US networks, Congress whitewash media role in 2000 election

First of a two-part series

By David Walsh
14 March 2001

In January and February the role of the US television networks in the events of election night (November 7-8) 2000 came under scrutiny in a number of quarters. On January 4, CBS News issued an 87-page report on its election night coverage. The same day, NBC News released a much shorter study. At the end of the month CNN, the all-news cable network, issued a report on its own performance. In the middle of February, amidst a flurry of publicity, the House Energy and Commerce Committee held a hearing at which the heads of the major news networks were questioned by a congressional panel.

This official political and media discussion centers on the two “mistaken” calls of the presidential vote in Florida made by the television networks election night—first, for Vice President Al Gore, the Democratic Party candidate, at around 7:50 p.m. (retracted at about 10:00 p.m.); second, for Texas Governor George W. Bush, the Republican candidate, at approximately 2:15 a.m. (retracted around 4:00 a.m.). The Florida vote was declared “too close to call” early on the morning of November 8 and a five-week political crisis ensued, which ended with the installation of Bush in the White House on the basis of an anti-democratic ruling by five right-wing Supreme Court justices.

Television executives have engaged in a good deal of breast-beating over the past four months about their errors election night. But, as is often the case in contemporary American political life, the issues not discussed in the public debate are the most significant ones.

Long experience has demonstrated that those engaged in a whitewash are often advised, for the sake of credibility, to acknowledge errors, lapses, or even minor transgressions, so as to avoid revealing serious wrongdoings. Such is the case here.

The official examination of the networks' conduct has been so framed as to exclude any discussion of the more general and crucial questions—whether Bush gained office through fraudulent and undemocratic means, and whether the mass media were complicit. The CBS, NBC and CNN reports, as well as the Congressional hearings, are classic examples of a cover-up, and a rather shabby one at that.

The committees set up at both CBS and NBC to look into their performances and produce reports included two network executives, plus an academic; CNN hired three outsiders, including right-wing columnist Ben Wattenberg.

All three reports are both self-serving and superficial. In relation to the problem of mistaken “calls,” each of the studies makes more or less the same recommendations: more oversight, less pressure to be the first network to make a call, less reliance on voter exit surveys, no projection of a winner in a given state until all its polling places have closed (in 12 states polls do not close at the same time), clarification of language used on broadcasts (not “Gore wins Florida,” for example, but “CBS News, based on exit polls, projects or estimates that Gore will win Florida”), uniform national poll closings.

Each of the reports places the major onus for the Florida miscalls on the Voter News Service (VNS), the organization that collects vote totals and conducts exit polls for all the major television networks and for the Associated Press (AP).

VNS is a story in itself, and one about which, until the problems on November 7, very few Americans knew anything. (As the CBS report authors observe, VNS only began to be mentioned by name on air when it became apparent that mistakes had been made. The networks were content to claim sole credit for their calls as long as they proved correct. There is a further, political purpose for making VNS the whipping boy, which we will discuss below.)

Until 1964 ABC, CBS, NBC, AP and United Press International (which no longer exists) each carried out its own vote tabulations and analyses on election day. In the New Hampshire primary, for example, each of the television networks would have telephones installed at some 300 polling places. As a cost-cutting and efficiency measure, the three networks and two wire services created the News Election Service (NES) in the summer of 1964 to keep a running total of the vote on election day. In 1990, the same networks and AP, plus the newcomer CNN, established Voter Research and Surveys (VRS) to do the same with exit polls and estimates.

It would be naïve to imagine that such a joint effort by huge corporations in competition with one another for advertising dollars could be a smooth-running operation. Indeed, the CBS report hints at this, indicating that from “VRS's inception, there were heated debates among the members, the first occurring over whether the CBS or ABC election unit would be the core of the new pool.”

