Do Republicans Only Win Elections in Small Population States?

newhampshire-photodisc.jpg
Photodisc

Vox, a left-wing news-explainer website, recently dropped the headline announcing that "The Senate's 46 Democrats got 20 million more votes than its 54 Republicans." The tagline stated this is the case "because Republicans come from faller states." Author Dylan Matthews elaborated further

The problem isn't that the deck is stacked in favor of Republicans. The problem is that the deck is stacked in favor of small states, which receive equal representation in the Senate despite dramatic variance in population. The Senate is a profoundly anti-democratic body and should be abolished.

We don't argue the actual figure of a 20 million vote difference, but is the reasoning true and logical? The overly-simplistic analysis gives us a hint. What we find is that the author likely started with the conclusion - that "Republicans are from smaller states" - and upon trying to deliver actual evidence to prove that is the case, and being unable to, gave up. So, what is the reality?

Republicans represent smaller states, and so do Democrats

The 10 most populous states are (in order) California, Texas, Florida, New York, Illinois, Pennsylvania, Ohio, Georgia, North Carolina, and Michigan. When we look at those 10 states, which we are led to believe should be dominated by Democrats, we find that Republicans hold 10 seats and Democrats hold 10 seats. Four of the states are represented equally by one member from each party.

When we jump to the 10 least populous states, which now should be dominated by Republicans, we find the states of Wyoming, Vermont, Alaska, North Dakota, South Dakota, Delaware, Montana, Rhode Island, New Hampshire, and Maine have Democrats representing 10 of these seats, and Republicans also 10.

If you want to find out why the Vox analysis literally goes no further than to say "because Republicans come from smaller states" this is your answer. And if you are worried that the 30 states in-between are more lopsided, don't be. For every smallish conservative Idaho, Utah, and Nebraska there is an equally smallish liberal Hawaii, Connecticut, and New Mexico paired with them.

So, About that 20 Million Figure

We know that Republicans will enter 2015 with an 8-seat, 54-46 majority. (To put this in a little more perspective, since it seems like a decent majority, we are really only talking about the difference of  a 4-seat swing out of 100 possible seats.) The Republican majority is quite a bit smaller than than the 54-46 lead indicates. If we agree that the 20 million voter difference is correct, and we now know that there really isn't a small-state advantage for Republicans and a large-state advantage for Democrats (as easily proven above), why is there such a big difference? That is also an easy one... if you care to look at the facts.

The single biggest factor in determining how many votes the candidates' received is when they got elected, not what states they are elected in. The three years in question include 2010 and 2014, two good years for Republicans, and 2012, a good year for Democrats. In the one year that the Democrats did extraordinarily well it happened to coincide with a Presidential election where turnout is many times higher than in mid-term cycles, where Democrats did well. That year, Democrats had 21 seats up for re-election while Republicans had just 10.

Democrats won 23 that year - representing exactly half of their 2015 congressional class - while Republicans won just eight. Therefore, Democrats received a huge bump of votes in a dozen safe and a few competitive seats simply because it was a Presidential year.

So, how much of an impact does this have in the 20 million figure? Of the 8 seats that Republicans won in 2012, thereby receiving a presidential-year bump, only Texas lands in the top 15 of the most populous states. So, Republicans received a healthy bump in a single state. They also received bumps in other states they won including Utah, Maine, Nebraska, Mississippi, and Wyoming, but all are miniscule due to their small size. The reality is that in the year where every elected US Senator would have a big bump in raw votes, there simply weren't many good states for Republicans and plenty of good states for Democrats (the reverse is true in 2016).

Meanwhile, Democrats had strong incumbents with token opposition and just so happened to have many of their high population states up in 2012. Incumbents had padded victories in top-15 states California, Michigan, New York, Florida, Massachusetts, New Jersey, Pennsylvania, Ohio, and Virginia. Even when races were somewhat close (VA, OH, PA) using the raw vote total of the victor alone paints a false impression of Democratic electoral dominance.

Impact of Presidential Years

Seeing how much impact a Presidential year has on vote collection is easy since all states have 2 Senate elections over 3 cycles, in this case 2010, 2012, and 2014. Every state that has a Presidential-year election will also have a non-presidential one. So if we look at California, we see that in off-year 2010, the Democrat won 5.2 million votes out of 9.4 million cast. Just two years later, an equally strong incumbent won 7.8 million votes out of 12.4 million case, an inflated bump of some 2.5 million voters, or more than 10% of the 20 million gap alone. (Lest you think it was a difference in candidate quality, the 7.8 million votes was up from the Senator's previous effort of 5.2 million in off-year 2006.

You find the same voter-inflation in every state. In New York, the 2012 Democrat earned 1.8 million more votes than the 2010 Democrat. In Texas, Republican Ted Cruz won 57% of the vote and a hefty 4.4 million votes in Presidential year 2012. In off-year 2014, John Cornyn was stronger at 62%, but won just 2.8 million votes.

Split states are even more telling. Republican Marco Rubio won by 20 points in off-year 2010 and collected 2.6 million votes while entrenched Democrat Ben Nelson won by 11 points in Presidential year 2012 and collected 4.6 million votes. So another 2 million difference towards that 20 million difference is based on the year of election, and little else. In Pennsylvania, Republican Pat Toomey took 2 million votes and 51% in 2010. Two years later, Democrat Bob Casey won a similar 53.6%, but received a full 1 million votes more than Toomey.

(Even the Republican won 500,000 more votes in losing in 2012 than Toomey took in winning in 2010.) And in Ohio, Republican Rob Portman won by an uncompetitive 18 points in 2010. In 2012, Sherrod Brown would win with a smaller 51%, but collected 700,000 more votes.

The Reality

Democrats won 50% of their seats in a Presidential election year, which greatly inflates the number of votes their candidates received and that are added to the "20 million vote gap." The large state/small state argument is based either on ignorance or willful dismissal of any and all facts. In 2016, Republicans will have 24 seats up for re-election. It is also a Presidential year and as a result every single victor will have a much higher vote total than they had in 2010. They have 16 seats that start off safe, meaning those alone could easily wipe out the 20 million voter advantage currently enjoyed by Democrats. While Democrats collected millions of bonus votes in 2012, Republicans will do the same in Georgia, Ohio, and North Carolina (all top 10 populations) and have a good shot to hold Florida, Pennsylvania, and Illinois.