Truck and Barter

E-Voting Issues At Wired

There are a couple of good articles on issues surrounding e-voting up now at Wired:

An Introduction to E-Voting

E-Vote Guidelines Need Work

The Introduction covers a number of things that have been brought up around these parts before.

One of the most alarming things from the "Guidelines Need Work" story is this bit:

"One problem with the Diebold code was that it had large, complex multi-logic statements with no comments (from the designers)," Rubin said. "That wouldn't pass this standard."

I know of few professional programmers that would ever let a serious piece of software out of development without proper commenting. The lack of it indicates sloppy work that was likely done by third-rate teams with more interest in getting out the door than getting it to work. Which means there is more than one significant problem. First off, that the code was allowed into use uncommented means that the writers of the original guidelines were beyond incompetent. Secondly, that Diebold allowed such shoddy work out its door indicates that the committee that is purchasing Diebold's services has chosen the worst example possible of "the lowest bidder". The question of whether or not the code was usable is entirely beside the point. More important is the fact that guidelines are being desigend by woefully uninformed political agents with little incentive to do anything other than "looking like they're doing something".

This strikes me as a political-committee created problem being addressed by yet another political committee. By adding paper-trails to electronic voting machines, you now add in the need to stock paper, reload machines, and, most importantly, require that the printout be easily verifiable by the voter. The original problem was that the ballots -- in a design approved by political agents unrepresentative of the entire population -- were often confusing (the troubles with lining up names and boxes in "butterfly ballots", for example). If the new design has to make both the machine and the new paper easily verifiable to someone for whom the paper ballots were confusing, then you've simply added a new level of useless complexity. Why not spend the money on making better paper ballots? The paper trail would exist, less money would be spent, less staff would be required at polling stations, and the question of hacking computers would be eliminated.

Meanwhile, the continued distrust of computer voting that results will certainly undermine voter confidence the only real leap I can see as worth the effort: the choice to vote online. (N.B.Yes, this is rife with problems, but the reduction in transaction costs, the spread of "voting places" to every library with a modem, while retaining paper ballots for those without access to a computer would achieve, I believe, enhancements in turnout that would be worth the cost of supplying something like unique bar-code printouts for each voter so that a quick recount could be achieved in the case of hacking/fraud. I still think the turnout we have now gives us a very good estimate of the true population, but there is certainly nothing wrong with more people voting.)

File Under: "Government, Proactiveness of"

Voting machines must include a verifiable paper trail and audit capability in time for the 2006 elections, according to a bill introduced this week in Congress.

The big spender wins again.

The Hindu, one of South India's most popular newspapers, reports that the Congress ended up spending Rs. 125,43,33,247 (or $27.9 million using exchange rate calculation) on the general election held earlier this year. This is the highest any party has ever spent (I adjusted for inflation and checked... its true) on an election campaign.

That kind of money is huge... the story goes on to list out the various cost breakups, and reading about these wastages just becomes a more disconcerting ordeal.

While few argue the almost unitary correlation between high spending and election results, I'm beginning to have my doubts on Democracy:

1. It costs too much.
2. The money comes from sources that are too few in number.
3. Refer to point 1.
4. Refer to point 3.

What worries me is that this trend is especially dangerous for a country like India, where such large amounts have far higher utility spent elsewhere. Using my own 'Big Mac' adjustment, I reckon that the money spent could well be worth $97 million dollars (A Big Mac costs about 54 rupees in India, and about $4 in the US). That is *still* one HECK of a lot of money.

You can read the original story here.

Update:
Yazad was kind enough to point out that I had got my Big Mac prices wrong. He makes the point that its not a Big Mac but a 'Maharaja Mac', and costs Rs. 54 Rupees not Rs. 20. I have corrected the values in the post. He also makes another point in his comment which I will probably write about in a seperate post. I still stick with the "one heck of a lot of money" statement though :-)

Rational Voters

It's been a few months since my last post here at T&B as I have been busy with this semester's work and the election was somewhat of a distraction. The results the other day were a pleasant outcome for me made all the more enjoyable by watching it in an auditorium with about thirty or forty other grad students. The beer and wine were great but some of the comments coming from other attendees were hilarious and made the evening that much more enjoyable( BTW, I am not one for giveng in to any peer pressure and proudly proclaim my preferences, so there might have been some rubbing it in at the end of the night). As you might expect, the phrase "how could these people be so stupid" was muttered once or twice.