CBS won out, but when VRS and NES were merged in 1993 into VNS, the difficulties apparently continued. In the 1990 and 1992 elections all calls had been made by VRS personnel, and communicated directly to the networks and by them to the public. In 1994, ABC, presumably still rankling from its organizational defeat within the combined service, formed its own decision team (using VNS data) and called several races before VNS and the other members. In response, all the networks formed their own decision teams, CBS and CNN creating a joint team. (Fox News joined VNS in 1996.)

On election day, VNS communicates its vote totals and exit poll findings to the networks and the AP (which maintains its own independent vote tabulation) and the latter organizations' “decision teams” make the determination when to call a state for a particular candidate. Not surprisingly, the calls are generally made within minutes and even seconds of each other.

For example, the projection that Gore had won Florida, based on VNS data, was made by NBC at 7:49:40 p.m., CBS at 7:50:11 p.m., Fox and VNS itself at 7:52 p.m. and ABC at 8:02 p.m.. An individual network's bragging rights, which undoubtedly have a cash value with advertisers, stem from how many “first calls” it makes.

The creation of VNS, logical from one point of view—why should there be six different organizations collecting voting data?—is, from another, bound up with corporate economic considerations and retrograde social trends.

The five networks and AP contributed a combined $33 million to fund VNS in the “election cycle,” including the primaries and culminating in the November 2000 election. Each individual network, in other words, spent a mere fraction of the cost of an individual episode of a successful television series such as Friends or ER on “Decision 2000,” or whatever other pretentious name each gave its coverage.

Such cost-cutting is part of a broader trend. Television coverage of the elections has declined sharply in scope and seriousness over the past several decades. This constitutes an element of the general decay of the electoral process—the narrowing of differences between the two parties, the coarsening of debate, the growing alienation of broad layers of the public, the increasingly predictable and pro forma character of election campaigns—all of this taking place within the context of growing social polarization.

What the network reports do not discuss

The various reports on the television networks' election coverage fail to make reference to two of the most remarkable, and interconnected, developments that took place on election night: the extraordinary governor's mansion press conference held by Bush, and related Republican efforts to pressure the networks into retracting their call for Gore in Florida; and the role played by Bush's first cousin, John Ellis, head of Fox News' decision team.

The projection made by all the networks by 8 p.m. of a Gore victory in Florida was considered a fatal blow to Bush's hopes, particularly when it was followed by the declaration of a Gore triumph in Pennsylvania. At this point panic reportedly set in within the Texas governor's camp.

Instead of continuing to watch the returns at the Four Seasons Hotel in Austin, Bush and his entourage abruptly moved to the governor's mansion. Florida Governor Jeb Bush, the candidate's brother, telephoned his cousin, Ellis, at Fox and asked him about the networks' claim of a Gore victory in Florida. “Are you sure?”, he asked Ellis, to which Ellis reportedly replied, “We're looking at a screen full of Gore.”

The Bush forces began a campaign to reverse the networks' call in Florida. Mary Matalin, a Republican media operative, raised doubts about the Gore call on CNN. At around 9:30 p.m. Karl Rove, Bush's chief political strategist, went on NBC and admonished the networks. “I would also suggest that Florida has been prematurely called,” he declared. “First of all, I thought it was a little bit irresponsible of the networks to call it [for Gore] before the polls closed in the western part of Florida. Florida is still split among two time zones, eastern and central. You all called it before the polls had closed in the central part of the country.”

This was the first reference to what was to become a minor right-wing rallying cry: the claim that the networks cost Bush votes by projecting a Gore victory before the polls were closed in the more heavily Republican northwestern corner of Florida. While there is an issue of principle here—candidates and voters have a right to expect that the networks will withhold their election calls until the polls in any given state have closed—in relation to the outcome of the 2000 election, the “early” call in Florida is essentially a red herring.

The polls in Florida's eastern time zone closed at 7:00 p.m.; at that point, only 5 percent of the voting-age population, according to the CBS report, had not voted. Moreover, the networks actually began calling the election for Gore at 7:50 p.m. in the east (6:50 central time), only 10 minutes before the polls closed in the Florida panhandle.

