Elections reform

For a Voting Rights Act of 2007, covering everybody’s right to vote and have that vote counted.

In comments on my “Democratic Agenda” post (scroll down two), Bernard Yomtov lays out an agenda of voting-related issues for the Democrats to push early in the new Congress:

*Require states to allocate polling places and equipment in such a way that voting is roughly equally convenient statewide, in terms of likely waits, access to polling places, etc.

*Prohibit the use of any voting system that did not provide a paper audit trail.

*Make it a felony to deliberately misinform voters about voting places or days.

*Extend the do-not-call list to cover political robocalls.

*Establish expedited methods for dealing with problems that arise on voting day or shortly before.

*Require a serious non-partisan analysis and testing, and appropriate certifications of voting equipment before it can be used.

In general, I think a Voting Rights Act of 2007 would be a great idea. As I understand it, there are Constitutional limitations on the capacity of Congress to dictate vote-casting-and-counting practices to the states, but almost anything can be achieved by tying it to Federal funding. (The lack of adequate numbers of highly competent pollworkers is a major problem; I’d propose to solve it by offering a $500 federal scholarship to any college student willing to undergo training in local voting procedures and work the polls at no fewer than three elections.)

I wouldn’t back Bernard’s plea for banning political robo-calls; instead, how about a ban robo-calls of any kind that call back on hang-up, unless the recipient of the call has consented in advance? (False-flag repeat robo-calls of the kind the RNCC spent millions of dollars on this year are already covered by laws against use of the telephone to harass and annoy and against the use of interstate communications facilities to defraud, and if the Justice Department isn’t aggressive about pursuing the cases from 2006 the authorizing and appropriations committees should ask why, and keep asking.

Given the distances rural voters have to drive, it would be hard to come up with standards that would really make voting equally convenient everywhere. An alternative would be to require that every precinct have paper ballots available for those who want to use them at any time when the wait to vote exceeds 30 minutes.

To the oft-heard query “Why don’t we just go back to paper ballots?” there are four good answers: they’re not adequately handicapped-accessible; they’re not actually hard to cheat with (although the cheating is mostly decentralized; they suffer from high rates of voter error, leading to many votes not counted; and they would extend the counting process for at least hours and perhaps days. Yes, lots of countries use paper ballots, but none of them elects dozens of officials at once or uses plebiscites routintely, as we do.

But all of the advantages of paper ballots except their being cheap can be matched by touch-screen voting, as long as the machine prints out a paper ballot for the voter to drop into the ballot box. The voter can review the ballot to make sure that the names the machine printed were the same as the names he voted for, and the machine-printed ballots can be made machine-readable. Yes, it would be possible to futz with the count, but the physical ballots would be right there, subject to spot-audit simply by running a random sample of precincts (selected after the first tally has been submitted) through a different counting machine located centrally, and subject to subsequent review in case of a recount. It would be very hard to alter a machine-printed ballot undetectably, and the rule should be that the ballot counts regardless of any stray marks, so every vote cast would count.

I have been told &#8212 I don’t know how reliably &#8212 that one voting-machine vendor actually offered such a machine, but didn’t make many sales, and that one of the big three vendors bought up the company and discontinued the product line. If that’s true, the patent covering that technology looks like a good candidate for taking by eminent domain and placing in the public domain, so that multiple companies could compete to produce touch-screen paper-ballot systems.

This is a case where the Democratic fundraisers (such as Steny Hoyer) have to be told that no amount of money the voting-machine companies could give deserves any weight at all compared to the other interests at stake here. To solve that problem more generally, we need to fix campaign finance, but that’s a topic for another post.

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact: Markarkleiman-at-gmail.com

24 thoughts on “Elections reform”

  1. Actually, we have a new (to New Mexico) system that uses optical scanners. You mark your ballot (with a provided felt tip pen), and when you're done you take it to the scanner. The machine reads the ballot, and if there's an error (overvote, stray marks it can't interpret, etc.) it spits it out as a spoiled ballot. If everything's okay, the scanner raises its vote tally by one and stores the ballot internally.
    The voter then has some security gyrations to go through, and at the end gets a clean ballot. Let's hope John D. Voter gets it right the second time around.
    To handle the handicapped accessibility problem, every precinct is equipped with a touch-screen machine that prints out a marked ballot to go into said scanner. Every ballot goes through the scanner.
    If the scanner's displayed count doesn't match the number of ballots in the machine, there's some form of fraud. More than than, the system is completely auditable by hand.
    It worked pretty well in New Mexico this time. We had a few problems with enough ballots in the Albuquerque area (someone ordered 140 ballots instead of 1400, and no one in the County Clerk's office caught it. Said County Clerk is our incoming SecState, and my head is already hurting…). To my knowledge, it was the first time since the introduction of machine voting that we've had a uniform system statewide.
    BC

  2. Thanks for the endorsement and kind words, Mark.
    You're right about the rural voters, I guess. I was thinking more in terms of having conveniently located polls in urban areas, and of delays once voters arrive at the polls. There really is no reason why voters in some precincts should have significantly longer waits than others because of an inadequate number of machines or similar issues. It's not a difficult problem to solve.
    As for robocalls, I frankly find them annoying even when they are for candidates I support. And of course the false-flag callers definitely need to be prosecuted vigorously.

