There was perhaps a time when much of the rest of the world viewed the United States of America not just as a militaristic crusader, but as a role model for advocates of liberal democracy everywhere.
These days, however, as former President Carter’s recent caustic comments about the likelihood of continued rigged voting in Florida this year have underscored, the American beacon of liberty is flickering in the wind.
Increasingly the US is viewed by the rest of the world -- and by some of its most astute internal critics -- as an arrogant, hypocritical, sclerotic plutocracy, whose own electoral institutions, as former President Carter correctly observes, are at risk of no longer meeting even minimal international standards for democratic elections.
Meanwhile, the US has the temerity to lecture other countries (as it did, for example, just this week with respect to Hong Kong) about the meaning of “democracy,” and to selectively intervene in the affairs of other countries in the name of democracy, whenever it suits perceived US interests.
As shown in Table A, this concern with democratizing the world may be a preoccupation of US policy elites, or it may just be pure rhetoric.
A recent poll by the Chicago Center on Foreign Relations shows that for the American public at large, the goal of bringing democracy to other nations ranks last on a list of fourteen major US foreign policy goals, with only 14 percent believing it to be “very important.” This made it less than one fifth as as important as securing energy supplies, protecting US jobs, or fighting terrorism. Even among policy elites, “democratization” ranked 12th on the list.
Once again, our aim here is to help Americans understand why their own electoral system has become so anachronistic and dysfunctional. We also hope to encourage them to take more interest in the rest of the world, be a little more modest, and understand that they too live in a “developing country” – one whose own particular version of “democracy” is in dire need of rejuvenation.
Campaign Duration/ Candidate Selection
Brazil. Like the UK and many other democracies, Brazil’s process for selecting Presidential candidates and the political campaigns that follow are relatively abbreviated, with most candidates selected by national parties caucuses, and campaign advertising and other electioneering activities limited to 60 days before the general election. (This is far from the shortest period. Indonesia just held a Presidential election, with 150 million registered voters, about 10 million more than in the US. It limits public campaigning to just 3 days in order to curb political violence, a long-standing Indonesian problem.)
The US. The US primary process for selecting Presidential candidates is a costly and tedious. It begins up to two-and-a-half years before the general election, wends its way through more than 37 state primaries and party caucuses by mid-March of the election year, and concludes with the main candidates already selected at least 6 months before their party conventions, and 8 months before election day.
The US primary process also gives extraordinary influence to a tiny fraction of voters who turn out for caucuses or primaries in “white bread,” states like Iowa (total primary turnout = just 9 percent of VAP) and New Hampshire (total primary turnout= 29 percent of VAP) that happen to have early primaries.
Several of these states also follow primary procedures that are only remotely democratic. In Iowa’s bizarre “house party” caucuses, for example, there is no secret ballot, so that one’s corporate bosses or fellow union members can easily observe whether one follows instructions.
The primary process also has a dramatic impact on campaign costs. With so many primaries bunched together, candidates are compelled to run national campaigns in multiple states. On the other hand, since the order of primaries has nothing to do with each primary’s weight in national totals, the country as a whole is compelled to sit through 9-sided “debates” between pseudo-candidates, over and over again.
All this probably reduces the supply of first-rate candidates who are willing to endure this grueling march. By the time the November election finally rolls around, most Americans are also probably sick and tired of the whole affair. On the other hand, campaign advisors, advertisers, and the news media like the current system -- it generates a prolonged “horse race” and recurrent employment.
Campaign Finance
While most leading democracies, like Germany, France, and Japan, have experienced serious campaign finance abuses, finance has much more leverage in systems like Brazil and the US, where Presidents are selected through expensive direct elections, not parliamentary polls. So far, despite numerous efforts, neither Brazil nor the US has been able to contain the excessive influence of campaign finance directly, but Brazil has made progress indirectly, by providing media access and some public financing.
In Brazil’s case, the campaign finance issue was highlighted by a recent series of scandals, including the 1992 impeachment and removal of President Fernando Collor, the 1994 “Budgetgate” scandal, the 1997 “Precartorios” scandal in Sao Paulo, the 2002 scandals involving illicit funds raised by Roseanna Sarney and Jose Serra, and the 2004 scandals involving senior Workers Party official Diniz and the “Blood Mafia.”
In the US, this year’s Presidential election will be by far the most expensive in history, nearly twice as expensive as the 2000 election, despite new legislation like the Bipartisan Campaign Finance Reform Act of 2002 (McCain-Feingold), which tried to limit the use of “soft money.”
“Supply-Side” Limits. Both the US and Brazil have repeatedly tried to legislate against the corrosive influence of money on elections by establishing such “supply-side” limits. In the US these efforts date back more than a century. Since the 1980s there have been a string of “reforms of reforms” in both countries to require disclosure of contributions by individual and corporate donors (Brazil), disclose receipts by candidates and parties (Brazil, US), limit the size of individual, union, and corporate contributions (Brazil, US), ban anonymous donations (Brazil, US) limit campaign expenditures (Brazil), and limit the amount that parties can raise (Brazil).
