Voting has gotten a lot more complicated in the last couple decades. In the first nationwide election I remember - 1994 - all votes were cast over a 12-or-so hour process on a single day. In person. At a polling place. There were a handful of stray exceptions, absentee ballots and the like, but these were a very small part of the overall tally and usually didn’t affect the outcome of the election. It’s easy to look at the election of 2000, decided on the thinnest of margins, as a turning point - and indeed it was. It’s equally easy to overlook the role the Internet played in complicating our voting lives.

There’s a somewhat tongue-in-cheek rule about bureaucracy known as Parkinson’s Law: work increases to fit the time allotted. As computer automation has reinvented modern work in the past half-century or so, software engineers have developed an interesting corollary: increased productivity from software supports increased bureaucratic complexity. Indeed, it certainly does seem that bureaucracy has become ever-more-complex in every direction you look.

Among other things, voting and election administration have become significantly more complicated in the two dozen or so years since that fateful midterm election. States have begun to support early voting, same-day registration, online voter registration, and so forth. Some states have actually worked to restrict the franchise via voter ID laws, restricted polling sites, and other measures. One key turning point in this process of revisiting the mechanics of voting was the National Voter Registration Act of 1993, which allowed voters to register to vote while renewing their vehicle registrations. A more significant turning point was the Help America Vote Act (HAVA) of 2002, passed in the wake of the 2000 election crisis. It set standards for election administration and required states to update their machinery, among other provisions; it also, arguably, complicated the registration process.

The ensuing two decades or so have been an arms race of voting reforms and regressive measures, meant in some cases to shape the electorate in a manner favorable to one party or another. This year, the coronavirus has created so much turmoil that its effects dwarf perhaps all of those previous efforts put together. A combination of new state reforms, voter enthusiasm, and social distancing measures mean that the 2020 early vote has blown the 2016 early vote out of the water.

At the same time, it’s hard to deny that new technologies have made this complexity and acceleration of reform possible. HAVA mandates statewide voter registration databases, and it’s difficult to imagine that such things could have existed 50 years ago; in some places, they would have been difficult to imagine even in 1994. At the same time these databases enable further reforms. Imagine how difficult it would be to administer an election in which voters could vote either early, or in-person, if the administrator did not know who was registered to vote and whether those people had voted early! Little wonder that early voting was first implemented in Washington, DC - which is so small that voter activity can be readily communicated to all precincts in relatively short order, even by sneakernet. Vote-by-mail and same-day registration entail still more challenges, and would be nearly unthinkable in a paper-and-pencil-only environment.

Not all bureaucratic complexities are created equal: some tend to expand the franchise while others restrict it. If we’ve learned anything this year, it’s that some complexity is necessary simply to keep the pace with all the challenges of modern life! So in a way we have Parkinson’s Law to thank: without the complexity enabled by modern databases and the Internet, it’s difficult to imagine just how we would have pulled off the 2020 general election, let alone kept pace for what appears to be record turnout. Of course it’s equally true that voter ID laws and other such restrictive measures are a dark side of this same law. But it’s certainly too simplistic to bemoan the complexity increase which software automation has given us.

All that being said, with greater complexity comes the opportunity for greater confusion. A huge part of the progressive tech sector, and a corresponding chunk of the political sector at large, is concerned with helping citizens navigate voter registration regulations and to ensure that they can take full advantage of the franchise they have. Private groups like BallotReady, Vote411, as well as heavyweights like Google and Facebook have released tools along these lines. States and municipalities have joined in the act as well, with efforts ranging from Pennsylvania’s Online Voter Registration API to the multi-state Electronic Registration Information Center (ERIC).

If all of this complexity and anti-complexity is sounding a little familiar, that’s because it should. There’s another process which is universally accessible and infamously complex: taxpaying. There’s an entire cottage industry of tax preparation professionals who seek to make paying taxes easy. Tax prep services are estimated to be worth around $10 billion, which is coincidentally about the amount of money that will be spent on politics this year. There are also numerous efforts on the part of government to make taxpaying easier, such as the 1040EZ form and IRS electronic payments. Indeed this similarity helped inspire the name of civic tech innovator TurboVote, which guides voters through the registration process in the same way TurboTax guides taxpayers through the tax filing process.

The experience of tax simplification has a few interesting lessons learned, which are hopefully useful to progressive tech innovators:

  • We’re not done with tax simplification by a long shot; indeed, Elizabeth Warren once suggested that in many cases, the government should be able to write your tax return for you. If that’s the gold standard for simplification of the tax process, what’s an equivalent in the voter registration process - automatic enrollment? Is that possible?
  • Simplification is a double-edged sword, and it can have a progressive or regressive impact. Infamously, the purportedly-simple flat tax would actually wind up raising taxes on the poor. Put another way, complexity in the election administration process may have an enfranchising impact.
  • Acts of government can have an impact which dwarfs that of the private sector. Automatic tax filing could extend the Earned Income Tax Credit to millions of families; programs like ERIC have touched tens of millions of voter registration records.
  • At its best, private industry can be a kind of proving ground or laboratory for more universal programs adopted by government - as Pennsylvania’s online voter registration API surely was. At its worst, regulatory capture can stall or even stop enfranchising mechanisms.

Taken together, I think it’s clear that there’s a lot of work for progressive tech innovators to do in expanding the franchise; and moreover, in enabling and nudging governments to do the same. Let’s hope that we’ve moved the needle a good way in that direction by the time the next general election comes around!