A Case Against Electronic Ballot Counting


In many ways, technological advances in ballot counting are solutions in search of problems. Think about it. Counting votes is easy. Any literate person is capable of counting a ballot. Over the past twenty-five years, states have increasingly made use of mechanical and electronic voting equipment to mixed reviews. As the use of electronic systems has increased, so has distrust in the integrity of our elections increased.

Private companies generally apply rudimentary costs/benefit analysis before spending their money. If it won’t increase their profits or advance their position in the marketplace, they won’t spend the money. Companies who spend money without regard to profit do so at their own peril.

In government, such analysis rarely occurs. Since the source of funding is treated as if it were limitless, there is rarely an economic justification made for technological expenditures. The justification invariably used is "efficiency." The mere fact that technology exists is reason enough to apply it whenever and wherever it can be argued to make a process or agency more efficient.

This absence of true market accountability has detrimental effects on taxpayers and the companies that provide goods and services exclusively to government. Since there is only one entity providing requirements and funding, the competition is limited to price alone. Innovation and common sense do not come into play.

Here’s a somewhat hypothetical example of how a market is created by government. This won’t be perfectly accurate but will be accurate enough for our purpose.

The legislature of a state, based on complaints and requests from voting officials or lobbying efforts by commercial interests, issues a bill requiring that all precincts convert to electronic systems by some date. They allocate 10 million dollars to be used by various counties to lease or purchase such equipment. The requirements for the equipment may be written by congressional staffers who possess dubious technical expertise, or have dubiously interpreted the advice of technological experts who testified before a committee convened to determine best practice. Sometimes, the legislature establishes a "blue ribbon" commission to then report back or use some of the funds to provide requirements.

In this case, there isn’t an external market demanding special voting machines. The examples available to them for study are other state governments who have successfully implemented electronic voting or purely theoretical claims about how wonderful it would be if such systems were employed. (When a government agency claims success, it is not the same thing as success in the commercial market. Total failure can and will be represented as success.)

Once the bill passes and the money is allocated, companies begin "competing" for the contracts. Sometimes these companies have formed in anticipation of the bill’s passage. The companies must meet the requirements as agreed by the legislature or panelists. If any advances in technology occur between the times the bill is passed and the technology is created, all the worse for the taxpayers.

For instance, if security wasn’t fully considered when drafting the requirements, the voters will have to wait until the legislature addresses it. In a real market, the market itself would drive these requirements. Companies who don’t meet or exceed requirements simply go out of business.

Government, in an effort to hide its bad planning will simply ignore a problem until such time as the problem can be blamed on the contractor.

Rick’s first law of government contractors:

Companies or individuals commissioned by government to provide a service or product will never produce anything which exceeds the requirements, even if the original requirements do not describe a working system.

Companies in contract with government don’t have to exceed expectations in order to win the contract and it isn’t profitable for them to do so. They have no incentive whatsoever to make their solution better than what was requested.

What is finally produced, if it works at all, is likely already technologically out-of-date by the time precincts receive it. It will be expensive, since the companies producing it do not have any external market pressure and it will be some years before a new cycle of upgrades can be justified, legislated and funded. Rinse and repeat.

But don’t take my word for it.

In “Hacking Democracy," (an HBO documentary which has been nominated for an Emmy this year), Bev Harris and associates demonstrated how easily election results could be changed without any trace of tampering. It was so easily done, California, Iowa and Pennsylvania rushed to remove voting machines of the same make and model prior to the May, 2006 elections. Recent California and Florida investigations into ballot scanning machines have resulted in suspension of certification for multiple manufacturers until proper security patches can be delivered. While there isn’t any evidence that elections have been tampered with, it is demonstrably possible to do so without leaving evidence.

The systems in question have been on the "market" for some time. Due to government’s reluctance to re-consider past decisions, flaws are only now coming to light.

There is also the practicality of this technology to consider. With even the largest precincts processing no more than 5,000 votes, the use of electronic systems to count ballots is essentially overkill. It’s akin to dropping a bomb on an anthill as a pest control measure; a far less advanced form of technology would be just as effective and would cost much less. Furthermore, it makes no sense to purchase or lease expensive equipment only to have it stored until the one or two elections per year take place.

The companies vying for government contracts routinely involve themselves in what look to be conflicts of interest. For instance, In San Diego County, the current registrar of voters, Deborah Seiler, was the very person who sold the county its voting machines prior to the time she gained office. In numerous states, election officials appear on voting manufacturers’ marketing brochures. Confidence in the process suffers. In some cases, the very politicians who push electronic voting initiatives receive donations from manufacturers of voting systems. Not only does this raise concerns of propriety, one has to wonder if there isn’t more than meets the eye. If there is anyone with an interest in manipulating an election result it is the incumbent politician. Even if it weren’t possible, this alone would create distrust.

There are areas where technology can improve the voting process. Ballots can be hard to produce. Some mechanical systems require expensive, perforated cards (hanging chad anyone?). Machines which create the paper ballots — as long as the ballot is human-readable, can be useful in significantly reducing printing costs and waste. On-demand ballot printing can save enormous costs and planning since there is no guesswork involved as to ballot quantity vs. voter turnout. It can also shorten the time when ballot changes can be made. There have been many cases where due to errors or passed deadlines, candidates who should have been on the ballot didn’t show up on the ballot by election day.

Unless they are paperless, machines which allow voters to select their choices are not inherently untrustworthy since they can’t affect outcomes. They merely assist the voter in making selections. (However, a simple laptop and a web browser can do that too).

The fundamental problem with electronic voting systems is not the ballot but the counting of those ballots. What occurs inside the electronic ballot scanner or counting system can’t be verified by anyone but the manufacturer. It’s simply impossible for the average voter to know if the votes are being counted accurately. They are relegated to depending on the word of government officials and their suppliers to tell them the truth. ("Yes, we fooled you all that time but we’ll be good from now on.")

So what’s the solution?

Go low tech on the counting side of the equation. By manually counting paper ballots, integrity and trust is restored. The time savings and convenience don’t outweigh the costs when you factor in the distrust a closed, unverifiable system creates. For almost 200 years, most elections in the U.S. were handled this way. No, this doesn’t alleviate fraud. It does potentially save billions of dollars to the taxpayer by eliminating unnecessary technology purchases while restoring accountability in the electoral system. Without accountability and transparency in our electoral system, technology additions do not provide any value no matter how persuasive are their advocates.