In the US, virtually all elections at all levels of government are privately funded, with no spending limits unless candidates opt for matched public funding. In federal races, contributions limits are now quite high -- $4,000 per election cycle. And these may be increased if opponents spend their own money. Corporations and unions are prohibited from making campaign contributions, but they have found ways to make huge indirect contributions, by way of Political Action Committees (PACs), state and national political parties, unregulated “in-kind” contributions (like Cisco’s $5 million of free networking services to each party convention) and most recently, virtually-unregulated “527” committees and 501-C4s.
So far, these “supply side” efforts to limit campaign contributions have proved unsuccessful in both countries. Donors and recipients are simply too creative, laundering contributions through “independent” fronts, providing in-kind contributions, spreading contributions across multiple family members, and so forth. Ultimately the reality is that in advanced capitalist societies that have very powerful private interest groups, highly unequal income distributions, sophisticated lawyers, and important government policies up for grabs, the most one can hope for from supply-side regulation is window-dressing – and that the insiders occasionally cancel each other out.
Demand-Side Limits. A much more effective approach is to limit the demand for private campaign funds with a combination of free media, direct controls on campaign spending, and public funding.
In the US, efforts to limit campaign expenditures directly have been throttled by the 1976 Supreme Court decision in Buckley vs. Valeo, which determined that “money is speech” – e.g., that any direct limits on campaign spending by candidates or parties violate the First Amendment.
US law does allow limits to be imposed on candidates who accept matched public funds, which have been available since 1974. But public funding has been limited, and is getting scarcer – the US matched public funding system is now on the brink of insolvency, partly because voluntary taxpayer contributions on tax returns have plummeted. (Among other reasons, the $3 check-off limits has not recently been increased.)
So while lesser candidates like Ralph Nader and Al Sharpton relied heavily on public funding for their campaigns, well-funded candidates like Bill Clinton in 1996, George W. Bush in 2000 and 2004, and John Kerry in 2004 have either refused public funding completely, or have limited their use of it to the final few months of the campaign. A few states – Maine, Minnesota, and Arizona, for example – have experimented with providing public funds for state candidates, with positive results, including a sharp reduction in the amount of time politicians devote to fund-raising. But these programs are not expanding, partly because of state budget crises.
As noted, no US-like constitutional limits on campaign spending controls exist in Brazil, or for that matter, in most other democracies, like the UK or Canada, for example. Brazil has adopted some spending limits for political parties and individual candidates, which seem to have been somewhat effective. Brazil has also provided some public funds for all registered political parties to defray campaign and administrative costs.
Free Media
Brazil. By far the most effective measure that Brazil has introduced to level the campaign finance playing field, however, has been the provision of free access to TV and radio for political candidates. TV advertising, in particular, is otherwise by far the most important ingredient in campaign costs. Furthermore, at least in principle, the airwaves are owned by the public. So it is not surprising that Brazil and more than 70 other democracies around the world have adopted such provisions, including most Western European countries, South Africa, India, Russia, and Israel. (Among First World countries that provide free political airtime: the UK, France, Germany, Spain, Italy, Sweden, and Japan.)
In Brazil’s case, ironically, the early adoption of this provision is partly due to the fact that the ownership of radio and TV broadcasting networks in Brazil are even more concentrated than they are in the US, dominated by the Marinho family’s Globo empire and 2-3 other private owners. On the other hand, a majority of Brazilians are poor and semi-literate, and depend heavily on TV and radio for all their news.
Under Brazil’s free media law, the only forms of paid political advertising permitted are newspaper advertising, direct mail, and outdoor events. Sixty days before an election, 1.5 hours of programming per day - the “horário eleitoral gratuito” - are requisitioned from broadcasters and divided up among political parties in proportion to their seats in the Camara, with a minimum allocation for parties that have no seats. Newscasters and debate broadcasters are required to give equal time to all candidates during this period, punishable by fines. Brazil’s political parties are also entitled to free public transportation, the free use of schools for meetings, and special tax status.
Some broadcasters have complained that these provisions are very costly (to them), that they may reduce news coverage for controversial subjects, and that the viewing public will simply tune out. However, it appears that most Brazilians continue to tune in as usual, and that they actually welcome the concentrated dose of campaigning, as compared with prolonged agony of the US approach.
The US. The US is one of a minority of leading democracies that still provides no free radio or TV access for political candidates. (In addition to Brazil, others that do include the UK, France, Italy, Israel, Hungary, India, Germany, Spain, and Sweden.)
This is especially important, because in many ways, TV advertising is the key factor underlying the whole campaign finance issue. From 1970 to 2000, US campaign spending on TV ads increased more than 10-fold, faster than any other campaign cost element. This year, the political ad revenues of US broadcasters will exceed $1.4 billion. This makes political advertising second only to automotive advertising as a revenue generator for TV networks. This year, most of this revenue will flow to broadcasting conglomerates like ClearChannel, Cox, and Viacom that are especially well-positioned in the battleground states.
In theory, since 1971 the Federal Communications Commission – now chaired by Michael Powell, son of the Secretary of State Powell -- has required US broadcasters to sell airtime to political candidates at the “lowest unit rates” available, but in practice this requirement has not been very effective. The 2002 BCRA also tried to limit corporate and labor funding for broadcast advertising to the last 30 days before primaries, but it did nothing about so-called “Section 527” or 501-C4 advertising by issue-oriented groups that are theoretically independent of campaigns.
