This paper examines four enduring and fundamental shortcomings of the American majoritarian/Presidential system: 1) the impeachment process and its cause, namely head of government/head of state confusion; 2) persistent divided government; 3) costly and seemingly perennial political campaigns; and 4) the lowest voter turnout among all of the leading OECD countries. The paper suggests certain features of parliamentary or consensus systems might overcome these persistent flaws of American politics.

This debate is shaped and informed by the constitutional choices of a number of countries embracing democracy since the early 1990s. It also takes place within the context of significant changes in political parties and party systems in many democratic states.

Thus, the broader goals of the paper are to start a serous dialogue on the respective strengths and weaknesses of majoritarian and consensus systems with scholars of American politics. From a comparative perspective, much scholarship on American politics can be limited with an insularity, or the "N of 1" problem. From a public policy perspective, how do we know if the paper’s hypothesized reforms might improve each of these four pathologies if they never make it to the debating table? Ultimately, the hope of this paper is to engage American and Comparative scholars in a fruitful debate that includes American politics in a Comparative orbit.

Introduction

This paper’s origin is over a decade old. It first took the form of an op-ed piece in a major US daily newspaper in the context of two significant – and taken together – puzzling events. In the late 1980s the United States was celebrating the 200th anniversary of the Constitution at the same time as Iran-Contra hearings stopped strangely short of completion. What explained the strange unwillingness of the US Congress to press seriously the Reagan administration’s deep involvement with this scandal? And how was the failure to thoroughly investigate Iran-Contra in keeping with constitutional provisions to investigate apparent executive branch misdeeds? Together with three other points – listed below – the op-ed piece argued that the US would be better served to reform the constitution and move to a parliamentary system rather than simply celebrating the constitution reflexively after 200 years. The response to this opinion piece was both predictable and paradoxical; predictable in that critics resolutely defended the constitution, yet paradoxical in that they simultaneously offered sharp criticism of the government!

Yet it’s the constitution that is responsible for the form of the American government. Why, when relentless criticism of the U.S. government from many groups has continued for over a decade, (Uehling, 1991) is the constitution seen as such an untouchable document? After all, the founders of the United States created the amendment process and there have been 27 of them in 213 years.

The Clinton impeachment spectacle of 1998-1999, however, prompted this author to revisit more extensively the issue of the American constitutional structure. At the end of the twentieth century, American politics confronted at least three other seemingly intractable problems: perennially divided government; out of control costs of running for public office; and shamefully low voter turnout. More seriously, none of these four problems is of recent origin, as all four have been working to erode the functioning of the American government for a period of between 25 and 50 years! This paper suggests that the core features of these four problems are:

• Confusion of the roles of head of state and head of government, of which the impeachment issue – from Watergate through Clinton’s impeachment – is merely symptomatic of a much larger problem.

• Persistent divided government, dating back to the "do nothing" 80th congress elected in 1946. The result is a serious accountability problem: who gets the credit and who gets the blame?

• Costly and perennial campaigns for all offices producing "the best politicians that money can buy." This problem that had its origins with the breakdown of the party caucus system and the growth of primary elections in the 1960s; and

• The world’s lowest voter turnout among all of the leading OECD countries, a phenomenon that began in the 1960s and has steadily intensified.

Why not? What exactly is the problem with a parliamentary system? Do those who cavalierly dismiss a parliamentary system even understand how one actually works? With the institutional erosion of the Congress, Presidency, party system, and commitment to participatory democracy by large segments of the population, maybe it’s time to take institutional redesign seriously? Perhaps Americans would criticize their government no matter how it was elected or behaves in office for historical or cultural reasons. The relevant point for this paper, however, is that things like trust in government, sense of efficacy, approval ratings for branches in government have all declined in recent decades. This helps to explain declining voter turnout, and highlights what is arguably a more significant trend toward a crisis in confidence among Americans concerning their governing institutions. This paper examines these 4 institutional blockages of the American majoritarian/Presidential system and suggests certain features of parliamentary or consensus systems might overcome these persistent shortcomings of American politics.

The paper is framed by three concepts central to understanding and shaping public policy in advanced industrial states with democratic constitutional structures. First, is the issue of comparability and ’American Exceptionalism’. (Lipset, 1996) The paper’s goal is to initiate a long-delayed dialogue on comparative constitutional structures with scholars of American politics. From the perspective of this comparativist, many Americanists are far too insular and often plagued with the "N of 1" problem. How do we know if these hypothesized reforms might improve each of these four shortcomings of American politics if they never make it to the debating table? Second, the paper hopes to participate in the active discussion among comparativists on the respective strengths and weaknesses of majoritarian and consensus systems. (Crepaz, 1996b; Crepaz, 1996a; Birchfield and Crepaz, 1998) As newly democratizing societies struggle with constitutional forms, why should longstanding democratic regimes be immune from comparison? Third, scandals surrounding money and politics in a number of democratic states (Barker, 1994) should prompt a comparison of parties and party systems and the context within which they function.

Clearly this paper does not argue that we could expect a quick call to a constitutional convention in the United States to move toward a parliamentary form of government. Nor does it underestimate the quite significant problems associated with "institutional transplantation" (Jacoby, 1999) from one country to another. Ultimately, the hope of this paper is to engage American and Comparative scholars in a fruitful debate that includes American politics in a Comparative orbit.

In this context, this paper inquires whether all four of the above- listed maladies of the American political system could be eliminated or fundamentally repaired with the introduction of an American parliament with multiple parties and a proportional representation electoral system. The paper is organized in 5 sections that address: 1) the cumbersome tool of impeachment; 2) the stalemate of divided government; 3) the costly, never- ending campaign process; 4) the continued deterioration of voter turnout, and 5) a critical analysis of the obstacles that a parliamentary remedy to these problems would face.
 
 

1a. Head of State vs Head of Government

Impeachment is merely a symptom of a larger problem. Its more fundamental flaw is that it highlights the constitutional confusion between the two functions of the US presidency: head of state and head of government.

Impeachment: Antiquated and Unworkable

Americanists have delved deeply into the minutiae of the impeachment process, as well they should. (Rhenquist, 1992; Berger, 1973) Comparativists looking at this issue, however, might ask a different question, namely how would other democracies handle similar accusations against its political leaders? It took over a year from the onset of the Clinton-Lewinsky saga to the completion of the impeachment process. Comparativists and citizens of other democratic polities find this astounding, since in a parliamentary system such a challenge to the executive would take the form of a vote of no confidence, (Lijphart, 1994) and the process would be completed within a month or two.

