Bell Curve The Law Talking Guy Raised by Republicans U.S. West
Well, he's kind of had it in for me ever since I accidentally ran over his dog. Actually, replace "accidentally" with "repeatedly," and replace "dog" with "son."

Sunday, October 17, 2004

Recount This: The Trouble With Florida

In Florida, the the certified vote had George Bush leading by 537 votes out of a total of 5,956,253 cast. (after the recount, the Bush lead was 193 before the United States Supreme Court ruled). That is leading by less than 1 in 11,000 votes (one in nearly 30,000 with the recount). That is less than 100th of 1%. What kind of a process is accurate to 1 in 11,000? Ivory soap is only 99 44/100 pure. This is just too high a standard of accuracy to expect from a voting system. I suggest that, even taking the vote totals as given, the vote in Florida was simply too close for any counting mechanism to have rendered an accurate count. Certainly not one run by an ad hoc mixture of part-time government officials and senior citizen volunteers. Introducing internet technology does not help, either. In other words, every recount would have given a different tally. Aside from the fact that there was massive fraud in other areas of the Florida election, keeping lpeople from voting, and so forth, such close numbers themselves can re-occur without the election tampering. We need to think of how to handle errors that small in ballot counting. Justice Scalia was accurate on one point: there is no reason to think that a human recount will finally "get it right." Error terms are part of life. It is too bad, but predictable, that in census counting, the intellectually dishonest Justice Scalia had forgotten about math, and proclaimed that an "actual enumeration" was really possible, and that statistical sampling was somehow "cheating"

Any thoughts, O statisticians and mathematicians?

25 comments:

Dr. Strangelove said...

OK… If we think of an election as a measurement process, then we know that measurement processes have different kinds of error. The most basic kind of error is that associated with attempting to take a single data point. This is the hanging chad or spoiled ballot: an individual vote that cannot be tallied with 100% confidence. This is something new technology can definitely help reduce.

The entire recount process in 2000 was based on trying to reduce this particular kind of error. When Scalia says no human recount will "get it right" he is technically correct since some ballots will never be discernable with 100% confidence--but if we view the purpose of the recount as simply getting a more accurate measure, then a more careful examination of each questionable ballot will reduce the individual vote error.

It turns out that a full, statewide recount (which Gore did not request) would have produced, using any standard, a slim victory for Gore--but any of the partial recounts Gore actually asked for would have resulted in a slim victory for Bush. There's irony for you.

The next kind of error is statistical error or sampling error. If (and this is a big IF) an election is supposed to be a measurement of the popular will on a specific day, then we have to ask how good the resolution of the measurement is—how much “noise” is in the system. Though the actual resolving power of an election is just about impossible to determine, I still think it's pretty safe to say that couldn’t possibly be good enough to discern a difference of 100th of 1% in the popular will.

And then there’s systematic error: given the butterfly ballot, disfranchisement, and allegations of fraud all nibbling at the edges, any statistician would have to conclude that the Florida 2000 election "measurement" could have been shifted one way or the other quite a lot, so again we can’t trust it to resolve a difference of less than 100th of 1% in the popular will.

But I actually don’t think our elections are meant to be purely a measurement of popular will. Elections are the agreed-upon decision-making process for a democracy; they are a set of rules for determining our next leader. So long as all sides perceive the process is fair and so long as the process can be concluded, an election will succeed in its task. We learn from Florida that randomness is inherent in this process, but given how closely divided the electorate must have been, no feasible measurement process could have produced a result other than a tie. And a decision had to be made.

For me, the trouble with Florida was: (a) it wasn’t fair—or at least a lot of people felt that way, on both sides, which amounts to the same thing since fairness is partly subjective; (b) it had no obvious way of achieving finality; and (c) it was obvious to anyone that the result was so close that a little cheating here or a few mistakes there was enough to tip the result, so the little stuff mattered. So as a democratic decision-making process that was supposed to provide fairness and finality, the election in Florida was a total failure. And I believe the biggest indictment of the Supreme Court was that while the majority understood that finality is indeed a requirement of a good decision-making process, they did not also see that in a democracy one also needs fairness. By stopping the recount the way they did, they achieved finality but not fairness.

I can think of a few decisions the Supreme Court could have made, if they wished to provide both finality *and* fairness.