Among its many other impacts, this policy compels incumbent US politicians to spend a huge portion of their time – at least 28 percent, in one recent study -- raising campaign funds. It also invites incumbents to be unduly empathetic to the media conglomerates that now dominate the US cable, satellite, broadcasting, and publishing markets.
Its key beneficiaries include:
ClearChannel(1200 radio stations, 37 local TV stations, 103 mm listeners);
Cox Enterprises(43 newspapers in Florida, Ohio, Texas, North Carolina, Georgia, Colorado; 9 TV stations; 75 radio stations;
Disney (ABC, E!, A&E, History Channel, Lifetime, Tivo (partial), Miramax, ESPN, 10 local TV stations, 66 radio stations);
GE(NBC, PAX TV Network, MSNBC, CNBC, Universal Pictures, Telemundo, Bravo, Sci-Fi, Vivendi Entertainment, 28 local TV stations);
News Corp/Rupert Murdoch (Fox Network, 20th Century Fox, National Geographic, HarperCollins, The NY Post, Sunday Times (UK), The Times(UK), multiple other Australian and UK newspapers, Avon Books, TV Guide (part), The Weekly Standard, BSkyB, 34 local TV stations);
Tribune Company(27 TV stations, Newsday, Chicago Tribune, Baltimore Sun, LA Times, 11 other newspapers, 1 radio station, etc.)
Time Warner(CNN, HBO, Warner Bros., Court TV, TNT, New Line Cinema, AOL, Netscape, Time Inc. (People, Time, Fortune, Sports Illustrated, Bus 2.0, Life, Popular Science, etc.) Mapquest, Little,Brown);
Viacom(CBS, BET, MTV, Comedy Central, 13 other cable channels, Paramount, Simon & Schuster, Free Press, Scribners, Infinity Radio (185 radio stations), 39 TV stations, 5 other radio stations).
Not surprisingly, under the influence of this tidy little interest group, federal regulation has become more and more lax on a wide range of broadcast- and cable-related issues in the last twenty years, including acquisitions, “lowest unit rate” regulations, “equal time,” digital spectrum, and pricing guidelines for cable network franchise agreements.
Meanwhile, the detente established between incumbent politicians and this “Sun Valley” alliance has stymied proposals for free political media in the US – and helped to perpetuate the role of money in American politics.
Vote Fraud/ Absentee Ballot Abuse/ Missing Ballots
Brazil. Outright vote fraud and vote buying are long-standing problems in Brazil, as in many other developing countries. One recent survey indicated that up to 3 percent of voters – almost 2 million people - may have sold their votes in Brazil’s 2002 election. A new law was adopted in 1999 to increase penalties for buying votes, but evidently there is still much work to do.
On the other hand, one beneficial side-effect of Brazil’s particular approach to electronic voting – which only provides printed copies for a sub-sample of voters – is that most voters aren’t able to prove that they have voted as instructed.
Absentee ballots are also not a problem in Brazil or other countries with centralized voter registration, because they don’t exist – people can vote anywhere, once. Those who happen to be outside the country on Election Day are permitted to cast ballots at Brazilian consulates.
The US. Outright vote buying is also an ancient American tradition – at least since George Washington bought a quart of booze for every voter in his district when he first ran for the Virginia legislature in 1757. Vote buying still occasionally turns up in the US today, and the Internet has added a new dimension to traditional vote buying, with the appearance in 2000 of sites like www.voter-auction.net and www.winwin.org that permitted voters to swap or even sell their votes in online auctions.
As one of these site’s authors rationalized it, this approach is much more efficient and direct than the standard practice, in which political middlemen like campaign consultants and advertising gurus routinely take 10-15 percent off the top of a campaign’s media spending in exchange for delivering votes.
However, the main areas that are ripe for massive abuse in the US now are probably absentee ballots and “missing/ faulty votes.”
Absentee Ballot Problems. The liberalization of absentee ballot laws since the late 1970s, combined with increasingly lax voter registration laws, and the absence of a national identity system or central voter registration list, have made this a growing problem in the US.
This year, absentee ballots may account for as much as 20 percent of the total US vote, up from 14 percent in 2000. This enables a panoply of specific malpractices, including (1) fraudulent registration and voting in the names of phantom, relocated, or deceased voters; (2) voting by the same individuals in multiple states; (3) gross violations of voter secrecy, including pressuring senior citizens to fill out their absentee ballots in a specific way; and (4) fraudulent voting by Americans civilians or military personnel who are located offshore – that also played a key role in the 2000 Presidential election.
It is not clear which major party has benefited the most from such practices. Bogus registration may have had its finest hour in the notorious 1960 “Chicago miracle,” when thousands of dead people allegedly voted for President Kennedy. This year, however, with the Republican Party in control of all three branches of the US federal government, 28 out of 50 state governorships (including 5 BG states), and 17 state legislatures (including 9 BG states), the Republicans may bear the most watching.
