Does the Electoral College system really favor low population areas?

Question

I was reading some questions on this site, and I stumbled up this comment. I'll transcribe the relevant parts of it here in case it ever goes away:

You could throw in something about how Hillary actually won the national popular vote and how the electoral college system is seriously broken.

...

And, technically, the United States is a republic, and the electoral college isn't representative (low population states are favored). -- @barrycarter

(emphasis added).

Is this claim really true? Thinking through it logically, it would make sense that in a hypothetical situation where a state has only a single resident, clearly that persons vote is a lot more important that someone who lives in California. But since states with smaller population have less electoral votes, that should counter this affect.

But I don't know how large this compensation is. Is it enough to fully counter the higher influence of lower population states? And if it's not, is the higher influence of low population areas relevant to the state or to regions within the state (such as counties)?

And another thought occurs to me. Is the electoral college intended to give smaller states more influence? It wouldn't seem unusual to me if it was designed so that each state has a certain amount of influence regardless of population so that extremely large states can't exert their will over all of the other states.

Answer 1

Yes it does, but there's a specific reason for it.

It's important from a historical context to understand that the framers of the Constitution had 3 major concerns in mind when they created the Electoral College, all of them talked about one way or another in the Federalist Papers:

1. To prevent politics from devolving into political parties. It obviously failed utterly in that regard. In the original Constitution, the President and Vice President did not run together as a team. Electors cast one vote each. The hope was that if the two were from differing political ideologies, they would have to work together to get things done. After two miserable electoral crises happened in the Election of 1796 and 1800, the 12th Amendment abolished that system.
2. To prevent a populist from getting elected. Again, current evidence shows us that this is yet another failure. The Framers were worried about the "tyranny of the majority" as they called it. The Electoral College was supposed to be completely independent of the electorate, being able to override them if a completely unqualified candidate got the popular vote. What actually happened is that states began passing laws that pledged ALL their electoral votes to the popular winner in their state -- even if the split was narrow, thus creating the "winner-take-all" system we have now.
3. To "handicap" large states so that small states could still have a voice. This is where your question becomes relevant. The Framers were concerned that the largest, most populous states would "crowd out" the voice of smaller states, such that they would never be able to participate in the elections process in any meaningful way. Indeed, if we look at America today, 54% of the U.S. population live in just 9 states: California, Texas, Florida, New York, Illinois, Pennsylvania, Ohio, Georgia, and North Carolina. The next 9 most populous states (Michigan, New Jersey, Virginia, Washington, Arizona, Massachusetts, Tennessee, and Indiana) take you to 75% of the population. Basically, a straight popular vote would mean that just 20 states (give or take) would be deciding all presidential elections. If you lived in Montana or Wyoming, you may as well not even vote. The Electoral College forces candidates to care about states like Iowa where they normally wouldn't.

To the point of your question, the system is set up so that smaller states get a small edge so they can still be competitive in an election against bigger ones. Can it be done better? Well that's a whole other question.

Answer 2

Is the electoral college intended to give smaller states more influence? It wouldn't seem unusual to me if it was designed so that each state has a certain amount of influence regardless of population so that extremely large states can't exert their will over all of the other states.

The expressed intention was to do the latter. But the method they chose automatically does the former. I.e. they deliberately shifted power to small states to avoid having it concentrated in large states.

Note that in reality, the modern winner-take-all system doesn't favor small or large states but politically balanced states. In fact, we normally call these swing states. This is because swaying votes in either Wyoming or California doesn't generally matter. The Republicans win the presidential vote in Wyoming and the Democrats win in California. So neither side spends much time campaigning in those states. Instead, the campaigns spend the most time in places like

1. Florida -- both big and contested.
2. North Carolina: average size and contested.
3. Pennsylvania: above average and contested.
4. Ohio: above average and contested.
5. Colorado: below average and contested.