1. The Supreme Court could have issued its own clear standard, ordered a full, state-wide recount under this standard, and called on all experienced poll workers across the country to converge on Florida, along with the National Guard perhaps, to supervise and manage the process. I think this would have been the best way, though it would have taken much longer. Unfortunately, it would not solve the systematic errors (fraud, disfranchisement).

2. They could have required a second election and done all of the above to ensure fairness. This might have eliminated some of the problems with fraud and disfranchisement as well. Unfortunately, it creates a new problem since having a “second” election on a different day than the one everyone was working toward wasn’t in the original rules as a possibility, so many people might feel that this is not a fair option.

3. They could have actually called it a tie and allocated the votes evenly between the candidates. This avoids the error problems altogether, but again, it’s not in the original rules, so it might be perceived as unfair. Note that when the Florida legislature voted to give their Electoral Votes to Bush regardless of the vote count—similarly trying to settle the issue by simply allocating the Electoral Votes by fiat—this wasn’t considered “fair” by the other side, even though technically it was allowed under the original rules. This is because people hate technicalities, and since almost no-one except the legal scholars even knew it was part of the original rules, it might as well have been a new rule.

4. They could have said that Florida's vote was too unclear to produce a final result, so they would simply not allow any electors from Florida to be seated at all. This is effectively the same as #3, but by simply disqualifying votes rather than actually allocating them by fiat, it might not be perceived as being quite as “outside of the rules.”

5. A coin flip. Think about it: even a coin-flip would have been better than what they did--at least that way the result would have been perceived as a fair (albeit arbitrary) decision-making process.

Anyhow, I think we do need a process for deciding elections that are too close, and I don't know what it should be, but in order to make it feel fair, there is one thing that we absolutely are going to have to do:

Announce the method ahead of time.

Raised By Republicans said...

Master of Time and Space,

RE: your assertion that elections are not measurements of popular will but rather an agreed upon set of rules for decision-making is 100% right.

I couldn't have said it better.

The Law Talking Guy said...

Great comment by the Master of Time and Space. I am not sure, however, that I agree with the distinction between "measuring popular will" and "set of rules for decisionmaking." When the rule is "determine the popular will" -- as election laws are often so generally written -- we have issues. The other problem is that any set of rules must be premised on basic values, the core of which is determining the popular will, and valuing each vote (and voter) equally. I agree that a second election, runoff, or division of electoral votes was the only rational solution. But a set of arbitrary rules is not acceptable, even if it satisfies the predictability requirement.

USWest said...

I think this is an interesting discussion considering that there are once agin reports of trouble in Florida in early voting. I also like the point of "finality" vs. fairness.

At the end of the day, I have to agree with MTS that "Our elections are not a measurement of popular will. Elections are the agreed-upon decision-making process for a democracy."

If we wanted elections to be a measurement of popular will, we would have direct elections, no electroal college. The fact is, we are given a voice that weighs heavily in the the final decision, but that does not determine it. I remember making this point way back when we were discussing the merits of the electoral college system.

The bottom line is that we agreed to a contract that told us how elections would be determined. It would never have been fair for everyone. Even today, many of us say it was unfair. Could it have been more fair? Yes. So perhaps the real quesiton is how much unfairness can we as a society tolerate.

And the idea of refusing the Florida's electors is not new. It was condisered the back in 1976, but rejected then as well. For fun, check out:
http://www.historyhouse.com/uts/tilden_hayes/

Dr. Strangelove said...

Retraction: I was not correct when I said that "under any standard" Gore would win with a Statewide recount, and Bush would win with the 4-county count.

The Florida Ballots Project is the only exhaustive examination of all 180,000 uncounted Florida Ballots. It was conducted over a six-month period in early 2001 by a consortium of news groups and academics. If you're a data nerd like me, or you simply want to relive the pain, then browse on over to their free data page and download the Florida Ballots Tabulator (Access 2000) database. You can then play around in the "scenario manager" and pick your standards, then run the scenario and look at the "report manager" to see the outcomes.

After checking through many scenarios, I have come to realize three things. First, either candidate could have won under reasonable sets of rules. Second, the less restrictive rules over the wider area usually favor Gore; the stricter rules favor Bush. So Democrats need to learn how to vote better. Third, I'm still pissed at Ralph Nader.

Raised By Republicans said...

Re: Florida. Two things had far more influence on the Florida outcome than the hanging chad/recount issue.