Missing/ Faulty Ballots. Of course the recent US track record with respect to mechanical vote processing is also not encouraging. During the 2000 US Presidential election, the whole country was forced to agonize for months over Florida’s mechanical voting machines, until a highly-politicized US Supreme Court awarded the election to President Bush by a 5-4 vote. CNN and other mainstream media organizations later hired the National Opinion Research Center to conduct a six-month audit of the 2000 Florida vote. Amazingly, when it was concluded, they declared that it showed that “Bush would have won anyway.” In fact a careful reading of the report shows that precisely the opposite was the case.
Furthermore, subsequent analysis of the 2000 Florida outcome shows that the balloting problems that figured in the Supreme Court decision were just the tip of the iceberg – compared with other glaring problems, like the bogus exclusion of thousands of black “pseudo-felons” from Florida’s voter rolls, and the intentional spoiling of thousands of black ballots.
According to the nonpartisan election monitoring group Votewatch, in the 2000 election Florida’s “hanging chad” problem was dwarfed by other counting problems. An estimated 4 to 6 million voters simply had their votes wasted because of faulty equipment and confusing ballots (1.5- 2 million), registration mix-ups (1.5 – 3 million), and screw-ups at polling places (up to 1 million).
This year, the US State Department has responded to concerns about a repetition of such behavior by inviting international observers from Vienna’s Organization for Security and Cooperation in Europe to monitor the US Presidential election – for the first time ever.
Other Measures
To prevent public officials in the Executive Branch from abusing their powers to manipulate voters, Brazil’s Constitution also requires them to resign six months before elections if they or their immediate family members intend to run. The US has no such disqualification period for political candidates who have worked in the executive branch.
National Administration of Elections
Brazil. Like many other emerging democracies, Brazil has an entire separate judicial branch, the Justicia Eleitoral, that is responsible for implementing all campaign finance and voting procedures all levels of government. (1988 Constitution, Articles 118-121).
This system is by no means perfect, but it is far more objective and “party-neutral” than the US system, which is heavily influenced by state and local politics. Brazil’s approach has also contributed to the rapid nation-wide adoption of reforms, like electronic voting.
The US. Since 1974, campaign finance for US federal elections has been administered by an independent regulatory body, the Federal Election Commission. But US voting procedures remain under the control of state and local authorities, even in the case of federal elections. While the National Institute of Standards and Technology makes recommendations concerning voting machines and registration procedures, these are voluntary.
The result is a hodge podge of inconsistent, incompatible and often out-dated paper ballot, machine, and electronic recording procedures that vary enormously among states, and often even within the same states.
Direct Elections/Run-Offs/”Electoral College”
Brazil. In the case of Brazil’s Presidential elections, there is no “electoral college” that stands in the way of the popular will. Article 77 of Brazil’s 1988 Constitution explicitly provides that Brazil’s President is that political party candidate who obtains an absolute majority of bona fide votes. If no candidate emerges from the first round of voting with an absolute majority, a run-off election is to be held within 20 days – as it was in October 2002. Elections for Brazil’s 26 state governors and the mayors of its large cities are also subject to similar runoffs.
The US. Presidential elections are decided by the notorious “electoral college.” As discussed below, this is really just a throwback to the protection of Southern slavery. Currently it serves mainly to protect the influence of “strategic minorities” in a handful of smaller so-called battleground states. (See the discussion in Part I of this series. ) In today’s highly partisan environment, it also (temporarily) gives minority candidates like Nader tremendous destructive power – without, however, affording them any representation at all between elections.
Indeed, the US’ “winner-take-all” system systematically discourages third party power and minority representation. In a more democratic system with proportional representation, preference voting, or even Presidential run offs (see below), voters could freely vote for candidates like Nader without fear of electing their “third choice.” It is no accident that parties like the Green Party, the Libertarian Party, and the Reform Party are in deep crisis, even as Nader’s stubborn crusade continues. Would that more of his spite were directed at “winner take all” and the electoral college, which has helped to institutionalize a political duopoly.
Technically, under the US Constitution (Article II), Americans don’t even have the right to vote for their President or Vice President. This right is delegated to Presidential electors, who may be selected by state legislatures any way they see fit. Indeed, popular vote totals for President and Vice President were not even recorded until 1824, and most state legislatures chose the electors directly. Since then, most states have chosen them according to the plurality of the popular vote in each state. But South Carolina – always a hotbed of reaction, with the highest ratio of slaves to whites -- continued to ignore the popular vote until the end of the Civil War.
Nor is the number of Presidential electors per state even determined by the relative population or voters per state. It is fixed at the number of Senators and Congressmen per state, with a minimum of 3 for each state and the District of Columbia. This guarantees that small rural states are over-represented. Because of the winner-take-all system, it also effectively disenfranchises everyone in a state who votes for candidates who lose that state, and partly disenfranchises all those in large states. At least four times in US history – 1804, 1824, 1876, 1888, and 2000 – this anti-democratic system produced Presidents who failed to win a plurality of the popular vote.
In the context of other democracies around the world, the electoral college is a very peculiar institution indeed. Kenya’s former dictator, Daniel Arap Moi, adopted something similar to control the influence of the country’s dominant tribe, the Kikuyu. The Vatican’s College of Cardinals has something similar.
