Our electoral system is broken. Just look at the U.S. House of Representatives. Only a tiny fraction of seats are competitive, and the results leave out far too many voters. Case in point: almost two years out, we can project the results in over 80 percent of seats for the 2020 elections.
Monopoly Politics 2020, the latest iteration in our biennial report, lays bare the deficiencies of the winner-take-all system used to elect the House of Representatives. The project began in 1997 with the insight - new at the time - that the partisanship of a district (measured by how the presidential candidates performed in the district relative to their national vote shares) was far more predictive of upcoming election outcomes than campaign-specific metrics. That insight now forms the fundamental baseline of data used by researchers and analysts who predict election outcomes, include the Cook Political Report, FiveThirtyEight and others.
Monopoly Politics demonstrates how much even today’s seemingly volatile political world remains predetermined. We make “high confidence” projections in the very safest seats for the least vulnerable incumbents two years in advance of each congressional election cycle. Although 2018 was an extremely unusual year, the high confidence projections we made in 2016 for 379 of the House’s 435 seats proved 97 percent accurate. Over four election cycles, our predictions have been 99 percent accurate, including getting every such projection correct in 2012 and 2016.
We released our 2020 high confidence projections on November 9, in 359 seats. Republicans have 190 such seats, and Democrats have 169, where they can rest comfortably, practically immune from general election competition. This lack of competition in over 80 percent of the House feeds our epidemic of partisan polarization.
Our full projections for all 435 seats demonstrate how an electoral system based entirely on geography favors Republican candidates specifically. In 2018, Democrats took the House thanks to a historic wave year - about 54 percent of the country favored Democrats over Republicans. In comparison, 2010 was a 53.8 percent Republican year, 2012 was a 52.0 percent Democratic year, and 2014 was a 51.9 percent Republican year.
Even though a majority of incumbents will be Democrats in 2020, our projections show that Republicans can take back the House even in a year that favors Democrats by as much as 51.6 percent. In 2012, Republicans won the House even when most voters favored Democrats, and that could happen again in 2020. That is largely because Republican voters are more efficiently spread out across most states, whereas Democratic voters are packed into smaller parts of the state (generally cities). In some states, Republicans exploit that distribution to exaggerate their influence with partisan gerrymandering, but Republicans also hold advantages in states that are not considered gerrymandered as well.
The newest Monopoly Politics also features new interactive options, letting users adjust the nature of the year on our interactive map and see which districts swing from blue to red or vice versa. The corresponding interactive spreadsheet also has a new look, and it includes a tab for state-by-state analysis.
The flaws in the winner-take-all system revealed through Monopoly Politics are not inevitable, nor required by the constitution. A bill introduced in 2017 - the Fair Representation Act, - would end winner-take-all politics by introducing multi-winner districts with ranked choice voting, in turn creating competitive races with winners who reflect the diversity of each district. An updated edition of our Fair Representation Act Report will be released in 2019 to accompany Monopoly Politics 2020.