I don’t see how is that undemocratic. You still end up with a government that runs on an agenda of popular compromises, be it the compromises are inter-party or not.
Wooly - I said “Under a 2-party, 1st-past-the-post system, both parties compete to appeal to more than 50% of the electorate.” You said “Not correct. They are appealing to win 50%+ of the seats.”
True. Guess I should have said “seats”. But my point stands. They have an overriding incentive to occupy the nationwide middle ground. Hence all the opinion polls and focus groups. Doesn’t such a system subject politicians to market forces effectively?
Urban Ranger - to me, that’s profoundly undemocratic. It’s what gives PR a bad name - some nuts sell their parliamentary votes in exchange for laws that 95% of the people don’t want.
So far, the discussion seems to be about the supposed pros/cons of PR. I’d be interested in hearing some concrete examples of how the FPTP system has damaged the quality of governance in countries that use it.
In other words, how have countries with 2 predominant broad-church-style parties fared less well than those with a multitude of ideologically-driven groups each representing only 10-20% of the population? It seems to me that the US and the UK have systems that deliver the goods (assuming, let’s say, that watered-down PR systems like Ireland’s might do equally well).
Every system has its weaknesses and there is no such thing as a perfect or even a best system. I remember reading an article showing it is possible for the voters to prefer candidate A to B, candidate B to candidate C and yet prefer candidate C to candidate A. Preference is not a transitive property. I am sure someone can find the detailed explanation which was quite complex.
At any rate, different circumscriptions may suit different conditions and cultures. The American system where small geographical circumscriptions elect one representative assumes voters are most grouped and bound by geography which might have been true 200 years ago when people did not move too much, blacks did not vote, etc but today it is not so true and a voter here may feel he has more in common with a voter far away than with the guy next door. This system has the obvious flaw that 49 % of voters can vote for candidates of party A and yet not get a single seat.
OTOH, electing candidates at large has its own shortcomings which have been pointed out. Having several chambers elected in different manner is a way to alleviate this as the laws have to satisfy all chambers (which could be more than 2).
That regarding the practical aspects because then we have the ethical and moral aspects. In a winner take all system, is it ethical that the minority goes unrepresented? etc… It is a very complex subject. Ask that guy D’hont.
has the obvious flaw that
First past the post seems fundamentally unjust to me.
Here in Ireland, in the middle of an election! I can rank the candidates in order of MY preference, choose from one or all of the candidates for a particular party and know that if my first choice doesn’t get elected my vote isn’t wasted as the second choice will be taken into account. I love the fact that I may have three candidates in my chosen party to chose from and also that my constituency will be represented by three, four or five people. This means that there is much fairer representation of the people and that an individual’s local work can be recognised.
The Irish have taken this to an art form and people can vote tactically to get the most from their vote.
Having lived in England, where the party chose the single candidate, I found myself in the terrible position of having to vote for someone I detested because I had no choice. As some areas are predominantly infavour of one party, people who prefer the opposition have no hope of having a say. In Certain places, you could put forward a monkey, and as long as it is in the right party, it will get elected. where’s the democracy in that???
I thought after I posted that that I should have made an exception for the Progressive Democrats (who, by the way, are neither) - except I think that in this case they really do prove the rule. The only smaller party that isn’t virtually irrelevant, and - guess what? - there’s also very little to distinguish them from Fianna Fáil and Fine Gael. So the Irish system still does not justify the argument that PR gives extremist parties more power.
In “first-past-the-post” the electors are consulted about the specific candidate - if the party leadership parachutes a monkey into a safe-seat, they may well end up with egg on their faces.
It would be much harder to defeat a monkey put at the top of party-list.
With respect to Ireland’s system, my concern is the cost of running: a much more practical impediment to democracy than the specifics of electoral mechanics. The US experience suggests that there are no economies of scale in running for office - getting your message out to 200,000 voters is four times as expensive as getting it to 50,000.
The topic of voting systems is extremely complex. Some sources:
ALTERNATIVE VOTING SYSTEMS by Steven J. Brams, Department of Politics, New York University, and Peter C. Fishburn, AT&T Bell Laboratories.
