How would the electoral college be if the 435 representative rule never happened

[muon2]

The increase in the size of the House in decades before 1920 was driven by political considerations, not merely population increase. The best approximation to the politics going forward from 1920 would be to set the total number of reps at each apportionment to the minimum number such that no state loses seats in the coming decade.

[muon2]

Here's how I would set up the alternate history. As IRL Congress could not agree to a reapportionment after the 1920 Census, in part over a dispute about the use of divisors. Also as IRL Speaker Longworth commissioned a panel of experts to determine the best method of apportioning representatives and they recommended the Huntington-Hill method (which wasn't adopted IRL until after the 1940 census). However, instead of the Reapportionment Act of 1929 which fixed the number at 435, Speaker Longworth pushed through the Apportionment Act of 1929 along with a spending bill to remodel the House chamber to accommodate more members.

The 1929 Act provided for the following:
States shall be apportioned no later than one year following each Census.
States shall be apportioned using the method of Huntington and Hill, with the number of seats set at the minimum number needed so that no state has fewer representatives after apportionment.
New states admitted shall have one representative until the next apportionment.
Districts drawn by states with more than one representative shall be of a contiguous and compact territory, and containing as nearly as practicable an equal number of inhabitants (from the 1911 act).

First applied in 1931, the apportionment resulted in a House of 537 members in 1933 with the following distribution, and an average district size of 228 K inhabitants.

In 1932 FDR wins the EC by 563 to 70.
In 1936 FDR wins the EC by 623 to 10.
In 1940 FDR wins the EC by 540 to 93.

[muon2]

Continuing my version of this alternate history, the Huntington-Hill method was already adopted in 1929, so no changes are made for 1941. The weak growth in UT sets the pace for the size of the new House at 586 members. The average district has 224 K inhabitants.

In 1944 FDR wins the EC by 555 to 127.
In 1948 Truman wins the EC outright with 396, compared to 236 for Dewey and 50 for Thurmond.

[muon2]

In response to the beginning post-war boom, Congress is concerned about the potential size of the post-1952 Congress. In 1947 they pass a new Apportionment Act that changes one provision:

States shall be apportioned using the method of Huntington and Hill, with the number of seats set at the minimum number needed so that no state that has increased in population has fewer representatives after apportionment.

In the 1950 Census four states lost population: AR, MS, ND, and OK. Under the 1947 revised act, AR, MS, and OK each lose one seat. Nonetheless, the House still increases to 642 members representing an average of 234 K inhabitants.

In 1952 Eisenhower wins the EC by 616 to 122.
In 1956 Eisenhower wins the EC by 636 to 101 (with 1 rogue elector).

AK and HI are admitted to the union in 1959 with one representative and 3 electors each.

In 1960 Kennedy wins the EC with 425 to 299, with 11 MS electors on an unpledged slate, 7 AL electors for Byrd, and 1 OK elector unpledged.

[muon2]

In the 1960 Census three states lost population: AR, MS, and WV. Under the 1947 act, AR and WV each lose one seat. Congress knows that population will force a big increase in the House, but they are shocked when apportionment requires 737 seats. New president Kennedy works with Speaker Rayburn to add a new House chamber that can seat at least 1000 and it's ready for Congress in 1963. The average district now has 242 K inhabitants.

Though DC would qualify for 5 EV based on its population, it is limited to 3 since it can't have more than the least populated state (AK).

In 1964 Johnson wins the EC by 761 to 79.
In 1968 Nixon wins the EC with 465, to 305 for Humphrey and 70 for Wallace (including 1 rogue from NC).

[muon2]

In the 1970 Census three states lost population: ND, SD, and WV. Both ND and WV lose one seat. The apportionment now requires 797 seats, and an average district has 254 K inhabitants.

In 1969 ME passed its law to divide its electors based on CD with 2 for the state winner. For this alternate history I made an approximation based on 4 districts with whole counties that were compact and contiguous with as equal a population as permitted without chops. The state voted for Pubs all three elections in the decade, but in 1976 and 1980 the Dems would get 1 elector.

In 1972 Nixon wins the EC by 872 to 27 (with one Libertarian vote from VA).
In 1976 Carter wins the EC by 506 to 394.
In 1980 Reagan wins the EC by 822 to 78.

[muon2]

The 1976 results saw ME split its electoral votes by CD. That raised the concern that other states might follow suit and if they gerrymander their districts, then that gerrymandering will skew the electoral vote. During that same decade IA was codifying its rules on drawing congressional districts, and in this timeline Congress adopts a version of the IA rules before the 1980 Census.