There have been many calls for the electoral college’s abolition, most recently by the New York Times in August 2004. But this would require a Constitutional amendment that would have to be ratified by three-fourths of the states – at least a quarter of whom have disproportionate power under the existing system.
It is course possible for a state to provide that its own electors are awarded to Presidential candidates in proportion to the popular vote within a state -- as Colorado is considering this year. However, unless all battleground states go along with this, which is unlikely, it is unlikely to gain national momentum.
Electronic Voting/ Registration
Brazil. To the surprise of many, Brazil has recently become the world pioneer in electronic voting and registration. When it held national elections in October 2002, 91 million out of its 115 million registered voters turned out – more than 70 percent of those of voting age, and 3 million more than voted in the US elections that same fall. In terms of global electoral history, the number of votes received by the winner, Luis Ignacio de Silva (“Lula”), was second only to Ronald Reagan’s total in 1980.
To handle this heavy turnout, Brazil relied heavily on electronic voting. The Tribunal Superior Eleitoral (TSE) had been experimenting with electronic voting systems since the early 1990s, becoming a real pioneer in the use of “DREs” (“direct recording electronic”) voting machines. Brazil first used DREs on a large scale in its 1996 elections, with 354,000 in place by 2002. For that election, it deployed another 52,000 “Urnas Eletronica 2002,” a state-of-the-art DRE that had been designed by Brazilian technicians with the help of three private companies – Unisys and National Semiconductor, two US companies, and ProComp, a Brazilian assembler that has since been acquired by Diebold Systems, the controversial American leader in electronic voting systems.
Because Brazil has been willing to commit to such a large-scale deployment, each Urna costs just $420, less than 15 percent of the cost of the $3000 touch-screen systems that Diebold features in the US. The Brazilian system lacks a touch screen; voters punch in specific numbers for each candidate, calling up his name and image, and then confirm their selections. The numerical system was intended to overcome the problem of illiteracy, which is still a problem in parts of the country. To handle operations in remote areas like the Amazon, the machine runs on batteries up to 12 hours. Initially there were no printed records, but the Electoral Commission decided to retrofit 3 percent with printers, to provide auditable records.
Like any new technology, Brazil’s approach to electronic voting is by no means perfect. Indeed, significant concerns have been voiced about the system’s verifiability and privacy – especially about the TSE’s recent move to eliminate the printers, supposedly because they slowed voting.
Among the most important proposed improvements are a requirement that all voting machines produce both electronic and paper records, in order to leave an audit trail and increase voter confidence in the system; that system software be based on “open” standards and available for audit; and that the system for identifying eligible voters be separated from voting, to insure privacy.
Despite these concerns, most observers agree that Brazil’s system performed very well in 2002. In Brazil’s case, within a couple days, winners were announced almost entirely without dispute, not only for the Presidential race, but also for the 54 Senate races, 513 Congressional races, 27 state governorships, 5500 mayors, 57,316 councilmen, and many other local contests – all told, races involving more than 315,000 candidates.
Given this success story, many other countries that have traditionally had serious problems with paper ballot fraud are also considering its use, including Mexico, Argentina, the Dominican Republic, India, the Ukraine, and Paraguay.
The US. This country, where the choice of voting technology is localized and highly political, still relies very heavily on paper ballot-based and mechanical voting system – while 42 states will have new voting machines in 2004, only about 29 percent of US voters (at most) will cast electronic ballots this year. And with each state free to adopt its own local variant on the technology, it is not surprising that implementation has been problematic -- recent experiments with electronic voting in New Mexico, California and Florida have all been riddled with problems.
In 2002, the US Congress passed the “Help America Vote Act” (HAVA), which authorized $3.9 billion to be spent by 2006 to help state and local governments upgrade their election equipment. Given the fragmented nature of US election administration, however, this has produced a competitive scramble for government contracts among 3-4 leading US electronic voting firms, including Diebold Election Systems, ES&S, and Smartmatic, the Florida-based company whose 20,000 electronic voting machines successfully handled Venezuela’s Presidential Recall referendum this summer.
But with no mandatory national standards and the numerous operating problems that we’ve already seen at state and local levels, it is not surprising that there is widespread, perhaps exaggerated, concern about the potential for hacking and manipulation. In this highly-politicized context, with no adequate checks and balances over the procedures, some observers have argued that computerizing US voter registration and electronic voting will actually make things worse.
Relative to more basic problems like absentee ballots, phony registration, malapportionment, and misrepresentation, we believe that these concerns are exaggerated. However, like many other elements of our “pseudo-democratic” political system, it is very hard to argue that the US track record with respect to electronic voting is an achievement.
Proportional Representation
Brazil. Brazil’s 81 Senators are elected by simple plurality for eight-year terms. But Article 45 of Brazil’s Constitution provides that the 513 members of Brazil’s House of Deputies are elected for four year terms according to a voting system called “proportional representation” (PR). Brazil also uses PR to elect city councils and state legislatures.
