Showing posts with label 2022 election. Show all posts
Showing posts with label 2022 election. Show all posts

10 November 2023

Election Rules And Gridlock In American Politics

This post restates an answer I made at Politics.SE where I am a moderator and participant, with only one minor modification to remove a reference to another answer at the site, adding a link, adding some discussion of the situation in Alaska, making minimal formatting modifications, and quoting a comment to the answer. I quote the question I am answering piecemeal during the answer.

In the course of answering this question, I was surprised to learn just how undivided and not gridlocked the state political process in the United States has become, in stark contrast to the federal political process in the United States.

A false premise
What kind of amendment clause can encourage many parties to represent the voters in the way originally intended by the founding fathers?
The Founding Fathers intended the constitution to result in a no party system within many disorganized majority and minority factions (see, e.g. The Federalist Papers, especially Federalist No. 10 written by James Madison), not a multi-party democracy. They predicted wrong, in part, because the world known to them had very little experience with national level democracies in early 1789 when the U.S. Constitution was drafted.

In particular, they were unaware that political parties arise spontaneously, more or less automatically, whether formally recognized or not by the electoral system, in any governmental system with a group of multiple legislators, even when legislators are appointed, rather than elected. But political parties profoundly change the dynamics of the legislative and electoral process in ways that they didn't anticipate.

Also, to slightly nitpick the title question:
What kind of amendment can oblige multiple political parties, to fix the unintended two-party malfunction of the constitution?
You can't oblige people to form multiple political parties. Some electoral systems favor the creation of multiple political parties, but this only happens if someone decides that it a worthwhile thing to do and creates and runs them. Often at the outset of a new political system, the people who caused the system to come into being are so united from that struggle that multiple political parties aren't established right away.

Securing a multi-party system through electoral reforms

But securing a multi-party political system, which is the international norm for democratic countries, rather than a two party system, in the United States wouldn't be that hard to achieve in terms of the formal constitutional process to do so. It wouldn't even require a constitutional amendment. Politically, however, there would be big barriers to achieving this goal.

Duverger's Law, a law in the social sciences sense, says that first-past-the-post election systems tend to cause dual party systems (there is an exception to this law for geographically segregated political movements, such as nationalist parties for a region like the nationalist parties in Quebec and Scotland). So, all one needs to do in order to end a two party system is to change the election law system. This doesn't require a constitutional amendment.

Reforms possible with a change to federal law without a constitutional amendment

The U.S. Constitution does not direct that members of Congress be elected from single member district plurality election system. This system exists in the U.S. because it is mandated by 2 U.S. Code § 2c, which currently mandates single member districts for U.S. Congressional elections in states with more than one representative in the House.

Historically, before 2 U.S.C. § § 2c was adopted in 1967, multi-member Congressional districts were common in U.S. states, although not by a proportionate representation system. The reason for this reform in 1967 is that the majority in a multi-member district would elect all members of Congress from that multi-member district from the party with a majority in that multi-member district and this approach was used to suppress the black vote in the South, and this legislation reinforced the mandate of the Voting Rights Act of 1964 to prevent the suppression of the black vote in the South.
This statute states, in its entirety, that:
In each State entitled in the Ninety-first Congress or in any subsequent Congress thereafter to more than one Representative under an apportionment made pursuant to the provisions of section 2a(a) of this title, there shall be established by law a number of districts equal to the number of Representatives to which such State is so entitled, and Representatives shall be elected only from districts so established, no district to elect more than one Representative (except that a State which is entitled to more than one Representative and which has in all previous elections elected its Representatives at Large may elect its Representatives at Large to the Ninety-first Congress).

Even with this statutory limitation, some U.S. states (e.g., Maine and Alaska) already elected members of Congress by ranked choice voting. Several other states require a majority vote to elect a member of Congress with a runoff election held if no candidate receives a majority of the vote in the first round (e.g. Louisiana), which like an instant runoff voting system eliminates the spoiler effects associated with third-party candidacies.

There is no reason that you can't have proportional representation in a country with a strong Presidential government like the U.S. with its current constitution. This is what is done in almost every other country with a strong President system countries (e.g. Turkey and Russia).

A conversion to a proportionate representation method of selecting members of Congress in states with more than one member of Congress could be accomplished without a constitutional amendment simply by repealing 2 U.S.C. § 2c and replacing it would a statute that mandated either rank choice voting or majority required to win with a runoff voting in states with a single member of Congress, and mandating proportional representation with a party list system of some kind in states with more than one member of Congress. This would also, simultaneously, end the problem of gerrymandering.

A statute, needless to say, is much easier to enact than a constitutional amendment which requires a two-thirds majority of both houses of Congress (or identical proposals from two-thirds of the U.S. state legislatures, or by constitutional conventions in two-thirds of U.S. states) to propose, and ratification by three-quarters of U.S. states to approve.

The relevant statute would require a majority of the U.S. House, a majority of the U.S. Senate (after overcoming a 60% of the U.S. Senate filibuster or using the nuclear option to abolish the filibuster in this situation with a majority of the U.S. Senate) and the signature of the President (or an override of a Presidential veto by two-thirds of the U.S. House and the U.S. Senate), with no state legislative action before the law in enacted.

The main barrier to this kind of change (which would be even greater to enact a U.S. Constitutional Amendment) is that at least naively, it is not in the interest of either of the two major political parties that control all but a handful of seats in Congress to create a multi-party system that would reduce their control of the process. But reforms of the electoral process contrary to the seeming narrow interests of those currently in office aren't unprecedented, usually because the political party backing the changes sees partisan gain from doing so, even though it technically makes their current voters less powerful.

For example, if the Democrats had a thin majority in both houses of Congress and held the Presidency, and abolished the filibuster in the U.S. Senate with the nuclear option, they might decide that proportional representation was the lesser of two evils to end the disadvantages that they face due to gerrymandering at the state level, and this might cause them to enact a statutory reform like this one.