Neither campaign made any stops in Wyoming, Montana, the Dakotas, or Vermont, states with three electoral college votes. The Clinton campaign made two stops in California, possibly for fundraising purposes. The Trump campaign made one stop in Texas. Maine, New Hampshire, and Iowa received multiple visits despite only having fourteen electoral college votes among them. Maine and Nebraska award electoral college votes by congressional district (two to the state winner for the Senate districts).

The campaigns made a number of stops in New York, a large uncontested state, but that was likely driven by the fact that Donald Trump and Hillary Clinton both lived there. The media market (many national channels have offices in New York, so it is easy to get attention there) and fundraising may have helped. There are also parts of Pennsylvania that watch the New York television channels (also New Jersey and Connecticut which are uncontested).

Answer 3

Wyoming has 585,000 people, and 3 electoral votes.
California has 39,250,000 people and 55 electoral votes.

In Wyoming, it's about 200,000 people per Electoral Vote.
In California, it's about 710,000 people per Electoral Vote.

If given the choice, would you rather work to win over 100,000 people, or 355,000 people to your cause?

---

On the other side, it is a lot easier to campaign in California: A well organized campaign can speak to many thousands of voters at once in dense areas.

In Wyoming, reaching voters requires long drives speaking to small groups in small towns.

Answer 4

HUGE OVERSIMPLIFICATION WARNING:

USA is not a republic as in Res Publica, something owned by the people. USA is a federation, it's "owned" by the states, not by the people. Every state is then owned by its people. The democracy of people is at state level, federation is a democracy of states.

That's the rough idea. It was polished to be more people-oriented, but the underlying framework was indeed intended to treat states more equal than their population alone would suggest (which boils down to "favoring the small ones").

It's not uncommon, eg the EU works the same way: it's a struggle between smaller countries "one state means one vote!", the larger ones "it's about people after all" and everyone in between proposing something very complicated that's superficially "fair" but actually intended to favor them specifically.

Answer 5

The answer depends on what you mean by 'favors'.

As many have pointed out, there's a big difference between CA and WY in terms of how many votes go towards each electoral vote.

It takes 3.5 times as many votes in CA for each electoral vote than it does in WY.

As such, one way to word it is: the college favors low population states in that their citizens' votes are each individually 'worth more'.

However, there is the fact that urban areas tend overwhelmingly vote for the democrat, rural, the republican. Higher population states tend to have more and larger urban areas. As cities get more population and rural areas, fewer, one could say: the college favors democrats in high population states, republicans in low population states

The reality is, however, that most states are clearly red or blue when it comes to presidential elections. Only a handful are purple. These swing states are where a lot of campaign effort is put. As such, another valid argument is the college favors swing states with higher populations in terms of influence each individual has on the final outcome of the election.

Finally, we should look at history. And look at who it was originally designed to favor. Initially, it was designed to accommodate the fact that this was a large nation (area wise) with large space between people in terms of access to information and time. The thought was that it would be difficult for people to become adequately informed and it would be best to have representatives vote for them.

The debate around the 12th amendment brought up the subject again and eliminating the college was again brought up (as the original reasons for it were becoming less relevant). But it was retained primarily to deal with the north/south divide over the issue of slavery.

At the Philadelphia convention, the visionary Pennsylvanian James Wilson proposed direct national election of the president. But the savvy Virginian James Madison responded that such a system would prove unacceptable to the South: “The right of suffrage was much more diffusive [i.e., extensive] in the Northern than the Southern States; and the latter could have no influence in the election on the score of Negroes.” In other words, in a direct election system, the North would outnumber the South, whose many slaves (more than half a million in all) of course could not vote. But the Electoral College—a prototype of which Madison proposed in this same speech—instead let each southern state count its slaves, albeit with a two-fifths discount, in computing its share of the overall count.

So one last way to state things is that historically, the electoral college was retained so as to not favor the North or the South on the issue of slavery.