  3. The machine I voted on printed a paper "receipt" that didn't come out, but was visible beneath a clear plastic shield. The machine told me to check it before finalizing my vote.
    I think VA must use those in some precincts, b/c a pic of the Senate canvassing showed a guy with one of them.
    Is that as good, do you think, as the paper printout that you call for?

  4. Re robocalls:
    It would help if all political calls had to have the "paid by" sentence at the beginning of the call.

  5. San Francisco also has the optical scan in the precint ballots. I haven't heard one complaint about them, except that it's a pain when you deliberatly undervote.
    I personally undervote rather than randomly choose a person when I have no information about the candidate. The poll workers ask you if you meant to do that. You say Yes, and they put it through anyway.

  6. I undervoted my ballot. I didn't vote for SecState, because I knew that the D candidate was an idiot who's screwed up every election she's ever run, and I couldn't vote for any Republican for SecState. Not after the crap other R SecStates have pulled.
    The machine took it happily, didn't make an issue of the undervote at all. They must be able to tell it to check for undervotes.
    BC

  7. NY is the last state to go electronic. I voted on an old mechanical machine. I'm also a rural voter & would say that the best solution is to say you have to have a polling station in every population center of over, say, 300 people. In any case, I agree about establishing a paper trail. As NY finally comes into compliance with current law, I for one will be agitating for a recountable paper ballot in some form. This is fundamental.

  8. The problem with printing out a paper "thing" lies in the fact that printers are pretty unreliable mechanisms, particularly when they are used as heavily as a voting-machine's printer would be. This is probably the reason most voting machines don't have printers.
    The New Mexico system (with paper ballots that are then optically scanned), gets around this problem while still achieving the goal of having an auditable piece of paper. We also use these in Cambridge, Massachusetts (and I assume throughout the state).

  9. Absolutely our system of voting needs a substantial overhaul. But the love affair with paper really overstates its reliability. What you are really calling for is for keeping the transactions, the sum of which comprises the totals.
    In my recent voting experience, the biggest problem was the check in process. This was hands down the cause of the lines. Add to this, the inexperience of the poll workers (no surprise), the physical segregation of each precinct check in (so 100 people are waiting at one table and none or few at the others) and the lookup and sign in lists, and you have the makings of a long line. This process needs to be redesigned. If I could think of a reason to create a voter id card, it would be to swipe, perhaps enter a pin and sign electronically.
    As to the receipt issue, I seriously doubt the monster ballot I voted on Tuesday would lend itself to an effective receipt.
    Ironically, one of the hated names in the voting machine industry, Diebold, has another product line ATMs, that have to conform to a lot higher standards. However, ATMs might be a good model to start with in terms of building integrity and controls into a system. I was a bank operations and systems auditor for 8 years during the '80s and am familiar with the technology and controls.
    Elections need to be auditable and every election should be routinely audited. However to do this, there's a far better chance to do the job systematically if the data is in electronic form. When I hear people talk about pulling random samples of paper ballots, I snicker. To pull a paper based random sample, the ballots have to be uniquely numbered, and then a list of random numbers needs to be generated. You then get to go through all the ballots to find the ones with the lucky numbers. If you are really lucky, they may actually be collated in a workable sequence. Trust me, it is a PITA.
    We as a people are going to have to be willing to foot the bill for substantial changes. There really should be complete reviews of voting technology, and the certainty of effective audits. We are also going to have to find ways to prevent the government entities from hiring the "usual suspect" consulting and private firms to create the systems.
    And YES, ELECTION DAY NEEDS TO BE A HOLIDAY!

  10. The real answer, for 5 of 6 of the mentioned problems:
    Vote by mail, with a human verifying the signature on the ballot. Oregon has offered to help you figure out how to do it, if you find it a difficult concept to grasp.
    Saves money, provides an audit trail, easy, low-pressure, etc. It's a no-brainer.

  11. The standard objection to voting by mail is that it allows coercion by someone who demands to see your ballot before you mail it in.