On the other hand, if Congress simply repealed 2 U.S.C. § 2c, the laboratory of democracy in the states could reopen and some U.S. states with more than one member of Congress, perhaps by citizen's initiative following the model of Maine and Alaska adopting ranked choice voting, could adopt proportional representation, and if this system was popular, other states might follow suit and this could become the national norm, or provoke a national level re-enactment of 2 U.S.C. § 2c that mandated proportional representation.

Proportional representation at a national level, rather than state by state, would require a constitutional amendment. But since the practical difference between a state by state proportional representation system and a national one would be slight (basically just a rounding error scale effect), and because this would also require amendment to the Presidential election contingent election system at the same time, it probably wouldn't be worth the additional effort to attempt to achieve one.

Furthermore, you don't want to have too many seats at state in any proportional representation system because then it becomes possible to elect fringe candidates backed by just a tiny percentage of the vote (less than 0.25% in a single national proportional representation system) which if the majority coalition is thin, gives outsized power to fringe groups. This has historically been a problem in Weimar Germany which dealt with it by adopting a 5% threshold to get any seats post-WWII, in Italy where unstable coalitions of centrists designed to keep the extreme left and the extreme right out of power led to a very fragile democracy after WWII until recently, and in Israel. In contrast, only a handful of states have enough seats to allow a party with less than 5% support in the voting public to gain a legislative seat because they have 20 or fewer members of Congress.

Reforms possible without federal statutory reform or constitutional amendments

Short of this kind of statutory amendment, it would be possible to enact more limited reforms that don't so strongly lead to a two party system without amending 2 U.S.C. § 2c, either a state by state level (possibly with reforms enacted via citizen's initiative to overcome the biases of legislators elected in a two party system in many states where this is possible), or by enacting an additional law mandating certain reforms for the conduct of U.S. House elections from single member districts in addition to 2 U.S.C. § 2c.

Any single member district system inherently distorts the results in a state with more than one member of Congress relative to proportional representation except in the most idealized distribution of voters. But one can (and a number of states, such as Colorado and Ohio have) imposed standards implemented on a non-partisan basis for how Congressional districts are drawn that reduce the impact of partisan gerrymandering relative to the worst case scenario. This doesn't weaken the two party system, however.

But if you mandate, either nationally or in a single state at a time, that states used ranked choice voting, or require the winner to have a majority of the vote to win (with runoff elections if necessary) or have a California style "jungle primary" with the top two finishers in the primary facing off in the final round vote, you can eliminate the spoiler effect of first past the post voting systems.

In a first past the post voting system, a vote for a third-party or independent candidate always helps the candidate from the political party least similar to the third-party or independent candidate that you vote for.

This inherent bias against third-party candidates in first-past-the-post voting systems prevents viable third-party candidates from emerging. But if you have a system that eliminates the spoiler effect, voters are free to support a third-party or independent candidate in a first choice or first round vote, without worrying about helping their least favorite candidate. And, free to do so, third-party candidates would receive more support and stronger candidates would be more comfortable running on third-party tickets, and gradually, a multi-party system would emerge, although probably not one with as many viable political parties as a pure proportional representation system, because any candidate elected would have to receive at least second choice support from a majority of voters in a single Congressional district. Still, it might make three or four political parties, rather than just two, viable and would allow more unaffiliated candidates to be elected.

Any of these systems (but especially "jungle primary" systems) also inherently favor more moderate candidates relative to a system in which each of two major political parties holds a primary election and those two candidates then compete in a general election (which is the status quo in most, but not all, U.S. states). Partisan primaries favor farther left of center candidates in Democratic primaries and farther right of center candidates in Republican primaries, and in the process make it harder for moderate or centerist candidates to make it to the general election.

First-past-the-post elections with closed political primaries in many states are among the multiple reasons that Congress currently has fewer moderates and a larger partisan divide between the two political parties than at any other time in U.S. history since the eve of the U.S. Civil War. This is trend that is likely to continue to become more extreme with West Virginia's U.S. Senator Manchin, the most politically moderate U.S. Senator right now, announcing that he will not seek re-election, and independent U.S. Senator Kyrsten Sinema, from Arizona, who was elected as a Democrat and left the Democratic party in December of 2022 (who is the second modest politically moderate U.S. Senator U.S. right now) is unlikely to be re-elected in 2024.

In 2024, the most moderate U.S. Senator who caucuses with the Democrats will probably be the independent U.S. Senator from Maine Angus King, and the most moderate U.S. Senator on the Republican side will probably be U.S. Senator from Maine Susan Collins, since there are no strong candidates campaigning for U.S. Senate seats in the 2024 election right now who are more politically moderate than either of Maine's Senators, and they are quite far apart. In contrast, for much of recent U.S. history, there has been some overlap in political ideology at the moderate end between the Democratic party and the Republican party in the U.S. Senate and in the U.S. House.

On the other hand, these election laws clearly aren't the only reason for the currently highly divided partisan breakdown in Congress. More states have reformed their election laws incrementally to reduce the tendency of those laws to elect less moderate candidates than had done so a couple of decades ago, but Congress has nonetheless grown more divided on partisan grounds. These institutions have only led to the current status quo with the help of additional factors not so directly related to U.S. election laws (most notably the fact that "realignment" of the two main U.S. political parties has run its course as described below).

Reducing partisan voting with legislative rule reforms

It would also be possible to reform legislative rules in the U.S. House and U.S. Senate respectively, in ways that would reduce the pressure on federal legislators to vote with their political party, weakening party discipline and the hold of the majority party on each house of Congress.