Answer 6

It really depends on how exactly you define a state as being “favored” or not. The conventional argument that the EC unfairly favor's low-population is how the Electors are assigned.

Electors = Representatives (proportional to state population) + Senators (2 for every state)

(Plus the special 23rd Amendment rule that gives DC three Electors even though it's not a state and has no voting representation in Congress.)

The addition of the two extra “Senatorial” votes has negligible impact on the EC representation of large states — California having 55 electoral votes instead of 53 is only a 3.77% “bonus”. But Wyoming getting 3 electoral votes instead of 1 triples its representation.

Using a simplistic population per Elector metric based on the 2010 Census numbers gives:

• California: 37 253 956 residents ÷ 55 Electors = 677 345 residents per Elector
• Wyoming: 563 626 residents ÷ 3 Electors = 187 875 residents per Elector

And based on the above numbers, a Wyoming voter is “worth” 3.6 times as much as a California voter.

BUT the electoral college is a winner-take-all system in each state. If you look at measures of “voter power” within such a system, like the Banzhaf Power Index, you get a much different picture.

Based on my own calculations, California is 20.8 times as likely to cast a “deciding” block of electors as Wyoming is. So, if you consider the metric of Banzhaf power divided by population, a Wyoming vote is still “worth” 3.17 times as much as a Californian's.

But let's consider the probability that an individual voter will swing the election. Using a “coin flip” model of voting in which votes are treated as independent Bernoulli trials with p=0.5, the probability of an exact tie (breakable by a single voter) is O(1/√n). So, the probability that any single Wyoming voter will swing Wyoming's vote is only 8.13 times a high as the probability that any single California voter will swing California's votes. And the probability that one California voter will swing the whole national presidential election is 2.56 times greater than one Wyoming voter swinging the same election. (Though, both probabilities are still miniscule compared to say, that of winning the Powerball lottery jackpot.)

BUT this “coin flip” model of voting is questionable, especially in “safe states” like WV (68.5% Republican) or DC (90.5% Democrat). In practice, voters have virtually no “power” (in the academic, probability-based sense) unless their state's polls happen to be close to 50-50.

This favoritism towards swing states has a practical effect on where Presidential candidates campaign, with FL, OH, NC, and PA having gotten the focus in the last election. These are not only swing states, but high-population swing states.

BUT, no matter how much we talk about states' relative influence, keep in mind that this is different from high and low population areas. Many states have a mix of urban (and Democrat-leaning) and rural (Republican-leaning) areas. Even if the EC does have a systematic bias towards low-population states, it doesn't help the deep “Red” rural parts of Illinois, which consistently get outvoted by Chicago. (Inversely, “Blue” Philadelphia got outvoted by the rural parts of Pennsylvania in the last election.)

The current partisan divide about whether the Electoral College is good or bad comes mostly from the outcomes of the 2000 and 2016 presidential elections, in which the Republican won the election despite losing the national popular vote. What both sides tend to forget is that these were both close elections that could just have easily gone the other way.

Answer 7

Low population states do get more electoral votes per person, as the other answer explained. They also get a smaller absolute number of electoral votes. Whether that is an advantage, a disadvantage, or neither is open to interpretation. For one thing, the vast majority of the 3-vote states are one-sided contests, where there is no question who will win. Furthermore, in a close election, it's much more likely to come down to a large state, or a few large states, than to a small state. Against all that, it's obviously an advantage to have more votes per person.

The framers apparently thought that the system they adopted was a reasonable compromise. Since neither large states nor small states seem to complain about it more often than the other, that was likely correct.

Answer 8

When measuring power in a weighted voting system (such as the electoral college) one should note that voting power is not proportional to the number of delegates. Two ways of measuring voting power are the Banzhaf and the Shapley-Shubik power indices. These tend to show that larger blocs tend to have even more power than simple proportionality would suggest.

An analysis has been made using the Banzhaf power index to measure the electoral power of each state and considering the likelihood of any individual voter being critical to the result in a state. The result of this analysis is that a voter in has a Banzhaf power that is roughly 2.4 times greater than a voter in Wyoming.

