Colorado Proposition 113: A popular vote for the popular vote

To my Fellow Coloradans — On Tue 20 Nov 2020, an issue both interesting and important will be put to vote in our state. No, not that. Something else. Proposition 113 is a referendum to uphold or veto Colorado SB19–042 which was voted into law by the Colorado General Assembly and signed by Governor Jared Polis on 19 Mar 2019. Read more HERE for how this law was put to a referendum.

Under SB19-042, Colorado joined the National Popular Vote Interstate Compact (NPVIC), a mechanism whereby a state agrees to cast its Electoral College (EC) votes for the winner of the national Popular Vote (PV) for president. So far, 16 states (if you include Colorado) with 196 total electoral votes have agreed to join the Compact. SB19–042 has no impact on the 2020 election, but once states totaling 74 more electoral votes join, the NPVIC will go into effect. Recall it takes 270 electoral votes to win the U.S. Presidency.

This article is a bit long, so here’s the TL;DR (and spoiler): I believe data and history tells us we should all vote Yes on Proposition 113. But I hope most readers will learn why by putting in the time to read the rest of this article.

A brief recap of the U.S. Constitution

I recently began learning more about the history of the EC, assuming there are strong, logical arguments for why it was created by the constitutions’ framers and why it has remained in place for so long. In fact, most historians agree the EC was born from last minute horse trading at the end of the First Continental Congress. It was not something the framers settled upon early and easily. This is true for much of the constitution: after all, the constitution has been amended 27 times, so it’s pretty obvious the constitution wasn’t / isn’t perfect.

The first 10 amendments (aka the Bill of Rights) were approved in 1791, just 3 years after the constitution was ratified by the states:

  • 1st: Freedom of religion, speech, press, assembly and petition
  • 2nd: Right to bear arms
  • 3rd: Right to deny lodging to government troops
  • 4th: Protection from unreasonable search and seizure
  • 5th: Protection from double jeopardy, right to due process and right to just compensation under eminent domain
  • 6th: Right to a fair, speedy and public trial with an impartial jury and legal counsel
  • 7th: Right to a jury trial for federal cases
  • 8th: Reasonable bail and fines plus protection from cruel and unusual punishment
  • 9th: Unenumerated rights (a catchall for stuff the framers forgot to put into the constitution which over time the U.S. Supreme Court has interpreted to include the rights to travel, to vote, to privacy and other pretty big ideas)
  • 10th: Limitation of federal government powers

That leaves 17 additional amendments over the next 229 years or about 1 change every 13 years. So, not frequent, but also not rare. Many will remember some of the more famous amendments:

  • 13th (1865): Abolished slavery
  • 18th (1919): Prohibition of alcohol
  • 19th (1920): Gave women suffrage
  • 21st (1933): Repealed the 18th! Yay!

But you may not recall some lesser known amendments like:

  • 12th (1804): Revamp of the EC
  • 17th (1913): Changed the selection of U.S. Senators from election by state legislatures to election by citizen PV
  • 22nd (1951): Created a two term limit for U.S. President

What did the 12th Amendment change?

But flaws in this system were revealed during the 1796 election. Washington bowed out (there were no term limits at the time, he’d just had enough of the job) and the election was up for grabs with our “two party system” beginning to emerge at the same time. In 1796, Adams (a Federalist) won narrowly over Jefferson (a Democratic-Republican… yes, that was a thing) which meant the President and Vice President were from two different parties. Despite having worked closely since the Second Continental Congress (1775), their friendship fractured during Adam’s term and the 1800 election campaign between Adams and Jefferson was what we would call today a “shit show.”

By the 1804 election, the 12th Amendment had been passed, rebooting the EC such that each elector cast one of their votes for President and the second vote for Vice President, which effectively enabled “party ticket” EC voting.

Since that time, there have been hundreds of attempts to further amend the constitution’s EC mechanics (TIME magazine noted in 1969 “more than 500 proposals” since the constitution’s ratification and author Jesse Wegman noted in his 2020 book Let the People Pick the President that over 700 attempts had been made).


A lot of people think this is a “Democrats for EC reform vs Republicans against EC reform” argument and I suppose there’s a kernel of truth in there as Republican candidates have lost the PV, but won the EC four times (1876, 1888, 2000, 2016), while no Democratic candidate has benefitted from this quirk of the Presidential electoral system.

Based on Wikipedia List of United States presidential elections by popular vote margin; * indicates EC winners / popular vote losers

So there’s an argument that “no Republican will ever win again if we move to a popular vote”. This is a myth dispelled by the fact that 20 Republican candidates have won both the PV and EC (on top of the 4 mismatch elections outlined above). Additionally, from the table above, one can see the PV margin of victory has been <2.5% on 12 occasions with Republicans barely winning 3 and Democrats barely winning 4 (including Kennedy’s 0.2% PV win in 1960). With the shift of a few votes in a few key places, EC votes could have easily gone the other way, creating even more EC vs PV mismatches.

How is the EC created?

The constitution itself specifies a basis (the U.S. Census taken every 10 years) and that “The Number of Representatives shall not exceed one for every thirty Thousand, but each State shall have at Least one Representative.”, but doesn’t specify an apportionment value.