1. Congressional districts shall each have a population as nearly equal as practicable to the ideal district population. No congressional district shall have a population which varies by more than one percent from the applicable ideal district population.
2.  To the extent consistent with 1, district boundaries shall coincide with the boundaries of political subdivisions of the state. The number of counties and cities divided among more than one district shall be as small as possible. When there is a choice between dividing local political subdivisions, the more populous subdivisions shall be divided before the less populous, but this statement does not apply to a legislative district boundary drawn along a county line which passes through a city that lies in more than one county.
3.  Districts shall be composed of convenient contiguous territory. Areas which meet only at the points of adjoining corners are not contiguous.
4.  Districts shall be reasonably compact in form, to the extent consistent with the standards established by subsections 1, 2, and 3. In general, reasonably compact districts are those which are square, rectangular, or hexagonal in shape, and not irregularly shaped, to the extent permitted by natural or political boundaries.

In the 1980 Census only NY lost population and loses seats accordingly. As in the previous decade PA just barely gained population and was the last state to be awarded a seat, setting the total number. The apportionment now requires 885 seats, and an average district has 255 K inhabitants.

Since the smallest state now has 2 seats and 4 electors, DC also gets 4 electors. ME does not split its EV this decade.

In 1984 Reagan wins the EC by 967 to 22.
In 1988 GHW Bush wins the EC by 787 to 202.

Edited to correct an error for FL population in the spreadsheet.

[muon2]

"the more populous subdivisions shall be divided before the less populous"

This strikes me as a bit odd. I wonder why Congress did that? Is a city a "subdivision" for this purpose, since cities in the text are on the same level as counties? If so, they are more protected against chops than subdivisions down the food chain, like towns or townships. In Columbia County, if the county needed to be chopped, it would force the chop to be in the town of Kinderhook, and that might affect the CD lines in a more dramatic manner, since all things being equal, depending on the size of the chop, one might need to design a CD that was a tad short of people to append Kinderhook.

Also odd is the cliff mechanism. If NY did not lose population, it would lose no seats. If it lost 1 person, it loses 3 seats. But then, if limited to just a one seat loss, we have an ever greater number of representatives.

Finally, I notice that that way the 1% deviation rule is phrased, the population range is 2% rather than 1%, since the deviation benchmark is the quota figure, not the difference between the least populated and most populated CD.  SCOTUS might not like that when the case comes before it.

Finally, this whole packing of the House screws less populated states, since the percentage of electors attributable to the two Senators from each state, is a lower and lower percentage of the electoral college pie. I wonder why more lightly populated state US Senators did not filibuster against the passage of the law that let the genie out of the bottle in the first place.

All lawyers on the forum should probably be banned. They are a pain in the ass.

The redistricting language I used is straight from IA statute, passed in response to a 1972 IASC case. SCOTUS might not like the deviation language, but Karcher wasn't decided until 1983, so Congress wouldn't have known how equal practicable meant to them. In this time line Congress did not strike away their 1911 redistricting language in 1929, and here they would be adding additional clarification to the 1911 language. I suspect the Karcher opinion would read differently had Congress maintained some direction over redistricting.

From 1870 to 1920 only once did states lose seats during apportionment, NH and VT in 1881. By 1929 the notion that there should be no losers was pretty well ingrained with both parties. As I noted the Speaker even set up a commission to study apportionment, but they did not adopt the findings. The only twist I applied was for Congress to apply the findings of that commission in 1929 instead of capping the size at 435. I don't know that there would have been a filibuster since I doubt any would have seen this as a departure from normal practice, and could not have foreseen WWII and how the post-war migration and Baby Boom would impact the size.

[muon2]

The 1976 results saw ME split its electoral votes by CD. That raised the concern that other states might follow suit and if they gerrymander their districts, then that gerrymandering will skew the electoral vote. During that same decade IA was codifying its rules on drawing congressional districts, and in this timeline Congress adopts a version of the IA rules before the 1980 Census.

1. Congressional districts shall each have a population as nearly equal as practicable to the ideal district population. No congressional district shall have a population which varies by more than one percent from the applicable ideal district population.
2.  To the extent consistent with 1, district boundaries shall coincide with the boundaries of political subdivisions of the state. The number of counties and cities divided among more than one district shall be as small as possible. When there is a choice between dividing local political subdivisions, the more populous subdivisions shall be divided before the less populous, but this statement does not apply to a legislative district boundary drawn along a county line which passes through a city that lies in more than one county.
3.  Districts shall be composed of convenient contiguous territory. Areas which meet only at the points of adjoining corners are not contiguous.
4.  Districts shall be reasonably compact in form, to the extent consistent with the standards established by subsections 1, 2, and 3. In general, reasonably compact districts are those which are square, rectangular, or hexagonal in shape, and not irregularly shaped, to the extent permitted by natural or political boundaries.

