If you voted using an electronic voting machine in November’s elections, there’s a sixteen to twenty-eight per cent chance your vote either didn’t count or counted for a candidate you didn’t want it to. Sixteen to twenty-eight per cent is the “failure rate” acknowledged by the Illinois State Board of Elections. Electronic machines that leave no paper trail were used in record numbers around the country this year, and almost everywhere problems arose. In Scurry County, Texas, the Associated Press reported, a faulty computer chip reversed the outcomes of two commissioner races. The problem was only discovered when poll workers became suspicious of the Republicans’ landslide victories and recounted votes by hand. They found that the Democrats actually won by wide margins.


In Broward Co. Florida, a Democratic stronghold and the epicenter of the 2000 presidential election controversies, a computer glitch lost 103,222 votes on election night. When the votes resurfaced the next day, according to county officials, they didn’t change the outcomes of any races.


In Davison Co. South Dakota, an auditor found that the machines were double counting certain votes, even though tests run before the election had shown no sign of error. The Senate race there between Democrat Tim Johnson and Republican John Thune was eventually decided by only 524 votes.


All over the country, similar quirks of the voting machines came into play, but despite these inconsistencies, the national news media reported a “smooth” election. Many experts, however, feel that electronic voting machines represent a serious threat to the democratic process.


Dr. Rebecca Mercuri is a professor of computer science at Bryn Mawr and the nation’s leading expert on electronic voting systems. She writes, “In the rush to correct problems exposed by the 2000 presidential election debacle in Florida, many municipalities were pressured or required to procure new voting systems. The most vulnerable of these systems are the fully electronic touch-screen or kiosk (DRE) devices because of their lack of an independent, voter-verified audit trail.” She adds, "Any programmer can write code that displays one thing on a screen, records something else, and prints yet another result." The basic role of election judges – to count and verify the accuracy of the vote – has been usurped and compromised by the private companies that operate the voting machines.


Nebraska candidate for governor Paul Rosberg described going to see his votes being counted this way: “The optical scan ballots were run through the machine. There is no counter on the machine, so you can't see what it counts. Then, the votes are added up on a computer in another room. I couldn't see how they transferred the votes from the machine that ran the ballots through over to the main computer -- maybe on a disk, maybe a modem? Anyway, I went in the private room where the main computer counts the votes. You can't see anything, just a dark screen. After standing there watching a computer on a table, with a dark screen and nothing to watch, I finally went home. This is NOT ‘watching your votes being counted.’ It's not right.”


Many of these touch-screen devices generate absolutely no physical evidence of the vote, and this is the problem. Voting systems are becoming less and less auditable. The computer code that runs the program to tabulate votes is neither infallible nor secure, but it is proprietary which means the public can’t see it. The companies that operate and test the voting systems have essentially taken a “trust us” stance on the issue, but very little is actually known about who owns these companies or what conflicts of interests the owners might have.


The two biggest election technology companies count roughly two thirds of the votes in the United States. Election Systems and Software (ES&S), founded by the brothers Todd and Bob Urosevich in the early 1980s, is by far the biggest company, responsible for about 56% of the vote counting in the country. Global Election Systems, now part of Diebold, is the second largest. Todd Urosevich still runs ES&S, and his brother Bob is now the CEO of Global Election Systems.


Ownership of these companies is not disclosed. From 1984 to 1987, sixty-eight per cent controlling interest in ES&S was owned by the billionaire Ahmanson family, who were instrumental in making the Republican Party take a hard turn to the right, contributing money to conservative Christian candidates and right-wing agendas. The Ahmansons are heirs to the Home Savings of America fortune, which was the largest savings and loan association in the world during the S&L scandal-ridden 1980s, and Howard Ahmanson is a major benefactor of the Christian Reconstructionist movement, whose followers wish to turn certain tenets of the Bible into national law.


