The Electoral College

[Lidless]

So Obama wins 53.3% of the popular vote yet over 67% of the Electoral Votes? Not to mention the 2000 election. This seems strange for a country that advocates democracy and equal representation. I thought I’d investigate.

There are a few quite good articles that discuss the pros and cons of the Electoral College system better than I could.

http://www.uselectionatlas.org/INFORMAT ... procon.php
http://usinfo.state.gov/journals/itdhr/ ... ulness.htm
http://en.wikipedia.org/wiki/U.S._Elect ... al_College
http://en.wikipedia.org/wiki/U.S._Elect ... al_College

So in summary the presidential election was never designed to be a popular vote (one leg of the ‘checks and balances’ does that – sort of) but a mechanism to garner wide support across diverse geographical areas.

Never mind the election now being dominated by swing states now as opposed to urban areas, I did some analysis on the 2008 results. Oh boy does the present system come at a cost!

In the numbers below:
EC = the number of Electoral College votes

Voters = the number of registered voters as at 2004 – the latest reliable source (http://www.statemaster.com/graph/gov_20 ... red-voters)

Republican / Democrat = the number of votes cast (as of now)

EC Weighting = (Electoral College votes) / (Registered voters), compared with the national average. So in Wyoming a voter has in effect 3 times the effective strength compared with the US as a whole. Wisconsin has the weakest.

Code: Select all

State             EC       Voters  Republican    Democrat     EC Weighting

Alabama            9    2,418,000   1,264,879     811,764             0.98
Alaska             3      334,000     136,585      80,505             2.37
Arizona           10    2,485,000   1,131,790     948,066             1.06
Arkansas           6    1,328,000     632,672     418,049             1.19
California        55   14,193,000   4,144,693   6,765,199             1.02
Colorado           9    2,307,000   1,020,135   1,216,793             1.03
Connecticut        7    1,695,000     614,584     973,302             1.09
Delaware           3      415,000     152,356     255,394             1.91
Florida           27    8,219,000   3,939,380   4,143,957             0.87
Georgia           15    3,948,000   2,048,244   1,843,452             1.00
Hawaii             4      497,000     120,309     324,918             2.13
Idaho              4      663,000     400,989     235,219             1.59
Illinois          21    6,437,000   1,975,801   3,293,340             0.86
Indiana           11    3,031,000   1,341,667   1,367,503             0.96
Iowa               7    1,674,000     677,508     818,240             1.10
Kansas             6    1,338,000     685,541     499,979             1.18
Kentucky           8    2,231,000   1,050,599     751,515             0.95
Louisiana          9    2,413,000   1,147,603     780,981             0.98
Maine              4      824,000     296,195     421,484             1.28
Maryland          10    2,676,000     938,671   1,579,890             0.99
Massachusetts     12    3,483,000   1,104,284   1,891,083             0.91
Michigan          17    5,364,000   2,044,405   2,867,680             0.84
Minnesota         10    3,080,000   1,275,400   1,573,323             0.86
Mississippi        6    1,510,000     687,266     520,864             1.05
Missouri          11    3,336,000   1,444,289   1,439,321             0.87
Montana            3      519,000     241,816     229,725             1.53
Nebraska           5      918,000     446,039     324,352             1.44
Nevada             5      965,000     411,988     531,884             1.37
New Hampshire      4      716,000     316,937     384,591             1.48
New Jersey        15    4,085,000   1,545,495   2,085,051             0.97
New Mexico         5      936,000     343,820     464,458             1.41
New York          31    8,624,000   2,576,360   4,363,386             0.95
North Carolina    15    4,292,000   2,109,698   2,123,390             0.92
North Dakota       3      412,000     168,523     141,113             1.92
Ohio              20    6,003,000   2,502,218   2,708,988             0.88
Oklahoma           7    1,781,000     959,745     502,294             1.04
Oregon             7    2,049,000     699,673     978,605             0.90
Pennsylvania      21    6,481,000   2,586,496   3,192,316             0.86
Rhode Island       4      522,000     157,317     281,209             2.02
South Carolina     8    2,238,000   1,034,500     862,042             0.94
South Dakota       3      425,000     203,019     170,886             1.86
Tennessee         11    2,739,000   1,487,564   1,093,213             1.06
Texas             34    9,681,000   4,467,748   3,521,164             0.93
Utah               5    1,141,000     555,497     301,771             1.16
Vermont            3      354,000      96,458     203,952             2.24
Virginia          13    3,441,000   1,726,053   1,958,370             1.00
Washington        11    3,133,000   1,098,072   1,548,654             0.93
Washington, D.C.   3      293,000      14,821     210,403             2.70
West Virginia      5      935,000     394,278     301,438             1.41
Wisconsin         10    3,225,000   1,258,181   1,670,474             0.82
Wyoming            3      265,000     160,639      80,496             2.99