Unlike the “plurality/single member district/ “first past the post” system that is used in most US federal and state elections, this approach insures that the overall mix of elected representatives more accurately reflects voter preferences. It also represents a wider range of opinions, since party candidates compete with each other. Indeed, various forms of PR are now used for the election of “lower houses” by all other countries in Latin America, almost all European countries, all the world’s largest democracies, the new democracies in Iraq and Afghanistan, and, indeed, the vast majority of democracies in the world today -- except for the US, Canada, the UK, and former British Caribbean colonies like Jamaica and the Bahamas.
Indeed, many of the world’s leading corporations have also turned to proportional representation, cumulative voting, or other forms of preference voting for purposes of electing their corporate boards.
Brazil uses what’s known as the "open list d’Hondt” version of PR. Here, political parties or coalitions register proposed lists of congressional candidates with the Electoral Court. There are no single-member congressional districts – rival candidate lists are drawn up by each party for each of Brazil’s 26 federal states, with the number of representatives per state determined by population, subject to minimums and maximums. Voters can opt for a party’s entire list, or select among individual candidates – unlike a “closed list” PR system, where candidates don’t compete with each other.
Of course Brazil’s proportional representation system is by no means perfect. Some have argued, for example, that using its 26 states as electoral districts has made Brazil’s legislators too remote from voters, and that smaller districts, or even a “mixed” PR system like that used in Mexico, Germany, or Venezuela may be preferable. Some voters may also find it too difficult to choose among so many different candidates, leading them to opt for party slates.
However, when it comes to buying shampoo or automobiles, most consumers believe that increased variety is a good thing – why should politics be any different? And it is hard to imagine an elected official who is more remote from minority voters than one who is elected – as more than 95 percent of US Congressmen now are – in a district that predictably goes either Democrat or Republican, year after year. This system, in turn, encourages the country to divide into polarized sectarian camps, where even majority voters feel less well-represented because incumbents can take them for granted.
Some have argued that the Brazil’s open list version of the PR system has contributed to a weaker party system, and to competition among a party’s own candidates. For example, any political party in Brazil can put forth candidates without having to obtain a minimum of the national vote. Party weakness may also be encouraged by the fact that incumbents can change parties without losing their spots on the ballot, one of several measures that favors candidates over parties. This may indeed have encouraged a proliferation of political parties – in Brazil’s 2004 Camara dos Deputados and Senado Federal, for example, 16 different ones are represented. Even with stable coalitions among the top 3-4 parties, the concern is that all this can substantially increase the negotiation costs of legislation, block Presidential initiatives, and lead to deadlock.
On the other hand, such negotiations may also produce outcomes that are more reflective of the popular will. And recent studies of the actual operation of Brazil’s Congress suggest that the concerns about fragmentation and policy deadlock have been overstated. Furthermore, the fact that delegate turnover in Brazilian elections since democracy has averaged more than 50 percent may be viewed as a good thing – especially compared with the “dynastic legislature” that the US has acquired.
The US. There is actually a long history of preference voting and proportional representation in the US at the state and local level. For example, a majority of the original 13 colonies’ legislatures were elected using multi-member candidate slates, and more than 60 percent of city councils in the US are still elected with at-large slates. There have also been numerous proposals to expand their use at the federal level, especially for the US House, which could be done without a Constitutional amendment.
However, currently, delegates to the House and Senate and almost all state legislatures are chosen in single-member-district “first one past the post” contests, where voters each district choose just one candidate for each office. Those voters whose candidate wins a plurality of counted votes (e.g., less than or equal to 50.1%) are awarded 100 percent of a district’s seats; all other voters get zero representation. This implies two kinds of inefficiency – “over-represented votes,” those that a winning candidate didn’t need in order to prevail, and “under-represented votes,” those cast for losing candidates.
Compared with a PR system like Brazil’s, the US is monumentally inefficient – in two-way races, 49.9 % of all votes in “winner take all” races are always wasted, in the sense that they are either more than winners need to prevail, or are votes for the loser that receive no representation.
There is a huge literature on the merits and demerits of proportional representation, which contrast its increased representation for minority interests with its alleged tendency to produce unstable coalition governments (the classic cases being Italy and Israel.) At the end of the day, there are undoubtedly some empirical trade-offs to be made. But the fact is that the vast majority of the world’s democracies, as well as many of the world’s leading private companies, have opted for various forms of PR systems for representation – with the exceptions of the US, Canada, and the UK. And among those that have opted for PR systems are all of the world’s newest democracies, including Brazil, South Africa, Indonesia, East Timor -- and even our own favorite new Middle Eastern democracies, Afghanistan and Iraq. Is there no content in this signal?
Apportionment/Gerrymandering/ “Safe Seats”
Like other countries with federal systems, Brazil and the US have struggled to (1) provide representation that is proportional to the actual number of voter-aged residents in all regions of the country, while at the same time (2) providing at least a minimum degree of representation for all regions, no matter how heavily populated.
In Brazil’s case, significant “malapportionment” – departures from representation that is strictly proportional to population -- continues exist, mainly in the form of overrepresentation for rural states in the National Camara and Senate. This is because Brazil’s Constitution guarantees each of its 26 states, plus the Federal District, at least 3 Senators regardless of population, and because representation in Brazil’s Camara is a truncated function of population, with each state guaranteed a minimum of 8 representatives and a maximum of 70, regardless of population.
