If the electoral college trend were to hold for next 100 years

[Padfoot]

Split Cali in three.

At least three if not four or five.  If LA County (the most populous in CA and the US) were to secede and become its own state it would be the 9th largest in the nation (based on 2005 Census estimates).  As its own state it would probably elect 15 or 16 congressman and have 17 or 18 electoral votes.  What happens to the new LA-less California then?  Not only would it still be in the top ten most populous states, it would also remain at the number one spot with about 4 million more people than Texas (the next largest state).

I would actually argue for splitting New York, Texas, and Florida as well.  These states weild too much power in the House and Electoral College.  The fact that these states are some of the worst gerrymandered in the country makes me even more uneasy.  The governments of only 4 states should not be in control of drawing districts for nearly a third of our congressmen.

[Kevinstat]

One interesting thought: What if Congress drew Congressional districts?  Subject to the President's veto (which could be overrided by a 2/3 majority in each house) of course.

On first glance that would seem to make the potential for gerrymandering worse since if one party controlled the federal government it could gerrymander every state's districts to it's advantage (currently gerrymandering by different parties in different states can at least somewhat cancel itself out - Republicans did really well in the 2000 state legislative elections so that didn't happen as much, but at least New York state couldn't be gerrymandered to their benifit).  If a nationwide Maine/Nebraska method was used, the governing party could thus arguably cement its hold on the Presidency as well as the lower house of Congress.  Now that the Supreme Court has ruled that redistricting can occur more than once a decade (I don't think there's any limit now to how often a state can redraw its congressional districts; if I'm wrong can somebody tell me), the governing national party could "re-gerrymander" (a term I'm sure has been used to desribe Texas Congressional redistricting) district lines in a state if and when their earlier plan seems not to be working for them anymore given unexpected political trends.  Or a party could draw a plan which they are only fully confident will ensure their majority through the next election, and then draw another plan if and where necessary to ensure their majority though the next election, and so on.  But such nationwide gerrymandering by one party would likely be very unpopular, especially since the party couldn't point to the other party doing it too at the same time in other states, so the costs to the national majority party of exteme gerrymandering could outweigh the benefits.  And if the constitutional amendment giving Congress authority to redistrict itself included an explicit restriction of redistricting to between the xyz0 and xyz2 elections, a national political party trying to ensure its majority for the remainder of the decade would have to make guesses about what the partisan ballance throughout each state they were gerrymandering would be throughout the ensueing decade.  If, by [xyz+1]0, the party which was the victim of the gerrymandering in xyz1 or early xyz2 was preferred over the other party by a greater margin than say, 52% to 48% nationwide, it would likely have control of at least one house of Congress or the Presidency at the end of that year.  It would then be able to protect itself against gerrymandering for the next decade, or exact revenge on the other party through gerrymandering of its own if it controlled the Presidency and each house of Congress and, perhaps a less extreme gerrymandering which would teach the opposition party not to take so great an advantage of its power again without being so bad as to continue the cycle of extreme nationwide gerrymandering whenever it is possible.  Perhaps I'm a dreamer, but it's an interesting dream to have (not that I had an actual dream about this, but you know what I mean).

One final thing.  I think a Congressional district method of choosing electors to the electoral college would be better if each state had only as many electoral votes as it had U.S. Representatives.  That way there would be rough equality in voting representation and power both in the lower-level math sense (the number of electors each voter can vote for divided by number of voters voting for that elector or those electors) and the higher-level math sense (the number of electors each voter can vote for divided by the square root of the number of voters voting for that elector or those electors, which is a better indicator of how much chance voters in different states or other units of election of Presidential electors have relative to each other of affecting the outcome of the election).  In the lower-level math sense the larger states are currently cheated, but in the higher-level math sense they have it really good as their voters gain more power as a whole by their larger number of electors (then the smaller states) being grouped together, and that benefit would generally, assuming random political ballances throughout the several states (which I know it isn't random, but it is still a good key), outweigh the benefit to the smaller states of having more electoral votes per capita.  Having a Congressional district method without the two statewide "Senatorial" electors would eliminate each of these two benefits certain states have over others based on their size.