Unlike a parliamentary system, in the strong President system of the United States, you don't structurally need stable partisan majorities in either house of Congress to have a stable, functioning government. As long as there are majorities to pass the necessary appropriations bills every year to keep the government running, and to approve enough Presidential nominees to executive branch and judicial positions to keep the government running, you don't need bills that are passed to have the same partisan majority backing them for every bill.

The U.S. already has relatively weak party discipline compared to parliamentary systems, but it could be weakened further.

We know this from experience.

In the 1970s and 1980s, the United States political party system was in the middle of a transition of the policy positions of the two major political parties (called "party realignment"), which basically traded ideological positions on most issues in a process that began with Presidential candidate Goldwater's "Southern Strategy" for the Republican Party in the 1960s.

This produced a de facto three party system of Northern Democrats (who are similar to today's Democratic party), Republicans (who are similar to today's Republican party), and Southern Democrats, who were conservative on issues of national defense, foreign policy, and social issues, but liberal on economic issues. So, the working majority in Congress on economic issues was made up of Democrats, and the working majority in Congress on national defense, foreign policy, and social issues was made up of Southern Democrats and Republicans. But this lack of a consistent one party majority of every policy subject matter didn't seriously interfere with the operations of the government.

This de facto three party system eventually ended as the two major political parties completed their realignment into one fully liberal Democratic party and one fully conservative Republican party. But if electoral reforms produced a new groundswell of mode moderate legislators in Congress, this could resurface.

To make this work, however, partisanship in the procedural rules of each house of Congress would have to change.

In the U.S. House, the main reforms that would be necessary would be (1) to remove the control of the partisan leadership over the committee assignment system (e.g. by seating members on the committees of their choice by seniority with a proportional partisan balance in each house), and (2) to remove the authority of the House Rules committee to decide on a partisan basis what legislation is taken up on the floor once bills clear committees, for example, by granting every bill that clears a committee a right to a floor vote with significant right to make amendments, in the order adopted, without a supermajority vote to consider urgent legislation out of order. The allocation of U.S. House resources, like office space, would also have to be vested in some non-partisan decision-maker.

In the U.S. Senate, partisan control is already weaker, but rules that strongly empower the minority acting in a partisan fashion, like Senatorial holds and the filibuster would have to be repealed.

Is the system actually not working as intended?
The founding fathers did not foresee that the constitution would foster a two-party political system that is radical, inflexible and deadlocked. There is probably little political will to reform it for many reasons, at the cost of prosperity, security, efficiency and peace.

The U.S. federal government is undoubtedly very prone to deadlock. A majority of representatives in the U.S. House, a large minority of U.S. Senators, the President, or the courts applying constitutional rules that invalidate federal laws, each have de facto veto power over new legislation. And, the power of the two major political parties in the U.S. is equally balanced.

One of the other well known political science laws is that in a two party system (or generalizing it, to a two coalition of parties system), where the parties (or coalitions) are ideologically coherent are prone to continually attempt to tweak the composition of their pre-election coalitions by changing their policies or political tactics, in a way that tends to bring them close to a 50-50 balance of power in the governmental institutions that they most strongly wish to control (the federal government in the U.S. case).

The U.S. federal level political system has been quite evenly balanced between the two major political parties for most of the period since the 1980s (more than four decades) with only brief periods when a single party had "trifecta" control of both houses of Congress and the Presidency, and courts that have been disinclined to thwart their agenda.

In contrast, during the Great Depression and into World War II, the federal government was a dominant party system united under FDR's Democratic Party, and the initially combative Lochner era U.S. Supreme Court eventually backed down when faced with court packing legislation that was likely to pass.

Also, the most recent Congressional elections in 2022 produced outcomes in terms of legislative seats held in the U.S. House that quite closely tracked the overall popular vote for each political party nationwide, with factors that favored Republicans in some places balanced by factors that favored Democrats in other, ultimately balancing out for the nation as a whole. And, the increased partisanship and ideological distance between the political parties seen in Congress mirrors increased partisanship and ideological distance between the political parties in the electorate itself.

Legislation that has bipartisan support (e.g. aid to Israel in the wake of recent events there) can still get passed. But it has been the exception rather than the norm pretty much since the 1980s (before which there was a short period when the Democratic party was dominant at the federal level) that a single party without bipartisan support can pass federal legislation leading to major policy changes. These windows of opportunity have appeared now and then and been utilized by the parties having them. But these windows of opportunity have generally been short-lived and resulted in only moderate and somewhat incremental, rather than decisive and bold reforms.

Arguably, a system that produces gridlock and little federal legislative activity in time periods when the nation is deeply and fairly evenly divided on most political issues, is exactly what the Founders intended when they created a federal political system. When the nation is united, major new national political policies can be enacted, but when the nation is divided, real major policy innovations become much easier to achieve at the state and local level than they are to achieve at the federal level.

The two party system doesn't necessarily have a lot to do with that. Ultimately, exercising political power to change the status quo is about securing legislative majorities to take action, which requires majority support from legislators as measured by the standards of the political system to pass legislation.

In a two party political system, one assembles coalitions that the organizers of the respective political parties hope will be sufficient to secure the majorities necessary to pass legislation before the election into two pre-negotiated coalitions.

In a multi-party political system, coalitions that their member political parties hope will be sufficient to secure the majorities necessary to pass legislation are negotiated after the election.

Lots of countries, especially non-federal "unitary" political systems, make it easier to secure a legislative majority than the U.S. does. But, it isn't manifestly obvious that the task of organizing a multi-party coalition (probably balanced close to 50-50 anyway) with sufficient backing to pass legislation after Congressional elections are held would have an easier time passing legislation than the task of building one of the two major party coalitions before the election to achieve this goal in the status quo.

Also, countries with these unitary systems where the majority threshold needed to pass legislation is lower are often more politically homogeneous than the United States is politically. This lack of political consensus at a national level was a fact that was apparent even back in the time when the Founders set up the current U.S. Constitution, and it shaped how the U.S. Constitution was written.