In general, this analysis suggests that voters in large states still have greater power than those in small states.

In practice, it is understood that many people don't change their voting behaviour from year to year, so a floating voter in a swing state has much greater chance of being influential than one in predictably red or blue state.

Answer 9

Technically speaking, a state with no population would receive 3 electoral college votes. This is because every state has 2 senators and at least one representative. Each state chooses electors, equal in number to that state's combined total of senators and representatives. Since number of representatives is dependent on population, bigger states receive more electors. The average size of a congressional district is 700 000 citizens, and as many states have much less than a million citizens, the guaranteed 3 electors are very significant in terms of electoral math.

Additionally, several high population states are mostly very partisan, i.e. NY, TX, CA, GA and WA. With some notable exceptions like Florida and the Midwest, most of the remaining states, which have a remaining electoral vote of 337. (Ohio was not included as midwestern, as it is a swing state.)

Answer 10

So I've been doing some math because I have seen some bad math here.

To calculate the voting power of any individual electoral vote, it's a bit more complicated than the Population (P) divided by the total electoral votes (E).

This is because each state gets 2 electoral votes automatically. All remaining votes are given to match the number of House of Representatives allocated to that state. So, it's incorrect to say the electoral vote is equal to P/E because 2 of those votes represent the entire population of the state.

It is better to say that when someone votes for the president, their votes are used to determine 3 votes, not one. So if we have a unit called Representative Power (R) of a single Electoral vote for a given state, then our formula becomes:

R = (((P/(E-2) + (2P))/(E-2)))

Thus given the numbers set by @LeeDanielCrocker, let's first see what R of Califorinia would be:

Let P = 39,250,00 and E = 55

Thus our equation comes to:

R(CA) = (((39,250,00/53) + 2(39,250,000))/53) R(CA) = ((740,566 + 78,500,000)/53) R(CA) = 79,240,566/53 R(CA) = 1,495,105

So, with that in mind, we can now calculate Wyoming using the same numbers:

Let P = 585,000 and E = 3 R(WY) = (((585,000/(3-2)) + 2(585,000))/(3-2)) R(WY) = 585,000 + 1,1700,00 R(WY) = 1,755,000

So the R(WY) is definitely greater than the R(CA).

In fact, one California electoral vote equals 82% of one Wyoming vote. But this is on a one to one voting weight only. No matter how you slice this though, 3 votes at 100% power is still less than 55 votes that represent 82%.

So there is the math behind the idea of the smaller the total electoral votes the more powerful your vote becomes. This would be true of the force behind a Congressional seat as well.

This is a result of what is called the Great Compromise. As a Representative Democracy, figuring out how citizens would be represented in Congress needed to be a balancing act. If each state got an equal number of votes, small states could impose their will on large states because their vote is equal to the hundreds of thousands, while the larger state's vote is in the 10s of millions. Conversely, if they were given out by population, then the large states would be more powerful than the small states, because their vote is worth the same, but they have 55 times more votes than the smallest ones.

The balance here was to allocate 2 votes to every state and additional vote for each division of population. This makes it that small states have more powerful votes but large states have more numerous votes.

To understand the electoral college's design, think of each state as an individual country and the Federal Government as a big treaty organization between all of the member states. The leader of this organization would in effect be the leader of all the states. So it was necessary to pick the leader that showed he had the best interests of the individual states in mind. Hollywood actors and Oklahoma corn farmers have very little in common and very different needs from government. Each state does have its own interests to look out for.

Since each state is entitled to pick its representation in the way it sees fit, provided that it is a Republic (read representative democracy which was how the framers understood it), then the balance is that each state is afforded a share of votes equal to their congressional delegations (2 + the House Delegation). They are free to distribute them however they see fit (mostly winner take all; two states delegate the districts to the winner in that district and the remaining two to the overall state winner, and South Caroline historically voted on the full delegation in the legislature; and a number require by law that electors vote for the candidate the state chose). However, the states are only given so many votes in accordance with their congressional power.