I'm always interested in how voters make decisions and what incentives are in place to guide them. Are voters really as stupid as the people last night and those around the world like to proclaim? Or do they in fact make rational decisions that work in their best interest?

As education is the achilles heal of the Republican Party, it seems that anything which could be done to mitigate it as an issue and, also, weaken the education unions which are part the Democratic establishment, is in its best interest. Charter schools are a step in that direction, however, G.O.P. voters in Washington don't seem to agree. Jim Miller has interesting analysis why(via Joanne Jacobs):

Rural Areas And Charter Schools: Washington state voters again rejected charter schools, voting against Referendum 55 by 58.5 percent to 41.5 percent. The reason is not hard to see. Many Republican leaders backed it, but rural Republican voters did not. Take a look at the current results, by county, and you will see what I mean. In little Garfield county, just 1,077 voters had voted on the issue, when I checked this afternoon, and just 35 percent of them had voted yes. The same county was giving 735 votes to the Republican candidate for governor, Dino Rossi, who supports charter schools, and just 387 votes to his Democratic opponent, Christine Gregoire. If you look at other rural counties, you will see the same pattern, strong opposition to charter schools and strong support for Republican candidates, many of them favoring charter schools. If rural voters had given the same support to charter schools that they did to Republicans, Referendum 55 might have passed.

I hope that I do not disappoint some of my friends by saying this, but I think this combination of attitudes by rural voters is rational. They are right to support Republicans, who are far more likely to listen to their concerns. And they are not wrong to oppose charter schools, which offer little positive to most rural areas.

The great advantage of charter schools, in my view, is that they introduce competition into education. But to have competing schools requires a density of population not found in most rural areas. My little high school had a mere 130 students when I graduated — and several very long bus routes. (For some from big cities, I should add that 130 was the number for the entire school, not the senior class.) And it was that large only after two towns consolidated while I was in junior high school. It was difficult for the two towns to support a single school system; it would have been terribly difficult for them to support two systems. And if a charter school tried to serve a large rural area, it would find that many of its students had to spend half the school day traveling.

And that isn't all. A charter school in a rural area would threaten the existence of an existing school system far more than it would in an urban area. A large suburban school can lose hundreds of students to a charter school and adapt; a rural school would be badly damaged by the same loss. This possibility would dismay most rural communities, where the schools are, more often than not, their psychological centers.

One for the reading list.

A friend of mine, with whom I have engaged in a number of debates on this very topic, pointed me to a new book on the electoral college: "Why the Electoral College is Bad for America," by George C. Edwards III.

Now on order from Amazon, rest assured I'll have some thoughts on it once it arrives.

From the review alone, I am skeptical on one point: the reliance on the anti-EC arguments from James Madison. While I have a great and enduring respect for the framer's of the Constitution, and in particular the brilliance of Madison, I think there is a tendency to look to these men as some form of oracles, from whom never ushered any poor judgements or miscalculations. These were men as given to political ambition and desire as anyone. In fact, it is precisely because they were flawed that I am even more in awe of their achievement.

In considering the justifications and decisions of the framers, I tend to give more weight to the early work of Madison, during the creation of the basic institutions of the US than to his arguments during and after Washington's first term as president. While the inciting incidents were surely the programs suggested by Hamilton, it was Madison the organizer that adroitly split the political sphere into warring factions and drove hard to oppose anything that seemed to displease Jefferson. The focus of his efforts shifted heavily, it appears to me, from the success of the foundling nation to the political career of his mentor. This just a few years after having written The Federalist letters in support of the very things Hamilton was pushing for. Jefferson's student in almost every respect, Madison quickly moved into the role of a political power-broker, working tirelessly for the support of Jefferson above all. That he came out in repudiation of the electoral college, after having given it such eloquent support strikes me as a sort of political maneuver designed chiefly to get and keep Jefferson into office. After all, the election of 1800 provided probably the first real test of the election system, with Adams and Jefferson having to rely on deals in the House of Representatives to decide the winner. Meanwhile, if Madison were truly honest about his push for a truly "democratic" election, why not attack the appointment of Senators by the House of Representatives? Obviously, I'm no historian, so there could be other sound reasons that Madison chose that time and that subject to attack; I just harbor suspicions about the reliance on Madison as clear evidence that even the "father of the Constitution" was against the electoral college.

