Wednesday, December 17, 2014

Winner-take-all Providing Disproportionate Outcomes in Japanese Elections

On December 14 2014, Japanese voters  well, about half of those eligible to vote — headed out to the polls to choose a new House of Representatives. The incumbent Liberal Democratic Party (LDP) - Komeito alliance emerged victorious, picking up an additional 20 seats to take their total to 326 seats in the 475 member House.

There are three main narratives of this election. One is of an unnecessary and uninteresting "snap" election that merely maintained the status quo. Prime Minister Shinzo Abe and his coalition were returned to power with another parliamentary supermajority, which means Abe can continue with his program of "Abenomics".   A second narrative is that of historically low turnout. Turnout fell 6 percentage points from 2012, to 53% of the voting age population. (In Japan, voter registration is automatic and administered by local authorities, who add people to the register using the Juki Net register of Japanese residents.)

The third narrative is of the disproportionality between votes cast for the LDP and the constituency seats won by them. Japan uses a parallel voting system, in which every voter gets two votes: one candidate vote for a single member in their local constituency; and a second vote for regional seats, elected using party-list proportional representation in multi-member districts.

In the 180 seats elected by proportional representation, seats were -- predictably -- allocated reasonably proportionally. The LDP won 33.1% of the party list vote across Japan and received 37.8% of the seats.

However, in the 295 constituency seats, LDP candidates received 48.1% of the vote across Japan... but won 223 of the seats. That's 76% of the seats from less than half of the vote. 

And so, overall, the LDP ended up with 291 (61.3%) of the 475 seats in the House of Representatives from 41% of the votes cast.

The disproportionality is due in part to malapportionment:  The 295 constituencies range in size from just over 200,000 registered voters to almost 500,000. The smaller constituencies tend to be rural, where support for the LDP is strongest - and so the LDP can win more seats from fewer votes.

More importantly, the disproportionality is a consequence of the use of single-member districts with winner-take-all in the constituency seats. Single-member districts with winner-take-all are inherently susceptible to disproportionality, especially in heterogeneous populations spread out unevenly (as in Canada) or in multi-party systems (as in Japan).

Unlike in Germany and New Zealand (which use a system similar, but superior, to parallel voting called "mixed member proportional"), the Japanese system makes no attempt to achieve proportionality in the legislative chamber overall. Thus, the unfortunate effects of winner-take-all dominate Japanese election outcomes.

Senate Confirms 3 Commissioners to the EAC

After four years without a quorum (and three years without a single commissioner), the Elections Assistance Commission (EAC) now has a quorum of three confirmed commissioners

The U.S. Senate acted yesterday, December 16, 2014, to confirm the new commissioners: Thomas Hicks (D), Mathew Masterson (R), and Christy McCormick (R). The EAC now has an opportunity to fulfill the duties it has fallen short of for the past few years while it has essential been inoperable. Frankly, this is cause for celebration!

FairVote, among many other organizations that want to see better elections, has been calling on Congress to confirm a quorum of commissioners to the EAC and allow the commission to accomplish its duties described in the 2002 Help America Vote Act (HAVA). In a 2013 email calling on supporters to contact their congressperson, FairVote stated that "creating an active EAC is a simple positive step--and one that happens to be the law." After several years of waiting, that step has finally been taken, and the EAC will be able to function once again with a quorum, and a group of commissioners that have generated optimism among members of the voting community.

The bottom line is that this is exciting. While national standards are not a fix-all solution to the electoral issues plaguing our nation's democracy, they do provide a base of uniform best practices for each state. In an age where thousands of voting jurisdictions run elections in thousands of different ways, and American voting technology falls far short of what it could be, this development is definitely something for voters and fans of democracy across the country to be excited about. This is a huge step forward, and is hopefully just the first step in many by this newly empowered commission.

Monday, December 15, 2014

Taking Election "Selfies" - Fun, but Potentially Illegal

Taking a photo of your completed ballot paper is, it seems, illegal in 44 of the 50 U. S. states. NPR reported on developments in New Hampshire, in which several voters who shared digital images of their completed ballots online are currently under investigation by the New Hampshire attorney-general's office. The offense of "taking a digital image or photograph of [your] ballot and distributing or sharing the image via social media or by any other means" carries with it a fine of up to $1000.

