Skip Navigation
Analysis

Now Is the Time to Replace Our Decrepit Voting Machines

We can’t wait until something goes wrong in the next election.

November 17, 2016

Cross posted with Slate

Although more than half the coun­try may be unhappy with the results, Amer­ica dodged a bullet on Elec­tion Day. That is, our voting machines gener­ally held up. The tabu­la­tions they produced were not so close as to throw the elec­tion results in doubt, and there’s no legit­im­ate indic­a­tion that any were hacked.

In the next pres­id­en­tial elec­tion, we may not be so lucky. With anti­quated voting devices at the end of their projec­ted lifespans still in wide­spread use across the coun­try, the U.S. is facing an impend­ing crisis in which our most basic elec­tion infra­struc­ture is unac­cept­ably vulner­able to break­down, malfunc­tion, and hack­ing. It’s not just an incon­veni­ence. If the machinery of demo­cracy is called into ques­tion, so are its found­a­tions.

Those of us who can recall the pres­id­en­tial elec­tion of 2000 know exactly what can happen when faulty tech­no­logy meets a razor-close elec­tion. The Bush-Gore contest came down to just a few hundred votes in Flor­ida, and butter­fly ballots and faulty punch card machines left us arguing about hanging, dimpled, and preg­nant chads. It left wounds that still afflict the coun­try. In today’s hyper­par­tisan envir­on­ment, such a scen­ari­o—or even unfoun­ded accus­a­tionsof a “rigged” elec­tion that gained postelec­tion trac­tion—­would be far more conten­tious. Just imagine what it might be like in 2020.

Absent a whole­sale replace­ment of our outdated elect­oral equip­ment, this scen­ario is becom­ing increas­ingly likely for our future elec­tions. The prob­lem of aging voting tech­no­logy reaches nearly every corner of the United States, as we docu­mented in a report released by the Bren­nan Center for Justice in 2015. Unlike voting machines used in past eras, today’s systems were not designed to last for decades. Although it is diffi­cult to predict how long an indi­vidual machine will reli­ably func­tion, the experts we spoke with gener­ally agree that machines purchased since 2000 have expec­ted lifespans of only 10 to 20 years. (And for most systems, it’s prob­ably closer to 10.) This makes sense: No one expects a laptop to run reli­ably for more than a decade. Yet on Elec­tion Day 2016, 42 states used voting machines that were at least 10 years old, and 13 of those states used ones more than 15 years old. If replace­ments continue to stall before the next pres­id­en­tial elec­tion, many more will surpass their recom­men­ded retire­ment age.

Perhaps even more troub­ling, these aging machines are partic­u­larly vulner­able to hack­ing. Although the coun­try has made import­ant advances in secur­ing our voting tech­no­logy in recent years, these older devices often rely on unsup­por­ted soft­ware (we found machines still oper­at­ing on Windows 2000) that does­n’t receive the regu­lar secur­ity patches that help protect against modern meth­ods of cyber­at­tacks and hasn’t been through the relat­ively rigor­ous federal certi­fic­a­tion program that exists today. What’s more, many of these systems don’t have a phys­ical paper trails or ballots to back up the results, mean­ing there’s no way to inde­pend­ently verify how voters inten­ded to cast their ballots in the case of a suspec­ted hack. Our coun­try’s patch­work of juris­dic­tion-by-juris­dic­tion voting systems would make it diffi­cult to manip­u­late results on a national scale, but hack­ers could still do consid­er­able damage by tamper­ing with votes in a swing district, steal­ing records to under­mine voter privacy, or just sowing suspi­cion of a larger conspir­acy.