That said, I'll suspend any judgement until after reading it. At which point I'll feel free to be as judgemental as I please.

Game Theory of Voting

In perusing web writing on issues about the Electoral College, I ran across two pieces at Slate, both by Jordan Ellenberg, I thoroughly enjoyed.

Part I: Vote!

Part II: Game Theory for Swingers

The first piece makes, in part, the point I was driving at below: that one of the effects of using the Electoral College is to make a voter, both in actuality and personal expectation, more likely to be decisive in determining the outcomes of an elective process.

Where to allocate campaign efforts as computed through game-theoretic strategies is the subject of the second piece. In a roundabout way, this continues to make my point. As the probability of winning in certain states approach 50-50, one would expect to see more campaign resources allocated to precisely those states. If we make the effect of campaigning dynamic (rather than fixing a certain percentage point gain or loss tochances of winning the state) as well, then it demands that a candidate further divide efforts among states. Concentration of electoral power into a few cities reduces the number of places that would be of interest to candidates. Additionally, assuming that current trends in voting patters for the major cities continue, the candidate that has a high probability of winning the largest cities need spend little time campaigning at all. All the time and money spent going to midsize cities could easily be drowned out by the relatively cheaper efforts made by a candidate in big cities.

Read 'em all, as they say.

Electoral College Questions

As painful as it is to relive, it seems like people are preparing themselves for another round of "electoral v. popular" vote issues. This time, however, the positions are reversed: Zogby, among others, suggests that Bush may lose the electoral college but win the "popular vote."

(NB: I still don't believe there is such a thing as a true popular vote in this country. Since we still operate under the electoral college, and since people understand this is the case, the representativeness of the "popular vote" as derived by simply totaling up all the people who voted is questionable, since the incentives to vote would -- in my view -- significantly different.)

I have to admit, I'm starting to think the possibility of a 2000 repeat is entirely possible (though I think Ohio, Wisconsin or Nevada might be the place to look this time, depending on the electoral votes needed to hit the magic number). I differ from a number of people that I've spoken to, however, in that I don't find this a terrible problem. What I worry more about is the unrelenting press coverage, the unending debates about lawyer tactics, and snide comments from both sides about vote-tampering.

More importantly, this could well bring a referendum on the electoral college itself. And on this, I happen to be partisan: I like the electoral college, and think tthat it should be retained at best, and only slightly modified at worst.

Among the arguments I hear on scrapping the system (and there are plenty others -- for another time) is that the current system forces the candidates to play favorites among the states, driving them to the places where things look tied up, rather than looking across the entire country. Which is true, I suppose. In every election I can recall, the importance of national polls has always taken a backseat to the question of performance of "Battleground States". As we can see by just turning on the news, the candidates are once again in Ohio, Wisconsin, Nevada, Pennsylvania, and others where the polls show nearly dead heats.

What are we offered by the alternative? The immediate reaction would be for presidential candidates to focus solely on population centers where the cost of reaching the next voter is smaller because of TV access, transportation costs, size of audience issues, etc. Effectively, a directly popular vote would turn us into a single nation of 100 million voters, rather than subdivisions of far smaller groups. While we all know that voting makes no rational sense since it takes time and money to do so, and the chance of the vote being decisive in any way is vanishingly small, people vote anyway. It's an emotional choice, not a rational one, and I'm fine with it being so. I like to vote, no matter how many times people show me that it's a waste of time. Apparently, there are millions of folks who agree with me.

If this is the case, imagine what the effects are of saying to the middle of the country that their votes are now being pooled with the major population centers of the US. When Miami, New York, LA and Chicago are orders of magnitude larger than numerous states combined, there would be no reason to even feel good about voting, since candidates would spend the majority of their time talking about issues of concern mainly to urban populations. I can imagine entire swaths of the country becoming even more disaffected in relation to national politics since they could see that their future is tied to whatever the large cities choose.

Taking the process as slightly more rational, removing the electoral college still acts as a disincentive to voting. The chances of a single vote being decisive are chiefly a function of the size of the voting population (as well as choices available; here I'm still just assuming two, though with more the effect is roughly the same) . You have a higher likelihood of casting a deciding vote when you are one of three than when you are one of a thousand. Separated into states, voters are part of a much smaller population than they would should the country vote as one large mass. If more people vote the more likely their vote to be decisive (or even the more they believe their vote could be decisive), then moving to a mass vote would tend to drive down the number of people voting since that marginal person went from being one of 3 million to one of 100 million.