By 1790 and the first U.S. Census, this “equation” had already blown-up: the U.S. had ~4m people across the original 13 states which worked out to ~60k people per Representative. At the time, the largest state was Virginia with ~750k people and the smallest was Delaware at ~59k. Virginia had 10 Representatives in U.S. Congress that year and Delaware had 1 or about 1 Representative for every 75k people in Virginia vs 1 for 59k in Delaware, a ratio of 1.3 (75k/59k). Not perfectly balanced, but not too bad.

Over time, this gap has fluctuated, sometimes increasing, sometimes decreasing. From 1789 until 1929, the number of U.S. Representatives steadily increased in-line with population growth and the admission of new states.

Pop / US Rep before 1929

However, that changed with the passage of the Reapportionment Act of 1929 which fixed the number of U.S. Representatives at the now familiar 435 members. If we had stuck with the constitution’s 30k people per Representative, today there would be ~11k U.S. Representatives! Instead, the current average is about ~765k people per U.S. Representative.

Pop / US Rep after 1929

And in 2020 this representation was very much in-line with where we were in 1790: about 1 Representative for every 746k people in California vs 1 for 579k in Wyoming, a ratio of 1.3 (746k/579k).

BTW… The charts above are based on data extracted from:

Okay, so what’s wrong with the EC?

Pop / EC before 1929


Pop / EC after 1929

one can see the gap ratio in representation is really 3.7: about 1 EC vote for every 718k people in California vs 1 for 192k in Wyoming (718k/192k).

Before you jump and say “As it should be! Rural states need more oomph to make our voices heard in Washington!”, let’s compare the 4 Corners States against our northern neighbor:

Pop / EC for WY vs 4 Corners

So clearly not all rural states like those of us in the Rocky Mountain region are treated equally and that Colorado’s EC “clout” is closer to California’s than Wyoming’s.

This is not meant to bash Wyoming, but to simply debunk another myth surrounding the EC: it does not necessarily give smaller states more power during elections.

EC Design

  • The EC is based on the composition of the U.S. Congress.
  • That composition isn’t baked into the constitution: it’s something the U.S. Congress has revised over time with laws, not constitutional amendments.
  • The fact every state, large or small, is guaranteed at least three members of U.S. Congress means every state gets at least three EC votes.
  • The constitution has been amended once so far to “fix” the EC and there have been hundreds of failed attempts to introduce other “fixes”.

Where the EC becomes really skewed has nothing to do with the U.S. Congress. Rather, it’s because of another element the framers included in the constitution: each state has the right to decide for itself how to appoint EC electors and how those electors should cast their vote.

Since 1790, the states have collectively changed their methods many times. At various points, different states have selected EC electors with such methods as:

  • Voters selecting EC electors by PV
  • Voters selecting an EC elector for the congressional district in which they lived
  • State Legislators selecting EC electors

Sometimes a state would use one method for an election, then shift methods for the next election.

Regardless of the way EC electors are chosen, at the current time, there are only two methods in use for how EC electors cast their votes:

  1. Cast their president vote for the Presidential candidate who won the plurality of votes in their state’s election, then cast their Vice President vote for whomever is on the same ticket. This is known as “winner take all”.
  2. Same as the above, but with some votes cast at the at the U.S. Congress district level rather than the state level. Only Nebraska and Maine use this method. For example, Nebraska has two U.S. Senators and three U.S. Representatives. The EC votes associated with the U.S. Senators are cast according to the state-wide plurality while the EC votes associated with U.S. Representatives are cast according to district-wide plurality.

Since Nebraska and Maine represent just 5 and 4 EC votes respectively (9 of 538 EC votes or ~1.7%), their method doesn’t have much impact at the national level. It’s really “winner take all” which sometimes drives elections to deviate from the “will of the people.”

Under “winner take all”, a single vote could theoretically change the outcome of a state’s EC votes. I imagine many readers are thinking: “Yep. Makes sense. Whoever gets the most votes, gets the EC votes.” Note I said “most votes” instead of the “majority of votes” since U.S. elections are typically based on “plurality” and not “majority”. So, this is clearly a flaw: a person who the majority of Coloradans didn’t vote for could garner all our EC votes. In fact, it happens regularly: in the last 10 elections (stretching back to 1980), Colorado has given their EC votes three times to candidates with <50% of the state’s PV (all campaigns with strong 3rd party candidates grabbing votes):

CO Winners with less than a Majority of Popular Vote

EC Overweighting of Swing States

For example, California is considered today to be a Blue state (BTW… it wasn’t always that way: Republican candidates have won California 23 times, but haven’t done so since George H.W. Bush won in 1988). As such, candidates spend very little effort on this state: the Democratic candidate will do some perfunctory stuff and the Republican candidate will almost ignore the state altogether. Similarly, a state like Texas is in exactly the same situation, but with roles reversed as a long-time Red state (a Democrat hasn’t won Texas since Jimmy Carter in 1976).