Laws prohibiting voters from copying or revealing the contents of their ballot hark back to the 1890s and the adoption of the Australian ballot. As part of the Australian ballot, voting became secret. Secrecy, long fought for in states like Massachusetts and California, was intended to prevent the bribing of voters (often with liquor). If no one could know how you voted, no one would want to bribe you to vote for them. After all, they might provide you copious amounts of free whisky, only to have you retreat to the secrecy of the ballot booth and vote for a different candidate.

The secrecy of the ballot could only be guaranteed if the state printed and provided the ballot AND voters were prevented from taking their ballot papers (or proof of how they voted) out of the polling place. It is for this reason that, in the age before voting machines, any vote in which the voter identified himself (or, after 1920, herself) was invalid.

Of course, no one in the 1890s could have foreseen the current trend of voters proudly sharing images of themselves with their ballots... and without any expectation of free whisky.

Thursday, December 11, 2014

Lessons from the Anniversary of Women's Suffrage in Wyoming

On this date, 145 years ago, Wyoming became the first state in the union to extend voting rights to women. At that time only about 1,000 women lived in the state and benefited immediately from the groundbreaking legislation. However, in 2014, women of voting age across our nation enjoy full suffrage rights, and will hopefully be equally represented in elected office sooner rather than later. Today, it is important to remember that this crucial--and now widely accepted--expansion of voting rights began when a group of legislators had the audacity to expand the electorate beyond the status quo, and allow the American public to recognize the value in a more inclusive democratic process. Without such action, we remain stagnant as a democratic society.

As we approach a similar juncture in our democracy, we should remember the lessons that Wyoming taught us 145 years ago. The debate around lowering the voting age to 16 is reaching a crescendo in the United Kingdom, and is well under way here in the United States. In June of 2013, Takoma Park, MD, became the first U.S. city to extend voting rights in local elections to 16 and 17 year-olds, and FairVote played a significant role in advancing the idea. There was certainly initial skepticism around the idea, however, the city council moved forward, Takoma Park's 2013 local elections took place, and the world kept turning. In fact, 16 and 17 year-old voters participated at higher rates than their older counterparts.

Opponents argue that these young potential voters might not be able to make sound decisions, or that they will vote like their parents--the same arguments that were made when women fought for suffrage. However, just like 145 years ago in Wyoming, Americans are beginning to understand that this expansion of the franchise makes sense, and has some very positive implications for local civic engagement, voter turnout, and the health of our democracy. Takoma Park's neighbor, Hyattsville, is considering adopting the same policy, and localities across the nation should do the same. Here at FairVote, we are confident that one day Americans will look back--as we look back at Wyoming today--and appreciate the innovation of cities like Takoma Park and Hyattsville for their role in expanding the right to vote to America's 16 and 17 year old citizens.

Wednesday, December 10, 2014

Electing the 114th Senate: Votes and Seats

With Louisiana’s Senate runoff election complete, the makeup of the Senate in the 114th U.S. Congress is now fully determined. Republicans netted nine seats to take a 54-46 majority*, returning to control of the chamber for the first time since their defeat in the 2006 midterms. 

As a body designed to represent states rather than citizens, the Senate’s partisan makeup tends to bear a fairly loose relationship to the raw numbers of votes that were cast to elect its members. With the final election results in hand, let’s take a look at how votes cast for Senate candidates translate to seats in the world’s greatest deliberative body. 

In all, Americans cast 202.5 million votes to elect the current Senate, spread across three election cycles in 2010, 2012, and 2014. Of these, 49% were cast for Democratic candidates and 46.6% for Republicans. Here's how votes cast equated to seats won for the two parties in each of the past four election cycles, with the far-right bars showing this relationship for the entirety of the currently elected Senate (click the image to expand):

Source: The Green Papers, FEC.

In the aggregate, Democratic voters are underrepresented in the Senate and Republican voters are overrepresented compared to their respective strengths in the electorate, although Democrats outperformed their raw vote totals in two of the past four individual elections. 

The 46 Democratic caucus members in the 114th Congress received a total of 67.8 million votes in winning their seats, while the 54 Republican caucus members received 47.1 million votes. This is in part a reflection of differences in turnout across elections and partly a reflection of the slight tendency for small states to elect Republicans and large states to elect Democrats. California Democrats Barbara Boxer and Dianne Feinstein, for example, won over 13 million votes combined to win their seats, while Wyoming Republicans Mike Enzi and John Barrasso combined for just over 300,000 (2.3% of the California delegation’s total). 