I'm not convinced that, on grounds of efficiency alone, we want to depress the number of people voting. We don't get much more "democracy" when more people vote than we have currently, but I don't see the need to induce flight if we get little benefit from it. The objection I hear, of course, is that the president should be focused on doing the most for the greatest number of people, which is made possible by letting population centers drive elections.

I tend to recoil from this idea simply on historical grounds alone. The arguments made by the framers of the constitution make it abundantly clear that there are real problems with not only mob rule, but rule by minorities as well. Republicanism versus outright democracy isn't a distinction without a difference in my view. The electoral college, it seems to me, was devised as a balancing act between these two potential outcomes. While areas like Miami, NYC, Chicago, LA, Atlanta and a few others do house the most people, they don't necessarily cover a full range of concerns for te US. Urban populations face very different issues, and thus have very different preferences on domestic policy, than do rural populations. The weighting system of the electoral college attempts to find a point where the size of "large" (population-wise) states doesn't entirely swamp the "smaller" states. Voting through the electoral college strikes me as a way to get a representative sample of the preferences over various issues of concern to the entire US.

While it may seem like we're all talking about Florida once again, the makeup of battleground states tends to change over time. Colorado was of no real concern last time around. But what I find more compelling is that the characteristics of the battleground states represent a vast array of concerns that may be an accurate reflection of the country as a whole: the concerns over the manufacturing sector in Ohio and Pennsylvania, farming and taxes in Wisconsin, immigration and the war in Nevada, etc. and so on. No, it's not perfect. But the alternative means presidential elections predicated on almost purely urban issues. Candidates being forced to deal with a small subset of states strikes me as as a better (though still not perfect) outcome than dealing with only a few important cities.

In whatever kind of split may occur, I take the electoral vote tally as the only significant outcome, and have no problem with a president who did not win the "popular vote". At least, no problem in terms of how he achieved office...

I will say, however, that I never did understand why electors aren't held to voting the way their states request. The "faithless elector" problem strikes me as real and unnecessary -- though if someone has a decent explanation, I'd be glad to hear it.

That depends on your definition of "disproportionate"...

In a lot places, ex-felons can't vote. I'm not sure of the strength of the arguments for and against, really, since I don't see it being of very much importance at all.

Aside: Take a look at who votes. Now take a look at the demographics for those who commit crimes. How many likely voters are we really losing?

But I did think this NYT article on how felony voting rules affect the black population was a bit odd in its discussion of the issue. Not that loose association with definitions and associations is anything new for the media...

The studies, the first to look at felon disenfranchisement laws' effect on voting in individual cities, add to a growing body of evidence that those laws have a disproportionate effect on African-Americans because the percentage of black men with felony convictions is much larger than their share of the general population.

Uhhhmmm...if the law says anyone that has a felony conviction can't vote, it falls on each felon in direct proportion to their criminality. One felony, one revocation of the ability to vote. How can it fall "disproportionately" on a group when it's directly in line with the number of people in the group that have been convicted of a felony? Now, the issue of incidence of felony convictions among a certain group of people is another matter entirely.

Ex-felons also have to report being convicted of a crime in greater numbers than those people who have not, in fact, been convicted of committing a crime. This doesn't have any bearing on the distribution of race, gender, age, educational attainment, or income of those people. The problem, to me, is the erroneous conflation of a result with a cause.

Voting Incentives Question

When I read articles like this one on the numerous and subtle shifts in leads among candidates, I'm always left wondering if the writers (and the pundits and campaign pros they discuss) are really getting what's happening when a poll starts to show a "shift".

So I'm torn. Are people changing their minds as new things come to light? Or are polls picking up a solidifying base of people that were either 1) motivated to answer a poll since it's close enough that folks aren't going to respond with the "how early do we start celebrating Christmas" sensation, or 2) getting a larger and more accurate response rate from "likely voters" because people are becoming solidified in decisions that were not yet set? Or some factor not mentioned?

(The above is referencing an amount of people big enough to register as a shift in repeated polling. Certainly, some people are changing their minds.)