  12. A ban on political robo calls is a very bad idea. The cost of a call is around $0.05. A piece of mail in the voters hand is about $0.60, depending on production costs. The penetration of the robo call is substantially higher than the mail piece. Robo calls are the most cost effective means of indirect campaigning. Low budget campaigns would be very difficult with out that tool. Keeping the cost of campaigns down is an urgent public need if grassroots politics is to survive. The annoyance of a phone call every couple of years is a rather small price to pay.

  13. Rick G wrote, "I was a bank operations and systems auditor for 8 years during the '80s and am familiar with the technology and controls."
    But voting and banking are not similar, because voting is conducted anonymously, and banking is not.

  14. >>The standard objection to voting by mail is that it allows coercion by someone who demands to see your ballot before you mail it in.
    And those so cowed they would give in to such demands are going to suddenly grow spines in a voting booth?
    People have been voting by mail for quite a while, and the "standard objection" doesn't seem to have anything to do with reality…

  15. Paper ballots
    Weekend voting
    Any interference with voting integrity is a felony, including misinforming voters, discarding registrations, changing registrations, discarding ballots, intimidating voters.
    It's time we take voting seriously. As we've seen, election results are a matter of life and death.

  16. "But voting and banking are not similar, because voting is conducted anonymously, and banking is not."
    All true. However, to maximize anonymity, we should theoretically shred ballots as soon as they are counted.
    What the ATM exemplifies is an automated process which handles money, that by design and transaction trails can largely be trusted. For voting the transaction trails would need to be designed to support anonymity. The point of the original post is that those advocating the paper trails do so for a transaction trail and that machines can be designed to support that aim.
    BTW, I also agree with those posters advocating mail in ballots. They certainly deserve consideration.

  17. I thought the standard objection to mail in ballots was the possiblity that a substantial fraction of them might bypass the registered voter entirely. And then have their signatures checked by the very people perpetrating the fraud.
    Of course, checking to see whether this has happened, by visiting a random sample of people recorded as having voted absentee, and asking them if they really did vote, seems to be regarded as a form of voter harrasment… I suppose because under our current system of partisan election management and challengers, it would, of course, always be members of one party going around checking in precincts dominated by another party.
    There's the real weak underbelly of the system: Elections run by partisan officers of politial parties.
    I think what we really need is a kind of "election corps", recruiting people who would be trained in running elections, and randomly assigned to precincts on a rotating basis. This would make fraud by elections officials extremely difficult to pull off, as most forms of it require colusion among poll workers.
    Of course, it would also require some form of voter ID, as you'd no longer have your ballot handed to you by local senior citizens who recognize every last voter in the precinct by sight… So, secure biometric voting ID, provided at no cost to particular voters. We can afford it.
    Finally, any form of absentee voting pretty much sacrifices the secret ballot. Absentee voting should be minimized, and we should spend the money to send polling teams around for a week before the election, physically visiting the home bound with portable voting machines.
    And, yeah, election day a holiday, to quiet the whining about having to visit the polls in person.

  18. Make Voting mandatory:
    As well as the controversy over vote technology if we HAVE to vote the turnout would blossom. Tie it to an income tax penalty if you don't. Public transportation for everyone and as suggested above a national holiday on election day. Combine this with federally funded campaigns and losing the K street crooks altogether and we might achieve a government 'of the people and for the people'.
    What a concept!

  19. American readers may be relieved to hear that the international observer mission from the OSCE found the elections reasonably fair – but they are worried about the lack of paper trails too. http://www.osce.org/item/22010.html
    The mission was unfortunately a token operation, far too small to be serious. 18 observers visited poling stations in 15 states, not including Florida (I wonder why). In contrast, the joint mission for the spring parliamentary elections in Ukraine fielded 900 observers.
    Well-run and professional observer schemes can make a difference and can easily be included in election reform. I would myself go for a hemispheric approach rather than Atlantic, bu that's a detail.

  20. I don't know how other states do it, but here in Oregon keeping the paper ballots indefinitely doesn't compromise anonymity. We mail our ballots inside two envelopes. The outer one contains the voter's identifying information, including signature. Once the signature is verified the internal envelope is removed and separated from the identifying info. After that the ballots are counted by machine. At that point and thereafter there is no way to tell who the ballot originally came from.
    Sure, if you have massive fraud by the people handling the ballots it would be theoretically possible to rig an election but no system is immune to that.

  21. "At that point and thereafter there is no way to tell who the ballot originally came from."
    And therefore no way to correct the count if fraud IS discovered after that point. I think we need ballot tracablity, and we can use some kind of third party key registry to prevent the usual problems you might expect from the government knowing who voted how.

Comments are closed.