The portrayal of the Clinton impeachment and trial is characterized as historic. For only the second time in American politics, an American president has been impeached in the House and put on trial in the Senate. And unlike Andrew Johnson – who became President at the death of Abraham Lincoln and was put on trial three years later – President Clinton is the first elected American executive to be impeached. Yet, the idea of using impeachment has been much less rare, having been raised three times in the past 25 years.

The first of these events of course was Watergate. Compared to the Clinton impeachment, the proceedings of 1974 Nixon impeachment investigation seem ennobling and Solomon-like. Many observers, then and now, wax eloquent about how the system "worked" in Watergate. (White, 1975) Worked? Compared to what? From the initial break-in at the Watergate complex in June 1972 to Nixon’s resignation in August 1974, a full 26 months passed in which the country was consumed with this scandal. And Nixon resigned before he was impeached and put on trial. How much longer would a full trial have taken? During that period, President Nixon was either distracted from or not fully engaged in, or precipitated such world events as: the first oil crisis, the first major postwar recession, an Arab-Israeli war, the overthrow of the democratically elected Allende government in Chile, continued negotiations with both the USSR and the People’s Republic of China, and the end of the Vietnam war. For more than two years, the United States government was preoccupied by constitutional crisis and not engaging all of its faculties to deal with these major events. And the lengthy and cumbersome impeachment process was the primary culprit.

The second of these events was the Iran-Contra scandal. Here impeachment was not seriously considered precisely because Tip O’Neill and the Democratic majority knew the lengthy and cumbersome process would not likely be completed before President Reagan ended his second term. (Timberg, 1995) And because the memory of Watergate was so fresh, few in the Democratic majority had the stomach for putting the country through another impeachment process. (Walsh, 1994) But the possible charges in the Iran-Contra scandal – trading arms to an outlaw government to secure the release of hostages, then diverting the funds to Central American rebels in direct violation of an act of Congress – were far more serious than either Watergate or the Clinton imbroglio. What, then, is the point of even having an impeachment provision if members of Congress are afraid to use it in such a serious case? Clearly, impeachment hasn’t "worked" at all. It has either not brought to fruition (Watergate), not used when it should have been (Iran-Contra), or completely trivialized and dredged up (Clinton-Lewinsky) when another path was clearly needed. But impeachment itself isn’t the real problem. A larger constitutional design flaw is the underlying reason.
 
 

Executive "Confusion"

The United States has a constitutional structure based on a separation of powers, while most parliamentary systems have a "fusion" of powers in that the Prime Minister is also the leader of the major party in parliament. However, within the American executive itself, there is a "fusion" of functions, which is the exact opposite of Parliamentary regimes.

More specifically, the US is the only developed democracy where head of state and head of government fused in one person. The President is the Head of State and, effectively, the Head of Government. In Parliamentary systems these two functions are performed by two different people. (Linz, 1993) This means impeachment of one person removes two functions in one and likely explains the dichotomy of popular desire for Clinton’s retention on the one hand, but also for some form of censure on the other.

Beyond the impeachment issue, when American presidents undertake some action as head of government for which they are criticized, they then become invariably more remote and inaccessible. For example, Presidents Johnson (Vietnam), Nixon (Watergate), Reagan (Iran/Contra), and Clinton (the Lewinsky Affair) all reduced their appearances at press conferences as criticism of their policies mounted. In short, when criticized for actions taken in their head of government capacity, they all retreated to the Rose Garden and sometimes created the impression that criticizing the President – now wearing the head of state hat (or perhaps, crown) – was somehow unpatriotic.

1b. Toward a Politically Accountable Prime Minister and a Ceremonial President

A parliamentary system with a separate head of state and head of government would produce two "executive" offices instead of just one. It’s odd that a country so fearful of centralized power allows the executive to perform functions that no other leader of an OECD country (save perhaps France) performs alone. Of course, the US does have an office that serves many of the functions of heads of state in other countries, namely the vice presidency. But the United States has an comparatively odd way of dividing constitutional functions. One office, the Presidency, does everything while the other, the Vice Presidency, does virtually nothing and simply waits until the president can no longer serve. An American parliamentary system would redefine these 2 offices so that one person (the head of state) would serve as a national symbol and preside over ceremonial functions. The second person (the head of government) would function much like a prime minister (chancellor/premier or some other suitable title) does in a parliamentary system, namely as the head of government who could be criticized, censured and held accountable for specific political actions without creating a constitutional crisis.

Thus were it necessary to censure or otherwise take action against the head of government (i.e. prime minister), the solution would be a vote of no confidence. This procedure would likely last no more than a month and produce one of three possible outcomes: the prime minister would survive the vote and continue; a prime minister would fail and step down and perhaps retire; or a prime minister would lose the vote but remain in the parliament and live to contest for political leadership at a later date. In other words, a vote of no confidence would solve the problem and move on and let the country address its political business. And unlike impeachment which is the political equivalent of the death penalty, a vote of no confidence does not preclude a politician’s making a comeback and returning to lead a party or coalition. Impeachment and removal from office, on the other hand, is much more final.

In fact, even after leaving office normally, it’s rare for American presidents to remain politically active. Former prime ministers, on the other hand, often return to the opposition benches after losing an election and continue to serve their party and their country. For example, rather than retiring from political life many former German chancellors retains a seat in the Bundestag as a member of their party’s fraktion in the German lower house. (Peterson, 1999) Because the US presidency is seen as such an "imperial" office, becoming a pedestrian member of the House or even of the Senate is seen as a political comedown. Viewed differently, Presidential "retirement" denies the American political system the experience and wisdom that US presidents have accumulated in their careers of public service.

Prime Ministers, unlike US presidents, are seen much more as active politicians not remote inaccessible figures. In a parliament, the prime minister as the head of government is required to engage - and be criticized - in the rough-and-tumble world of daily politics. In short, the head of government must be accountable. The British prime minister, for example, is required to participate in a weekly "question time" in which often blunt and direct interrogatories are pressed by the opposition. There is no equivalent forum for the American president to be institutionally questioned as a normal part of the political process.

Some Americans might object that such a power might be used in a cavalier fashion. Suppose that the head of government could be removed easily after debilitating scandals? This is unlikely in a well-designed parliamentary system. Such cynicism would likely produce a backlash that would constrain partisanship. In fact, the Germans have institutionalized such constraints in the "constructive vote of no confidence" requiring any removal of the head of government to be a simultaneous election of a new one. The context of such a parliamentary system lowers the incentives to engage in the politics of destruction. The political impact of destroying any particular individual in a collective body such as a cabinet or governing party or coalition is much less significant than removing a directly elected president.