Total            538  142,072,000  57,838,800  66,056,046             1.00

[/size]

OK, a bunch of numbers. What does it tell us?

First Past The Post
Firstly, the first past the post system where winner takes all in each state is deeply flawed. Only Maine and Nebraska use a form of proportional representation. In California over 4m people voted Republican but were not represented by any EC delegate. Across the country it added up to 52.3m people being effectively disenfranchised. That’s 42% of the people who voted. In the infamous 2000 election it was 'only' 47.7m people, but that represented 47.0% of all who voted.

Over 40% of the electorate disenfranchised? Whatever the pros and cons of the system, this is too much of a con, in more than one sense of the word.

Of course, there is a systematic flaw in the concept. Technically, and I stress technically, you could have had 11 people vote for one candidate (one in each of the 11 biggest states with no other vote cast), and over 64.7m vote for the other candidate in all the other States – and the 11 people would have won. Somewhat a reduction ad absurdum approach since it simply wouldn’t happen in real life, but it seems strange to have a system where technically it could.

We all know that in 2000 if just 500 voters had voted differently (Florida), Gore would have won the election. What about this year? Well, the minimum number of voters it would have required to vote differently for McCain to have won is just 604,000 - Colorado, Florida, Indiana, Iowa, New Hampshire, New Mexico, North Carolina, Ohio and Virginia. A majority of the EC, but only 47.7% of the popular vote. If instead of pure votes, we go by easiest swings – ie marginal seats – it’s not that much more. Only 694,000 votes – take New Mexico out of the list and replace it with Minnesota. This time that’s a whopping 47.8% of the popular vote.

Proportional Representation
If each state had apportioned the EC proportionally, Obama would have received 288 EC votes (53.5%) compared with 53.3% of the popular vote. Pretty close considering there is weighting for each state. And what about the 2000 election? Well no candidate would have received a majority (and Nadar would have received 7), it would have gone to the House of Representatives which was Republican at the time so Bush still would have won. Even taking the third party candidates out and apportioning the EC votes based purely on Rep / Dem votes incredibly it would have been a tie (269-269) and would again have gone to the House of Representatives.

Interesting to see that in both years, despite different weightings given to different states, using proportional representation within each state is very close to the overall popular vote.

Rural weighting
I did wonder if the weighting towards rural areas automatically skewed the EC in favour of the Republicans. Rural areas, with the population more spread out, tend to be more self-sufficient, less ‘community-minded’ and therefore tend to lean more to the Republicans. A simple analysis shows there is no significant correlation between the EC weighting and the Republican % in both 2008 and 2000 – even after taking out the anomalous Washington DC.

Of course, the US Senate, since it is just two per state no matter what the population in the state, has even more of a rural weighting and so is even worse.

I make no comment about using such weightings other than to say, well, it feels wrong. I reserve the right to form a more considered opinion later!

Voting Power
I did wonder whether the weightings gave even more power than one might initially think to any one state. Consider the following hypothetical: the US is made up of just two states, Virginia (3,441,000 voters and 13 EC votes from the above table) and Massachusetts (3,483,000 voters and 12 EC votes). Although the latter has more voters, it has no voting power whatsoever since whatever the Virginia voters decide will end up with the majority of the EC votes (13 out of 25). The voting power is Virginia 100% (13 votes), Massachusetts 0% (12 votes).