[Padfoot]

Eliminating the two "extra" EVs each state gets for its Senators would directly undermine one of the biggest compromises forged by the founding fathers.  The House was created to represent the will of the people.  The Senate was created to ensure that the views of states with small numbers of people were not ignored due to their low populations.  Thus, by combining these two principles in the EC, everyone is happy.  States with high populations are still given a lot of power in the EC but the small states don't get ignored.  Plus if you eliminate the "extra" Senate votes, that means the nearly one million people in Montana only get one electoral vote.  If 1/100 of the population is only getting 1/435 of the vote in deciding who is president you have a very lopsided Democracy.

I think it makes the most sense for states with 10 or less House seats to be on the Congressional District method.  For states with more than ten House seats you use the proportional method with a twist.  I'll use the 2004 election results and Ohio as an example to help me explain.  If you use the pure proportional method each candidate gets 10 of Ohio's EVs because the race was so tight here (50% Bush 49% Kerry).  However if we only divide the "House" EVs based on vote percentage and award the two "Senate" EVs to the winner of the popular vote Kerry gets 9 EVs and Bush gets 11 EVs.  Thus the winner of the state still comes out with more EVs but the overall result is still fairly equal to the popular vote.  Also, this mirrors the CD method in which the winner of the popular vote wins the two "Senate" EVs and it lessons the affect of gerrymandering since it is harder to gerrymander when you only have 10 CDs.

[Kevinstat]

Montana does not have 1/100 of the United States population.  As of the 2000 census, it had just under 1/311 or 1.40/435 of the United States resident population (not counting DCs resident population in the denominator) and just over 1/311 or still just under 1.40/435 of the United States apportionment population, which includes oversees population.  It had 1.51/435 of the United States resident population (which was eqivilent to apportionment population back then, as overseas population was only included in the apportionment population once before the 1990 census) as of the 1980 census, and had 2 representatives in Congress, so they were overrepresented in the U.S. House of Representatives by 32.39% back then while they are underrepresented in the House by only 28.54% now based on their apportionment population or by 28.44% based on their resident population.  Montana is currently overrepresented in the Electoral College by 73.69% based on its percentage of what the United States apportionment population would be if the District of Columbia was included and by 73.94% based on its percentage of the United States resident population.  Of course now it likely has less real power than average as it has a smaller number of electors who are elected in a slate, but my plan would get rid of that as well and would each congressional district one electoral vote.  Congressional districts wouldn't likely be ignored based on the size of the state they belong to, but based on how solid they seem for one candidate or another.  It would be like the U.K.'s Westminster system in a way.

I live in the less competitive of Maine's two congressional districts, although I would still hear advertising for the second district as I currently do as that district stretches well into the Portland media market which includes the Augusta area where I live, but my vote would be slightly less likely to affect the choice of an elector or electoral slate than it is now in a close election (two of Maine's four electors are still elected statewide now, you know).  That's okay with me as I know that inequality vis-a-vis more competitive congressional districts would be based on the different voting preferences of voters in my congressional district compared to those in routinely closer congressional districts, not systematic as the current method of allocation and election of Presidential electors is (well, I guess each state could hypothetically have the same population and number of voters, but that's kind of different in my opinion).

[True Federalist (진정한 연방 주의자)]

Here's another map for you, assuming the percentage rate of change for each State between 1990 and 2000 holds steady each decade until 2100.

Green states increase theiev EV's in comparison to 2000, blue hold steady and red decrease.  Nevada with a population of 322,643,950 has the most EV's at 89.  I guess a lot people will be staying in Vegas.  Obviously this extrapolation is ridiculous, but so were the others.

[jimrtex]

One interesting thought: What if Congress drew Congressional districts?

This is certainly within their Constitutional authority.

You must be logged in to read this quote.

Or they could farm it out to an independent agency.  This is what they have done in the case of apportionment, where they have embedded the formula in law, and simply have the Census Bureau report the apportionment numbers.  This avoids debates over apportionment methods and number of representatives.

You must be logged in to read this quote.

There is a possibility that if conducted at a national level, any abuse would be more visible.

You must be logged in to read this quote.

Congress has directed that the representatives be elected from districts.  Since the manner of electing representatives is vested in the States (subject to Congressional override), the States have authority to modify their districts, just as they modify their other election laws.

In the Texas redistricting case, the plaintiffs argued that it was illegal to use 2000 census numbers for redistricting in 2003, if the redistricting was done voluntarily, based on an argument that it violated the equal protection clause.