First, over 90 thousand people voted for Nader. If only 1 in 20 of these people would have voted for Gore instead, Gore would have won Florida. Naderites are fond of arguing that the Nader voters didn't draw from Gore because they were largely new voters who were only voting because Ralph Nader energized them. However it is extremely unlikely that 99.5% of the Nader voters were first time voters who would not have voted otherwise. It is far more likely that most of the Nader voters would have voted for Gore had Nader not been in the race, and so Gore would have won Florida by a considerable margin.

Second, something like 30 thousand African Americans were "mistakenly" told they could not vote by Florida state officials. 90% of African American voters voted for Gore. Had these citizens been allowed to exercise their right to vote, Gore would have won Florida by a considerable margin.

To sum up the Florida story. We have one factor that resulted in variations of hundreds of votes in ways that subsequent studies show are more or less random. We have two factors that resulted in voting distortions of tens of thousands of votes that consistently biased the vote in favor of the Governor's brother. Yet, the "big story" is hanging chads. Why? Perhaps the journalists covering the story were incompetent (certainly). Perhaps Gore screwed up by not playing the "race card" forcing more national attention on the real story (almost certainly).

Raised By Republicans said...

RE: Elections and the Common Will.

Rouseou and other Enlightenment thinkers were fond of talking about the Common Will or similar concepts. But nearly every study of collective dicision making has shown that identifying a single, collective will is nearly impossible. Groups simply don't work like that.

Groups are really masses of individuals each with their own idiosyncratic preferences. When we talk about the "Common Will" of the "Common Good" we are making the mistake of assuming that all people - or even most people - agree on a sufficient collection of policies to justify talk of a single, "Common Will." If we try to make each decision by some idealized polling of the group as a whole we can easily fall into cycling from policy to policy as new coalitions constantly form, collapse and re-form.

Electoral systems are the rules by which we constrain this cycling. All of them have problems. None of them are perfect refelctions of the "Common Will" (if, indeed there is such a thing). However, they all have the advantage (in the absence of fraud) of being predictable and offering finality and some semblence of legitimacy.

Anonymous said...

A comment from the Australian interloper. To be completely honest, I was simply stunned with how undemocratic your voting system is. Regardless of what you think about Australian compulsory voting and the possibility of electoral roll fraud, our system is simple and effective. the Australian Electoral Commission sets the electoral boundaries and the voting rules, runs the booths and does the count. Companies can even hire them to run internal elections: our department uses the AEC for any wage agreement votes.

A friend who worked on one the booths mentioned how they would have someone on the door checking people as they went out to ensure they place their ballots in the correct box. They go so far as to go through the rubbish bins outside the exit for ballots, and those ballots are counted.

County by county rules, different ballots, returning officers with political affiliations, politicians setting electoral boundaries? No wonder you need international observers. :-)

The Law Talking Guy said...

Hey, Koala-boy, don't be so proud. First, I'll bet your Australian Electoral Commission is political, because it's full of human beings with power. Second, we have elections in most places twice a year, on a whole variety of topics. We don't just mark a couple boxes once in a blue moon. Things are more complicated. Third, I suspect there is a lot more fraud in Australia than you realize. After all,Chicago elections officials have been finding ballots "in the garbage" for more than a hundred years.

Dr. Strangelove said...

Re: Law Talking Guy's comments. Far be it from me to let the facts get in the way of your opinions… but I must rise to the defense of Koala-boy.

1. To assume that the AEC couldn't possibly be fairly independent and impartial is far too jaded. Why not? Consider who runs it: the AEC consists judges, retired judges, and civil servants—including my favorite, the “Australian Statistician” (love that title!) And unlike the US, the AEC also has complicated rules it must follow when drawing up districts:
(a) They never have global redistributions. They redistribute districts in a given state when, “the number of electors in more than one third of the divisions in a State deviates from the average divisional enrolment by over 10% in three consecutive months, or a period of 7 years has elapsed since the previous redistribution.” They even have mini-redistributions sometimes as part of the continual tweaking.
(b) By law they must give consideration to the, "community of interests within the proposed Electoral Division, including economic, social and regional interests; means of communication and travel; the physical features and the boundaries of existing Divisions."
(c) The process is conducted openly, with public comment and suggestions. It must be approved several times, by increasingly larger groups of the AEC.
(d) Thousands of temporary staff are recruited for each federal election, from a range of backgrounds such as teachers and bankers. One of the requirements of the AEC's recruitment process is that persons wanting to be employed as temporary staff must declare they are politically neutral. The AEC argues that their staff represents a, “reasonable cross-section of society, and consequently, of privately held political opinions. This very diversity of opinion, experience and background is a strong deterrent to the development of any overall political bias within the AEC over time.”