Rove's appearance on NBC was followed by an impromptu press conference held by Bush in the governor's mansion at around 9:50 p.m., an event unprecedented in US election night annals. Bush chastised the networks for their calls in Florida and Pennsylvania, saying both states were too close to call. This extraordinary intervention by a presidential candidate has been all but forgotten (or passed over) in the media coverage of election night. None of the networks' reports even refer to it, and a supposedly hard-hitting article in Brill's Content by Seth Mnookin (“It Happened One Night”) simply mentions in passing “a defiant appearance by Bush.”

Yet on November 8 an article in the Washington Post was relatively forthright:

“All the turmoil in Florida produced an extraordinary bit of television drama, with four networks abruptly backing off their projection that Gore would win Florida's crucial 25 electoral votes. They did so after Bush allowed cameras into the Texas governor's mansion so he could insist that the Florida contest was not over.

“At 10 p.m., CBS, ABC and CNN all said they were moving Florida into the undecided category, more than two hours after they had used exit-poll data to call the state for Gore. NBC followed 15 minutes later.

“The networks' flip-flop came about 10 minutes after they aired an unusual videotape in which Bush, with his father, mother and wife, challenged the television projections in Florida and Pennsylvania. ‘The people actually counting the votes have come to a different perspective.... I'm pretty darned upbeat about things,' said Bush, undoubtedly with an eye on turning out his supporters in western states.

“The network reversal quickly changed the commentary, which had increasingly been saying it would be very difficult for Bush to beat Gore after having lost Florida, Michigan and Pennsylvania.”

The networks have since claimed there was ample data flowing in to justify their retraction of the Gore call in Florida. Whatever the case, it would be hard to argue that the unprecedented pressure applied by the Bush camp had no impact on television network executives, who are generally indifferent to the sentiments of the population at large, but extremely sensitive to the demands of the corporate and political establishment.

Telephone conversations between the Bushes and their cousin at Fox continued into the night. These remarkable facts—that the presidential candidate's brother was in charge of the government (and vote tabulating) apparatus in the contested state, and that their first cousin held a critical position on the decision team of a major network charged with calling the election—have provoked little debate or even comment in the media or, for that matter, from the liberal establishment or the Democratic Party. This, despite the fact that Rupert Murdoch's Fox News, with Ellis leading its decision desk, was the first network to declare Bush the victor in Florida at 2:16 in the morning.

Notwithstanding the indifference of the media and the political officialdom, the role of Ellis and Fox News raises some obvious and pointed questions:

What, if anything, did George W. Bush and his brother Jeb know about the vote in Florida that the public did not? Why were they apparently so certain Florida would end up in their camp? What discussions took place between Bush operatives and John Ellis of Fox News?

Did, in fact, George W. Bush or his associates, through Bush's cousin, call the election for George W. Bush?

On numerous occasions, through all the largely manufactured scandals of the Clinton years and beyond, Whitewater Independent Counsel Kenneth Starr or Republicans in Congress have subpoenaed witnesses and impounded records either to obtain information they thought might incriminate or embarrass Clinton, or to generate the appearance of criminal behavior, or simply to harass.

It is, however, a remarkable fact that in this case, concerning the result of a presidential election, no one has demanded that the Bush team's phone logs or notes be produced.

The psychological and political importance of having Bush declared the winner in Florida, and hence nationally, was enormous. It was no doubt an important element in the calculations of Ellis and the Republican camp. From that moment onward, a section of the public, encouraged by the Republicans and the media, viewed Bush as the legitimate winner and Gore the “sore loser.”

The failure of the CBS, NBC and CNN reports even to refer to Ellis or to Bush's election night press conference is sufficient to brand them as travesties.

What the network reports disclose

Each report takes as its starting point the legitimacy of the final outcome of the election crisis: the suppression of votes in Florida and installation of Bush.