I'm rather adamant in my belief that polling is highly flawed largely because of their prominence. They tend to affect the people their meant to sample, the responses are far from accurately given, and the questions are often far from neutral.

I'm just not convinced that much of the campaigning is "swaying" voters from one side to another, so much as it might be inspiring enough to get people to answer "yes" when the monotone voice says "are you likely to vote in November." If this is true, then it might indicate that negative campaigning plays a bigger part than we'd like to admit. Maybe I'm off-base, but I tend to think outrage over an issue might be a bigger motivator than mild indifference. And if that is so, then the best place to start investing money in the campaign is wherever it appears you could motivate a group by giving it something to vote against. (For instance, announcing pro-choice candidate votes in highly conservative areas. Or highlighting spending profligacy to those who consider themselves highly fiscally responsible.)

The old adage is that the only way to make the choir sing is to preach to them. In this election year, when the outcome is expected to be close, each vote is more and more dear (as the decisive group gets smaller and smaller in expectation, each vote nears the point where it is viewed as pivotal, and indeed may well be), a lot of money is being spent on "shoring up the bases" on both sides. But perhaps the real gain is on those people who haven't yet heard something pro or con about a candidate in their particular area of interest?

I know commenting isn't frequent here (and that's perfectly fine as long you come back -- we like having you, loud or quiet) unless you want to talk about Iraqi Dinar, but if anyone has some thoughts on this, I'd love to hear them...

Bush or Kerry: Who's the market better for?

Most investment shows I've seen (and yes, I watch a lot of cable news, so I happen on these on weekend mornings), are studiously trying to predict the results of the coming presidential election. The thought is that, for varying reasons, one candidate might be better for markets (or certain portions of "the Market") than the other.

Barry Ritholtz, chief market strategist at Maxim Group, suggests, in a BusinessWeek Online interview, these folks may well have the causality reversed:

Q: How do you think the upcoming Presidential election is affecting the stock market? A: The markets are impacting the [race]. The capital markets act as a future discounting mechanism, anticipating economic conditions 6 months to 12 months into the future. Politics matters far less to asset managers than whether the economy is expanding or contracting. Markets may not always get it precisely right, but they come close enough that it's wise to pay attention to what they're saying.

I think this is clearly one of the implications of findings from such things as Ray Fair's model for predicting elections: it's not only asset managers for whom the economy matters more. I do think there is some circularity in the whole issue (not in Barry's argument). The markets may drive who ends up in the oval office, but aside from the folks doing the "heavy lifting", portions of those markets are driven by people investing based on who they believe will win (and thus are attempting to capture the best future returns based on their belief about who will be best for the market):

Conversely, when investors expect a slowing economy, which means weaker revenue and lower earnings, they become less willing to "pay up" for earnings. That's what we have seen since January of this year -- the markets became less willing to pay a premium for future earnings.

The difference, obviously, is that the people Barry mentions often have far better information than the average investor.

Since so many people "vote their wallets", it's no small issue. But this could be changing. This may be one of the few times an issue of foreign policy trumps the economy in November. More from Barry:

Q: So you think Kerry will win? A: Here's where things get tricky: Once all the quantitative data is in -- and assuming there's no "October Surprise" -- I look for an analogy with another, historically similar period. This includes economic data -- interest rates, taxes, unemployment, inflation -- as well as geopolitics.

What makes the 2004 election such a challenge to forecast is that we have never seen a Presidential term with a burst market bubble, a recession, a major terrorist attack on U.S. soil, a big tax cut, and not one, but two, wars. So without an analogous comparable, making a prediction with a high degree of confidence becomes quite problematic -- it's just a crapshoot.

If you won't let me weasel out of giving an answer, then I'll fall back on quant work. All four data points [from earlier in the article] suggest the incumbent gets defeated in November.

Those Stupid Voters...

Election cycles bring out a few certainties: campaign ads that call the other side names for having used negative campaigning, incessant prophesying about 'electoral math', and journalists (re-?)discovering the oddities and issues inherent in the simple act of voting, among many others.

This last category, however, tends to grate on me the most since it inspires articles such as this one: Do Americans know enough to vote intelligently?