2. Australian elections are not “once in a blue moon,” but are frequent and at least as complicated as ours. The Australian House of Representatives and Senate are both elected for three-year terms. They also have national and state referenda. Furthermore, each state has elections that occur at different intervals (3, 4, or 8 years in different states for different bodies). And then there are local elections too. In all cases, voters must fill out the ballot paper by numbering ALL the candidates in order of their preference. Failure to number all the candidates (or an error in numbering) renders the ballot invalid (called "informal"). There are frequently 10 or 12 candidates in a seat—sometimes dozens. Over 95% of Australians vote at elections (it is compulsory), and yet in most elections only 2%-3% of votes cast are invalid. (An amusing point: their laws specify that “Antarctic electors” are exempt from the compulsory voting requirement.)

3. Your "suspicion" is just that. The official government line is that, “It has been concluded by every parliamentary and judicial inquiry into the conduct of federal elections, since the AEC was established as an independent statutory authority in 1984, that there has been no widespread and organised attempt to defraud the federal electoral system; that instances of multiple voting that do occur show no pattern of concentration in any Division, marginal or otherwise; and that the level of fraudulent enrolment and voting is not sufficient to have overturned the result in any Division in Australia. That is, there is no evidence to suggest that the overall outcomes of the 1984, 1987, 1990, 1993, 1996 and 1998 federal elections were affected by fraudulent enrolment and voting.” They do acknowledge fraud in elections decades ago, however. Take it as you will.

Raised By Republicans said...

Most of the things you said in defense - sorry defence - of the Australian electoral system were remarkable for their vague nature and broad similarity to the rules here in the US.

I've never met an Australian who didn't think Australia was better than practically every country on Earth at practically everything. It that too, they are very much like Americans.

The Law Talking Guy said...

Hey, MTS and Koala-boy, here's my response.

The AEC is made of "judges, retired judges, and civil servants" so it's apolitical? Let's see, Antonin Scalia, Katherine Harris, gee... There may be advantages to removing gerrymandering from the legislature's direct control, and some American states do that (e.g., Iowa and Arizona) but it's far from perfect.

American election laws are antiquated, to be sure. What we are beginning to realize as a nation is just how true this is. But we are also realizing that the solutions are not all obvious. Compulsory voting would result in one thing only: a movement to repeal it, supported by every politician under the sun. Paper ballots have problems, just like other ballots. Fake I.D.s are easier to get past poll volunteers than even liquor store clerks.

Check the bibliography of this article http://www.econ.usyd.edu.au/drawingboard/digest/0409/sawer.html for insights.

Dr. Strangelove said...

Let me see if I have this right: the plethora of facts I provided is not only "vague," but somehow "remarkable" for its vagueness, even when compared to the heap of unsubstantiated assertions that preceded it; the AUS method of continuous apportionment by an independent, non-partisan committee is "broadly similar" to the way we gerrymander in the US; most judges are as notoriously partisan as Scalia (such as Souter, Stevens, O'Connor, Ginsburg, Kennedy, Breyer, Thurgood Marshall, Earl Warren--and presumably the nice fellow LTG clerked for); and Katherine Harris was an un-elected civil servant. Apparently, the problem with my post is that I actually took the time to research it rather than just loudly spew out my opinion. So now I’m just going to spew.

Look, after spending many hours with Koala-boy discussing the US electoral system, I've come to realize that the US electoral system has many undemocratic flaws and, gosh, maybe some of the newer democracies might have better solutions…? Why does this rather obvious point raise such blind ire in RbR and LTG? Just because these other systems aren't perfect doesn't mean that some of their methods aren't better. I'm so tired of that mode of reasoning where you point out one potential difficulty in every new idea and then smugly dismiss it. It's a lot harder to build a bridge than shoot at it safely from the sidelines, you know.

The redistricting battles in Texas, Pennsylvania, Colorado, and even the lame redistricting here in California all show that the system has serious problems. When half the Texas legislature flies to New Mexico in order to stop the redistricting, you know it's gotten ridiculously partisan. Do you really believe that an independent commission would do worse? That it would be less fair? Isn't it much more likely that it would be better?