The NBC report is perfunctory. Of the two longer studies, CBS's is the more informative. Typically, while feeling no need to respond to the widespread sentiment that the Bush election was fraudulent and his administration illegitimate, the CBS authors are sensitive to every allegation of the Republican right wing, no matter how far-fetched.

The CBS report, for example, disputes the accusation that early calls by the television networks affect the outcome of voting (i.e., that residents of areas where polls are still open will be discouraged from voting by projections of a national or state winner). The report's authors expose several so-called studies of the Florida vote as the work of “Republican partisans, not unbiased observers.”

They reject, albeit diplomatically, the charge made by Republican Congressman W. J. “Billy” Tauzin of Louisiana that the networks demonstrated bias for Gore by delaying calling states for Bush. This absurd allegation, originally made by Tauzin, the chairman of the Energy and Commerce Committee, on November 9 and repeated in a letter to the network news organizations December 11, formed the initial justification for the February hearing in Washington where the television news executives testified.

The authors of the CBS report observe that the mistaken calls of the networks pale in comparison to a more serious issue, the “dirty little secret” of American politics: in their words, that US elections are “prone to human error, mechanical error, confusion and disorganization,” with some two million votes being thrown out for every 100 million cast.

They comment: “We have heard much about the punch-card ballots in Florida. But we now know that a third of the country votes by punch-card ballots. In Cook County, Illinois [Chicago], in this election, more than 120,000 punch-card ballots were discarded.... In Detroit, some polling places did not have enough electronic voting pens to service the voting booths. In Massachusetts, 30,000 votes were left uncounted in 51 precincts because of human error. In New Mexico, election officials thought that a handwritten notation about absentee votes from one precinct indicated 120 votes for Gore, when the actual number was 620.”

The CNN report, about whose “independence” the network made a great fuss, is the more explicitly reactionary in its political outlook. As opposed to the other networks' commentators, the authors of the CNN study express definite concern over the extent to which the networks' mistaken calls inflamed a volatile political situation.

In their introduction, they write: “The uncertainty about who had won Florida, engendered by the closeness of the Florida contest, but exacerbated by the mis-reporting, turned out to play an unhealthy role in the subsequent tense and potentially dangerous post-election controversy until the final determination of the race after more than a month in a climate of public rancor.”

In the section entitled “Recommendations,” the CNN report returns to this theme: “There is no shortage of angry Americans who at any given moment believe something ‘unfair' has happened in the world's model democracy. The weeks following the Florida election led to complaints about bias and/or lack of competence in the broadcast media, and helped set an angry tone in the country concerning the outcome of the election. One would think that only great benefit might tempt major news organizations to risk placing even an extra twig on that fire.”

Here the authors of the CNN report barely conceal their contempt for the American people, whom they disparage for presuming to believe that the “world's model democracy” could possibly be unfair. It does not occur to them that a contradiction might exist between the admission that a great many Americans “at any given moment” think something “unfair” is happening, on the one hand, and their glowing characterization of American democracy, on the other.

Concerning the actual events of November 7-8, the CBS and CNN reports paint similar pictures. Taking into account that both reports leave entirely out of the picture the machinations of the Bush camp, what follows is the official version of the mistaken calls in the Florida vote.

The call for Gore

Between 7:00 and shortly after 8:00 p.m., all indicators strongly suggested to VNS analysts that Gore was heading for victory in Florida, and by a considerable margin. At 7:48 p.m. NBC became the first television network to declare Gore the projected winner in Florida, followed by the other networks, AP and VNS itself within the next 15 minutes.

At 8:10 p.m. the CNN-CBS decision team reviewed the Florida data and concluded that the exit polls had underestimated Gore's victory margin by nearly 4 percent. The team was more convinced than before of a Gore win. Between 9:00 and 9:45 p.m., however, the projected Gore lead failed to materialize. Faced with a Bush lead in the actual tabulated vote, all the networks began to consider pulling back from the Gore call. They did so at 10:00 or shortly after. At 10:16 VNS officially retracted its call for the Democratic candidate.