I tend to think the people who write this kind of thing must have had a particularly bad day when they hit on the topic. Someone cut them off in traffic, someone in front of them carried 11 items into the 10-items-or-less line, or they had to be at the DMV that morning. I wonder this because I tend to think that only people with a burning sense of self-satisfaction and superiority could sit down and write an article that basically asks: why should we let people that seem pretty stupid to me, vote? The article is a breathless ping-ponging from "people don't know what they're talking about when they vote" to "people can't get information" to "people are too dumb to sort good information from bad when they finally do get it".

Well, gee. It's amazing we all get up in the morning and put our pants on the right way, isn't it?

For the life of me, I've never understood what the problem is with partially informed people. I don't see the direct line from "more informed" to "a better democracy." In fact, this sense that people should be more active in their civic duties, lest we all fall into a den of iniquity strikes me as the opposite of what we're all on about here. It's a push back to the Platonic philosopher-king, where we hope like hell the really really smart people also happen to be benevolent, since only they get to make decisions. (No, I know this isn't entirely what Plato's getting at in The Republic, but the point is that the character is drawn in this way to get at the larger attempt to elaborate on "justice".)

If we believe that people, of their own accord, will invest their own time and resources into those things that return the greatest value to them (monetarily and emotionally), then perhaps its simply that getting overly involved in politics isn't what drives most folks. Why is this a bad thing? To me it indicates that the ultimate choices made at, in this case, the federal level don't hold massive sway over the lives of the individual. If every four years these people had to figure out if we all are going to be able to travel freely or be registered like cattle, be able to start families or have reproduction limited by law, pay 20 or 75% of our salaries in taxes, then perhaps work to seek out information would grow. Isn't this why the economy turns out to be the most important issue in almost every election? Despite the presidency being one of the least effective institutions for affecting the overall economy, people view the outcome as directly impacting their lives in significant ways. Is this a sign of ignorance or stupidity? Only if we value knowing certain facts as being "intelligent".

Honestly, what good is it to be able to name the Secretary of State? Did we vote for him or her? If we don't like the job the person is doing, can we vote them out? Even if you can rattle off the names of every cabinet member in the last 5 administrations, does this make you somehow more worthy to vote?

One of the many benefits of the kind of aggregation system that we have, both in terms of voting and the market, is that it necessarily evacuates the need for any one individual to be fully informed on every subject. It simply can't happen. But articles like the one linked to above function as a sort of elitist wringing of the hands about just how poorly informed everyone seems to be. Witness the ridiculous idea offered by the Stanford prof: getting people together to discuss topics for an afternoon by paying them money. Clearly, the 364 other days of the year people are too busy picking their noses to pay attention to what the prof thinks really matters.

Again, I don't see the causal line between more information and a better outcome. (By the way, that seems an implied statement in the article: how can we all do better, since the only way the US could have gotten to where it is now is through collective stupidity?) More choice can be a highly confusing thing, after all. The ability to meaningfully discern differences between each political position falls as the number grows. Even the learned press can't tell the difference between tax burden and tax rate. Same thing with increasing voter participation. What's the functional difference between a 54 and 59% turnout? If it's a sample of the population, both of them are more than sufficient.

Anyway. I'll stop the ranting now, and get back to work. That is, if I can figure out how to use the funny metal box on my desk with the TV thingy attached to it.

Chicago-Style Voting

In another example of the kind of uncertainty I mentioned around that seemingly sacrosanct number of 537 votes that Bush "lost" by in Florida the "popular" election (I hate even referring to it as an election, save that it works as a reference since it seems to resonate as real to some folks despite there being no such thing), there's this article on dual registrations of some voters.

Thousands Registered to Vote in 2 States-Report

About 46,000 people are registered to vote in two states, New York and Florida, a violation of both states' laws that could affect the outcome of the November presidential election, according to an investigation by the Daily News.

Throwing further doubt on the question of which way the vote would have gone in subsequent recounts:

Of the 46,000 registered in both states, 68 percent are Democrats, 12 percent are Republicans and 16 percent didn't align themselves with a party, the newspaper reported on Sunday.

There's just no way to say that it was exactly 537 votes, or that even if 600 more votes had been cast and agreed on to make Gore the winner, that it would have resolved the issue. If the dual votes were tossed out (and not just in one state, since this would constitute fraud the votes would be eliminated from the count altogether), the majority of votes lost would have been for Gore.

Note this also:

The duel registrations have gone undetected because election officials do not check voter rolls across state lines, the newspaper said.