And consider the radical notion of having national standards and national funding for federal elections. What's wrong with that? Do you really believe that the current US "system" where every state (or county!) has a different voting system, where lots of local registrars are overwhelmed and underfunded, and where there are no uniform standards for counting votes (remember the original subject of this thread?) is better than organizing it at a national level, like every other major democracy does? If Bush proposed a piecemeal system like that for Iraq or Afghanistan, where votes were tallied by the local "warlords" without supervision, you'd throw the book at him--and you know it.

When Koala-boy says that he was startled by the undemocratic electoral system we have here, don't shoot the messenger. Remember that he entered our country during the California Recall election (that showcase of democracy...) and then had to live through the bulk of the most vicious, moneyed, litigious presidential campaign in history. And frankly, he's right--and it’s not just him. Don’t forget that the whole country was similarly appalled 4 years ago by the Florida fiasco, when we all began to learn just how bad it was. Subsequent efforts to fix it have been poorly funded and often lame, sometimes causing more confusion.

We need electoral reform in this country, and I think we all agree on that. Don't just throw your hands up and say that this is the best we can do in America. The knee-jerk defense of US election practices that I've read here will never fix anything. If you don't like the some of the AUS ideas--which solve some problems but can cause others--then suggest better ones. Australia has some good ideas, and so do some other democracies. Let’s all put on our humble caps for a while, look around, and allow ourselves learn from others if they have something to teach.

Anonymous said...

Interesting that you say comments in defence, RbR. After reading your response and that by LTG, I was reminded of Republican tactics from both of you: attack presented information (some vague comment about comments being vague), present no facts ("I'll bet your Australian Electoral Commission is political"), provide unsubstantiated opinions ("We don't just mark a couple boxes once in a blue moon"), and degenerate to name calling. I take it all in jest, but the irony of your arguing style was too much to ignore. For a blog critical of Republican methods, they seem to be high on the uptake. I take it RbR you never met me then, as I have never argued that Australia is better than every other country. Maybe you inferred that inappropriately, as is your want.

I made a personal comment about how stunned I was that the US system seems so ad hoc -- as I was -- and drew a parallel to the system I know best for comparison. You can hammer the Australian system as much as you like. Yes, all political appointments have the possibility of becoming political. A difference between our countries is that in Australia the expectation is that they will be apolitical, and mostly they are.

One fact remains: the USA will have some presence of international observers during the Presidential election. Most "First World" countries -- Australia included -- do not even have that suggested as a possibility. Aren't you embarrassed by that?

Raised By Republicans said...

I was kidding about the Aussies bragging about stuff (but then Americans do the same thing). But your reaction my good natured ribbing reminds me of joke I heard from an English comedian (so consider the source), "The Australians are the only people on Earth with a chip on each shoulder."

But getting to the point about what I meant about "vague defense"....Master of Time and Space said that Australian redisctricting rules were better than American ones because they required that district boundaries account for "community of interests within the proposed Electoral Division, including economic, social and regional interests; means of communication and travel; the physical features and the boundaries of existing Divisions." That's vague. What are "community interests?" What are "economic, social and regional interests?" How much cattering to those interest is sufficiently taking them into account? I am pretty confident that the US rules for district remapping are similar in spirit if not exactly like those in Australia and that in BOTH cases (not claiming superiority here for either system) it is a matter left for some interpretation. In the US it is the courts that interpret the rules and approve new boundaries. In Australia it seems to be the Australian Electoral Commission (AEC).

So who appoints the AEC? This is vitally important given their supposed impartiality. The majority of the members of the AEC are appointed by the State governments. The AEC's website is unclear about who appoints the three person leadership group but as near as I can tell they are all three appointed by the Prime Minister. Yes, they are usually judges or former judges and the Australian Statitician (a government appointment?) but their appointment by the Prime Minister would seem to be grounds for questioning their impartiality - that is if one were cynical about it. What's more I'm not certain how independent the Australian judiciary is from the Government anyway (I'm not an expert on Aussie judicial independence but if I had to guess I'd say they were more independent than the British judiciary but less independent than the American, Italian or German judiciaries).

What I want to get across here is that in the final analysis there is not a whole lot of difference in the way the the US conducts its elections and redisctricting and the way Australia does it. Sure there are detail differences but in both the US and Australia, these details are matters for interpretation (because they are vague in both countries). In both countries appointees of elected officials are in charge of making sure the elections are fair. In both countries the systems has aspects that make them vulnerable to abuse but on the whole both countries use more or less reasonable methods.