Federal level political gridlock is ugly and frustrating.

But political gridlock is not the norm in most U.S. states, despite the fact that they also have two party political systems, mostly because most U.S. states aren't so evenly politically divided and secondarily because most U.S. states have fewer barriers to legislating (like the minority veto powers created by U.S. Senate rules and unequal populations for U.S. Senate electoral districts, i.e. U.S. states).

enter image description here



Republicans secured a trifecta in Louisiana after the 2023 general election. Democrats gained control of both houses of the Virginia state legislature in the same election, although Virginia's Governor is still a Republican.

In the wake of the 2023 general election in the U.S., for example, there were 22 Republican trifectas (plus Nebraska where Republicans elected on a non-partisan basis to the unicameral legislature controlled it and the Governor was a Republican), 17 Democratic trifectas (plus the Democratic party controlled District of Columbia), and 10 divided governments where neither major political party had trifecta control. As a result of the 2023 elections, there are fewer divided governments across the United States than at any other point from 1992 to 2022.

In all of those ten states, except Pennsylvania and Alaska, both houses of the state legislature is controlled by the same political party. In theses eight states, one major political party controls the legislature and the other major political party holds the Governorship (with a Governor who is almost invariably a moderate within his or her own political party and who usually holds a line item veto that brings more give and take to partisan budget negotiations). 

Alaska's state legislative houses both have split party control after adopting ranked choice voting, with the state house also having a significant number of members who aren't members of the two major parties. This has produced a political culture which is "moderate and consensus based." A majority of the legislators in both houses of Alaska's state legislature are Republicans (as is Alaska's Governor), but each house has chosen instead to organize itself on a bipartisan basis.

Pennsylvania is the only U.S. state where the legislative process is even close to being as vulnerable to being gridlocked as the federal legislative process. It has a Governor who is a Democrat, Democrats control the state house by one seat out of more than two hundred, and Republicans control the state senate.

Also, legislative process gridlock on particular key issues can be overcome with citizen's initiatives in many U.S. states, an option that is not available at the federal level in the United States.

Federal government political gridlock has not been accompanied by corresponding state level political gridlock in the U.S.

Divided control and gridlock is the exception rather than the norm at the state level in the United States, and there are no states with legislatures elected from districts with unequal populations nor are there any states that give legislative minorities the privileges the receive in the U.S. Senate. So, getting clear majorities to pass legislation is easier at the state level than at the federal level in all but a modest minority of U.S. states.

Given the currently deeply divided state of political preferences in the U.S. at the national level, that has a strong regional geographic character, a system where state level policy reforms are much easier to achieve than federal level policy reforms where gridlock prevails, is arguably a feature and not a bug, of just the kind that the Founders intended.

One Comment

There were a number of comments to this answer, but one in particular deserves to be mentioned as it adds an important factual observation:
"not in the interest of either of the two major political parties" In the simplest case where no new parties get founded and actual voters' votes remain the same, PR would have denied the Republicans a majority in the House in 1996, 1998, 2000, 2004, 2012, and 2016; but there has been no case since the present system was introduced when the Dems won a majority under the present system but would have been denied it under PR. Hence, PR could be argued to be in the Dems' interest. – Daniel Hatton

06 December 2022

Democrat Wins Senate Runoff Election

The Democrats have secured one net U.S. Senate seat pickup for a 51-49 majority after incumbent Democratic Senator Raphael Warnock of Georgia defeated deeply flawed Republican candidate Herschel Walker, in a runoff election that concluded today in a roughly 51.3% to 48.7% race with 99% of the vote counted (a 2.6 percentage point margin of victory), which are results significantly better than in the first round race, which Warnock won by a 0.9 percentage point margin with about 2.1 percentage points going to a Libertarian candidate. Warnock apparently won about 90% of the third-party vote in the runoff election (holding all other things equal).

In the races of the U.S. House, taking the last outstanding race, CO-3, in which deeply flawed Republican incumbent Lauren Boebert is almost sure to survive the automatic recount currently underway, by roughly 0.16 percentage points, the closest federal election in the country this year (in which should have been a safe Republican seat). 

Democrats have lost a net ten seats and their majority in the House. Republicans will have four more seats than they need for a bare majority in the U.S. House (almost exactly 51% of the total number of seats outstanding in a mirror image of the Senate).

The marginal seats in both chambers were resolved in elections with less than two percentage point margins of victory in a nation poised on a knife edge of equal support for the two major political parties in the context of the current rules of the game.


So, with a Democratic President and Vice President, and a 6-3 conservative majority in the U.S. Supreme Court, the federal government will spend the next two years in a gridlock even deeper than the current one in which the lack of a Democratic filibuster-proof majority in the Senate has prevented almost all significant legislation from being passed. But Democrats will have an easier time getting President Biden's appointees confirmed.

Midterm Election Voter Turnout

Voter turnout in the U.S. midterm elections in 2022 as a percentage of eligible voters (of the 50 states and the District of Columbia) was as follows:

#1 Maine 61.5%
#2 Oregon 61.4%
#3 Minnesota 61.0%
#4 Wisconsin 60.1%
#5 Colorado 58.5%

. . .

#47 Oklahoma 40.1%
#48 Alabama 37.7%
#49 Mississippi 37.1%
#50 West Virginia 35.7%
#51 Tennessee 31.6% 
 
National average: 46.8%


Turnout was slightly lower nationwide than the record modern midterm election turnout of 2018.

About 88% of people who are eligible to vote are active registered voters in Colorado. In Colorado, about 5.8% of adults (i.e. people who are at least 18 years old) are ineligible to vote because they are not U.S. citizens and about 15,000 adults (0.3%) are ineligible to vote because they are currently in prison.