So far as I know (and this of course could be changing), that isn't going to change with electronic voting.

Florida 2000: It all seems so easy in hindsight

I'm not the most ardent fan of the Electoral College that you're bound to find, but I do usually end up on the side of keeping it around.

Most of the arguments I've heard about abolishing it hinge, to greater or lesser extent, on the feasibility, desirability, simplicity, or similar aspects of turning entirely to the popular vote. I've just finished reading Tim Noah's two-part (so far) dressing-down of the electoral college (Part I, Part II), "America's Worst College" and I have to say I remain not only unconvinced, but rather depressed by the attempt.

First off, I'm surprised at the ease with which Noah assumes the recount process went in Florida 2000. Those 544,000 votes, and the 537 votes in Florida, appear obvious to hindsight. But it's worth remembering that the massive legal battles raging in Florida and elsewhere were largely about the method used to count those ballots. At the time, it was not clear at all how many went to either side. And thus we still, to this day, have cries about disenfranchisement during the process.

The uncertainty isn't just a tale about badly-punched cards. It raises the very real, very problematic issue of measurement error. Just how certain are you that every vote was counted correctly, every time, everywhere? If you claim 100% for numbers in the millions, we need to schedule a trip to Vegas, you and I, where I plan to take you for a good deal of money. From what I know of the recount process, that 537 number is, at the very best, the center of a range of possible numbers that it could have been. But lest you think we're talking about a couple of votes in either direction, just remember that as the number of total votes climbs, so too does the absolute number of miscounted votes in the total population of votes. In 50 million votes, the range would include thousands in either direction, maybe more. (The precision is growing with computer voting, I should say, though even that is subject to problems.)

But this is almost secondary to the other fallacy Noah appears desperate to hang his argument on: that the popular vote makes each vote "count more":

Remember, also, that under a popular-vote system, all votes are equal. Under the Electoral College's winner-take-all allocation (in all states but Maine and Nebraska), votes in big states count more than votes in smaller states because they can leverage a lot more electors.

Eh, no.

What do we mean by a vote "counting"? Well, there is of course the notion that your vote registered as part of the overall tally. You got your vote in, and the person doing the counting got yours and put a vote in the column for your choice. Civic pride abounds! But then there's the idea of one vote counting "more" or "less" than another. How does this work? Well, think of it this way: you're town is going to vote on a building a new bridge over the local stream. It's a town of a cozy 10,000 people. You live on the other side of the stream from the town, so the bridge is a good thing (assuming it outweighs the costs of the new tax that will have to be levied) and you plan to vote for it. Now, your 1 vote is entered, along with all 9,999 others. Your vote has counted. But of course, most of the town lives on the other side of that stream, so it looked like the bridge was going up, no matter what you voted. Did your vote really "count" then? Couldn't you have just stayed home, grilled a couple of burgers, and been happy the next day that you got the bridge and invested no costs in voting? After all, it's not like your vote would have been decisive...

And there's the rub. When we talk of voting counting more or less, we implicitly refer to the probability that your vote will be decisive in, say, an election. As the number of voters increases, that probability decreases. Would you have a better chance in your town of 10,000, or in your state of 10 million? If everyone is split perfectly 50-50, your chances are about the same. But shift the percent a little, and it's a proportionately bigger group that gets to be decisive one way or another. And so it is with the states. Voting in Montana means being one of 902,195 people. Voting in Florida means being one of 15,382,978 people. Where do you have a better chance of being decisive?

Being decisive, in the case of a presidential election, means having the chance to throw an entire state's electoral votes in the direction you prefer. Do you have a better chance of swaying Montana, or California? With the electoral college, the small states do get over-represented to some extent, since they all the get the same +2 for senators as the rest. But this is to help make sure the state's interests aren't avoided in federal policy making. By moving to the "popular vote", we simply become a nation of some 105,405,100 voters.

A couple of things happen. First, your chances of being decisive drop dramatically. The psychological effects of this are akin to those that prompted news outlets to stop forecasting results too early on election night: if you don't think your vote matters (well, less than generally for this case) you just won't go to the polls. Second, the areas with the highest population become far over-represented, since they have better information-sharing abilities. (Newspapers reach more people, television stations have more viewers, a single place holds a larger number of people at once, etc.) Politicians will simply work on those areas that are densely populated, rightly believing that they will get a much higher return on their investment in campaigning. Why spend $1000 a head to get a message out in Montana, when you can reduce your marginal cost per person in California and hit $10 per (I'm making the numbers up; the effect is what I'm trying to highlight)? In effect, NYC, Miami, Chicago, and LA will be electing presidents. Which means that policy will naturally follow.