What's more, the differences in outcomes between the two electoral systems (Single Member Districts Plurality in the US and Alternative List in Australia) are small. Both systems are biased in favor of large "catch-all" parties and against small parties. Sure the Australian system allows for more complex voting (a good thing in my opinion) but in the end, small parties have their biggest influence by encouraging the larger parties to examine new issues (much like Ross Perot did with the deficit in 1992). Political scientists regard the US and Aussie electoral systems as being much more like each other than either is like the systems used in Continental Europe (Proportional Representation).

Am I upset about what happened in 2000? You bet!
Do I think the US would be better off with Australian observers - or observers from any other established democracy - watching the election in 2004? YOU BET! (But not if the observers they send are appointed by the Australian PM, a close ally of G.W. Bush.)
Am I embarrased about the need for international observers? Frankly, I just want them here. The stakes are far too high for me to worry about how it looks.

Dr. Strangelove said...

RbR writes (I quote verbatim): "...in the final analysis there is not a whole lot of difference in the way the US conducts its elections and redistricting and the way Australia does it. Sure there are detail differences but in both the US and Australia, these details are matters for interpretation (because they are vague in both countries). In both countries appointees of elected officials are in charge of making sure the elections are fair."Nice try, RbR, but no dice.

First off, you conveniently dodged the original redistricting issue. The US and AUS do it about as differently as a democracy can tolerate. Please consider:

· The US does a complete redistricting every 10 years, conducted behind closed doors, by the separate (and highly partisan) legislatures of each individual state. And as you can clearly see by examining some of the remarkable gerrymanders that have resulted from this, there are no meaningful rules they must follow that require them to consider community or transportation or anything else (allow me to offer as Exhibit A the new Texas Congressional District #25--which could generously be described as being in the shape of a mutilated staircase viewed edge-on).

· Meanwhile, AUS does continuous, piecemeal redistricting, conducted as an open process, by a single, independent national commission. They are bound not only by statute (come on, RbR, of course the highest-level directive is general—have you read the US constitution lately?) and also by more complicated lower-level rules and 20 years of precedent. You will find no mutilated staircases on their maps.Second, the only specific assertion you made of equivalence between AUS and US elections (and it's pretty vague, actually) is that in both countries, "appointees of elected officials are in charge of making sure the elections are fair." And even this is wrong. For example, you might recall that in CA our Secretary of State (Kevin Shelley) is an elected politician. You might also remember that in Palm Beach County, Florida, Elections Supervisor Theresa LaPore recently lost her re-election bid because she had designed the butterfly ballot in 2000. We put the foxes in charge of the henhouse over here.

Third, you also deliberately ignored the larger point that AUS elections are conducted by national standards, using national funding--whereas ours are conducted district-by-district, using whatever meager local funding the cash-strapped town council can spare, using different standards, different voting equipment, different timetables, etc.

Fourth, you dismissed even the differences that you did deign to acknowledge--e.g. multi-member districts, full preference voting, compulsory voting--as "details" that are merely "matters of interpretation" so you could sidestep the whole issue of whether one or two of these AUS reforms might work in the US. So you pretty much took your humble cap, threw it on the floor, and jumped all over it.

Finally, you cited that the "similar outcomes" of the elections in AUS and US as proof that there was no real difference. Outcome does not equal process. I know you know that's a fallacy. What the "similar outcomes" do suggest is that making these reforms in the US would not skew the elections significantly in favor of any political party--so actually, this means that the AUS reforms pass the first test of fairness.

Look, I'm not "sold" on the AUS system by any means, and there are many things about the US system I like. And I am proud of our strong democratic tradition. But I can see that the AUS system uses different processes, some of which might be helpful in ameliorating some of the undemocratic (or at least unfair) aspects of our own. It's a really simple point, actually.

Raised By Republicans said...

At the risk of engaging in post-mortem equine flagelation allow me to clarify.

Somehow, Master of Time and Space and "Annonymous" (from Australia) got very upset at me. I'm not certain what I've said to offend them so deeply. After all, it wasn't me who started calling people cute, fuzzy, tree dwelling mammals.

I am not claiming that the US system is either perfect or even significantly better than the Australian system.