Let’s add another state into the mix – Rhode Island (522,000 voters and 4 EC votes). This time Virginia has a say in the matter, but no more than the other two! Any combination of two out of the three will produce a majority. The voting power is now Virginia 33% (13 votes), Massachusetts 33% (12 votes), and Rhode Island 33% (4 votes).

So how does it look like for all of the US? Well I ran the model and it turns out that there is no undue influence as a result. In fact the correlation between electoral votes and voting power is 99.8%. Only California has any meaningful voting power over and above the votes (11.4% of the power, 10.22% of the votes).



Conclusion
The Electoral College system certainly needs a fresh rethink. Personally, I’d go with proportional representation within each state, albeit a more direct version than that in Maine and Nebraska. As to the disproportionate weighting by state as a whole, I still have to think about that.

[sauronsfinger]

Excellent post Lidless. And all that data you researched and provided for us shows us just why it would be tremendously difficult to get 3/4 of the states to ratify a Constitutional Amendment changing or outright aboloshing this system. States like Wyoming, Alaska, Vermont and Idaho are simply not going to change the system which today gives them disproportionate power.

When I taught high school Government and US History for 34 years we always studied this system and I would point out the elections where the popular vote went one way and the Electoral Vote went the opposite. I used to say that we will change this someday but since it has not happened in 100 years or so, we will only change it when it happens in our lifetimes.

So then we get the election of 2000 and I am proved to be a liar. The American people sat on their collective behinds and did nothing about it. People care more about reality TV and stupid pop culture figures than they do issues like were raised in the 2000 election. Sad sad sad.

There is a move afoot where individual states pass a law which pledges the Electors of their own states to which ever candidate finishes first in the national popular vote. It would take effect in the election after enought states join in the movement and reach the equal of 270 electoral votes.

Of course, that would be a collection of individual state laws and you have to wonder how long before a scenario would arise would arise where one or more states who agreed to do it that way simply rebel because their own voters are so opposed to the popular vote winner that they go back on their word.

Can anyone say US Supreme Court?

here is a column in which the movement is mentioned

http://www.washingtonpost.com/wp-dyn/co ... 00808.html

[Ellienor]

Well, to get rid of it requires amendment of the U.S. constitution (very unlikely) or each small state passing a law changing to proportional electors, as Nebraska and Maine have. However, the small states really have no incentive to do it, since they retain outsize influence relative to their size (as you so ably point out). The whole thing (and the Senate rules) were specifically incorporated during the writing of the Constitution because of the small states' fears of being overrun and ruled effectively by the wishes of the big states.

I'm from Colorado, and I like that I saw my new president-elect once in person. If it was popular vote, campaigning would be limited to urban areas.

But I recognize the problems with vote disenfranchisement, but that is the way the Founders set up the Union--with mechanisms specifically designed to increase the power and influence of the small states.

[Holbytla]

Well there is this http://en.wikipedia.org/wiki/Presidenti ... Reform_Act and what I and this
http://www.america.gov/st/elections08-e ... 71936.html
and this
http://www.ncsl.org/programs/legismgt/e ... 7-2008.htm

As you can see from the last link, there is a lot of move afoot to try and change things, but also as you can see most bills are stuck in "pending" status. The effort to change this type of legislation is tough with having to garner 2/3rds vote and then there is the constituionality of it all.

The ideal of the system is sound and rarely are things as off kilter as they were in 2000. But certainly a move to a more democratic primary type system would serve better. Just that the amount of effort to change is probably more than the desire at this point.

[Lidless]

I'm from Colorado, and I like that I saw my new president-elect once in person. If it was popular vote, campaigning would be limited to urban areas.

As opposed to swing states under the current system.

[Holbytla]

We never get visitors. Fait accompli.

[Voronwë the Faithful]

No time to comment substantively right now, but I just wanted to say that I was glad to see this thread. I think it is an important and interesting topic to discuss.

Thank you, Lidless!

[vison]

Cool, Lidless. Thanks.

[Lidless]

I have a serious interlektyooal...inturlectool...whatever side too, you know.

[Voronwë the Faithful]

[sauronsfinger]

If 2000 had happened again this year with Obama taking the popular vote but losing the Electoral vote, I suspect, I strongly suspect, things would not have been as quiet as they were with the shenanigans of 2000. Maybe then something would have been done about it.