Most of the U.S. population is represented by two Senators from the same party. The below chart breaks down the country’s population (2013 Census estimates) by representation in the Senate: 

Thirty-four states with a combined population of 191.8 million are represented by at least one Republican Senator, compared to 30 states and 213.1 million represented by at least one Democratic Senator. 

Stay tuned for more FairVote analysis of the 2014 midterms and their implications for American democracy. 

*Independent Senators Bernie Sanders (VT) and Angus King (ME) caucus with the Democratic Party. Lisa Murkowski (AK) was elected as a write-in candidate in 2010 but caucuses with the Republican Party. All three are classed here as members of the party they caucus with. 

Thursday, December 4, 2014

Canadian Parliament Votes on Mixed-Member Proportional Voting

On December 3, 2014, the New Democratic Party (the Official Opposition Party in the Canadian Parliament) moved:
That, in the opinion of the House [of Commons]: (a) the next federal election should be the last conducted under the current first-past-the-post electoral system which has repeatedly delivered a majority of seats to parties supported by a minority of voters, or under any other winner-take-all electoral system; and (b) a form of mixed-member proportional representation would be the best electoral system for Canada.
An impassioned debate broke out. In support of mixed-member proportional (MMP), members of the Canadian Parliament spoke about the "extremely unfair" winner-take-all system and the "false majorit[ies]" it creates. They endorsed MMP as ensuring "no one who goes to vote will be able to say that his vote will not count." In reflecting on the low voter turnouts that blight Canadian elections, member of parliament (MP) Alexandrine Latendresse said, "[w]hen people realize that their vote can make a difference [under MMP] and that they can influence what their government and Parliament look like, they will vote."

The complete transcript of the debates is available on the Canadian Parliament's website.

The motion was ultimately defeated -- though it garnered 110 votes in favor (and 166 votes against). Voting in favor of the motion were most members of the New Democratic Party, half of all Liberal Party MPs, the Green Party, the 2 members of Bloc Quebecois and three Independent MPs.   

Democracy for America's Creative Use of Ranked Choice Voting

The organization Democracy for America recently conducted an internal, online "pulse poll" for likely 2016 Democratic Party presidential candidates.

Unlike "vote-for-one" polls, which often serve only to show fractured results, DFA let its members rank all candidates in order of preference.

On its results page, it shows the results if you look only at first-choices. However, you can click on any candidate to see how polling might look without that candidate in the race. It accomplishes this by recounting that candidates' votes for their second choices.

In a ranked choice voting election as conducted in a number of elections today, including several California Bay Area cities and the largest cities in Minnesota and Maine, a candidate wins by achieving a majority of the active ballots, with weaker candidates removed round-by-round until the strongest winner can be identified.

Instead, this poll allows you to see the results if any candidates are removed in any order, which cleverly allows you to see what polling might look like if only a certain subset of candidates were in the race.

This poll clearly shows Elizabeth Warren winning, with Bernie Sanders coming in second. An even more interesting use of this method would be in Republican internal polls, like those conducted at CPAC. The Republican field currently is particularly fractured and would benefit greatly from the use of ranked choice voting polls.

Illinois Extends Pro-Suffrage Practices

The public policy organization, Demos, has reported that the General Assembly in Illinois passed Senate Bill 172 (SB 172), expanding pro-suffrage practices such as same day registration, grace period registration for university students, and online registration. Illinois joins thirteen other states and the District of Columbia who have extended pro-voting measures that make the process of voting and registering as accessible as possible. 

FairVote's Promote Our Vote program applauds practices that improve voter turnout by making voting easier and more accessible. Practices like the ones Illinois has enacted can increase turnout on all levels of government and create a more civically engaged community. To continue the momentum, more localities should push for more pro-suffrage policies such as a Right to Vote Amendment in the Constitution, extend voting rights to felons, and allow 16 and 17 year-olds to vote in local elections. To learn more visit

What Was the National Partisan Sentiment in November 2014?