I am saying that Australian superiority on this matter is not as obvious as Master of Time and Space and Austalian Anonymous assert.

Now on to the details:

Yes, I ignored the national rules versus local rules bit because the US is a federal system of the type that allows constitutional, legislative and administrative autonomy at all levels of government. Having a single, unified electoral rule in the United States is probably unconstitutional. So why bother debating that point?

You continue to insist that the AEC is independent and non-partisan. But there is a lot reason to think that is not the case. Its leaders are political appointees of the Prime Minister and the rest of it are political appointees of the State governments.

When I said that political appointees are responsible for interpreting redistricting rules in both countries I was referring to the AEC in the Australian case and the requirement for judicial approval of redistricting in the US case. You may remember that the Texas case you mentioned went to the Supreme Court? They ruled that it may stand (which I don't like because I'm a Democrat but probably would like if I was a Republican) but there was due process and transparency of a sort. Actually, redisctricting plans have been struck down by courts in the past so there are constraints on the partisan whims of the legislaturs - they are just very vague and subject to judicial interpretation. (Secretary's of State are not elected in all states by the way, also they don't get the final approval power over redistricting).

True: Outcomes do not equal process. But outcomes are a funciton of process (which is why are having this debate in the first place). If we see VERY similar outcomes in both systems it is reasonable to suspect, at least, that the processes are not radically different. Indeed, detailed studies of the two electoral systems we're talking about have determined that they operate by very similar mechanisms with regard to the incentives to voters.

Regarding the shape of the districts: This is an obvious difference between the two systems. And its one that I now think is connected to the multi-member district size. What I'm thinking (and this is merely educated speculation on my part) is that the fact that the Australians have multi-member discritcts may mean that the shape of those districts is less salient. When you can get a seat by being second in the voting fine tuning the voting roles to ensure a first place finish through redistricting is less important.

Dr. Strangelove said...

Here is what I believe we can agree on.

1. The AUS system for redistricting is different from the US system in several ways:
(a) AUS redistricting continuous and piecemeal. US is all at once, every 10 years.
(b) AUS redistricting is done by a single body at the federal level. US redistricting is conducted by many bodies at the state level.
(c) The AUS redistricting authority (the AEC) is at least designed and supposed to be as independent and nonpartisan as possible. None of its members are elected, and many are drawn from the judicary. The US redistricting authority (state legislatures) is highly partisan, and all members are politicians.
(d) AUS redistricting is conducted as an open process with public comment. US redistricting is often done behind closed doors--the maps are not drawn up with public input.
(e) AUS redistricting is at least supposed to follow rules to keep the districts sensible. US redistricting has few rules. The courts may strike down a plan for gross violations (usually racial prejudice) but rarely do so.
(f) Outcome: AUS districts (which are multi-member for senate only, other house is single-member) are not visibly gerrymandered. US districts are obviously and hideously gerrymandered in many states.

2. The AUS system for running and certifying the elections themselves is different from the US system in several ways:
(a) AUS voting is compulsory. US voting is not compulsory--in fact, it requires special registration that causes some headaches every year. Attempts to make it easier to register (e.g., the "Motor-Voter" act) are highly politicized.
(b) All standards for AUS federal elections are set uniformly across the nation by a single national authority. Standards for US federal voting vary widely from state to state concerning time, manner, place, and even how invalid votes are determined.
(c) AUS election results are certified by a single national authority that does not consist of politicians. US election results are certified piecemeal by state or county, often by elected politicians.
(d) AUS election funding is provided by the federal gov't. US election funding comes piecemeal by state or county--resulting in many of the differences in the quality of pollworkers and equipment.

3. The overall AUS electoral setup is different from the US setup in several ways:
(a) AUS have multi-member districts for the senate (but single-member for house). US districts are never multi-member.
(b) AUS has full preference voting. US has only winner-take-all voting.
(c) AUS employs the "one man, one vote" principle. US has the Electoral College, where the outcome has defied the popular vote on several occasions.
(d) AUS election outcomes have considerably more minority-party represenation in the (multi-member) senate than we do.

What I have listed in these 3 points is as accurate as I can find. You may consider that these differences are irrelevant, but by that standard, I have to wonder if you would consider the differences between the US system and any other electoral system to be relevant.

