide that decisions to increase expenditures would automatically increase the levels of all current federal taxes by the percentage necessary to pay for the extra expenditure. These parallel changes in the various tax rates would not, of course, produce any substantial change in tax structure, the percentage of total revenues resulting from each individual tax. Footnote 27

The Committee on Tax Structure would be prohibited from making any changes in the total revenues collected by the federal government. It would operate under a rule stating that any change it proposes in the rate at which one kind of tax is assessed will be automatically offset by a change in the opposite direction in the rates of all other federal taxes. Thus if a tax accounting for 25% of federal revenues is abolished, the other taxes previously generating 75% of revenues would all be increased by 33% so there will be no net reduction of total receipts.

One may well wonder why congressmen who are unwilling to enact separate spending and taxing measures that will balance the federal budget would support enactment of rules producing the same result. This is a good question going to the heart of the differences between retail actions, A ---> X + Y, and wholesale actions, R ---> X + Y. A congressman lacking the courage to stand up and insist the budget be balanced in the context of a specific spending proposal may be all too happy to do so in the abstract. Individual actions deduced from a rule we have adopted are not necessarily the same as they would be if we made each individual decision on its own merits. Otherwise, there would be no use in having rules. When a hijacking or kidnapping has occurred, the best action in the specific case may seem to be to capitulate to the terrorists' demands--otherwise life may be lost. However the best rule for dealing with terrorists may be to refuse to deal, because dealing simply encourages more of the same bad type of behavior, increasing insecurity and risk to life in general. The same people who would be unwilling to resist such demands on the merits of a specific case may therefore be quite willing to deduce such a resistance from a rule that they have adopted. Rule making and enforcement force us to consider the broader picture and ramifications of our individual actions and, as such, it expands the rationality with which we act. And what is true in dealing with hijackers may be applicable also to budgeting and taxing decisions.


Beginning in 1962 a series of Supreme Court decisions regarding legislative apportionment shook some aspects of American politics almost to their foundations. Declaring legislative districts that do not give voters equal representation unconstitutional, the Supreme Court imposed a standard of "one man, one vote" on elections to all American legislatures except the U.S. Senate. [Footnote 28] The decisions have forced redrawing of many lines in order to equalize the population in districts from which state representatives, state senators, and U.S. representatives are elected. But why has there been so much shouting about "malapportionment" when power is so much more skewed by virtue of arrangements within Congress?

As a representative institution Congress often gets low marks compared to the presidency. Having to run for reelection in one specific district that is a small part of the country, congressmen may have a very parochial orientation on policy matters. Present arrangements in Congress provide abundant opportunities for confusing the public on who is responsible for what: There is the basic obscurity of individual congressmen. There is the complex, confusing nature of the procedures and outcomes of legislation, shot through as it is with poorly assimilated compromises. There is the committee system, bicameralism, and the presidential veto. There is the fantastic growth of staff working for individual congressmen and for their committees. And there is the vast scale of legislative output.

Although it may be politically impossible to make basic changes in Congress, we may find it instructive to think about what kinds of change we might want to make if we could. In physics, gedanken (thought) experiments, those which cannot be performed, have long been very productive of scientific insight. Perhaps thinking about the impossible may also be rewarding for political observers. It may help us to discover things that are possible that we never would have thought of otherwise. And some things that would be politically impossible if they had to be enacted by Congress just might prove feasible if a new constitutional convention were to meet. It is exactly this possibility, of course, that might make Congress reluctant to convene such a convention.

The Challenge

The following proposal is put forward for your consideration, discussion, and evaluation. It challenges certain things now generally taken for granted:

1. The assumption that the same type of organization is ideal for enacting both laws and bodres.
2. The assumption that laws and bodres should be enacted by the same institution.
3. The assumption that laws should be enacted by a group of legislators.
4. The assumption that the ombudsman function is best performed by the same people who make laws.

A Proposal for Reform

Let us wipe out Congress as presently constituted. Supreme power to make laws (as distinguished from bodres) would be vested in a one-person legislature, elected directly by the whole country for a term of four years. The legislator general would be provided with a bureaucratically organized staff. All laws--general rules enforceable by sanctions--made by any branch or agency of the government would be appealable to the legislator general and subject to his absolute veto. The legislator general himself could enact laws, subject to an absolute veto by a one-house chamber (perhaps called the House of Representatives to emphasize historical continuity) of any reasonable size. Members of this chamber would be bound by contract or oath of office not to communicate with any officer of government about individual cases. That is, the discussions of this chamber are to be conducted solely at the level of rules and hypothetical cases that might arise under these rules.