As part of its Monopoly Politics report, FairVote has calculated that, controlling for candidacy effects, the underlying national two-party preference of the nation this November was 52% for the Republican Party and 48% for the Democratic Party.This paints a picture of a closer election that the nation-wide two-party vote tally of 53% (R) to 47 % (D). FairVote's underlying national two-party preference is a better measure of the partisan sentiment of the nation in 2014 because, unlike the two-party vote tally, it controls for external factors unrelated to partisan sentiment like the number of incumbents running and the number of uncontested seats.

Our methodology is explained in more detail on the Monopoly Politics website, but it can be briefly outlined. We begin with the an average election, in which the two-party vote is split 50%-50% between the two parties. Then we calculate the 2014 election's deviation from that.

Firstly, we exclude uncontested races and open seats. These races could unduly skew our calculations about voter sentiment.  Then we compare each district's two-party vote tally in the US House in 2014 against the district's two-party vote tally in the most recent presidential election (2012). The two-party vote for president is an imperfect indicator of a district's underlying partisanship, yet it is the best we have available at the district level. Other measures, such as partisan affiliation from registration records or survey data, are less reliable because (in the former case) they include scores of registered voters who did not vote --more than half in 2014 -- or (in the latter case) they have high margins of error associated with small, district-level, sample sizes.

After we do this calculation, we are left with the raw performance of the candidate relative to the district's partisanship.  For example, in Maryland's 5th District, Steny Hoyer (D) received 64.0% of the two-party vote in his district in 2014. In 2012, President Obama (D) received 65.0% of the two-party presidential vote in MD-5. And so, Hoyer's performance was 1% less than the partisanship of his district.

The median Republican candidate's raw performance was 4.64 percentage points better than Mitt Romney's two-party vote in the district. The median Democratic candidate's raw performance was 0.91 percentage points better than President Obama's two-party vote in the district. The midpoint between these two medians provides us a measure of the national partisan swing - 1.9% toward the Republicans. And so, we end up with an underlying partisan preference of 52% (rounded to the nearest whole number) to the Republicans and 48% to Democrats.

In comparing the Republican incumbency advantage over the Democratic incumbency advantage we get a better  measure of how the American electorate viewed the two parties. This is because our measure of partisan preference controls for the effects of incumbency, the fact that there were more Republican incumbents and the greater number of uncontested races for Democratic incumbents. The raw two-party vote numbers are affected by all those factors, which are unrelated to partisan sentiment.

It is interesting to note that 2014 is almost a mirror image of the 2012 election. In 2012, the underlying national two-party preference of the nation was 52% Democrat to 48% Republican. However, a key difference is that in a 52% year for the Democrats (2012), the party won only 46% of all US House seats.  By contrast, in 2014, in a 52% year to the Republican Party, Republicans picked up 56% of the House seats. This bias against the Democratic Party in the US House is alarming,  and is discussed in more detail here.

Congress Member Highlights the Importance of Participation

U.S. Congressional Representative Matt Cartwright (D-Pa.) published an op-ed in The Hill this week, explaining the importance of political participation in a democracy. He begins by citing several statistics from the 2014 midterm elections. "[J]ust 36.3 percent (approximately 82 million) of Americans participated in this year's election," he writes. "Only in the 1942 midterm election was turnout lower." He argues barriers to voting are the primary problem and calls for increased access for voters.

FairVote's Promote Our Vote focuses on pro-suffrage practices and policies, supporting concrete changes that ensure fair and equitable voting rights. We seek changes that will improve voter turnout, protect voter access, and expand suffrage.

Wednesday, December 3, 2014

Latinos Lack in Power Positions in Boston Area City Government

A report commissioned by the Greater Boston Latino Network found that while the Latino population in the Boston area has boomed in recent years, the faces in city government do not reflect this shift. The report asserts that "This is not just about Latinos. This is about the quality of government...Inclusive government is better government."

Here at FairVote, we encourage local leaders to consider the impact that voting structures have on the ability of people of color to be represented in local government. Boston itself is currently considering ranked choice voting (RCV) for it's four at-large city council seats, as a way to improve its local democracy. Cities can also find local solutions to address the often low and racially unequal voter turnout that leads to unrepresentative city governments. Learn more at

U.S. Senate Results Show Continued Rise in Predictive Power of Partisanship

What Election Day showed about the changing face of American democracy

Republicans took the Senate in a wave this November, but the real story of the Senate this year is one of partisan entrenchment: a continued increase in the rise of party affiliation above all other factors as the most important driver of election outcomes.