The most balanced option is to distribute like Maine and Nebraska (congressional districts determine all but 2 of the electoral vote, and the 2 remaining are given to the winner). This would get us closer to the popular vote but we would still run into an occasional popular/electoral split.

Now, before you say this is undemocratic, consider this:

Switzerland is the world's only Direct Democracy and is, with notable exceptions, modeled after the United States Federal System. Switzerland does give its people the right to directly write and repeal their laws and even amend their constitution (with exception to citizens rights, which cannot be repealed from the constitution). In order to pass in this method, the law must pass with Double Majority. This means that popular vote alone does not get the job done, the law must achieve popular support in a majority of the Cantons (basically the same thing as States) in order to pass. A popular/Canton split means the law is not valid. This check is intended to avoid the tyranny of the majority like the electoral college was in the United States. Double Majority is a little hard to do in the United States Presidential elections because there is no proposal as to what would happen if a Double Majority is not achieved on a leadership position. Swiss got around this by making an Executive council of seven co-equal executive members, with each one rotating into the Council President and Vice President, who would be the de facto leader in a diplomatic settings. This committee is appointed by the legislature, not election of the people, so even in one of the most Democratic nations in the world, the people do not get to directly pick their top leadership by popular vote.

Answer 11

It was the intent of the Electoral College to give an extra voice to sparsely-populated rural states. But the founding fathers got their math wrong. In fact, the EC gives populous states even more power than they would have under a strictly proportional system, especially if they are politically diverse. This is because the effects of the "unit rule" totally overwhelm the minor effect of disproportionality.

Wyoming has 585,000 people and 3 electoral votes (that's one EC vote for 195,000 people). California has 39,250,000 people and 55 electoral votes, for one EC vote per 714,000 people.

Let's say that swaying a voter costs $1 in advertising. Without the unit rule you could buy a CA EC vote for $714,000 or a WY EC vote for a bargain $195,000. But you can't buy single EC votes: you have to win the whole state to get any. So if both states are close and that one district full of voters puts you over the top, then your $714,000 in CA bought you 55 EC votes, for less than $13,000 each, while your $195,000 in Wyoming only bought you 3, for $65,000 each. So I as a candidate will ignore WY and spend my money in CA, TX, NY, PA.

This is especially true if the big states are reasonably close-polling, which they usually are. A democratic candidate might choose to give up TX to spend more in CA, and republican vice versa, but neither of them is going to make any but a perfunctory appearance in WY.

Answer 12

Using census numbers from wikipedia, which allows also conveniently sort table by any column:

In the House, the situation is less skewed (you can sort states by "census population per house seat, 2010").

Top five states where one house representative represents least number of population are: RI, WY, NE, VW, VE, NH, where only WY and NE could be called "low population" states.

On the opposite side, top 5 states with highest population represented by one House representative are: MT, DE, SD, ID, OR. I was quite surprised, because Delaware and Oregon stick out, rest are "low population" states. I guess Montana is there because it does not have enough population for two representatives.

In senate the discrepancy is more obvious: More than 50% population live in just 9 states: CA, TX, FL, NY IL, PA, OH, GA, NC, with total population as of 2010 census of 156,878,852, which is 50.9% of the total census population of 308,156,338.

These 9 states (representing majority 50.9%) have just 18 senators, remaining 41 states have 82 senators.

Then there is another spin on the data: gerrymandering which allows redistricting favoring the party which held the power when redistricting is made, allowing representative to pick the voters which s/he will represent.

WaPo has interesting article about gerrymandering showing computer generated district (with bias for compactness) vs the reality.

But I would guess that gerrymandering is based not about density of population, but by expected voting preferences of the population, where rural areas (with low population density) have different preferences as suburbs and urban areas.