The adult population of Colorado in 2022 was 4,631,586, of whom 4,346,493 were eligible to vote.

Nationally 7.3% of the adult population is not eligible to vote, because they were not citizens (in most cases), or as collateral consequences of a criminal conviction 1.2% (or both). About 7.5% of people in the U.S. are not U.S. citizens, but some of those non-citizens are not adults or are also disqualified from voting due to criminal convictions in the state where they live. Disqualification from voting as a collateral consequence of a criminal conviction differs materially from state to state.

02 December 2022

The 2022 Midterm Election Turnout And Colorado State House Results Analyzed

Turnout

Turnout in the 2022 midterm elections in Colorado this year was 2,540,666 voters who voted ballots (not necessarily in every race), out of 3,833,468 active registered voters in the state, a 66.28% voter turnout rate. Turnout in the 2020 Presidential election in Colorado was 86.54% by the same measure.

Interpolating slightly from the state demographer's office's estimate, the voting age population of Colorado was 4,167,610 in 2019, and this probably increased by about 1.5% between 2019 and 2022. So, the voting aged population of Colorado at the time of the midterm election was about 4.23 million, so voter turnout was about 60.1% of the voting aged population in the midterm elections. Thus, about 90.6% of people who are part of the voting aged population in Colorado are registered to vote which, given disqualifications from voting for felons who are incarcerated and non-U.S. citizens, is exceptional. 

The turnout rate in Colorado was also well above the historical U.S. norm for midterm elections, almost double the long term national average turnout as a percentage of the voting age population. Voter turnout in Colorado has consistently been among the two or three highest of U.S. states in recent years, and the 2022 midterms were no exception to that rule.

The Colorado State House Race

There are sixty five seats in the Colorado state house, all of which are elected every two years. 

One of the seats won by Democrats (HD-7 in Denver) and three of the seats won by Republicans (HD-20 in suburban Colorado Springs, HD-51 in exurban Fort Collins, and HD-63 in the rural Front Range east of Fort Collins) were uncontested. The other 61 seats had at least a Democratic party candidate and a Republican party candidate. There was also a Libertarian candidate in 20 of those 61 seats, and in one of the 20 seats where there was a Libertarian candidate (HD-1 in Denver) there was also a fourth unaffiliated candidate in the race.

Democrats won 46 seats and Republicans won 19 seats in the Colorado state house in 2022, which is two more seats than the Democrats needed to have a two-thirds supermajority. Going into the election Democrats had held 41 house seats and Republicans held 24 for a net gain of five seats.

The map below from Ballotpedia is from before the election in what appear to be the newly drawn districts, but it does illustrate the geography of partisan politics in the State of Colorado.

In the Colorado state senate race, the results left Democrats with 23 seats to 12 Republican seats, a net gain of three seats and a single seat short of a two-thirds majority in that house of the Colorado General Assembly. Going into the election in which only half the seats were before the voters, Democrats had 20 seats and the Republicans had 15. That race is harder to analyze, however, because only half of the seats in the chamber were in play in this election.

Democrats also control all statewide elective offices in the state. 

The justices on the state supreme court are officially non-partisan and nominated by a blue ribbon commission in groups of three prospective candidates from which the Governor can choose. But, a majority of the current state supreme court justices (and state judges in Colorado more generally) were appointed by Democratic Governors.

Close Races

Few of the races were close, and the Republicans didn't win a single one of the six genuinely competitive races this year. Neither party won any of the other side's safe seats. If there is an overall theme is that the Democrats won in swing voting suburbs this year.

Each of the sixteen Republicans who were elected who won a contested House District race won by at least 13.6 percentage points, and the median margin of victory for a winning Republican in a contested race was 24.29 percentage points.

The Democratic Party's margin of victory in the median race in the state house, which it needed to win to secure its majority control of that house was 11.22 percentage points in HD-62 in eastern Pueblo and the San Luis Valley. 

Democrats won all 18 races where the margin of victory was less than the closest Republican win, all 11 races where the margin of victory was less than 8 percentage points (the next closest race had an 11.97 percentage point margin of victory), and all 6 races where the margin of victory for the winner was less than 5 percentage points (in which the Democratic margin of victory ranged from 0.9 percentage points to 3.37 percentage points).

The closest race was the 0.9 percentage point margin of victory in the two candidate HD-43 race in the Highlands Ranch suburb of Denver. The other two close house races that Democrats won which didn't involve potential third-party spoiler effects were HD-19, in suburban Colorado Springs, with a 3.37 percentage point margin of victory, and HD-61, which includes many mountain resort towns including Aspen and the surrounding area, with a 2.82 percentage point margin of victory.

Democrats secured an outright majority of the votes cast in 43 of the 46 house races that they won. They won three races (HD-16, HD-25, and HD-50) with pluralities that were smaller than the third-party Libertarian vote. Colorado does not have runoff elections at the state level if the plurality winner fails to receive a majority of the votes cast in the first round.

In HD-16, in suburban Colorado Springs, the Democratic candidate's margin of victory over the Republican was 2.17 percentage points and the Libertarian candidate received 2.66% of the vote. It is possible, although unlikely, that the Libertarian candidate was a spoiler in this race whose candidacy caused the Republican candidate's defeat.

In HD-25, in the mountainous portions of the Jefferson County western suburbs of Denver, the Democratic candidate's margin of victory over the Republican was 1.39 percentage points and the Libertarian candidate received 2.05% of the vote. It is plausible that the Libertarian candidate may have been a spoiler in this race whose candidacy caused the Republican candidate's defeat.

In HD-50, in Greeley, the Democratic candidate's margin of victory over the Republican was 1.89 percentage points and the Libertarian candidate received 3.51% of the vote. It is very possible that the Libertarian candidate may have been a spoiler in this race whose candidacy caused the Republican candidate's defeat.