The head of state would be above that kind of criticism generated in no confidence votes and simply serve as an apolitical symbol of national pride. In nation states that have disposed of their monarchies, ceremonial presidents perform many of the same roles as constitutional monarchs such as Queen Elizabeth do, but much more inexpensively. In fact, many of these ceremonial roles are performed by the American vice president (attending state dinners/funerals, cutting ribbons, presiding over the Senate, etc.) The problem is that the Vice President is often a political afterthought, chosen more for ticket-balancing functions and/or for inoffensive characteristics than for any expected major political contributions. (Light, 1983; Sindler, 1976) On the other hand, the type of individual usually chosen as a ceremonial president in a parliamentary system is a retired politician from the moderate wing of one of the major parties who has a high degree of stature and can serve as a figure of national unity. In effect, the office of ceremonial president is often a reward or honor for decades of distinguished national service, hardly the characteristics of an American vice president.

In retrospect, one might say that President Clinton was impeached not for abusing head of government functions, but for undermining the decorum and respect associated with heads of state. The separation of head of state and head of government would have a salutary effect on this specific point. Scandals destroying heads of state would have little real political significance since the head of state would not wield real political power. Similarly, scandals destroying heads of government would have significantly less impact than the current American system. The head of government role, once separated from the head of state role, would no longer attract monolithic press and public attention or be subject to extraordinarily unrealistic behavioral expectations.

2a. Divided Government

Who Gets Credit, Who Gets Blame?

From the "do nothing" 80th Congress elected in 1946 to the 106th elected in 1998, a total of twenty seven congresses, the United States has experienced divided government for more than two-thirds of this period. In only eight of those twenty seven congresses has the president’s party enjoyed majorities in both houses of congress. (Fiorina, 1992) Some might observe this phenomenon and praise the bipartisan nature of the American system. (Mayhew, 1991) But to justify such a conclusion, defenders of bipartisanship would have to demonstrate high public approval of governmental performance, particularly when government was divided. Based on the past twenty years of declining trust in government, (Hetherington, 1998; NES, 1999) such an argument is increasingly hard to justify.

One explanation for the American preference for divided government is the fear of concentrated political power. Building on a long tradition of support for the aphorism "the government that governs least, governs best" adherents of this belief do not dismay the apparent inability of government to accomplish significant tasks. Behind this belief is quite likely a preference for "passive" rather than "active" government. This has clearly been at the heart of Republican Party philosophy since the Reagan years, but its roots have their origins with the framers and the onset of its clearest recent manifestations beginning during the Truman administration. (Jacobson, 1990) Yet in a search for passivity, the result often turns out to be simply inefficiency.
 

		President	House	Senate	+/-
1946	D	Truman	Rep	Rep	-
1948	D	Truman	Dem	Rep	-
1950	D	Truman	Rep	Rep	-
1952	R	Eisenhower	Rep	Rep	+
1954	R	Eisenhower	Dem	Dem	-
1956	R	Eisenhower	Dem	Dem	-
1958	R	Eisenhower	Dem	Dem	-
1960	D	Kennedy	Dem	Dem	+
1962	D	Kennedy	Dem	Dem	+
1964	D	Johnson	Dem	Dem	+
1966	D	Johnson	Dem	Dem	+
1968	R	Nixon	Dem	Dem	-
1970	R	Nixon	Dem	Dem	-
1972	R	Nixon	Dem	Dem	-
1974	R	Ford	Dem	Dem	-
1976	D	Carter	Dem	Dem	+
1978	D	Carter	Dem	Dem	+
1980	R	Reagan	Dem	Rep	-
1982	R	Reagan	Dem	Rep	-
1984	R	Reagan	Dem	Rep	-
1986	R	Reagan	Dem	Dem	-
1988	R	Bush	Dem	Dem	-
1990	R	Bush	Dem	Dem	-
1992	D	Clinton	Dem	Dem	+
1994	D	Clinton	Rep	Rep	-
1996	D	Clinton	Rep	Rep	-
1998	D	Clinton	Rep	Rep	-

 

While the fear of concentrated government power is understandable for historical and ideological reasons, many of the same people who praise divided government also express concern regarding government efficiency. (Thurber, 1991) Yet divided government quite likely contributes to the very inefficiencies that voters rightfully lament. Under divided government, when all is well, each of the two parties claims responsibility for the outcome; when economic or political policies turn sour, however, each party blames the other. This condition leads to a fundamental lack of political accountability and the self-fulfilling prophesy that government is inherently inefficient.

Rather than being an accidental occurrence, divided government is much more likely to result due to the American constitutional design. For it is constitutional provisions that are at the heart of divided government; 2 year terms for Congress, 4 year terms for the Presidency, and 6 year terms for the Senate invariably produce divided government.

Were it only for these "accidental" outcomes of divided government, political accountability might be less deleterious. Exacerbating the problem, however, is the decline of parties as institutions. This has caused individuals to have weaker partisan attachments – despite the increased partisan rhetoric of many elected officials during the 1980s and 1990s – and has thereby intensified the fragmentation of government. (Franklin and Hirczy de Mino, 1998) Clearly, divided government is more problematic when partisan conflict between the two parties is greater as the sharper ideological conflict and the increased party line congressional voting of the 1990s would suggest. (Aldrich, 1995) Under these circumstances, divided government seems to be more problematic, since two highly partisan parties within the American political system seems potentially dangerous. Persistent divided government over time will likely produce a fundamental change in the relationship between Presidents and the Congress. Presidents are unable to bargain effectively with a hostile congress – witness the 1995 government shutdown – leading the former to make appeals over the heads of Congress directly and, hence undermine the legitimacy of the legislative branch. (Kernell, 1997) This argument parallels the one made in recent comparative scholarship (Linz, 1993) regarding the serious problem of dual legitimacy in presidential systems.

For those who lament the irresponsibility – or perhaps irrelevance – of the two major parties, an institutional design that would force responsibility should be praised. Quite simply, those who praise divided government because it "limits the damage" can not simultaneously complain about the resulting governmental inefficiency and political cynicism that ultimately follows when accountability is permanently clouded.
 
 

2b. Political Accountability and the Fusion of Government

A number of scholars have addressed the deficiencies of divided government, but have recommended something less radical than the adoption of a parliamentary system. This argument suggests that the problem is that the electoral cycle, with its "off year" elections, intensifies the likelihood of divided government in non-presidential election years. Such advocates propose as a solution the alteration of the electoral cycle so that all congressional elections are on four year terms, concurrent with presidential terms. Lloyd Cutler advocated such a proposal that would have all federal offices – Presidential and Congressional – on the same 4 year cycle as the president, thereby creating a clear majority. (Cutler, 1989) While this is an an interesting argument and an innovative proposal within the context of the American constitutional structure, it ultimately contains a fatal flaw. Because there is no guarantee that this proposal would alleviate the residual tension between competing branches of government, it merely sidesteps the accountability factor. For example, the last two times that Democrats had "unified government" (1977-1981 and 1993-1995) they failed to act in a cohesive or particularly responsible manner. This is less a judgment on the effectiveness of responsible government than it is on tensions built into the separation of powers in the U.S. that strongly discourages party unity across the executive and legislative branches of government.