It is necessary, however, to submit these events to closer examination.

Voter News Service is a well-tested organization. However one may feel about the practice of projecting winners in elections as early as possible, and the motives for doing so, VNS personnel have a considerable expertise in that sort of work. In determining the projected winner in a given state, the service's statisticians require that there be less than a 1 in 200 chance of error. The VNS analysis in Florida was based on a combination of exit polls (voters were queried in 45 precincts), the state's tabulated raw vote, county models, past voting patterns, current projections and unofficial results supplied to the service by poll workers in 120 precincts.

The following calculations are based on the CNN and CBS references to VNS's internal report, which was only made available to its member news organizations. Since percentages alone are mentioned by CNN and CBS, not actual numbers, some degree of approximation is required.

At 7:50, 50 minutes after most of the state's polls had closed, VNS projected Gore winning Florida by 7.3 percent, or approximately 52.5 to 45.2 percent. Taking into account that more people ultimately showed up at the polls in Florida (nearly six million voted) than VNS had projected, its 7.3 percent estimate was probably somewhere in the range of 400,000 votes. This is, relatively speaking, a huge margin. Bill Clinton won Florida in 1996 by 5.7 percent (or some 300,000 votes). Gore won Michigan and Pennsylvania, two of the other critical races, by 4 percent.

VNS apparently made a number of projections between 7 p.m. and some time after 8, but they all indicated a Gore victory. Louis Boccardi of AP, in his testimony before Congress, noted that “shortly after 7 p.m. Eastern time ... the VNS exit polling data were indicating that Mr. Gore could win by a margin of more than 6 percent.” This is confirmed by the CNN report, which refers to “VNS exit polling information supplied to the CNN/CBS Decision Team show[ing] Gore leading Bush in Florida by 6.6 per cent.” According to Mnookin in Brill's Content, at 7:40 p.m. VNS estimated a 51.1 to 46.5 percent victory (a 4.5 percent margin) for Gore.

At 7:45 VNS information suggested that the first returns indicated, if anything, that the exit polls had been overstating the Bush vote. The CBS report states: “The average error within those precincts suggested that the survey was actually underestimating the Gore lead by 1.7 percentage points.”

In his statement before the House Energy and Commerce Committee on February 14, Ted Savaglio, executive director of VNS, declared: “On Election Night our statistical models, based on our exit polls and actual vote from a number of sample precincts, showed Vice President Gore ahead—decisively it seemed—in Florida. Our decision team considered other variables, including absentee vote beyond that which already was accounted for in the models, and determined that the data clearly justified making a call, which we did shortly before 8:00 p.m..”

By 8:10 p.m., the data, including actual votes, pointing toward a Gore victory was so convincing that all the networks apparently felt confident in the projections they had made. (It must have been around this time that Ellis told Jeb Bush, the Florida governor, “We're looking at a screen full of Gore.”) The CNN-CBS Decision Team reviewed the Florida data and concluded that the exit poll had underestimated the Gore victory margin by nearly 4 percent. According to the CNN report, “That, along with other sets of data, makes the team more certain of a Gore win there.”

The decision team members later reported to the two networks: “Even if we had not made the Gore projection at 7:50, we surely would have made the projection looking at this data at 8:10. In our many years of examining decision screens we do not believe that there has ever been a single instance in which the leader changed in a race in which we had this much data from survey, VPA [Voter Profile Analysis], and county vote and ten estimators all showing a six point lead or more. Presented with this consistent data there was no reason to justify not calling this race. We would not have been doing our jobs if we had not called this race at this time when presented with this data. If we cannot believe data this convincing from VNS the entire purpose of our Decision Team is undermined” (emphasis added—DW).

The implication here is that the VNS data and projections were drastically wrong, but the comment is perhaps more suggestive than the authors suspect.