You must be logged in to read this quote.

The Texas legislature did not draw the plan used in in the 2002 elections.  It was drawn by a federal district court.  Under the US and Texas Constitutions, as well as federal and Texas law, the authority to district is vested solely in the Texas legislature.  At best, the federal court plan should be considered a remedy for 2002 only, subject of course to the continued dereliction of the Texas legislature.   In their 2001 decisions, the district court noted that they were incompetent to consider many issues that were legitimate considerations for the legislature.  The plaintiffs were told to go tell the legislature.  This was appealed to the US Supreme Court (in essence arguing that the federal district court could and should legislate from the bench).  The USSC refused to take the case.  The best term for the 2003 redistricting is simply "redistricting" or "reapportionment" (for historical reasons, districting is also referred to as reapportionment.

You must be logged in to read this quote.

There may be a bias here.   Many Democrats tend to be indifferent to such things.  They may think that they are merely redressing past grievances.

You must be logged in to read this quote.

No Constitutional amendment is needed.

You must be logged in to read this quote.

Historically, when electors were elected from districts, they were chosen from electoral districts (eg Maine would be divided into 4 electoral districts, each choosing a single elector).  The current system used by Maine and Nebraska of having a mixed system (district plus at large electors) is unusual.

[Padfoot]

Yet another map.  This one shows how things would look in 2100 if each state had the same net increase of people as they did from 2000-2005 every five years.  (ND was the only state to have a net loss from 2000-2005 according to Census estimates).


Grey=Unchanged
Red/pink=Loss
Blue=Gain

Darker colors indicate larger losses/gains and lighter colors indicate smaller changes.  Some of the Northeast states are hard to see  so here's some of them written out: DE-4 gain of 1, NJ-13 loss of 2, CT-6 loss of 1, RI-3 loss of 1.

[Kevinstat]

jimrtex,

Were some of the concurences in judgement of certain U.S. Supreme Court Justices on the 2005 case regarding the 2003 or 2004 Texas congressional redistricting, ... did any justices basically say that the second redrawing of the lines (not getting into the issue of TX-23) was only okay because the first drawing wasn't done by the Legislature?  I know certain Republicans have said that while fuming against Democratic attempts to redistrict Ohio by citizen initiave while justifying the 2003 Texas redistricting, but I'm curious if any Supreme Court Justices (like Scalia perhaps) came to that conclusion.

[Kevinstat]

By the way, jimrtex, thanks for the info on how the District method was previously done.  So those districts would have always been smaller than the state's congressional districts.  Perhaps Lewis Trondheim (or whatever he goes by nowadays) or one of our other map-happy members could start a thread on dividing each state into Presidential elector districts with no at-Large electors.

If each state elected all of its presidential electors by single-member districts, and assuming the same percentage of citizens voted in each state (obviously not true, but good for making my point), a voter in state with exactly 1/435 of the U.S. population would have 189% more representation in the electoral college per capita than a voter in a state with exactly 53/55 of the U.S. population (that ratio would not be changed by electing any or all of a state's electors by districts, only within a state if the districts were substantially unequal in population per elector), but only 70% more voting power using the higher-level math I described in my first post on this thread.  That is slightly less equitable than if both states used the Maine-Nebraska method, in which the voter in the smaller of those two states would have only 43% more voting power.  Currently, a voter in the larger of those two states would have almost 152% more voting power than a voter in the smaller state even if there was no correlation between what state an elector is from and who that elector votes for for President.  With that correlation and either of the two most commonly used power indexes for weighted voting (the states effectively having weighted votes), I'm sure the current mathematical disparity in power in favor of voters in larger states is even greater.  Voters in the larger states benefit not only from electing a greater number of electors in a group but from the fact that those electors can be expected, with an occasional exeption or two in any given election, to vote together on the only two votes they cast.  That same second factor would come in to play to a lesser degree with the Maine-Nebraska method, only this time to the benefit of the smaller states as an individual district elector from a state with few Congressional districts would be more likely to be an elector for the same ticket as the state's two at-large electors than would be an elector from a state with more Congressional districts.  So the increased power a voter from a state with exactly 1/435 of the U.S. population would have over a voter from a state with exactly 53/435 of the population would actually likely be slightly greater than 43%.  But such a plan would still likely be more equitable based on state size than having all of a state's electors elected by districts.