Cutler’s proposal is also flawed on another dimension as well. In the search for political accountability, this system could produce the opposite effect from divided government, namely exaggerated majorities common to parliamentary regimes with majoritarian electoral systems such as the UK. The most familiar examples of this phenomenon were the "stop-go" policies of post-World War II British governments, as each succeeding government tried to overturn the previous election. (Ashford, 1981) While creating governing majorities is important for political accountability, we argue below that the Cutler proposal’s failure to change the electoral system to proportional representation might create a different set of problems.

Under a fusion of power system, in which the current presidency would be redefined, the resulting parliamentary system would make the head of the legislative branch the executive, thus eliminating the current separation of powers. Under a parliamentary system the elected majority or coalition government would be much more clearly accountable for success or failure. The prime minister would remain in office as long as majority support of the government party or a coalition of parties could be maintained. Yet if a government should lose its majority between scheduled elections due to defection of its party members or coalition partners, the head of state then would ask the opposition to form a new government and, failing that, call for new elections. This avoids the constitutional crises that the clamor for impeachment seem to engender in the American system.

But what if coalition members try to spread the blame for poor performance to their partners? In theory, the greater the flexibility available to in shifting from one governing coalition to another (with a different composition), the greater potential for this kind of musical cabinet chairs. The potential for such an outcome is far less than in the American system, however. A century of experience in other parliamentary regimes shows that members of such a party capriciously playing games with governing are usually brought to heel at the subsequent election.

In other words, the primary advantage to such a parliamentary system is that it hightens the capacity for voters and citizens to evaluate government performance. Of course, many individuals might object to the resulting concentration of power. However, if voters are to judge the accomplishments of elected officials, the latter need time to succeed or fail, and then the voters can make a judgment on their tenure. The most likely outcome would be a governing party or coalition of parties that would have to stay together to accomplish anything, thereby increasing party salience. Phrased differently, such an arrangement would likely lead to an increase in responsible government.

Many Americans might react unfavorably at the mention of the word coalition due to its supposed instability. Here we need to make the distinction between transparent and opaque coalitions. Coalition government in parliamentary systems have the reputation of increased instability. That, or course, depends on on the substance of the coalition agreement and the willingness of parties to produce a stable majority. But in most parliamentary systems, these party coalitions are formed transparently before an election so the voters can evaluate and then pass judgment on the possible coalition prior to election day. It’s not as if there are no coalitions in the US Congress. There they take the opaque form of ad-hoc groups of individual members of Congress on an issue-by-issue basis. The high information costs to voters in understanding the substance of such layered bargains hardly is an example of political transparency.
 

Table 3 Comparative Coalitions
 

American	Parliamentary
Opaque	Transparent
Issue-by-Issue	Programmatic
Back Room	Open Discussion
Unaccountable	Election Ratifies
Unstable	Generally Stable

 

One recently published work goes beyond the Cutler position and actually suggests adopting a British-style parliamentary system. (Manuel and Cammisa, 1999) While this work addresses the divided government issue clearly and comprehensively, it is less explicit about the head of state/head of government issue, and does not address the campaign finance or voter turnout issues at all. In addition, it makes continued reference to the British system with scarcely a mention of other parliamentary systems.

Finally, for those concerned that the "fusion" of the executive and legislative branches – on the British majoritarian model – would upset the concept of checks and balances, a multi-party consensus parliamentary system produces them slightly differently. (Lijphart, 1984) The distinction between differtent types of parliamentary regimes (majoritarian and consesus) as expounded by Lijphart is important here. The former concentrates power and makes "checking" difficult, while the latter institutionalizes the process in a different, and more accountable form. A multi-party parliamentary system would also satisfy a major concern or political representation, namely the provision of minority representation and protection. Such a system would be an improvement over one of the major problems of majoritarian systems (excessive concentration of power). Moreover, such a consensus parliamentary system would also address many of the concerns of the framers such as the "tyranny of the majority" problem. The checking and balancing in a parliamentary system is performed by the voters in the ballot box. The multiparty system itself is a form of checking and balancing, since no one party is likely to dominate. Consensus systems thus represent a compromise between the current U.S. system and the sharp concentration of British Westminster systems. Americans who simultaneously favor checks and balances but decry inefficient government need to clarify what they actually want their government to do.

3a. Permanent and Expensive Campaigns

The cost to run for political office in the United States dwarfs that spent in any other advanced industrialized democracy. (Alexander and Shiratori, 1994; Burnell and Ware, 1998) The twin problems are time and money. More specifically a never-ending campaign "season" and the structure of political advertising that depends so heavily on TV money. (Gans, 1993) Listening to the debates about "reforming" the American campaign finance system are bizarre to students of other democratic electoral systems. Quite literally no other country does it this way. More than $2 billion was raised and spent (Mann, 1997) by parties, candidates and interest groups in the 1996 campaign. As the 2000 presidential campaign progressed, the influx of money continued apace.

With respect to permanent campaigns, the two year congressional cycle forces members of the House of Representatives to literally never stop running for office. The amount of money required to run for a Congressional seat has quadrupled in the past decade. Presidential campaigns are an several orders of magnitude beyond the House of Representatives or the Senate. They are simply off the scale both in terms of cost and duration. By themselves they are at least two years long, frequently longer. Unless a presidential candidate is independently wealthy or willing and able to raise $30 - $50 million it is simply impossible to run for this office.

How did the United States get into this difficulty? Many of the problems stem from the post-Watergate "reforms" that tried to limit the amount of spending on campaigns which then produced a backlash in the form of a 1976 Supreme Court decision (Buckley vs Valeo) that undermined this reform attempt. In essence, Buckley vs Valeo held that "paid speech" (i.e. campaign spending) has an equivalent legal status as "free speech". (Grant, 1998) Consequently, since then all "reform" efforts have been tepid measures that have not been able to get at the root of problem. As long as "paid speech" retains its protected status, any changes are dead in the water. Clearly paid "speech" should not have an effect equivalent to free speech in the context of campaign finance.