While large states have more electors, the current electoral system attempts (to use an analogy) to use the group of people that voted as a representative sample of the whole state. Contrary to Noah's view, this isn't an accidental process. Had the number of popular votes been more condensed in one place or another, the electoral votes might well have gone differently. But as it is, with such a closely divided populace, the uncertainty in measurement is purposefully traded in favor of greater representation of diverse interests.

3rd Party Toothpaste

Much has been made of the 3rd party “spoiler effect” this election cycle. For example, many Democrats are angry that Ralph Nader is running again. They fear his being a possibility will draw votes away from Kerry and hand Bush his second term. “A vote for Nader is a vote for Bush,” is a common phrase used by Kerry supporters. Of course that is based on the assumption that Nader supporters prefer Kerry to Bush, which may or may not be true. Let’s think about it.

It seems to me that the market for toothpaste is essentially a two party system. Crest and Colgate look like they have substantial market shares. Suppose I were Aquafresh man. When I check out with my Aquafresh, no one says to me, “You know that buying Aquafresh is like buying Colgate because you’re taking that profit away from Crest.” Sounds ridiculous, doesn’t it? In fact, buying Aquafresh is not like buying Colgate at all. They are not perfect substitutes. Neither are Kerry and Nader or Bush and Badnarik (the Libertarian Party candidate). I don’t care that Crest is not getting money from me because I don’t prefer Crest. I prefer Aquafresh. A vote is like a purchasing decision. It is an expression of preferences.

Those of us who vote for 3rd party candidates are not doing so because we don’t understand the consequences of our actions. On the contrary, I think I am acting very rationally. I get more value out of supporting whichever candidate I prefer than I would voting for the Democrat or Republican I dislike least. I understand that allegiances to political parties are stronger than allegiances to toothpaste companies. Nonetheless, the comparison is interesting to ponder.

As it turns out, I’m a Colgate man but I still plan on voting for a 3rd party. And what about the half of the Americans who don’t vote at all? Well, they just don’t care about their teeth.

Low Turnout for EU Vote

From BBC News:

Outgoing European Parliament President Pat Cox described the results as a "wake-up call" and warned European leaders that they had to demonstrate the EU's relevance to voters.

If the primary motive for voting is civic duty or group identity, it would seem rational to conclude that either one or both are down. But isn't it possible--even likely--that voters still recognize the enormous power, control, and relevance of the EU government, but believe that they are irrelevant to EU leaders?

More E-Voting: Am I Sounding Like Cassandra Yet?

I guess I'm not the only one to have considered the potential problems of electronic voting machines that produce no paper trail. Though, now it looks like people are hoping that the paper trail-enabled systems fail: E-Vote Printer's High Stakes Test.

"That's what we're hearing, that a lot of election officials hope we fail because they don't want to be bothered with paper ballots," said Steve George, spokesman for Nevada's secretary of state.

Several states, responding to public outcry for a physical record of votes, have mandated or announced legislative plans to demand that e-voting systems print a paper record so voters can verify that the machine registered their vote accurately before the record drops into a ballot box. The states include California, Illinois, Missouri, Nevada, Vermont, Washington and West Virginia.

But many county election officials oppose the idea, saying printers will create more problems for poll workers if they break down or run out of paper, and they will cause longer poll lines if voters take more time to check their ballots. The officials also don't relish the election recounts and lawsuits that could arise if paper records don't match final digital vote tallies.

Well, that's nice. Are they saying it would be much better if people just didn't hear about the discrepancy between voter intent and the machine result?

I'll say it again, just because I can: this new technology does not solve the problem. There have been no massive breakdowns in voting because of the cost of ballots, or the lines, or people taking time to verify their voting. In fact, the problem in Florida was precisely because people didn't verify their ballots. The only things that have been a problem with voting machines both in Florida and elsewhere is a lack of clarity of use, and no clear intent. That can be solved with a touchscreen and a machine-punched ballot. Attempting to change anything is simply creating more problems than it's solving.