[Holbytla]

Ah I knew there was something that bugged me about those numbers. Your proportions are based on voters rather than population. Whereas representation is based on population.

United States congressional apportionment is the redistribution of the 435 seats in the United States House of Representatives among the 50 states in consequence of the constitutionally mandated decennial census. Each state is apportioned a number of seats which approximately corresponds to its share of the aggregate population of the 50 states (populations of Washington, D.C. and federal territories are not included in this figure). However, every state is constitutionally guaranteed at least one seat.

The decennial apportionment also determines the size of each state's representation in the United States Electoral College—any state's number of electors equals the size of its total congressional delegation (i.e., House seat(s) plus Senate seats).

Federal law requires the Clerk of the House to notify each state government of its entitled number of seats no later than January 25 of the year immediately following the census. After seats have been reapportioned, each state determines the boundaries of Congressional districts—geographical areas within the state of approximately equal population—in a process called redistricting.

[Lidless]

I can easily redo the EC Weighting based on total possible voters, but that's the only analysis it will change. All other results and conclusions remain the same, as it does not effect anything in the analysis other than Wyoming is 2.99 and Wisconsin is the most diluted..

I suspect the result of that will not be significantly different.

[ToshoftheWuffingas]

Are there Federal laws on redistricting? Some of those boundaries look weird to me. Given that the districts elect Federal representatives I would have thought there should be. One can disenfranchise groups by the way you draw boundaries. Say you have a group strongly localised with a certain voting pattern. Cutting pie slices into that locality so each segment is diluted by a bigger opposing community removes that group's voting power.
How is that prevented?

[sauronsfinger]

Tosh - the rules for drawing districts are pretty loose and up the the State legislatures who have the responsibility to do so every ten years using data from the census. Basically, we use to follow three major rules
1- each district must have the same number of persons in it
2- each district must be in one piece, ie. contiguous
3- the district cannot be unfairly drawn (gerrymandered) to give unfair advantage to one party over another

#3 has gone by the boards in favor of each party getting equal advantage in the same number of districts that is reflected by vote totals. In other words, lets say a state has 60% vote for Republicans and 40% for Democrats. And lets say that the state has 20 Congressional districts.
The people doing the drawing might create only four districts where there really is a pretty fair and equal division where either party has a good chance of winning. Nine or ten districts might be drawn to intentionally produce a Republican winner while six or seven districts might be drawn to intentionally produce a Democratic winner.

Its not suppose to be that way but in practice that is exactly what it is.
The US Supreme Court has the job of reviewing any appeals for unfair gerrymandering.

[ToshoftheWuffingas]

During our Thatcher years the Tories had a showcase district in the centre of London, Westminster. They used it to show the wonders of reducing services, privatisation and low taxation. They sold off a cemetary to property developers for less than a dollar. To ensure they kept in power they rehoused low income residents out of their district thus diminishing their voting strength. In asbestos ridden tower blocks.
When the long arm of the law eventually caught up with them and demanded our public money back the leader fled abroad.

[Túrin Turambar]

I can think of some more urgent reforms that tbhe U.S. needs before the abolition of the Electoral College, but getting rid of it would probably be a good idea. At least then everyone would have an incentive to vote for President no matter where they lived.

[Lidless]

Tosh, Dame Shirley Porter, the Tesco heiress worth around GBP 200m was fined GBP 27m. She transferred almost all of her wealth to other members of her family and trusts in an effort to avoid the charge, and subsequently claimed assets of only GBP 300k.

The case was finally settled in 2004 where she agreed to pay GBP 12m, and now has a GBP 1.5m home in London - Westminster of course - as well as other properties abroad.

Amazing how GBP 300k can grow over time.

[ToshoftheWuffingas]

Ta for the update Liddy. I didn't know they had finally stung her though it seems she got off lightly by £15 million. I hope she had to pay £20 million to her lawyers to escape that. Moving voters out is one thing, moving them to properties which were known at the time to be an asbestos hazard is a whole different ballgame. She wasn't the only one guilty in all that affair, only the best known.


I return you from UK corruption to your normal discussion...