At its essence this issue is a debate between "citizens" and "consumers". What Buckley vs Valeo has done is to equate the citizenship function (campaigning, voting, civic education) with a market-based consumer function (buying and selling consumer goods as commodities). (Brubaker, 1998) Unlike the United States, most other OECD democracies consider citizenship a public good and provide funding for parties, candidates and the electoral process as a matter of course. The US – more specifically, the Buckley vs Valeo decision – conflates the concepts of citizen and consumer, the logical extension of which is there are weak limits on campaign funding and no limits on the use of a candidate’s own money. We are all equal citizens, yet we are not all equal consumers. Bringing consumer metaphors into the electoral process debases the very concept of citizenship and guarantees that the American political system produces the best politicians money can buy.
 
 

3b. Free Television Time and the Return of Political Party Dues

Any broadcaster wishing to transmit to the public is required to obtain a broadcast license. Why? The airways have the legal status of public property and to have access to such property, the government licenses these networks, cable channels, and stations to serve the public interest. In return for such licenses, broadcasters are able to sell airtime to sponsors of various programs. Unfortunately for campaign costs, candidates for public office fall into the same category as consumer goods in the eyes of the broadcasters. (Weinberg, 1993) What has always seemed odd to observers of other democratic states is that there is no Quid Pro Quo requiring the provision of free public airtime for candidates when running for election.

For any serious reform to take place, there needs to be a concession from from all broadcasters to provide such time as a cost of using the public airways. One golden opportunity about to be missed is the changed from analog to digital transmissions that all broadcasters are undergoing. With all the discussion of campaign finance reform, there has been almost no serious discussion of making the price of admission for these lucrative digital channels completely free TV for all representative candidates and parties. Ideally, this would not be for "commercials" but for one and two hour long debates and open fora where candidates and parties could directly question each other. Since the largest share of campaign money is TV money, this reform would solve the problem at its source. And by restricting the time window when these free debates would take place to the last two months before a general election, it would address the time dimension as well. Such practices are standard procedure in all developed parliamentary systems. Very simply, as long as "reform" efforts try to regulate the supply of campaign finance, it will fail. The task that needs to be addressed is the regulation of demand.

But what of the costs of running for office not associated with paid TV commercials? Here too the United States could borrow a page from parliamentary systems by changing the political party contribution structure from individual voluntary contributions (almost always from the upper middle class and the wealthy) to a dues structure common to political parties in most other developed democracies. (Duverger, 1972) Such an elimination of private contribution funding for candidates in favor of a party dues structure would perform the additional salutary task of rebuilding parties as functioning institutions. (Allen, 1999b) Rather than continuing in their current status as empty shells for independently wealthy candidates, they could become the kind of dynamic membership organizations they were at the turn of the century when American political parties did have a dues structure.

4a. Low Voter Turnout?

There are many conventional wisdom explanations for low voter turnout. By comparison the United States voter turnout is under 50% in presidential elections, 35% in congressional elections since the 1970s, and as low as 20% for municipal elections. By comparison, the leading OECD countries have voter turnout ranging from 67% to 90% of their adult population. (Powell, 1986) Among the most commonly raised explanations for the US deficiency are: registration requirements, the role of television, voter discouragement and voter contentment (although the latter two can not both simultaneously be correct). (Wolfinger and Rosenstone, 1980) None of these explanations are particularly convincing, and none of them offer any concrete suggestions as to how the abysmally low voter turnout might be overcome.
 

Table 4

Comparative Voting Turnout - 1945-1998 Average
 

Country	Turnout %
Italy	92.5
Germany	80.6
Great Britain	74.9
France	67.3
US (President)	48.3
US (Congress)	43.0
US (Municipal)	21.5

Source: Voter Turnout: A Global Survey (Stockholm: International IDEA, 1999)
 
 

One seldom-examined reason in mainstream American electoral analysis is the two party system and the electoral method that produces it: the single member district, first past the post, or winner take all system. With the rise of such new organizations as the Reform Party, the hegemony of the Democrats and Republicans is again facing increased scrutiny. But the problem of a third (or fourth) party gaining a sufficient number of votes to actually win seats and challenge the two party system is formidable. The electoral arithmetic would require any third party to win some 25% of the vote on a nationwide basis before it would actually gain more than a token number of seats. And failing to actually win seats produces a "wasted vote" syndrome among party supporters which is devastating for such a party. Two examples at the presidential level are most illustrative. In 1980, John Anderson ran as an independent candidate and gained near 20% support in many pre-election polls. Yet as the November general election loomed, the support plummeted like a stone to 6 percent. (Gold, 1995) In 1992, Ross Perot as the candidate of the Reform Party received some 20% of the vote, but of course got zero electoral votes. (Lacy and Burden, 1999) Most voters who become disillusioned with the electoral process refer to the "lesser of two evils" choices they face. In such a circumstance, declining voter turnout is not surprising.

The American system of primary elections only exacerbate the problem. Although few Americans actually realize it, primary elections are a device for a political party to select its candidates to run in the general election. Prior to the primary system, which only took on serious significance beginning in the 1960s, political parties relied on caucuses and conventions of party members to select the best candidates for office. "Reforms" in the 1960s and 1970s – both wise and unwise – have reduced the role of parties to flags of convenience, open to the highest bidder. Not only is party "membership" an ephemeral concept, recent legal rulings that prevent "closed primaries" (available, that is, only to party members) further erode the definition of a political party. This was followed by the "open primary" that literally allows any registered voter to cast a ballot in any party’s primary. What is the meaning of "political party" under these circumstances?

Moreover, the proliferation of multiple primaries and general elections – often with runoffs when no single candidate receives a majority of votes – has produced a sense of electoral fatigue. Americans are called to the ballot box far more often than voters in any other OECD country. In the process, the parties’ primary function of sorting out ideas and selecting candidates for voters – which they did for the first half of the century – has completely disappeared. With the destruction of parties as serious political institutions, there is one less signaling organization to provide cues and guidance for voters. Decreased voter turnout is not a surprising outcome. The US is a diverse country with many regional, religious, racial, and class divisions. So why should we expect that two "catch all" parties will do a particularly good job in appealing to the interests of diverse constituencies? The solution to lower voter turnout is a greater number of choices for voters and a different electoral system.

4b. Proportional Representation

Under electoral systems using proportional representation, The percentage of a party’s vote is equivalent to the percentage of seats allocated to the party in parliament. Comparative analysis shows that those countries with proportional representation – and the multiple parties that PR systems produce – invariably have higher voter turnout. (Grofman and Lijphart, 1986) In other works, PR voting systems provide a wider variety of political choices and a wider variety of political representation.