But under a PV, California and Texas matter. In 2016:

  • Clinton (Dem.) won 62% of the vote in California while Trump (Rep.) won 32%
  • Trump (Rep.) won 52% of the vote in Texas while Clinton (Dem.) won 43%

Said another way, there are 4.5m California Republicans and 3.9m Texas Democrats who’s votes didn’t really matter under EC “winner take all”. And perhaps worse, there were 8.8m California Democrats and 4.7m Texas Republicans whose votes were entirely taken for granted. It’s clear California is Blue and Texas is Red, so under the EC why bother engaging with those voters? Under PV, these voters (and voters like them across Colorado) will finally matter.

Drilling down a couple levels, let’s consider another common myth: “If we had a popular vote, big cities like New York City and Los Angeles would get to pick our President!”. Of course, since big cities tend to be liberal, the views of conservatives would be trampled! Not so fast. Let’s do a little 4th Grade math:

  • There are currently 317 cities in America with >100k people. This represents just 29% of the U.S. population.
  • That means 71% of Americans live in cities with <100k people.
  • And the biggest cities… >500k… represent just 14% of the total.

Do you really think New York City and Los Angeles with their combined 4% of the U.S. population can outvote the rest of us?

In addition, big cities are obviously not 100% Democrat. One estimate shows New York City and LA County are ~17% and ~21% Republican, respectively. If all 18m people in those places could vote? 3.1m Republicans votes would actually mean something, more than the entire population of Nevada and 18 other small states.

If you look at the numbers, this pattern repeats itself across the country and in any given election, there are typically only a small minority of states which really influence the election. Looking again to 2016, the following six states were won / lost by ~1.5% or less:

  1. Florida: Trump won by 1.2% and gathered 29 EC votes
  2. Wisconsin: Trump won by 0.8% and gathered 10 EC votes
  3. Pennsylvania: Trump won by 0.7% and gathered 20 EC votes
  4. Michigan: Trump won by 0.2% and gathered 16 EC votes
  5. New Hampshire: Clinton won by 0.4% and gathered 4 EC votes
  6. Minnesota: Clinton won by 1.5% and gathered 10 EC votes

Removing New Hampshire, the five remaining states represented 85 EC votes in a race which was won with 305 EC votes (85/305 is 28%). Where do you think the bulk of Trump and Clinton campaign time and resources went?

In 2016, Trump won Wisconsin, Pennsylvania and Michigan by a combined 78k PV worth 46 EC votes. If those 78k votes had been cast for Clinton, the election would have gone the other way. Where do you think the Trump and Clinton campaigns worked the hardest to engage with voters?

I will tell you: during the 2016 campaign, 365 of 399 (91.5%) total events held by the two candidates after their nominations occurred in just 11 states:

  1. Colorado
  2. Florida
  3. Iowa
  4. Michigan
  5. New Hampshire
  6. Nevada
  7. North Carolina
  8. Ohio
  9. Pennsylvania
  10. Virginia
  11. Wisconsin

Colorado was pretty lucky last election with 19 total visits, but it may not be the same during the next election or the one after that and so on.

Under the EC, campaigns focus their attention on places where they think they can win, not where they know they will win or lose regardless of what they do. To be clear, what’s important about the level of attention is not how much the candidates appear in front of voters, but rather the weight they allocate in their platforms and policy plans towards things that matter to those voters. There will alway be big, national platform issues like gun rights, healthcare, etc. It’s the regional and local issues which matter most to many voters and attention paid by campaigns leads to delivery on those issues.

While each election is a little different in terms of which districts in which states end up deciding the election, the fact remains the super majority of American voters are ignored during campaigns and marginalized by the election, with a handful of voters in a few key districts deciding the outcome.

Is this article ever going to end?

First, and as noted previously, the NPVIC will not go into effect until there are enough states in the compact to hit 270 EC votes, but once that happens, the states (not the Federal government) will have FINALLY fixed one the constitution’s major flaws. Until that time, Colorado’s EC votes will continue to be cast according to current state law.

Second, rely on solid info. This is REALLY hard with so much noise out there. Even the 2020 State Ballot Information Booklet (our “blue book”) perpetuates some easy to debunk myths:

  1. “a national popular vote may encourage candidates to focus their campaigns in large population centers where they can efficiently reach more voters.” As you’ve seen above, this is what happens today under the EC, but it’s even more hyper targeted to swing districts in swing states. Most political campaign experts, on both the right and the left, agree that under a PV, campaigns would have to take a more national and holistic approach because every vote would count.
  2. The NPVIC “attempts to sidestep the U.S. Constitution and could lead to disruptions in our electoral system”. On the last point, I think most would agree that EC votes over the past 20 years have not exactly been smooth. Regarding a “sidestep” of the constitution, the NPVIC relies on two original elements of the constitution: states’ rights on how to cast their EC votes and being able to enter into interstate compacts. Both of these have been features of the constitution from the beginning and states have changed their EC methods and entered / exited interstate compacts many times since 1789. It’s not a “sidestep” for a state to use powers it’s always had.

Don’t get sucked into the myths. Clearly, I’m a proponent of the NPVIC and in my view: Yes. Colorado should join the NPVIC. But digest the data, look at the history and decide for yourself: is America ready for “one person, one vote” to choose the one politician who represents all Americans?

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store