The other 17 Libertarian candidates in house races and the 1 unaffiliated candidate in a house race, were irrelevant to the ultimate outcome of the 17 house races in which the other 18 third-party candidates contended. Likewise, third-party candidates were not potential spoilers in any of the statewide partisan races this year, all of which Democrats won with an outright majority of the votes cast.

The combined third-party candidate performance in HD-1 in Denver which had two third-party candidates was 4.53%; the Democrat won that seat by a 32.29 percentage point margin of victory. The best performance by a Libertarian candidate alone in a house was this year was in HD-15 in suburban Colorado Springs, where the Libertarian won 4.14% of the vote; the Republican won that seat by a 17.66 percentage point margin of victory.

The HD-50 race was also the marginal race that gave Democrats a two-thirds majority in the house. They could have secured a two-thirds supermajority in the house even if the Democrats had lost HD-43 which they won by 0.9 percentage points and HD-25, which they won by 1.39 percentage points. 

The Popular Vote Compared

The popular vote (including votes cast for candidates in the four uncontested races) was 53.32% for Democratic house candidates, 45.84% for Republican House candidates, 0.82% for Libertarian candidates, and 0.02% for a single unaffiliated candidate. 

If the 65 seats in the Colorado house had been allocated on that basis, as the would have been in a proportional representation list system, for example, the Democrats would have won 35 seats, the Republicans would have won 29 seats, and one Libertarian candidate would have won a seat.

Of course, this is distorted by the fact that turnout as measured by votes cast for house candidates by party omits Republican and Libertarian voters in the one uncontested house seat won by a Democrat, omits Democratic and Libertarian voters in the three uncontested house seats won by a Republican, and omits Libertarian voters in the 41 contested races that didn't have a Libertarian candidate running. It also doesn't fully account for reduced turnout by Democrats in safe districts given that there were no close partisan statewide office elections in 2022 in Colorado. Still, many statewide ballot measures gave every voters some incentive to vote, and turnout was overall quite high. 

The percentages for the aggregated house vote races aren't far outside the range of the performance of Democrats in statewide partisan races in 2022 in Colorado, although they rival the best performances that the Republicans had in any of the statewide partisan races (in part because there was a smaller third-party vote in the state house races since there were no third-party candidates in 41 of those races).

In the U.S. Senate race, incumbent Democrat Michael Bennet defeated Republican candidate Joe O'Dea by a 55.88% to 41.27% margin, with the Libertarian candidate securing 1.74% of the vote, the next most popular third-party candidate getting 0.66% of the vote, and one more third-party candidate getting 0.45% of the vote.

In the Governor's race, incumbent Democrat Jared Polis defeated Republican candidate Heidi Ganahl by a 58.53% to 39.18% margin, with the Libertarian candidate receiving 1.15%, and the other two third-party candidates receiving 0.86% and 0.27% of the vote respectively.

In the Secretary of State race, incumbent Democrat Jena Griswold defeated Republican Pam Anderson, 55.1% to 42.08% (in a race Griswold was administering by virtue of holding that office), with the Libertarian candidate receiving 1.47%, and the other two third-party candidates receiving 0.71% and 0.18% respectively.

In the State Treasurer's race, incumbent Democrat Dave Young defeated Republican candidate Lang Sias, 53.67% to 43.03%, with the Libertarian candidate receiving 3.30%. The other two statewide parties didn't field candidates for this race.

In the State Attorney General's race, incumbent Phil Weiser defeated Republican candidate John Kellner, 54.74% to 43.04%, with the Libertarian candidate receiving 2.21%. The other two statewide parties didn't field candidates for this race.

In the State Board of Education At Large race, Democrat Kathy Plomer defeated Republican candidate Dan Malloit, 53.43% to 43.04%, with the Libertarian candidate receiving 2.84% and another third-party candidate receiving 0.68% of the vote.

Of course, as I have noted previously, Democrats won five U.S. House seats (CO-1, CO-2, CO-6, CO-7 and CO-8) all of which are in Denver, Boulder, and the suburbs of Denver, of which, CO-8 was a highly competitive seat that they won by only a small margin. Republicans won what should have been three safe Republican seats (CO-3, mostly CO-4 in the rural Front Range, and CO-5 in Colorado Springs), but won CO-3 covering the the mountain towns, Western Colorado, and Southern Colorado which is a heavily Republican district by only 0.16 percentage points (the race is currently undergoing an automatic recount to be completed by December 16, 2022), which was the closest Congressional race in the entire United States (although it didn't make a difference for control of the U.S. House which Republicans secured in the midterms by a thin margin). This race was absurdly close for such a safe seat because Lauren Boebert, the Republican incumbent in the race seeking re-election, was such a deeply flawed candidate.

23 November 2022

Another Midterm Update

Updated Results

According to the New York Times, the current tally for the U.S. House is 212 Democratic seats and 220 Republican seats, 

with the U.S. Senate at 50 Democratic seats and 49 Republican seats. Three U.S. House seats remain. 

Alaska will very likely go to the incumbent Democrat Mary Peltola who has 49% of the first choice vote in a ranked choice voting race with Republican Sarah Palin with 26% and Republican Nick Begich with 24%. Sarah Palin needs to win more than 95% of second choice Begich vote to win and that is unlikely to happen.

CO-3 was won by Republican Lauren Boebert by about 0.17 percentage points, the closest race in the county, and while it is going to an automatic recount, her margin is large enough that this is unlikely to change the result.

CA-13 is on track to be won by the Republican by about 0.45 percentage points.

So, the likely outcome for the House will be 222 R to 213 D, a loss of eight net Democratic seats.

Upon closer review, it looks like not a single House seat that touches the Pacific Ocean will be represented by a Republican in the House. Likewise, not a single Congressional District in New England will be represented by a Republican.