In the 1980 Census only NY lost population and loses seats accordingly. As in the previous decade PA just barely gained population and was the last state to be awarded a seat, setting the total number. The apportionment now requires 878 seats, and an average district has 255 K inhabitants.

Since the smallest state now has 2 seats and 4 electors, DC also gets 4 electors. ME does not split its EV this decade.

In 1984 Reagan wins the EC by 960 to 22.
In 1988 GHW Bush wins the EC by 780 to 202.

Interesting how the minimum number of electoral votes in a state is now 4 instead of 3. I wonder if it increase even further in later time now

Almost certainly. I'm working on 1990 and there are still three states with 2 seats, but it may be that 3 is the minimum by 2010.

[muon2]

The 1990 Census had IA, ND, WV and WY losing population. PA was just barely growing and set the apportioned size of the House at 971. Dems controlled Congress in 1991 and weren't going to consider any reduction to the apportionment since the losing states were mostly going to be ones in the northeast where they were strong. WV was the only state to lose a seat. The average district had 256 K inhabitants.

Since the smallest state still has 2 seats and 4 electors, so DC also gets 4 electors. ME now has 5 CDs, and does not split in either 1992 (sorry Perot) or 1996, but GWB gets 2 CDs in 2000. In 1991 NE passed a law to split electors like ME, but it does not come into use during the decade.

In 1992 Clinton wins the EC by 747 to 328.
In 1996 Clinton wins the EC by 769 to 306.
In 2000 Gore wins the EC by 543 to 532.

nb. The EC in this timeline is now double what it is IRL.

[muon2]

In the 2000 Census no state lost population, only DC. However ND barely grew and that drove apportionment to 1068 seats with an average of 262 K inhabitants each.

Legislation to cap the number at 1000 passed in 2001, but it was pointed out that only by growing the House over the previous century did Gore win the presidency, and Gore vetoed the bill. In the wake of the 9/11 attacks Congress passed an act allowing members to link by secure teleconference lines from the state capitols and reduce the risk of an attack wiping out Congress at once. This also reduced the need to build more space for the House in the US Capitol and the existing space from the early 1960's could be remodeled for ceremonial purposes when all were present.

Since the smallest state still has 2 seats and 4 electors, so DC also gets 4 electors. ME has 5 CDs, but does not split this decade. NE has 7 seats and in 2008 Obama wins 2 of them. In order to get the timeline back in order, McCain challenges Gore in 2004 and the states vote as they did IRL.

In 2004 McCain wins the EC by 612 to 560.
In 2008 Obama wins the EC by 807 to 365.

[muon2]

So let's bring this to the present. What if the 1929 act never happened to lock in 435 members in the House, but instead codified what had been common practice since the Civil War.

In the 2010 Census only MI lost population and loses 2 seats as NY is the last to be brought up to their status quo. AK moves up to 3 seats, leaving only VT and WY with 2, so DC continues to get 4 electors. There are now 1140 members in the House and the average district has 270 K inhabitants.

In 2012 Obama wins the EC by 780 to 464 (1 EV from NE).
In 2016 Trump wins the EC by 705 to 539 (Trump gets 3 EV from ME and loses 2 in NE).

Though the expanded EC matched the popular vote winner in 2000, it would not in 2016. The senate seats in the EC don't impact this result either. It looks like this the sort of case the EC is meant for - to work against a candidate that relies too much on a regional base, regardless of popularity.

[muon2]

If you'd like more to reference, this is my map of ME in 5 CDs following the rules set up for this TL. I drew it with block groups so that I could minimize the population deviation while chopping no towns and minimizing county chops. I got the vote for each district by adding the totals from the Atlas 2016 election results.

[muon2]

Here's what I used for NE. The deviations exist to keep the number of chopped counties at the minimum of 2. A chop into Sarpy (or Saunders) could be used to get all the deviations of CD 4-7 down below that currently in CD 3. CD 4 is D+0.4 and I know it went for Clinton in 2016, but I don't have the data to determine what it did in 2012.

[muon2]

Is the fourth district literally just Lincoln City?

There are a few adjacent townships to bring the population up to the quota. Otherwise it's about 98% made up of the city (pop 258,379). The rule in the 1970's act is not to chop it unless necessary for population reasons.