However, as shown in Table 6,the overall degree of “mal-apportionment” -- defined as the median ratio of a state’s share of representatives to its share of VAP – is only slightly higher for Brazil’s Camara than in the US House, while the degree of mal-apportionment for Brazil’s Senate is much lower.
Furthermore, unlike the US, Brazil permits its federal district – Brasilia – to elect both Senators and Congressmen. In contrast, the US has stubbornly refused to permit the 563,000 residents of the District of Columbia to have any voting representation in either the House or Senate.
Finally, the total number of elected national representatives in Brazil – 513 deputies for the Camara and 81 Senators – is more than twice as high per voting age resident in Brazil than the US. Since the US has been slow to adjust the total number of House and Senate members in response to increases in population, it now has one of the lowest ratios of members to voters among leading democracies – more than six times the ratios in the UK, Canada, and South Africa.
Compared with countries like Brazil that elect their Congressmen from multi-candidate lists for fixed states, the US political system’s focus on “single-member districts” is also far more open to partisan gerrymandering. There is a long history of dominant parties drawing district lines to favor their own candidates or create safe seats. Traditionally, every ten years, state legislatures in the US redraw their district boundaries to the fit the latest Census data.
Several other factors have also exacerbated this trend toward the creation of careerist legislators in the US. As one recent analysis concluded,
…(T)he incidence and extremism of partisan redistricting have escalated. Voting patterns have become more consistently partisan, enabling political mapmakers to better predict how voters will vote. And advances in computer technology and political databases allow cartographers to fine- tune district boundaries to maximize partisan advantage.
More generally, districting isn’t an issue that necessarily favors either main party. But one consequence of it is a clear trend toward more safe seats for incumbents of both parties, fewer competitive races, and a growing geographic polarization into “red” and “blue” districts – the so-called “Retro-Metro” divide.
In this fall’s US Congressional races, for example, out of 435 US House races, only about 30 will be decided by margins of less than 10 percent, and just 5-6 percent, or 20-25, that will be “effectively contested,” with a serious possibility that seats may change party hands. Many of these involve seats where incumbents are voluntarily retiring. For the US Senate, where redistricting is not an issue, the incumbency advantage is only slightly lower -- only 4-5, or 10-12 percent, of this year’s 34 Senate races will be effectively contested.
So regardless of who wins the Presidency this year, the reality is that the US House and Senate are both likely to remain under Republican control.
This is consistent with a trend toward increased advantages for incumbents in US elections. In 2002, for example, just 4 out of 383 US House incumbents lost their seats to non-incumbent challengers. This was the highest rate of incumbent reelection since 1954. Of 435 races in 2002-04 (including special elections), 81 percent were won by incumbents with margins of more than 20 percent. Of the 379 incumbents who were reelected in 2002, the median margin of victory was 39 percent. (See Table 7.)
Among the 56 non-incumbents who were elected, 11 won in newly created “safe” districts, by a median margin of 19 percent. Thirty-two were elected from the same parties that had represented their districts before, by a median margin of 15 percent. Just 13 were new non-incumbents who managed to oust the prevailing party in their districts, mostly by defeating other new non-incumbents. In fact, in 2002, more US Congressmen were elected in uncontested races, with 100 percent of the vote (n=31), than in “close” races where the winning margin was 10 percent or less.
Given these trends, it is not surprising that the average lifespan of Congressmen has been increasing -- apart from voluntary resignations. As of 2004, the median Congressman had served 8 years, and 20 percent had served at least 16 years or more. The main source of turnover in this increasingly entrenched, carefully-districted, careerist “people’s house” is now retirement, death, or incarceration, not voter decisions.
Similar trends are evident for the US Senate and the Presidency, although the advantages of incumbency are less than for House races. In the case of 2002 Senate races, 16 of 33 races (48%) were won by more than 20 percent, and 67 percent were won by more than 10 percent. Incumbents were reelected in 23 (88 percent) out of 26 races where they decided to run. As in the House, the main source of turnover was retirement or death. As of 2004, the median Senator had been in office 10 years, and the top 20 percent had median tenures of 28 years.
As for the Presidency, out of 42 Presidents through Clinton, 25 ran for reelection, and 16 (64 percent) were reelected. Whether or not President Bush can take comfort in these odds is less clear, however – among the ten post-War Presidents that preceded him, just four (Eisenhower, Nixon, Reagan, and Clinton) were reelected.
At least at the Congressional level, there appear to be strong interdependencies between incumbent advantage and the existing systems for financing campaigns, conducting elections, representing voters, and defining districts. Finance not only strengthens incumbent advantage; it also follows from it, in the sense that incumbency makes it much easier to raise money.
Having once established this interdependent system of interests, it is very hard to unravel. Is it any wonder that more and more Americans have simply decided that they have better things to do with their time than vote – even though the issues at stake have never been more important, not only for Americans, but for the world at our mercy?
All this adds up to an electoral system that is, on the whole, much more democratic than that in the United States.