The "lesser of two evils" choice characteristic of majoritarian electoral systems like the United States takes many potential voters out of the electoral process. It is particularly deleterious for lower income voters since expensively financed wealthy candidates rarely appeal to the interests of these individuals. (Skocpol and Fiorina, 1999) As a concrete example of the uncompetitiveness of many two party congressional races, the 1998 election was illustrative. In two party "races" in districts where incumbents have a huge financing advantage, incumbents win 95 percent of the seats. In other words, incumbents are almost guaranteed victors before the votes are cast. (Cox and Katz, 1996) In other words, in 1998 there were fewer than 30 seats that were actually "in play" on election day. (Shafroth, 1998) This is a mandate for many voters to simply stay home.

Eliminating majoritarian single member districts (SMDs) in favor of PR voting would have several immediate effects. First, it would increase the range of choices for voters, since parties would have to develop ideological and programmatic distinctions to make themselves attractive to voters. As examples in other countries have shown, it would lead to formation of several new parties representing long underserved interests.

Engaging in some speculation, it is likely the the change to a PR system in the United States would see the instant creation of at least 8 parties. There would be rump versions of Democrats and Republicans, of course. But it’s expected that there would be some serious growth of Reform, Libertarian, Labor and Green parties. The likely formation of a one or more minority-based parties could be envisioned and last but not least, a Christian party.

Such a change would force rebuilding of parties as institutions, since candidates would have to run as members of parties and not as independent entrepreneurs. The so-called Progressive "reforms" at the turn of the century and the 1960s introduction of primaries – plus TV advertising – plus the widespread use of referenda have all had powerful effects in undermining parties as coherent political organizations. (Dwyre, 1994) In trying to force market-based individual "consumer choice" in the form of high-priced candidates on the polity, the collective institutions that are political parties have been hollowed out and destroyed. And the SMD electoral system never gives a chance for renewal of third parties since the electoral system biases 3rd parties until they can get over 25% of the vote, a huge hurdle.
 
 

There are, of course, a wide range of standard objections to PR voting systems by those favoring retention of majoritarian SMD systems. The most common are: coalitional instability; "too many parties"; and "who’s my representative?" Let’s take up each one in turn.

The first of these, coalitional instability, was addressed briefly above, but it needs to be restated here. The US has coalitional instability in the Congress right now, but it takes a different form than the so-called unstable parliamentary coalitions. American unstable coalitions are issue-by- issue ones, usually formed in the House cloakroom with the "assistance" of lobbyists. Few average voters know with certainty how "their" member of Congress will vote on a given issue. With ideologically coherent parties, they would.

An American parliament with several parties would discipline these parties very effectively. Clearly there would have to be a coalition government since it is unlikely that any one party would capture 50% of the seats. The practice in almost all other coalition governments in parliamentary systems is that voters prefer a predictable set of political outcomes which force parties to both define their programs clearly and transparently, once entering into a coalition, to do everything possible to keep the coalition together.

The second standard objection to PR is the "too many parties" issue. PR voting has been practiced in parliaments for almost 100 years in many different democratic regimes. There is a long history of practices that work well and practices that don’t. (Norris, 1997) When this argument is used two countries are invariably chosen as bad examples of PR, namely Israel and Italy. There is an easy solution to this problem of an unwieldy number of parties, namely to to create an electoral threshold. Under an electoral threshold system, a party must receive a certain minimal percentage to gain seats in the parliament. The significant question is what should this minimal threshold be? The Swedes have a 4% threshold and have 6 parties in their parliament, the Germans have a 5% threshold and have 5 parties represented in the Bundestag. (Lijphart, 1984) When the New Zealand parliament recently changed its electoral system from majoritarian to PR, there was considerable debate about the threshold issue. (Denemark, 1996) A significant body of literature suggests that the highest acceptable threshold is approximately 7% or less. (Rae, 1967; Lijphart, 1994) Higher than that, for example in Greece (10%) or the French second ballot runoff (12.5%) essentially defeats the purpose of PR voting in the first place.

The third standard objection to PR voting is the "who’s my representative?" one. In a society so attuned to individualism, an obvious objection that many Americans would have to a party list system as opposed to an individual member of congress would be the lack of a district representative. This argument presumes, of course, that all Americans have a member of Congress that represents their views. However, a liberal democrat living in former House Speaker Newt Gingrich’s district northwest of Atlanta might genuinely wonder in what way he would represent that liberal’s interests? By the same token, conservative Republicans living in Vermont have the independent socialist, Bernard Sanders as the state’s lone member of Congress representing "their" interests.

Yet if Americans are still insistent on having "their own" individual representative, then the US could borrow the German hybrid system of "personalized" proportional representation. This system requires citizens to cast two votes on each ballot: the first for an individual candidate; and the second for a list of national/regional candidates grouped by party affiliation. (Allen, 1999a) This system has the effect of personalizing list voting because voters have their own representative but also can choose among several parties. Allocation of seats by party in the Bundestag, however, functions more like proportional representation. Specifically, the percentage of total seats won per party corresponds strongly with the party’s percentage of the popular vote. For example, if a party’s candidate wins a seat as an individual member, his or her party loses one seat from those won via list voting. In practice, most of the district seats are won by the two large parties, the Social Democrats and the Christian Democrats, because the district vote is winner-take-all. The smaller parties’ representatives are almost always elected through the party lists. Thus, the list system creates stronger, more coherent parties.

If there were PR voting in an American parliament, what would the threshold be? The dual purpose of a threshold is to balance representation and political accountability. We are suggesting here that the US threshold should be at least 6% and possibly as high as 7%. The goal is to devise a figure that represents all significant interests yet does not produce instability. At the first election, if the New Zealand experience in the mid-1990s is any guide, there would likely be an initial proliferation of multiple ideological and regional parties. (Denemark, 1996) At outset, the American predilection for "start-ups" would likely see anywhere from 15-20 parties trying to win and gain seats. Setting the threshold higher than 5% would insure that the only parties represented would represent significant constituencies and not produce so many parties that might produce problems of governability.

The "shake out" of parties would likely produce some strategic "mergers" – as in the corporate world – of weak parties which, as single parties, might not attain the 6%-7% threshold. For example, a separate Latino party and an African-American party might insure always attaining an 7% threshold by forming a so-called "rainbow" party. Similarly the Reform Party and the Libertarian Party might find it safer electorally to merge into one free market party. What are the primary arguments in favor of PR? There are four that merit discussion: simplicity, diversity, government representativeness, and the elimination of the decennial redistricting charade.

With respect to simplicity, several scholars, most notably Lani Guinier have argued against the 2 party system. (Guinier, 1994) But her initial suggestion was a complicated cumulative voting system that got her branded a as "quota queen" and the slighting by her former friend, President Clinton. As proposed, it was similar to "weighted" voting in municipalities such as Cambridge, MA or multi-member district system once practiced in various state legislatures and suggested by US Rep Cynthia McKinney (GA) in the mid 1990s. (Richie et al., 1996) The flaw with the Guinier/McKinney proposals is that, in attempting to retain concept of individual candidate representation, such plans shoot themselves in the foot by forgetting to keep it simple.