The Mexican border is touched by four Republican House seats and seven Democratic House seats.



The last U.S. Senate seat, in Georgia, was won in the first round by the incumbent Democrat who received 49.4% of the vote and will be resolved in a December 6, 2022 runoff election on Tuesday. If the Democrat loses, it will exactly match the status quo. If he wins, it will be a gain one of net Democratic seat in the Senate.

About 2.1% of voters in the first round in Georgia voted for the Libertarian candidate in the U.S. Senate race. If all of the same voters who voted in the first round in Georgia also voted in the second round, the Democrat would need only about a third of the Libertarian voters to chose him over the Republican to win, which is very likely. So, the Republicans need to reduce Democratic turnout and/or increase Republican turnout to win in the runoff election on Tuesday (and earlier due to early voting). 

In a ranked choice voting system, the Democrat would probably have won the GA Senate race.

What if the national tilt had been ± 1 percentage point different?

One of the big insights of the political race models at the 538 blog is that federal partisan races and to a lesser extent all partisan races in the same election are highly correlated with each other. Myriad factors from Presidential popularity, to wars (see "Wag the Dog"), to supreme court decisions, to current events, to weather than impacts partisan outcomes differently, can all shift this national partisan tilt.

Democrats lost by less than a percentage point in three other House seats where Republican were favored (including one, CO-3 where the Republican was strongly favored. Democrats won only one seat where Democrats were favored (NY-18 where a Democrat was "narrowly favored") by less than a percentage point. 

If Democrats had had a par for the course midterm outcome, with each party winning all of the seats where it was favored and splitting the 36 competitive seats evenly, Democrats would have lost eight more seats than  they did.

But, with just a single percentage point shift to the left in the national outcome across the board (or in the key federal race), would have required just one in two hundred voters to flip their preferences, Democrats could have kept their majority in the House and increased their Senate margin to 52 seats which would have been enough to overcome the two Democratic votes unwilling to abolish the filibuster.

On the other hand, a single percentage point shift to the right in the national outcome across the board (or in just the key federal races), which again would have required just one in two hundred voters to flip their preferences, Republicans would have had 226 House seats instead of 222 (giving Republicans room for eight defectors instead of four to pass partisan bills), and would have controlled the Senate with 51 seats.

The 7 House seats that Democrats lost by less than one percentage point were (or will be): CO-3 (0.17), CA-13 (0.45), MI-10 (0.49), IA-3 (0.69), NY-17 (0.82), AZ-1 (0.88), and NY-22 (0.98). If they had won all of those seats, the balance in Congress would have been 220 D - 215 R (a loss of just one net D seats).

Democrats lost one U.S. Senate race, WI (0.99), by less than one percentage point, and would have won Georgia's Senate race without a runoff if there had been a one percentage point shift in their favor, resulting in a Senate balance of 52 D- 48 R (plus the Vice President's Democratic tie vote).

The 4 House seats that Democrats won by less than one percentage point were: CO-8 (0.69), CT-5 (0.77), WA-3 (0.89), NY-18 (0.99). If Republicans had won all of those seats, the balance in Congress would have been 226 R - 209 D (a loss of 12 net D seats).

Democrats won one U.S. Senate race, NV (0.78) by less than one percentage point, but shift of one percentage point towards the Republicans would still have required a runoff election for the Republican in Georgia, although his odds in a runoff election would be much better if he had won it. Assuming for sake of argument that the Republicans won the GA runoff election, the Senate balance would be 49 D - 51 R.

Outcomes Compared To Expectations

Overall, and in most districts, Republicans did better than Trump did in 2020. 


The Senate races were a mixed bag compared to the 2020 Presidential results. Democratic Senate candidates did better than Biden did in five competitive U.S. Senate races and worse in three. Seven of those shifts weren't outcome determinative with a Democrat winning in a U.S. Senate race in five states that Biden won in 2020 and a Republican winning in a U.S. Senate race in two states that Biden won in 2020. But, incumbent Republican Senator Ron Johnson in Wisconsin did 1.6 percentage points there to win a state that Biden had won by 0.6 percentage points in 2020.

Simply adding up Republican and Democrats votes for Congressional races would be a less reliable way to indicate the shift in partisan voting in 2022, because turnout is suppressed in the many districts (and Senate seats) that were very lopsided, and because sixteen House races were uncontested: three Democratic party held seats (IL-7, MA-4, and NY-13) and thirteen Republican Party held seats (AZ-8, AZ-9, FL-5, LA-4, PA-13, PA-14, SC-3. SC-4, TX-6, TX-11, TX-25, TX-31, and WI-6).   

Still, the Democrats did remarkably well compared to expectations going into the election. Democrats won one House seat where a Republican was narrowly favored (WA-3), and won every House seat where a Democrat was favored. Democrats won 25 competitive House seats and lost 11 competitive House seats.

In the U.S. Senate races, all of the races turned out as expected, subject to the GA Senate runoff on Tuesday which Democrats were expected (and still are expected) to narrowly win.

Do Americans Want Gridlock Or Federalism?

The closeness of the ultimate partisan outcomes at the federal level really couldn't be any closer, and the result is gridlock at the federal level. Arguably, that is the right result when a country is so closely divided.

On the other hand, at the state level, partisan gridlock has declined. Democrats made gains in Governor's races, sometimes in open seats, and held onto closely contested races in red leaning states.

Both halves of the legislature flipped from red to blue in Michigan, albeit narrowly, for the first time in decades. Democrats won a trifecta in Minnesota; held both chambers in Colorado, Maine, Nevada and Oregon; staved off Republican supermajorities in the North Carolina House and Wisconsin State Assembly; and clawed back seats in the New Hampshire State House. . . .