Democrats defended Senate seats in seven states carried by Mitt Romney in 2012: Alaska, Arkansas, Louisiana, Montana, North Carolina, South Dakota and West Virginia. It appears likely that Republicans will take all of those seats. (Louisiana Democrat Mary Landrieu will face a tough runoff election in December; Republican candidates won the other six.) Out of all 36 Senate races held this year, only three—in Colorado, Iowa, and Maine—were won by a different party than the party of the winning presidential candidate in that state in 2012. And assuming a Republican victory in Louisiana, the new Senate will contain just 16 “crossover” Senators from states that favored the opposite party in 2012’s presidential election. The lesson? States and voters are aligning themselves by partisanship in increasing lockstep.

The fact that this year’s realignment worked to Republicans’ advantage is in large part because this year’s Senate races were concentrated in conservative states to a historic degree. The Monkey Cage’s Patrick Egan notes that the median state holding a Senate election this year voted for Barack Obama by 7.2 percentage points less than his national vote share in 2012, the largest such gap in over six decades. But the real story here is partisanship, regardless of party, and Republicans may eventually find the shoe on the other foot: eleven of the 16 remaining “crossover” Senators are Republicans in states that voted for Barack Obama in 2012, though not all are deep blue (Colorado, Florida, Illinois, Iowa, Maine, Nevada, New Hampshire, Ohio, Pennsylvania, and Wisconsin). Five Democratic crossover Senators remain, in Indiana, Missouri, Montana, North Dakota, and West Virginia.

It’s significant that many of these are perennial presidential battleground states: their partisan composition is close enough to an even split to allow their votes for Senate—like their votes for President—to go either way based on relatively small shifts in the partisan climate, so crossover Senators are an expected phenomenon. In the deep red and blue states, by contrast, the crossover Senators tend to be held tenuously by either vulnerable newcomers or corner-case incumbents. Heidi Heitkamp (D) will face an uphill battle in North Dakota in 2018, having won a razor-thin upset in a strong Democratic year in 2012. Montana’s Jon Tester (D) also scraped out a narrow win in 2012, bolstered by a Libertarian spoiler candidate; he will likely find it difficult to compete in a state that voted for Romney over Obama by 13 points. Meanwhile, blue Maine’s Susan Collins (R) is notably moderate and has held the seat for almost two decades, while red West Virginia’s Joe Manchin (D) is a uniquely conservative Democrat. Though each election cycle has its quirks, these crossover seats are not likely to be sustainable holds in the long term.

The rise of partisanship-as-predictor reflects the increasing ideological divergence between the two parties and the withering of the last vestiges of the parties’ heterogeneity in the twentieth century, such as white Democrats in the South and moderate Republicans in the northeast. This year is a further indication that partisanship is increasingly the dominant driver of federal election outcomes, dwarfing candidates’ attributes, voting records, or campaign tactics and spending. See this analysis from FairVote’s Monopoly Politics report for a fuller look at the dominance of partisanship in national elections—and why it matters.

Wednesday, November 19, 2014

New Zealand Labor Party Leader Announced

Yesterday (11/18/2014), the New Zealand Labour Party announced that it had chosen Andrew Little as the party's new leader.   Last week, FairVote reported on the four-way race for leader and the uncontroversial use of ranked choice voting (RCV) to count the ballot of the thousands of party members and affiliated union members, as well as the parliamentary caucus.

RCV proved crucial--if entirely uncontroversial--to the outcome. After the first round of counting was complete, Grant Robertson had a plurality of the votes but none of the four candidates had a majority of the vote. And so, under RCV, no candidate was elected in the first round.

Grant Robertson, the NZ Labour leader candidate who received a plurality of the votes (Photo Credit: NZ Labour)
Next, Nanaia Mahuta, who received the fewest votes, was excluded from contention. The second choices of voters who voted for Mahuta were taken into consideration at this point. However, still no candidate had yet achieved majority support.