I think that 1b, 1c, 1d, 1e, 2b, 2c, 2d, 3b, 3c are worth considering for the US. If they are unconstitutional, then I'm arguing we should change the constitution--I thought discussing this kind of stuff was germane to this blog. I will remember RbR's remark that it's not worth debating anything unconstitutional if he should ever advocate any change in our constitution :-)

Raised By Republicans said...

OK, I'll buy into most of those points (give or take). And I'll admit to rhetorical overreach on the dismissal of debates about difficult reforms that require constitutional amendment.

But I still contest point 1.C that the AEC is designed to be non-partisan. If it were intended to be non-partisan why would its top leadership be appointed by the Prime Minister? Also selection from among the judiciary is no guarantor of non-partisan behavior. You are assuming that the Australian judiciary is truly indpendent. It may not be. Finally, even if the judiciary as a whole is independent, the fact that the Prime Minister is in charge of hiring (and firing?) members of the AEC undermines that independence while serving in this capacity.

Anonymous said...

G'day. A couple of minor comments. I post Anonymously because I cannot be bothered signing up for my own blog. Your "good natured ribbing" was lost in what appeared normal American (Republican?) defence-by-attack. The joy of text is a lack of tonal information: maybe you should re-read some of what you write with an eye to other interpretations.

On the AEC, it would seem your only concern is that the heads of the AEC would be political in nature, as they are appointed by politicians. Well, below the three person commission, all employees of the AEC are public servants, hence go through a standardised job interview program that excludes appointments on the basis of any discrimination, whether that be sexual, racial, political or other. The chairman of the AEC is appointed from three presented judicial candidates by the Governor-General, the equivalent of the US President, which is a non-political figurehead position with no electoral mandate. The non-judicical appointment is an agency head: typically the Head Statistician (what a political power-house that is!). The Commissioner is appointed by the Department of Prime Minister and Cabinet.

Maybe the Commissioner could be political. However Australia follows the Westminster system and it is assumed that appointments are non-political; if they are political this gets criticised extensively. At the end of the day, the AEC decisions are made public, it is possible to dispute decisions, and there is a known and exacting mechanism for taking things to court. The judicary in Australia is extremely independent of government: invariably the High Court judges appointed by a party do not follow party lines at all.

Yes, the system is not perfect, but I think RbR's concern about the roles being political is more due to living in a system where everything is. There is a different expectation in Australia.

Raised By Republicans said...

I don't care if you post anonymously. I just had to have something to call you other than the possibly offensive "Koala Boy."

OK, I'll come clean. My concern about the politicization of the AEC is probably because my dissertation was in large part about how politicians in Parliamentary systems can enforce their will on administrative appointees.

The literature in the UK (and other "Westminster" systems) used to be dominated the image of the independent, professional civil servant maintaining administrative consistency and professionalism as politicians come and go. But resent research - on both sides of the Atlantic - is increasing aware of the power of the politicians to force the supposedly independent civil servants to do their bidding. Australia's bicameral legislature and federal structures probably allow for a bit more independence than is observed in the much more centralized Britain but I'm still not convinced by conventional assumptions about the independence and professionalism of civil servants. Those have always been overblown and political scientists and public administration researchers now see them as mostly unsubstantiated by emperical observation.

Your mention of the Governor General (still a Royal appointee?) does make me more inclined to see the AEC as independent. What would completely convince me is if the AEC triumvarate was unable to be fired by the same people that hired them. The AEC would have an enormous amount of independence if one part of the Australian government appoints them but another part removes them if they turn out to be corrupt or something.

Dr. Strangelove said...

Do we get a prize for the longest thread? :-)

Raised By Republicans said...

Yeah for the longest thread!

Aussie Aussie Aussie Oy Oy Oy!

Anonymous said...

Fair point, and no doubt there are issues. Department secretaries are now appointed directly by ministers. Maybe the difference is expectation, but that could naivety on my part. Still the mechanism is in place in Australia, and this is standardised. Regardless, the best system iun the world is not going help when the "left" party is centre-right and doesn't challenge the incumbent on any substantive issues.

And if I never had to hear "Aussie, aussie, aussie" again it wouldn't be soon enough.

The Law Talking Guy said...

What is the review mechanism for Australia's redistricting? I don't think that's been covered - maybe I missed it. I mean, the officials are appointed by partisan politicians and certainly have access to information about how to gerrymander.

One of the problems in the USA is the sheer size of disttricts, with so few reps for so many people. Districts are about 750K now. With more reps, gerrymandering becomes less of an issue (population is less stable and predictable).