Bodres will be enacted by an entirely separate organization, a 100-member Board of Directors, elected by districts. For simplicity and historical reasons, the Board of Directors might be called the Senate. The president, as Contractor General of the United States, would have a nonabsolute veto, including an item veto, over bodres enacted by the senators. Consent by the Senate would still be required for presidential appointments and ratification of treaties and other major contracts to which the U.S. is one of the parties. The Senate would handle appropriations, for these are bodres rather than laws, and the amended Constitution would provide that tax levels would be set automatically by the appropriation decisions. Tax structure, a matter of law rather than bodre, would be determined by the legislator general. Members of the Senate would also act individually as ombudsmen for their respective constituents who are having trouble dealing with executive departments.

Pros and Cons

As a mental exercise you may wish to jot down your own views of the possible advantages and disadvantages of the proposed reform of Congress before reading this concluding part of the chapter. You may well see some angles that are not explored here and your own list may become an agenda for class discussion.

Nine possible advantages of the proposed reform might include the following: 1. The one-person legislature would reduce transaction costs in legislating to an absolute minimum. There would be no quorum problem, no need to break into committees so everyone can have a say, no time-consuming arguments among equals as to the wording of a bill.
2. It focuses responsibility for laws on a given, visible, and interesting individual instead of diffusing it among many obscure, uninteresting legislators. Thus it facilitates democratic control of the general population over the law-making process.
3. It puts a complete end to the apportionment problem as far as law-makers are concerned.
4. The House's absolute veto restrains the legislator general from acting highhandedly.
5. It guarantees that the legislator general must have a national perspective if he wants to stand any chance of being reelected.
6. It educates people to the important distinction between laws and bodres by making these the responsibilities of separate institutions.
7. It guarantees a balanced budget, thus removing one basic way by which the general public is presently tricked into thinking government can give it something for nothing.
8. It makes possible the enactment of simple laws by eliminating horse-trading among legislators.
9. It separates the ombudsman from the law-making function, thus allowing the legislator and his veto chamber (House) to concentrate on rule making. But it still leaves the ombudsman function in the hands of prominent individuals with great power over the bureaucracy--the senators.

Four possible disadvantages of the reform include the following:

1. It may concentrate too much power in the hands of one individual. Or it may be regarded as doing this even if it does not in fact do so, thus reducing popular acceptance of the legitimacy of American political arrangements.
2. There is a danger that the legislator general will be a "prisoner" of his staff, putting all legislative power into bureaucratic hands.
3. It may streamline the legislative process too much. If anything, we already suffer from too many laws, and this might aggravate the problem.
4. There is nobody assigned to enact pseudo-laws (as distinguished from laws and bodres) under the proposal.

Pseudo-laws constitute a very high proportion of present government business (as we will see, especially in Chapter 10). This "weakness," however, is an entirely deliberate part of the plan. On the premises in terms of which the present analysis is written, pseudo-laws ought not to be enacted; if enacted, they ought not to be enforced; and if enforced, they should not be obeyed. To the extent that the proposed reform would heighten public sensitivity to the distinction between laws and bodres, it should also make it easier to see the difference between laws and pseudo-laws and to act on this perception.


Its size and work load force Congress to rely on committees to get the job done. This reduces the ability of Congress to coordinate government policies and increases inequalities in the power of different congressmen. But congressmen may be more interested in assured wielding of some power than in trying to maximize their clout, individual or collective. Thus congressmen spend great portions of their personal and staff time in politically valuable troubleshooting for constituents rather than participating in the rule making which is the task of Congress as an organization.

The present system of national taxes reflects the Congress, which created it. It is not a simple system of general rules reflecting a coordinated strategy of taxation. Instead, a tremendously complicated system has emerged from hundreds of particular decisions variously sought and opposed by thousands of particular interests. Yet critics pointing to sinister sell- outs to "monied interests" fail to note that the overall tax system is nearly flat rate, as required by the rule of law. The tax system is a disgrace, not because it is unfair, but because its complexity produces monumental social inefficiency, with thousands of tax lawyers and accountants.


1. Most bills introduced into Congress are not enacted into law (or bodre, or in pseudo-law). Is this bad? Why or why not?

