The Electoral College
Neutralizing the Small-State Advantage in the Electoral College
A common complaint about the Electoral College is that it gives some states disproportionately greater weight than others relative to the presidential elections. That lack of interstate parity would be eliminated if the number of Representatives were greatly increased.
As required by the Constitution, each state’s total number of presidential Electors is equal to the size of its congressional delegation: Its total number of Representatives plus two, for its Senators. Consequently, the less populous states benefit from greater shares of representation in the Electoral College than do the more populous ones (relative to their respective population totals).
Though this “small-state advantage” is by design, the founders also intended for this advantage to be diminished as the country grew. This is because the founders expected the size of the House to forever grow along with the nation’s total population, and therefore so would the Electoral College. However, because the size of the House has been arbitrarily fixed at 435 since 1913, the small-state advantage has not diminished since then. If the Electoral College (EC) were the size it should be, each state’s share of the EC would be almost exactly equal to its share of the total population.
Maximizing Parity in the Electoral College
In this context, parity is the ideal wherein each state’s share of EC votes is equal to its share of the total of all the states’ populations. Currently, there are great disparities, meaning that many of the states’ shares of total EC votes are disproportionate to their shares of the total population. For example, though Delaware has 0.30% of the nation’s total apportionment population, it has 0.56% of the Electors in the EC. This section explains how these disparities can be eliminated by substantially enlarging the size of the House of Representatives.
Interstate disparities: Two causes
The states’ lack of parity in the EC is caused by two completely unrelated factors. The first factor can be called the apportionment disparity: Because we have far too few Representatives, every state has either a deficit or surplus in representation relative to their share of total population. These disparities explain why the population sizes of congressional districts vary widely from state to state. For example, relative to the 24th apportionment, which will go into effect as of the 118th Congress, the largest (most populous) congressional district will be 82% larger than the smallest (least populous) one. As a result, the House is in egregious violation of the Constitution’s one-person-one-vote equality requirement. This apportionment disparity extends to the EC simply because the number of Electors is derived from the number of Representatives.
The second factor can be called the augmentation disparity, which results from adding two to each state’s total number of Representatives. For most states, especially the smaller ones, this results in a total number of Electors which is disproportionate to its share of the total population. The way this augmentation disparity is commonly understood is best explained with an example: For the five states that each only have one Representative, such as Delaware, the additional two Electors increase their number of Electors by 200% (from one to three). However, for the largest state — California with its 52 Representatives — the additional two Electors increase its number of Electors by less than 4%. And, of course, all the other states lie somewhere in between.
The net effect
Though that simplified analysis of the small-state advantage is an easy way to conceptualize it, the aggregate analysis reveals that the reality is quite different for two reasons. First, the addition of those two Electors can either compound or offset the underlying apportionment disparity. And second, the actual impact of the augmentation disparity on any particular state is a function of the aggregate impact of adding 100 Electors to all 50 states.
The easiest way to explain the foregoing is with the following three examples (illustrated in the adjacent chart).
Example 1: As of the 2020 census, Delaware had 0.30% of the total apportionment population. However, because they were shortchanged by the apportionment, their single Representative is only 0.23% of the states’ representation in the House. Adding two Electors gives Delaware 0.56% of the states’ representation in the EC. Though Delaware’s EC delegation is almost twice its share of the states’ total population, it is a much smaller advantage than expected from the simplified analysis
Note: For the purposes of this analysis, DC’s total population and three Electors are disregarded. Including them would change some of these percentages slightly, but would not change the overall state-to-state analysis which is the basis of the apportionment.
Example 2: New York is an example of a state which was double shortchanged. Though their census tallied 6.11% of the nation’s population, their 26 Representatives account for 5.98% of the representation in the House. And its share of the EC is further reduced by adding Electors to all of the states: New York’s 28 Electors account for 5.23% of the EC, which is 14% less than its share of total population!
Example 3 is a state which was double bonused: Though Oregon has 1.28% of the population, their six Representatives give them 1.38% of the House. And their eight Electors account for 1.5% of the EC, which is almost 17% more than its share of population.
All of these representational disparities would be substantially eliminated if the number of Representatives were sufficiently increased but, before explaining that, let’s zoom out to see the analysis described above for all fifty states.
In the chart below, each state’s percentage share of the total population is indicated by the small diamond (◆), and its percentage share of 435 Representatives is indicated by the vertical bars.
Though the small scale of the chart above makes it appear that the states’ shares of representation are nearly equal to their shares of total population, most are quite different. In fact, they are so different that the disparity between the largest and smallest congressional districts is 82%, and the relative standard deviation of all the district sizes is a whopping 2,986%. These apportionment disparities are a result of there being far too few Representatives relative to having over 330 million residents in a fifty-state republic.
Therefore, even if the states’ EC delegation sizes were based only on the size of their House delegation (excluding the additional two), the inequities resulting from the apportionment disparities would still be duplicated in the EC. Though those resulting disparities are not as significant as that caused by adding two Electors, it’s important to understand how that affects the final analysis.
The next chart is identical to the last, except that it adds a vertical bar to indicate each state’s percentage share of representation in the EC.
The chart above is the same as the initial chart (Figure 1) with the three examples, but with all 50 states included. It illustrates that adding two to the number of Representatives either increases or decreases any state’s share of total Electors. Note that, as with the examples in Figure 1, a state’s share of the EC may be more or less than its share of House representation, which may be more or less than its share of population.
The net effect of both the apportionment and augmentation disparities are indicated in the next chart.
For each state, the chart above indicates how much more (or less) representation it has in the EC than its share of the total population of all states. For example, as mentioned previously, Delaware has 0.30% of the total population, but 0.56% of the EC votes. Therefore, it has 87% more representation in the EC than warranted by its share of total population.1((0.56% ÷ 0.30%) – 1) = 87%
If there were perfect parity, all of these values would be equal to zero. As it is, the average deviation to perfect parity is 38%.2The average deviation is the average of the absolute values of all of the deviations.
These disparities are reduced by enlarging the size of the House of Representatives, as illustrated by two examples. The first example is to increase the number of Representatives to 692, a size suggested by the widely-cited cube-root “law”. As shown in the chart below, such a modest increase in the size of the House produces a correspondingly modest reduction in interstate disparities, reducing the average deviation to 24.5%.
With a disparity between the most and least populous congressional districts of 75% (and a relative standard deviation of 1,623%), the 692-Representative apportionment is still in egregious violation of the Constitution’s one-person-one-vote equality requirement. Therefore, it is not surprising that the apportionment disparity is still quite significant. And because each states’ number of Representatives is relatively small, augmenting it by two still results in a significant advantage for the less populous states.
The final scenario, below, is a much larger House with 6,692 Representatives, a size which is compliant with both Article the first and the Constitution’s one-person-one-vote requirement. The average deviation of this solution is only 3.0%.
In the case of the 6,692-Representative apportionment, the underlying disparity between the largest and smallest congressional districts is only 3.7% (with a corresponding relative standard deviation of 12%), substantially eliminating the apportionment disparity. And because there are many more Representatives for each state, augmenting those totals by two is of little mathematical consequence.
The supporting data for all of the charts in this section are provided in this table.
An Equitable Electoral College
For those who believe that the lack of interstate parity in the EC is a problem that should be eliminated, there is only one constitutional solution: Substantially increase the size of the U.S. House. If they’re not willing to do that, then there’s no point lamenting about this inequity, especially since the Electoral College is here to stay.
© Thirty-Thousand.org [published 8/31/22]
Explore More Topics:
- 1((0.56% ÷ 0.30%) – 1) = 87%
- 2The average deviation is the average of the absolute values of all of the deviations.