Though voting went relat­ively smoothly this year, a scat­ter­ing of issues that popped up during the elec­tion hinted at what prob­lems may await if we fail to replace aging equip­ment. Voters complained of touch­screen calib­ra­tion errors that “flipped” votes in North Caro­lina, Texas, Nevada, and Geor­gia and interfered with select­ing straight party tick­ets in Pennsylvania. Optical scan machines malfunc­tioned in parts of Michigan and Massachu­setts, and a few in Illinois had to be replaced because a “memory card blew.” Although all of these issues appear to have been resolved by delayed or altern­ate voting meth­ods, that does­n’t mean that glitches like these are unprob­lem­atic. It may never be clear how many people didn’t vote in this elec­tion because of the wait times. In the 2012 elec­tion, between 500,000 and 700,000 failed to vote because of long lines.

These voting machine issues aren’t a surprise. We have heard from dozens of elec­tion offi­cials who say they struggle to keep their aging machines running and that replace­ment parts are increas­ingly diffi­cult, if not impossible, to find. Some even said they have resor­ted to eBay to find anti­quated parts—­from analog modems to dot matrix printer ribbon­s—to keep their voting systems running. Prior to the elec­tion, we surveyed 274 county elec­tion offi­cials in 28 states. More than half of the offi­cials said that they would need new machines by the 2020 pres­id­en­tial elec­tion, and 80 percent of those said they did not know if or how they would be able to pay for the replace­ments.

At least accord­ing to data we collec­ted from four states—Vir­ginia, Ohio, Minnesota, and Color­ado—­Cortés’ sugges­tion proved troub­lingly true. In these states, counties whose elec­tion offi­cials purchased or had near-term plans to purchase new machines had an aver­age median house­hold income of $10,000 or more than those that did not. In Color­ado, we also found an urban and suburban versus rural divide—­counties that replaced machines gener­ally had a higher popu­la­tion dens­ity. If only some counties can replace aging voting equip­ment, it is possible that machine break­downs could dispro­por­tion­ately affect certain voter­s—­namely, rural or work­ing class and poor voters.

Our polit­ical discourse is full of talk of the need for invest­ment in infra­struc­ture such as roads and bridges but almost never includes mention of that infra­struc­ture most crit­ical to a func­tion­ing demo­cracy: our voting system. We estim­ate that the nation­wide cost to update voting machines could easily cost $1 billion—in fact this might be a low estim­ate since repla­cing machines will likely require the replace­ment of other incom­pat­ible systems. Consid­er­ing the size and scope of the federal budget, this is a paltry sum. If the expense is shared with the states, it should be a small lift. Lawmakers can start with a smal­ler, imme­di­ate invest­ment prior­it­iz­ing the aging elec­tronic devices that are, by far, the most insec­ure.

Unfor­tu­nately, to date, there has been a lot of buck-passing, with federal offi­cials arguing this is a respons­ib­il­ity of the states, and with state offi­cials arguing that the burden should fall on counties. But counties and towns have other press­ing budget­ary needs. The truth is that until there are prob­lems, most citizens don’t think about voting machines. They are far more likely to be concerned about whether their roads are paved, the snow is cleared, and their teach­ers are paid.

The good news is that at least a few federal offi­cials and experts are paying atten­tion. Last year, Rep. Hank John­son, D-Ga., intro­duced a bill that would alloc­ate $125 million in match­ing grants for states to replace outdated voting equip­ment. Some—in­clud­ing Secret­ary of Home­land Secur­ity Jeh John­son and a bipar­tisan group of secur­ity expertsthat included former National Secur­ity Agency director Michael Hayden—have stressed the neces­sity of secur­ing and invest­ing in our voting systems, as we do crit­ical infra­struc­ture like the elec­tric power grid and nuclear sites. Others, such as computer secur­ity expert Bruce Schneier, recom­mend that the govern­ment develop processes for detect­ing and respond­ing to malfeas­ance, includ­ing stand­ards for fair resol­u­tion of an elec­tion should tamper­ing be discovered.

Consid­er­ing all that could have gone wrong, Amer­ic­ans were lucky not to have a major contest­a­tion of the results on Nov. 8. We can’t rely on such luck next time. There’s four more years until the next pres­id­en­tial elec­tion—and we need to start think­ing about this prob­lem now, not just a few days before we cast our 2020 ballots.

(Photo: Wiki­pe­dia)