Republicans have made modest gains, however. They flipped the Virginia House of Delegates last year, though not the State Senate, while gaining seats in New Jersey. They may have broken the Democrats’ supermajorities in New York, while picking up seats in the Illinois Senate, New Mexico House and a host of red states. They took supermajorities in both chambers of the Florida Legislature, the Iowa Senate, the North Carolina Senate, the South Carolina House and the Wisconsin Senate. In races for governor, they notched commanding wins in Florida, Ohio and Texas, and gave Democrats a scare in Kansas and Oregon. . . .

But in 2022, not a single state legislative chamber flipped from blue to red. A party in power hasn’t achieved that result in a midterm election year since at least 1934, according to Post.

In 40 U.S. states and the District of Columbia, one of the two major political parties can pass ordinary legislation without any support from the other political party because that party controls the Governorship and either both houses of the state legislature in bicameral states, or the only legislative body in Nebraska and the District of Columbia. A few states even have partisan state legislative supermajorities in trifecta states.

Only ten U.S. states have the pattern of divided government that the United States federal government does.

As I noted previously:

Heading into the 2022 election, there were 23 Republican trifectas, 14 Democratic trifectas, and 13 states with divided governments where neither party held trifecta control. As of November 16, there were projected to be 22 Republican trifectas, 17 Democratic trifectas, and 10 divided governments where neither party had trifecta control.

[Nebraska has a unicameral nominally non-partisan state legislature that is controlled by Republicans and has a Republican Governor, which is equivalent to the trifecta but not included in the count above.]

There was one state (Alaska) where trifecta status remained unclear. Before the election, Alaska was under divided government. Trifecta status changed in six states. In Maryland, Massachusetts, Michigan, and Minnesota, divided governments became Democratic trifectas. In Nevada, the Democratic trifecta became a divided government. In Arizona, the Republican trifecta became a divided government. State government trifectas. (The map below is pre-2020.) 

The "divided" states are now (or could be when the final results are in): Alaska, Arizona, Louisiana, North Carolina, Virginia, Wisconsin, Kansas, Kentucky, Pennsylvania, and Vermont.

There are divided state legislatures in Alaska and Virginia. 

But, in Arizona, Kansas, Kentucky, Louisiana, North Carolina, Pennsylvania, Vermont, and Wisconsin, where the divided government comes solely from a Governor. In Vermont, Democrats control the state legislature and there is a Republican Governor. In the other seven states with undivided legislatures that are not trifecta states, Republicans control the state legislature and there is a Democratic Governor.

One could call the midterm election results a pro-federalism outcome: strengthening the ability of many state governments to pass sweeping partisan legislation, while restraining the power of the federal government to pass partisan legislative reforms even further.

All of the "trifecta" states except New Hampshire and Georgia voted for President in 2020 according to their trifecta partisan lean. Of the current ten "divided" states, Arizona, Vermont, Virginia, Wisconsin, and Pennsylvania voted for Biden, while Alaska, Kansas, Kentucky, Louisiana, and North Carolina voted for Trump. Maine and Nebraska each cast one electoral vote contrary to the rest of the state (balancing out). The 2020 Presidential election map, for comparison purposes, was as follows:

Following the 2020 election, there were fewer states with partisan splits in their U.S. Senate delegations than at any time since direct elections of Senators began, and by flipping the Republican U.S. Senate seat in Pennsylvania to the Democrats, this record will be pushed one state further from six to five if Democrats can hold onto the Georgia Senate seat in the December 6 runoff election.

Democrats currently have five U.S. Senate seats in Republican trifecta states: one each in Montana, Ohio and West Virginia (each of which voted for Trump in 2020) and two U.S. Senate seats in New Hampshire and Georgia, respectively (both of which voted for Biden in 2020).

Republicans have one U.S. Senate seat in a Democratic trifecta seat: Maine, which voted for Biden in 2020. It is held by Republican Susan Collins, one of the moderate Republicans in the Senate, who is the only Republican in the U.S. House or the U.S. Senate from New England. Maine's other U.S. Senator, Angus King, is an independent who caucuses with the Democrats.

Democrats have two U.S. Senate seats each in the divided states of Arizona, Vermont, Virginia, and Pennsylvania (each of which also voted for Biden in 2020). Arizona, Virginia and Pennsylvania have Democratic Governors, Vermont has a Republican Governor, and Virginia has a Republican Governor and a divided legislature.

A Democrat and a Republican split the U.S. Senate seats in divided Wisconsin which voted for Biden in 2020 and has a Republican controlled state legislature and a Democratic Governor.

Republicans have two U.S. Senate seats each in the divided states of Alaska (which voted for Trump in 2020, has a Democrat as its sole member of the U.S. House, has a divided state legislature by one vote in the state house, and has a Republican Governor), and in Kansas, Kentucky, Louisiana and North Carolina (each of which also voted for Trump in 2020 and have Republican controlled state legislatures but a Democratic Governor).


One can get pretty close to the concept of a "clearly red" or "clearly blue" state by concluding that if one party won both houses of the state legislature, voted for that party's Presidential candidate, and has two of the three people serving as its Governor and two U.S. Senators of the same party, it is still "clearly" aligned with that party, even though there is one outlier Governor or one outlier U.S. Senator.

But, a state is "purple" if this does not hold.

By that measure, the seven purple states are Alaska, Arizona, Georgia, New Hampshire, Pennsylvania, Virginia, and Wisconsin. 

Two clearly blue states and seven red states have one outlier statewide elected Governor and U.S. Senator.

Thirty-four states and the District of Columbia are completely solidly either red or blue states.

Maine is a clearly blue state with an outlier Republican Senator. Vermont is a clearly blue state with an outlier Republican Governor. 

Montana, Ohio, and West Virginia are clearly red states with an outlier Democratic Senator. Kansas, Kentucky, Louisiana, and North Carolina are clearly red states with an outlier Democratic Governor.