In 1987 the investment group connected to the Ahmansons sold their shares in ES&S, 35% to the McCarthy Group and 45% to the World Herald Company, Inc. According to Congressional Quarterly, recently elected Sen. Chuck Hagel (R.-Neb) lists among his assets a $5 million investment in the McCarthy Group and neglects to mention that he was CEO and Chairman of the Board of ES&S, the company responsible for counting his own votes.


In 1998, ES&S acquired Business Records Corp., a Texas-based election company. Twenty per cent of the stock of the merged company was given to BRC owners. Among the owners of BRC are Caroline Hunt, of the Hunt Oil family, Alex Sheshunoff, a financial data publisher, the late P.E. Esping, who founded First Data Merchant Services, C.A. Rundell, CEO of Integrated Securities Systems, Inc., and Ed Belanger, president and CEO of CDS Technologies. It’s confusing and uninteresting, and for all this, the names of the owners of ES&S are still unknown. For all anyone knows, the Ahmansons may still hold considerable sway. Basically, without disclosure, there’s no way to know who’s controlling the counting of the nation’s votes or what vested interests or political agendas they may have.


Such anonymity is more disquieting when taken with the “glitches” seen across the country. The seriousness of the computers’ vote mis-recordings has been played down, but experts say such problems, called data path errors, represent the single most serious problem a voting machine can have. Furthermore, industry standards for computer software programming that make code easy to audit, spot changes in the programming, and even provide a history of all the coding done are ignored by the election technology companies. According to programmers inside ES&S, the company’s own 25-point manual to prevent vote rigging is “not used,” and employees are actually actively discouraged from pursuing it as an issue.


Compounding the mystery are this year’s lack of Voter News Service (VNS) exit poll data and the degree to which election tallies deviated from pre-election polling. In some races, the race for Georgia governor or Colorado Senator for instance, final election results showed swings of ten percentage points or more from polls taken just days earlier. Comprehensive comparisons of polls and final results take a long time to compile, but early estimates seem to show that Republicans on average picked up about five points on election day, confounding many predictions and in spite of the fact that undecided voters typically vote two to one for Democrats at the last minute.


Journalist Lynn Landes recently interviewed veteran pollster John Zogby of Zogby International polling service. Landes wrote, “I asked him if over the years he had noticed increased variation between pre-election predictions and election results. Zogby said that he didn't notice any big problems until this year. ‘Things were very different this time. I blew Illinois. I blew Colorado and Georgia. And never in my life did I get New Hampshire wrong...but I blew that too.’”


One possible explanation is that the Democrats were simply unable to rouse their traditional support base. It’s difficult to know, as the VNS hasn’t released the exit poll information it collected which would shed light on how different demographics voted. Conflicting reasons have been offered for this. One is that the poll data was knocked out by a “massive computer glitch.” Another is that after two years and millions of dollars in funds, the VNS is still working out bugs in its software that caused problems in the elections of 2000. The simplest answer was offered by VNS media service director Lee C. Shapiro who explained, “We are looking into whether or not we will process the data.” Evidently, the VNS hasn’t even decided if it wants to do its job this time around.


The case that the Democrats’ loss of the Senate and key governorships was simply due to better organization and political rallying on the part of the G.O.P. is weakened by the fact that in many parts of the country, Democratic ballot initiatives like limited class size in Florida did better than the clearly identified candidates who supported them. In Florida, the class size amendment passed while McBride, who made class size a central platform of his campaign, lost to Jeb Bush by fifteen points. In addition, voter turnout was unexpectedly high for a midterm election, which is traditionally a boon to Democratic candidates.


Indeed, if November’s election results had been less clearly partisan nationally, one could conclude more comfortably that any fraud that took place cut both ways. Now, the far right is poised to assume near total control of all three branches of government. They will be looking to set the agenda on such issues as weakening environmental, health and safety, and antitrust legislation, deregulating big business, confirming conservative judicial appointments, and, of course, going to war. The facts surrounding the companies and the voting machines that put the Republicans into office are at best hazy and at worst dangerously undemocratic. The issues are complicated and not very sexy so it will be a long time before the national media ever covers the story, if it ever gets covered. In the meantime, how sure can you be that your vote counted?