[jimrtex]

Were some of the concurences in judgement of certain U.S. Supreme Court Justices on the 2005 case regarding the 2003 or 2004 Texas congressional redistricting, ... did any justices basically say that the second redrawing of the lines (not getting into the issue of TX-23) was only okay because the first drawing wasn't done by the Legislature?  I know certain Republicans have said that while fuming against Democratic attempts to redistrict Ohio by citizen initiave while justifying the 2003 Texas redistricting, but I'm curious if any Supreme Court Justices (like Scalia perhaps) came to that conclusion.

I don't think that there was any dissent as to whether authority to redistrict is vested in the state legislatures, and that there is no constitutional or legal bar to mid-decade redistricting.  Where there was dissent was in the manner such redistricting occured.  I don't think that it would have mattered whether there were multiple legislative redistrictings.

There were 3 parts to the decision.

III - VRA violation in South Texas, particularly TX-23.  That was 5-4;
Kennedy, Souter, Ginsburg, Stevens, Brever v. Thomas, Alito, Scalia, Thomas.

IV - VRA violation in DFW area, particularly TX-24, Rejected 4-5; Souter, Ginsburg, Stevens, Breyer v. Kennedy, Thomas, Alito, Scalia, Thomas.

II - Partisan Gerrymander.  This had two subparts.  One was whether, a voluntary re-redistricting done for partisan purposes denied equal protection to supporter of one party.  This was rejected on a 4-5 vote, Souter, Ginsburg, Stevens, Breyer v. Kennedy, Thomas, Alito, Scalia, Thomas.

Kennedy emphasized the point that such a standard would have left in place the 1991 Democrat gerrymander (because it was done during an effort to create equal population districts), while disallowing the 2003 Republican gerrymander (because equal population districts already existed).  In addition, Kennedy found that partisan districting wasn't the sole motivation behind the district plan.

Based on that, I don't see how Kennedy could object to one partisan redistricting plan drawn by a legislature being replaced by another 2 years later.

II - The second subpart was whether voluntary (eg. non-necessary) mid-decade redistricting violated one-man/one-vote.  This was rejected on a 2-7 vote, Stevens, Breyer v. Kennedy, Souter, Ginsburg, Thomas, Alito, Scalia, Thomas.   Basically the argument was made that it is OK to use census figures for redistricting, if you had to redistrict, but not if you were redistricting voluntarily. 

Scalia and Thomas would have rejected all of the partisan gerrymandering claims, and re-iterated that there was no justiciable standard.  Roberts and Alito, rejected the partisan gerrymandering claims for this case, but said that it was not at issue in this case whether such a standard could ever be found.

Basically you have the 4 activist justices who claim that there are standards that could be applied for partisan gerrymander cases (though they disagree on what those standards should be); Kennedy who insists that there might be such a standard, he just hasn't seen it yet.  And there are 4 conservative justices that say that there never is going to be a standard found.  So as long as Kennedy is around, you may still be getting these kind of cases.

[jimrtex]

Yet another map.  This one shows how things would look in 2100 if each state had the same net increase of people as they did from 2000-2005 every five years.  (ND was the only state to have a net loss from 2000-2005 according to Census estimates).

Interesting that SC would eventually gain a seat.  It has gone the longest without changing of any state with more than 2 seats without change.

1 - DE since 1820
2 - NH since 1880
3 - NE since 1960
4 - AR since 1960
5 - OR since 1980
6 - SC since 1930
7 - AL since 1970
8 - MD and MN since 1960
9 - MO since 1980
10 - MA since 1990
11 - VA since 1990
13 - NJ since 1990

[Padfoot]

Interesting that SC would eventually gain a seat.  It has gone the longest without changing of any state with more than 2 seats without change.

1 - DE since 1820
2 - NH since 1880
3 - NE since 1960
4 - AR since 1960
5 - OR since 1980
6 - SC since 1930
7 - AL since 1970
8 - MD and MN since 1960
9 - MO since 1980
10 - MA since 1990
11 - VA since 1990
13 - NJ since 1990

South Carolina actually doesn't increase its rank very much in my scenario.  It goes from currently being the 25th most populous stateto being 23rd in 2100.