Perl, Politics, and Pairwise Voting: Perl as the Activist’s Friend by
Rob Lanphier.
A document in Spanish.
And P. C. Fishburn, S. J. Brams. Paradoxes of preferential voting. Mathematics Magazine, vol. 56, no. 4, (1983), 207-214. (I could not find it online)
Thank you “sailor” - this looks like just the thing I want!
America has a problem in that few of its Congressional districts are genuinely competitive - Congress has managed to carve out ‘safe’ seats for most of its members, with the result that only a relative handful of seats can change parties in a given election.
Consequently, very few Americans are in a position to affect the outcome of a Congressional race, which is a pretty ridiculous state of affairs.
Proportional representation is one cure for that. But it could be done equally well by defining some measure of geographical cohesiveness of Congressional districts, and mandating that the most cohesive set of district boundaries be used. This would still produce a certain number of safe seats, but not nearly so many as under the current system: to get this many safe seats requires intelligent design, rather than mere chance. 
Sailor alluded to Kenneth Arrow’s Impossibility Theorem, and perhaps the narrower Condorcet’s paradox.
Nice cites. Thanks.
I recall reading a paper which studied a Union election in Britain, where the voters had ordered their candidate preferences in great detail. Using that data, the reseachers were able to test the extent whether different voting systems produced different outcomes.
Basically, however, all of the various systems produced similar outcomes with one exception. The exception was first past the post voting. This suggested that voting reform is an easier problem than it appears: merely avoiding FPTP will produce better results. So 1) we don’t have to get too bogged down in the technicalities and 2) we can choose a simpler alternative voting system such as Approval Voting.
Examples of Problems with FPTP
In the US, there is periodic discussions of “spoilers”, such as Perot, Ralph Nader and Jesse Jackson. In any three way race, the winner becomes the one whose side of the ideological spectrum has fewest candidates. This effect can be seen most clearly in primary elections: if a cluster of candidates cover a similar ideological space (not unlikely, nor undesirable) that sort of position loses. Winners therefore try to pursue an uncrowded ideological space, even if it isn’t especially popular.
One might argue that such an event occurred during the 1980 Republican primary, when the most conservative candidate (Reagan) beat a handful of moderates (including George Bush I). The policies that followed, labeled “Voodoo Economics” by candidate Bush, had (shall we say) consequences that were not predicted by its advocates.
I remember the following example from my History class at school (when I was 14) to illustrate why FPTP can end up being underrepresentative of the entire population’s wishes:
Imagine a country of only three constituencies. There are two parties, the Xenophobes and the Yentls. The consituency breakdown of the vote is thus:
Constituency A: 10,000 voters 75% X, 25% Y: An X seat
Constituency B: 10,000 voters 45% X, 55% Y: A Y seat
Constituency C: 10,000 voters 49% X, 51% Y: A Y seat
In this result, the Ys have a majority in two constituencies, and thus have a parliamentary majority to form a government. However, the actual numbers who voted accross the country are:
Y: 13,100 votes
X: 16,900 votes
The Ys have formed a government with a minority vote.
The flip side is, if you use one of the PR systems in this example, you may have to give an X candidate to one of the constituencies that voted for Y.
I guess that’s why, ridiculously complex though it is, the Irish system of having second, third, fourth, etc. choices is more fair.
Ah but you see jjim this has been remedied in both the US and the EU by the allocation of seats per constituency based on relative size. As pointed out in earlier posts this has other inherent problems, but it does safeguard us from ending up with a minority domination of parliament formed on the faux majority vote your example describes.
Now as we might remember this was debated hotly as a part of the end phase of the 2000 presidential in the US, were it was feared that once the smoke settled the winner would have an effective minority of the vote but a majority of the elector seats. As it turned out the two followed each other.
So is jjimm’s example - each consituency has one seat per 10,000 voters.
Indeed, I stand corrected and somewhat dumbfound at my own stupidity!
Sorry jjim and thank you sirjamesp.