The second advantage to PR is diversity, or literally changing the faces of congress. The experience of PR voting in other countries is that it changes the makeup of the legislature by increasing both gender and racial diversity. Obviously, parties representing minority interests who find it difficult to win representation in 2 person races, will more easily be able to win seats under PR. Very simply, because candidates will not have to run as individuals – or raise millions of dollars – the parties will be more easily able to include individuals on the party’s list of candidates who more accurately represent the demographics of average Americans. For example, many political parties in other countries will alternate male and female candidates on the party lists, resulting in a gender balanced group representing that political party in the parliament. In fact, in a competitive party environment of six parties or more, each party needs to be very attentive to representing the interests of its particularly constituency. Failing to do so could see the party drop below the electoral support threshold necessary for representation in parliament. What a multi-party list system would do would provide a greater range of interests being represented and broaden the concept of "representation" to go beyond narrow geography to include representation of such things as ideas and positions on policy issues that would be understandable to voters. Moreover, as for geographic representation on a list system, it would be in the self interest of the parties to insure that there was not only gender balance – if this is what the party wanted – on their list, but also other forms of balance including geography, ideology, and ethnicity among others.

The third advantage is one of government representativeness. Not only is a consensus-based parliamentary system based on proportional representation more representative of the voting public, it also produces more representative governments. (Crepaz, 1996b) This study finds that consensus-based, PR systems also produce a high degree of "popular cabinet support," namely the percentage of voters supporting the majority party or coalition.

The fourth advantage to a PR system in the US is that it would eliminate the redistricting circus that transpires every decade. Each decade as the census takes place there is the excruciating task of micro-managing the drawing of congressional districts. (Peterson, 1995) In essence, districts would be eliminated. Candidate lists would be organized statewide, in highly populated states, or regionally in the case of smaller states like New England. To insure geographical representation, all parties would find it in their own self-interest that the candidate list included geographical diversity starting at the top of the list.

Getting from Here to There: From Academic Debates to Constitutional Reform?

Of the four proposals, two of them could be accomplished by statute: campaign reform and the voting system. The other two would require constitutional change: head of state/government and divided government. While this paper suggests that to function optimally, all four would need to take place at the same time, this is, of course, extremely unlikely. Given the above caveats, it would be easiest to effect campaign reform (the Supreme Court willing) and to alter the voting system. The latter would require a crisis in which neither of the two major parties could command a majority in the House of Representatives. The only way this would happen is if independent candidates or minor parties win a small number of seats with the Democrats and Republicans dividing the remainder. The "price" of getting a majority of House members to vote for a speaker would be a concession of the larger of the two parties to change the electoral system to produce a "coalition" government in the Congress. Is it remote? Yes, but if the Reform or Green parties, for example, could win US House seats in an otherwise close election, such an event could transpire. A more likely event would be the same experience transpiring at the state level. The US has a long history of various national reforms first being developed at the state level (Skocpol, 1995) due to possibilities of political experimentation within the states.

Are there perils in keeping presidential system with PR. Yes there are, and they could be substantial. Mixing the 2 systems by introducing a multi- party parliament but retaining an American president with the powers that the office currently has could be fraught with difficulty. (Linz, 1993) There are two developed countries that have tried this combination, one producing a suboptimal outcome, namely France with its five republics and fourteen constitutions; and another a disastrous one, namely Chile and a violent military overthrow of a democratically elected government in 1973. The problem here is that the presidency is an individual – and non-proportional – office. Unless a president could be assured of a majority in parliament, unlikely with a multi-party parliament, then the divided government problem simply rears its ugly head again.

Clearly the largest obstacles to such a radical change in the American constitutional system are cultural and structural. Culturally, the ethos of American individualism would have difficulty giving up features such as a single all-powerful executive and one’s own individual member of congress, no matter how powerful the arguments raised in support of alternatives. Very simply, ideology and cultural practice change very slowly. A more serious obstacle would be the existing interests privileged by current system. All would fight tenaciously to oppose this suggested change. Clearly at the forefront would be the Democratic and Republican parties, the media industry, and organized special interests who currently back both parties to "cover their bets".

Finally, specialists in American politics may dismiss this argument as the farfetched "poaching" of a comparativist on a terrain that only Americanists can write about with knowledge and expertise. However, the durability of all four of the above-mentioned problems, stretching back anywhere from 25 to 50 years, suggests that Americanists have no monopoly of wisdom on overcoming these pathologies. More seriously, what this comparativist perceives is a fundamental failure of imagination based largely on the "N of 1" problem that all comparativists struggle to avoid. If a single observed phenomenon – in this case, the American political system, is not examined comparatively, one never knows whether prevailing practice is optimal or suboptimal. In essence, those who do not look at these issues comparatively suffer a failure of imagination because they are unable to examine the full range of electoral and constitutional options.
 
 

Aldrich, J.H. (1995) Why Parties? The Origin and Transformation of Political Parties in America. Chicago: The University of Chicago Press.

Alexander, H.E. and Shiratori R., eds (1994) Comparative Political Finance Among the Democracies. Boulder: Westview Press.

Allen, C.S. (1999a) ‘Proportional Representation’, in Krieger, J. (ed.), Oxford Companion to Politics of the World. Oxford: Oxford University Press, (in press).

Allen, C.S., ed. (1999b) Transformation of the German Political Party System: Institutional Crisis or Democratic Renewal?. New York: Berghahn.

Ashford, D.E. (1981) Policy and Politics in Britain: The Limits of Consensus. Oxford: Basil Blackwell.

Barker, A. (1994) ‘The Upturned Stone: Political Scandals and their Investigation Processes in 20 Democracies’, Crime Law and Social Change, 24 1, 337-373.

Berger, R. (1973) Impeachment: The Constitutional Problems. Cambridge: Harvard University Press.

Birchfield, V. and Crepaz M.M. (1998) ‘The Impact of Constitutional Structures and Collective and Competitive Veto Points on Income Inequality in Industrialized Democracies’, European Journal of Political Research, 34 2, 175-200.

Brubaker, S.C. (1998) ‘The Limits of Campaign Spending Limits’, PublicInterest, Fall, 33-54.

Burnell, P. and Ware A., eds (1998) Funding Democratization. New York: St. Martin’s.

Cox, G.W. and Katz J.N. (1996) ‘Why Did the Incumbency Advantage in the US House Elections Grow?’, American Journal Of Political Science, 40 2 (May), 478-497.