[muon2]

I bet doing a map of 137 districts for California would be a nightmare lol

It's not unlike doing a map for the lower house in a state legislature. CA has 78, but NY has 150 in its House of Representatives. I've worked on the real maps for IL (118) and OH (99) and they aren't that much harder than a congressional plan.

I think many people just draw subjective plans that look good and achieve reasonable equality and then 100 districts are a lot harder than 10. But if a plan is constrained by objective rules like minimizing chopped counties while getting the least inequality and reasonably low erosity, strategies that work for 10 districts will also work for 100, and then it isn't 10 times as much work. An example of this were the maps of the WI legislature (99) that jimrtex and I did separately did last year.

[muon2]

The 1990 Census had IA, ND, WV and WY losing population. PA was just barely growing and set the apportioned size of the House at 971. Dems controlled Congress in 1991 and weren't going to consider any reduction to the apportionment since the losing states were mostly going to be ones in the northeast where they were strong. WV was the only state to lose a seat. The average district had 256 K inhabitants.

Since the smallest state still has 2 seats and 4 electors, so DC also gets 4 electors. ME now has 5 CDs, and does not split in either 1992 (sorry Perot) or 1996, but GWB gets 2 CDs in 2000. In 1991 NE passed a law to split electors like ME, but it does not come into use during the decade.

In 1992 Clinton wins the EC by 747 to 328.
In 1996 Clinton wins the EC by 769 to 306.
In 2000 Gore wins the EC by 543 to 532.

nb. The EC in this timeline is now double what it is IRL.

I know this is like months and months old (but since I started this thread I don't feel so bad about reviving it lol) anyways, I am curious to know, if you still have this info, which district Perot at least comes closest to picking up, and that if even a slight universal swing towards him in any way could tip him at least one electoral vote in this time line

Of course the CDs would be different and smaller than IRL. So to that extent anything would be a guess. If there were any CDs in this scenario that might have gone for Perot in 1992 they would almost certainly have to include counties where Perot placed first. Finally there would have to be enough votes in those counties to out weigh counties in the CD where Perot was not first.

As an example, consider KS which had 3 counties plus a tie for Perot, all of which were in or adjacent to current CD 2. In this scenario KS has 10 CDs, so each is less than half the RL CD population. However all three of those counties were tiny and would only make up 12% of a CD, and Perot barely beat Bush in them. Adding enough counties to make a whole CD would wash out any Perot gains in those little counties.

The more obvious place to look is in ME, where Perot did best. In this scenario ME has 5 CDs and each would have 245.6K population. Perot won three counties (Piscataquis, Somerset, and Waldo) that would make up 41% of a CD with a net win of about 1600 votes over Clinton. That might not be enough unless one carefully selected adjacent towns. However, if there were a central ME CD it probably wouldn't take much of a swing to get Perot the win. And since ME does allocate EV by CD, Perot could get one.

[muon2]

So let's bring this to the present. What if the 1929 act never happened to lock in 435 members in the House, but instead codified what had been common practice since the Civil War.

In the 2010 Census only MI lost population and loses 2 seats as NY is the last to be brought up to their status quo. AK moves up to 3 seats, leaving only VT and WY with 2, so DC continues to get 4 electors. There are now 1140 members in the House and the average district has 270 K inhabitants.

In 2012 Obama wins the EC by 780 to 464 (1 EV from NE).
In 2016 Trump wins the EC by 705 to 539 (Trump gets 3 EV from ME and loses 2 in NE).
In 2020 Biden wins the EC by 713 to 531 (Biden gets 2 EV from NE and loses 3 in ME).

Though the expanded EC matched the popular vote winner in 2000, it would not in 2016. The senate seats in the EC don't impact this result either. It looks like this the sort of case the EC is meant for - to work against a candidate that relies too much on a regional base, regardless of popularity.

Decided to assume the district breakdown as 2016 for ME/NE rather than calculate manually, but this was inevitable.

Thanks! I look forward to the next iteration of the map at the end of the year.

[muon2]

With the release of the 2020 apportionment populations I can continue the series. In this timeline the 1929 act never happened to lock in 435 members in the House, but instead codified what had been common practice since the Civil War. Initially no state could lose a seat due to reapportionment, but that was modified by the 1947 Act such that no state that gains population over the decade can lose seats. In 2010 the total in the House was 1140 members.

In the 2020 Census IL, MS and WV lost population and both IL and WV lose a seat as OH is the last to be brought up to their status quo. VT and WY remain with 2 seats, so DC continues to get 4 electors. There are now 1193 members in the House and the average district has 278 K inhabitants.