Nor is Brazil alone in providing a potential role model for those in the US who are serious about revitalizing democracy at home:
In April 2004, South Africa held its third national election, with 15.9 million South Africans, or 76.7 percent of registered voters, turning out. While the ANC won a commanding 69.7 percent of the vote, 20 other parties also participated, and 11 of them won seats in Parliament. Compared with 1994, when South Africa held its first post-apartheid election, public confidence in the political system and the future have both risen substantially.
This 2004 turnout, while impressive, was below the record level set in 1994, when more than 19 million South Africans voted in the country’s first democratic elections ever. However, the difference may be explained by the fact that in that first election, no formal registration was required -- South Africa wisely considered it far more important to hold national elections as soon as possible, rather than worry too much about registration niceties. In subsequent elections, as it implemented formal registration, voter turnout has declined somewhat. (The new US-backed regime in Iraq and Afghanistan, which have devoted an inordinate amount of time to preparing for national elections and registering voters, might well have learned from this example.)
To reinforce confidence in its electoral processes, South Africa has adopted special procedures to insure that independent international observers are present at its elections; in 1999 its elections were attended by more than 11,000 observers, including representatives from the OAU and the Commonwealth.
India, the world’s largest constitutional democracy, also held a successful parliamentary election in April-May 2004. More than 387 million people, or 56 percent of registered voters, voted, choosing among 220 parties and 5400 candidates, using more than 1 million electronic voting machines to register their preferences.
Indonesia, the world’s largest Muslim country, and its second largest democracy, held its first-ever Presidential election, as well as parliamentary and local elections, in May-September 2004. More than 155 million people, or 75 percent of registered voters,turned out to choose among six different Presidential candidates, for the most part by driving nails through the pictures of their favorite candidates on paper ballots. With few exceptions, the voting proceeded peacefully, and although there was substantial vote-buying on all sides, about 9 percent invalid votes, and some intimidation, independent monitors like the Carter Center and the LP3ES-NDI found the election generally fair.
Venezuela’s President Chavez is undoubtedly a very divisive leader – not unlike some recent US Presidents that we have known. But at least his opponents have been able to make use of a right that no Americans have – the constitutional right to initiate a recall petition half-way through a President’s term, and if 20 percent of registered voters agree, to demand a recall referendum. 
After a prolonged efforts by the opposition to gather the necessary signatures, in May 2004 Venezuela’s Supreme Court – albeit, like the US Supreme Court, populated with supporters of the President -- certified that there were indeed enough signatures to require a referendum on whether Chavez could serve out his current term until 2007.
In August 2004, in a national referendum that was conducted with electronic voting machines, Chavez won by an overwhelming 59-41 percent margin. While diehard opposition leaders, as well as the Wall Street Journal and the US Government -- which had supported a 2002 coup attempt against Chavez -- expressed doubts about the margin, the referendum was validated by independent observers like President Carter, the OAS, and a team of Johns Hopkins/Princeton political scientists.
This means that Chavez has now held five free elections and referenda in seven years, more than any other Venezuelan President. He won them all by commanding margins.
We recall the fact that in 2000, the hapless Al Gore captured a plurality of the US popular vote (48.4 percent), despite all the games played with ex-felons, absentee vote abuses, and lost ballots discussed above, and even with third- party candidates taking another 3.7 percent of the vote.
Since it is hard to believe that President Bush’s absolute popularity has increased very much since then, one wonders what the results might have been if only the US were as democratic as Venezuela – e.g., if only Americans had been able to initiate such a Presidential recall referendum, and like our good Latin neighbor to the south, determine the outcome by popular vote of the nation as a whole, under the watchful eyes of international observers.
Instead, as we've argued here, the US is still captive to age-old anti-democratic contraptions, superstitions, and subterfuges, and its particular version of "democracy" still labors in the long dark shadows cast by venerable institutions like states rights, felon disenfranchisement, and white supremacy. 
We can try to impose our self-image on to the world if we like, but we should not be surprised if the world asks us to hold up a mirror.
**
©James S. Henry and Caleb Kleppner, SubmergingMarkets™, 2004
I am disturbed by an article that incorrectly bases it's argument on the assumption that we are a democracy, when in fact we have never been one. We have always been a representative republic, with the exception of electing govenors and local city and county representatives. The reason for this is because the reach of the federal government was supposed to be and should be general and limited to serve and protect our common needs as Americans while leaving the important decision making power to the state and local officials. Over the years, socialistic views of federal power has usurped the rights of detailed and necessary governing of states and local municipalities; as evidenced by the vast diversity in environmental needs from state to state that socialistic organizations demand federal authority over. How can the Federal governmant determine a national policy for one state with a bay watershed and others with great plains, totally different needs? The electoral college protects less populous states from being overrun by urban, populous states trying to federalize urban needs, which is not the job or responsibility of federal government. Serve and protect. Read the constitution; Take a civics lessen, better yet, attend the fabulous citizenship classes our immigrants take to become citizens as my neighbors have. One is voting for the 1st time as a new citizen, not for my guy, and I'm peoud of him. His wife is taking her citizenship test and they have command of our constitution and a great respect for it.
Posted by: Christie | September 29, 2004 at 10:43 PM