The system has worked reasonably well here in Ireland and generally the politics of compromise and consensus have been encouraged by the exigencies of inter-party post-election Government formation. Also, the argument that instability is an inevitable consequence of PR hasn’t held true in the Irish context. A two party Government supported by four independents has survived a full five year term (it was helped by the fact that none of the independents could be described as ideologically driven - I would have worries about the stability of a Government involving the Green Party or similar). I would also agree with Ruadh’s view in relation to the distinct lack of extremism in Irish politics - parties have increasingly clustered around the centre. This is obviously not necessarily a good thing - the current disillusionment with politics stems at least partly from this.
However, one negative consequence of the Irish PR system has been overlooked. Because candidates are competing against party colleagues, there is no room for differentiation on policy grounds (Irish parties are much more homogeneous than their US counterparts). This means that candidates must compete on non-policy grounds and this leads to a dangerous level of political clientelism. National legislators are reduced to making meaningless enquiries in relation to minor issues on behalf of their constituents just to be seen to be doing something. This is despite the fact that the representations, in the vast majority of cases, have no effect at all. They need to perpetuate the myth that its ‘who you know’ rather than rights and entitlements that secures services. At it’s most extreme, the system is a powerful impetus to corruption, albeit of a different kind to that suggested in the OP.
This focus on ‘parish pump politics’ means that representatives don’t dedicate enough of their efforts to national issues and influencing policy, examining the executive, monitoring public expenditure, shaping legislation etc. The national parliament suffers as a result.
There are ways of tackling this without changing the electoral system (e.g. strenghtening the role of local Government and local elected representatives) but it needs to be acknowledged.
I’m usually an apostrophe Nazi so apologies for my reversal of ‘its’ and ‘it’s’ in the last post. I’m off to get my petard (I think it’s in the garage under the ab isolater and the gutbuster I bought when I was drunk) upon which I shall be hoisted.
MWAP (who couldn’t just leave it alone)
So there I am in my shower, thinking to myself…how can I be so stupid? Obviously I can’t be that stupid. Or? Well…maybe my previous missive in reply to jjimm wasn’t exactly founded on clear and focused thoughts I’ll get back to my conjecture.
In the US and the EU the constituency break down of seats is over such wide geographies and hence focused on such different interest groups that I believe this remedies the problem. For these two cases it makes no sense to assume that neither the Yentls nor the Xenophobes will be identical from one constituency to the other. It is the elected representative that counts and it makes some sense to let each constituency decide what flavor this representative should have without interference from the general majority vote. Granted that once they assume their seat it is probable that they will vote and legislate along general party lines. But let’s face it, GOP representatives from California are very different from GOP reps from Texas. Just like Socialists from France are damned well different from from Labour reps in England. It gets even more complicated when we consider constituencies within the constituencies… But that too much for someone who just went from stupidity to addled thinking in the time span of a shower.
My point? An argument could be made for constituency based seats on the basis of regional autonomy in political bodies which rule large populations over large areas like the US Congress or the EU Parliament. I do however believe that it makes little sense within a smaller state like Ireland, where it will inevitably lead to misrepresentation and your current system seems better.
Toronto used to have two aldermen per ward, with municipal politics supposed to be non-partisan. Most wards would end up with one who focussed on city issues and one who specialized in barking dogs & pot-holes.
This system ended before I became interested in municipal politics and I have no opinion on how well it worked. Municipal politics is so basic-service and name-recognition driven that the analogy is not very good - but I thought I’d mention it.
Arizona has a good system for this. They created two minority-Hispanic districts, one in urban Phoenix the other hitting parts of Tuscon, Yuma and Phoenix, three suburban Phoenix districts, a Tuscon-based district, an entirely rural district and a suburban/rural district that’s basically what’s left over (the Navajo and Hopi reservations legally have to be separate. It’s a system I think all states should use, because it tosses partisan interests out for regional cohesiveness. Other states that have independent commissions aren’t as good (Iowa uses a computer to draw its districts, but they don’t keep communities of interest together well, and New Jersey basically has Dems and Reps submit maps and a professor decides which is the least gerrymandered). We should implement the Arizona model nationwide, but redrawing of lines is a state matter so it would be a long and slow process.