Crepaz, M.M. (1996a) ‘Consensus versus Majoritarian: Democracy, Political Institutions and their Impact in Macroeconomic Performance and Industrial Disputes’, Comparative Political Studies, 29, 4-26.

Crepaz, M.M. (1996b) ‘Constitutional Structures and Regime Performance in 18 Industrialized Countries - A Test of Olson’s Hypothesis’, EuropeanJournal of Political Research, 29, 87-104.

Cutler, L. (1989) ‘Some Reflections About Divided Government’, PresidentialStudies Quarterly, 17, 485-492.

Denemark, D. (1996) ‘Thinking Ahead to Mixed-Member Proportional Representation: Party Strategies and Electoral Campaigning under New Zealand’s New Electoral Law’, Party Politics, 2 3 (July), 409-420.

Duverger, M. (1972) Political Parties. 3rd Ed. New York: Wiley.

Dwyre, D. (1994) ‘Disorganized Politics and the Have-nots: Politics and Taxes in New York and California’, Polity, 27 (Fall), 25-47.

Fiorina, M. (1992) Divided Government. New York: Macmillan.

Franklin, M.N. and Hirczy de Mino W.P. (1998) ‘Separated Powers, Divided Government, and Turnout in U.S. Presidential Elections’, American Journalof Political Science, 42 1 (January), 316-326.

Gans, C. (1993) ‘Television: Political Participation’s Enemy #1’, Spectrum:the Journal of State Government, 66 2, 26-31.

Gold, H.J. (1995) ‘Third Party Voting in Presidential Elections: A Study of Perot, Anderson and Wallace’, Political Research Quarterly, 48 (December), 751-773.

Grant, A. (1998) ‘The Politics of American Campaign Finance’, ParliamentaryAffairs, 51 2 (April), 223-240.

Grofman, B. and Lijphart A. (1986) Electoral Laws and Their Consequences. New York: Agathon Press.

Guinier, L. (1994) The Tyranny of the Majority: Fundamental Fairness in Representative Democracy . New York: The Free Press.

Hetherington, M.J. (1998) ‘The Political Relevance of Political Trust’, American Political Science Review, 92 4 (December), 791-808.

Jacobson, G.C. (1990) The Electoral Origins of Divided Government: Competition in U.S. House Elections, 1946-1988. Boulder, CO: Westview.

Jacoby, W. (1999) Transferring Institutions in Germany: Alchemy, Fetishism, and Politics. Ithaca: Cornell University Press.

Kernell, S. (1997) Going Public: New Strategies of Presidential Leadership. Washington, DC: CQ Press.

Lacy, D. and Burden B.C. (1999) ‘The Vote-stealing and Turnout Effects of Ross Perot in the 1992 U.S. Presidential Election’, American Journal of Political Science, 43 1 (January), 233-255.

Light, P.C. (1983) ‘The Institutional Vice Presidency’, Presidential Studies Quarterly, 13 (Spring), 198-211.

Lijphart, A. (1984) Democracies: Patterns of Majoritarian and ConsensusGovernment in Twenty-One Countries. New Haven: Yale University Press.

Lijphart, A. (1994) ‘Democracies: Forms, Performance, and Constitutional Engineering’, European Journal of Political Research, 25, 1-17.

Linz, J. (1993) ‘The Perils of Presidentialism’, in Diamond, L. and Plattner M. (ed.), The Global Resurgence of Democracy. Baltimore: Johns Hopkins University Press, 109-126.

Lipset, S.M. (1996) American Exceptionalism: A Double-Edged Sword. New York: Norton.

Mann, T., ed. (1997) Campaign Finance Reform: A Sourcebook. Washington, DC: Brookings Institute.

Manuel, P.C. and Cammisa A.M. (1999) Checks and Balances? How a ParliamentarySystem Could Change American Politics. Boulder: Westview Press.

Mayhew, D. (1991) Divided We Govern: Party Control, Lawmaking, andInvestigations, 1946-1990. New Haven: Yale University Press.

NES (1999) The NES Guide to Public Opinion and Behavior. Ann Arbor: University of Michigan.

Norris, P. (1997) ‘Choosing Electoral Systems: Proportional, Majoritarian and Mixed Systems’, International Political Science Review, 18 3 (July), 297-312.

Peterson, K.J. (1999) ‘Graceless Exit’, World Press Review, 46 3 (March), 39ff..

Peterson, P. (1995) Classifying by Race. Princeton: Princeton University Press.

Powell, G.B. (1986) ‘American Voter Turnout in Comparative Perspective’, American Political Science Review, 80 (March), 17-43.

Rae, D.W. (1967) The Political Consequences of Electoral Laws. New Haven: Yale University Press.

Rhenquist, W.H. (1992) Grand Inquests: The Historic Impeachments of Justice Samuel Chase and President Andrew Johnson. New York: Morrow.

Richie, R., Anderson J.B. and McKinney C. (1996) ‘Proportional Representation’, National Civic Review, 85 1 (Winter), 45-51.

Rundquist, P.S. (1991) The House of Representatives and the House of Commons: A Brief Comparison of American and British Parliamentary Practice. Washington, DC: Congressional Research Service, Library of Congress.

Shafroth, F. (1998) ‘Voters leave Congress much as it was’, Nation’s Cities Weekly, 21 45 (November 9), 1-4.

Sindler, A.P. (1976) ‘Making the Vice Presidency an Appointive Office?’, Policy Analysis, 2 (Spring), 351-355.

Skocpol, T. (1995) Social Policy in the United States: Future Possibilities in Historical Perspective. Princeton: Princeton University Press.

Skocpol, T. and Fiorina M.P., eds (1999) Civic Engagement in AmericanDemocracy. Washington, DC: Brookings Institution.

Thurber, J.A. (1991) ‘Representation, Accountability, and Efficiency in Divided Party Control of Government’, PS, 24 (December), 653-657.

Timberg, R. (1995) The Nightengale’s Song. New York: Simon and Schuster.

Uehling, M.D. (1991) ‘All-American Apathy’, American Demographics, 13 (November), 30-4.

Walsh, L.E. (1994) Iran-Contra: The Final Report. New York: Times Books.

Weaver, R.K. and Rockman B.A., eds (1993) Do Institutions Matter?: Government Capabilities in the United States and Abroad. Washington, DC: The Brookings Institution.

Weinberg, J. (1993) ‘Broadcasting and Speech’, California Law Review, 81 5, 1101-1206.

White, T.H. (1975) Breach of Faith: The Fall of Richard Nixon. New York: Athenaeum.

Wolfinger, R.E. and Rosenstone S.J. (1980) Who Votes?. New Haven: Yale University Press.