2. How do you react, and why, to the following proposed change in our tax laws?
For every five pounds that a taxpayer is overweight, he shall pay a 10% surcharge on his federal income tax.

3. Explain why it is rational for individual congressmen to devote most of their time to non-legislative work.

4. What are the similarities and differences between the organization of Congress and bureaucracies?

5. What is the paradox of voting, and why is it no paradox?

6. Why can the filibuster help make the Senate more democratic?

7. Explain why Congress must enact rules if it wishes to maximize its own power over the government.

8. To what extent do you think the low quality of national legislation is the result of:

9. Explain why it is impossible for the government to spend money without passing the costs along to individual taxpayers.

10. Assuming that Congress would never go along with the proposal for a one-person legislature, by what means could such a change be implemented if public opinion were strongly enough in favor of it?


To access other chapters,
go back to the Table of Contents.


* F. A. Hayek, Rules and Order (Chicago: University of Chicago Press, 1973), p. 175.

1. Mark 0. Hatfield, Not Quite So Simple, New York: Harper and Row, 1968, pp. 107-115.

2. Deems Taylor, ed., A Treasury of Gilbert and Sullivan, New York: Simon & Schuster, 1941.

3. Toledo Blade, November 20, 1977.

4. The American College Dictionary.- "A person who takes up residence in a place, with no more property than he brings in a carpetbag, to seek special advantages for himself " Also, a contemptuous expression for "a Northerner who went to the South after the Civil War to seek political or other advantages made possible by the disorganized condition of political affairs."

5. Drawing boundary lines for election districts so as to maximize seats won by one's own party. The term is a combination of the exceedingly peculiar shapes of some of the resulting districts--one that looked like a salamander, notably--and the name of one of the early practitioners of this art, Governor Gerry of Massachusetts.

6. Graham Allison, Essence of Decision, Boston: Little Brown, 1971.

7. The amount a person would pay today in return for the right to receive a larger sum at some future time and feel that he has just broken even. If we impute an interest rate of 5%, $1 would be the present value of $1.05 a year from today.

8. J. Dunner (ed.), Dictionary of Political Science, Totowa, N.J.: Littlefield, Adams & Co., 1970, p. 270.

9. See Bertrand de Jouvenel, "The Chairman's Problem," 55 American Political Science Review 368-372 (1961).

10. Joseph Clark, Congress.- The Sapless Branch, New York: Harper and Row, 1964, p. 23.

11. David Klahr, "A Computer Simulation of the Paradox of Voting," 60 American Political Science Review 384-390 (1966), p. 385.

12. Ibid., p. 384.

13. Amold J. Heidemheimer (ed.) Political Corruption: Readings in Comparative Analysis, New York: Holt, Rinehart and Winston, 1970, p. 442.

14. See Herbert A. Simon, Administrative Behavior, New York: Free Press, 1965.

15. Government Finances and Employment at a Glance.Washington,D.C.:Bur eau of the Census,December 1977.

16. Donald A Lindow, letter to the editor of Detroit Free Press, November 23, 1977.

17. Detroit Free Press, February 23, 1978.

18. Civil servants might be an exception!

19. See Mancur Olson, The Logic of Collective Action, New York: Schocken Books, 1968.

20. H. L. Mencken, American Mercury, February 1925. Quoted in National Review, April 12, 1974, p. 411.

21. Government Finances and Employment at a Glance.

22. 1976 Federal Tax Course, New York: Commerce Clearing House, 1975, p. iii.

23. Since the payroll tax applies only to the first $X,000 of each employee's salary, no tax is paid on salary over and above this figure. It may therefore be cheaper to have an employee who is going over the top anyway work extra hours rather than adding new workers for whom additional taxes would have to be paid. The salary on which Social Security tax must be paid has been increasing rapidly, making this problem less important in the process.

24. With a progressive tax, capital gains in the value of a stock held for 10 years are taxed in a higher bracket in the year in which the stock is sold than they would be if the same income had been equally distributed across the 10 years and taxed partly in each of those years.

25. Adrian Daily Telegram, December 29, 1977.

26. Dewey B. Larson, The Road to Full Employment, Portland, Oregon: North Pacific Publishers, 1976.

27. Some changes in tax structure would result because taxes on some kinds of activities may discourage those activities more than taxes on other kinds of activities. Thus doubling the tax rate on something will probably not double the government revenue produced thereby.

28. Baker v. Carr, 369 U.S. 186 (1962); Reynolds v. Sims, 377 U.S. 533 (1964).