Nanaia Mahuta, the NZ Labour leader candidate who received fewest votes (Photo Credit: NZ Labour)

Finally, David Parker, who had the least support out of the remaining three candidates, was excluded from contention.  The second choices of voters who voted for Parker were taken into consideration and, of the two remaining candidates (Robertson and Little), Little has the greatest support. Little also had majority support, winning 50.52% of the vote against Robertson's 49.48%.

David Parker, one of the NZ Labour leadership candidates  (Photo Credit: NZ Labour)
Little was able to win the party's leadership only by reaching out to Mahuta and Parker's supporters and asking to be ranked second (or third).  In this way, RCV encouraged consensus building and elected the candidate with the broadest support within the party.

Andrew Little during the leadership campaign (Photo credit: Radio Australia)
Although the margin of victory was ever so tight (if one caucus member had voted for Robertson instead of Little, Robertson would have one) and a winner did not emerge until the third round of counting, no one is doubting the legitimacy and the fairness of the use of RCV to determine the outcome. Not even Robertson's supporters.

Tuesday, November 18, 2014

Michigan's Changes to the Electoral College as Applied on a National Scale

As announced yesterday, Michigan State Rep. Pete Lund announced a plan to change the way that Michigan allocates its Electoral College votes during presidential elections. The plan is distinct from other plans we have analyzed before. FairVote's initial analysis reveals that Michigan's plan, if applied nationwide, would have led to Obama winning a total of 287 electoral votes and Romney winning a total of 251 electoral votes in 2012.

Additional findings include:
  • If applied nationwide, a total of 78 electoral votes could be labeled as swing electoral votes.
  • If applied nationwide, 24 states would remain without any swing electoral votes.
  • Michigan would have awarded 12 electoral votes to Obama and 4 electoral votes to Romney in 2012.
  • Given the partisanship of Michigan, Michigan would have had 4 swing electoral votes in the 2012 elections.

We will release a full spreadsheet of our initial results shortly.

Originally posted November 14, and updated November 18, 2014, using the legislation's explanation of the formula for allocating electoral votes. Our original analysis was based on statewide votes, rather than the statewide vote for the top two candidates.

Friday, November 14, 2014

The Demise of Proportional Representation Proposals in Myanmar.

A few months ago, FairVote covered debates about the adoption of proportional representation (PR) in Myanmar in preparation for the 2015 Elections. The 2015 elections are a significant test of Myanmar's burgeoning democracy as Aung Suu Kyi’s  National League for Democracy is likely to oust the remnants of Myanmar's military junta (the Union Solidarity and Development Party). The PR proposal had the potential to mitigate the losses expected for the Union Solidarity and Development Party and may have kept them a viable political force beyond 2015.

Today (11/14), the Myanmar House of Representatives announced that it would not pursue PR, because it would require a constitutional amendment. 

Certainly, a constitutional amendment would appear to be legally necessary to introduce PR for the House of Representatives (Pyithu Hluttaw in Burmese). The official English translation of Section 109 of Myanmar's 2008 Constitution says that:
Pyithu Hluttaw representatives [shall be] elected ... in accord with law on the basis of township as well as population or combining with an appropriate township which is contagious to the newly-formed township if it is more than 330 townships.
Although a somewhat confusing translation, the section appears to require districts centered around townships. It's less clear that an amendment would be needed to reform the House of Nationalities, called the Amyotha Hluttaw (see section 141 of Myanmar's Constitution).

A constitutional amendment of section 109 would be a mammoth political task. Under the 2008 Constitution, seventy-five percent of all legislators would need to pass a bill endorsing proportional representation, after which a national referendum would be held on the question. The amendment would only be passed in the instance that more than half of all citizens eligible to vote voted "yes" to the change (s436). 

In the past, the legal requirements for changing the Constitution have been ignored where politically expedient.  Perhaps, in the demise of this proposal for PR in Myanmar in an instance when it would have helped the former military junta, we are seeing the rise of a constitutionalism and respect for rule of law in Myanmar -- things that have often been lacking in its post-colonial history. 

Thursday, November 6, 2014

Was 2014 a Republican "Wave" in the House?

Republican candidates for U.S. House substantially outperformed Democrats nationwide on Tuesday on their way to a historic GOP majority: the AP's current count has Republicans carrying 52.5% of the national popular House vote to Democrats' 44.5%. However, these raw popular vote counts are skewed by uncontested races, differences in turnout across districts and ballots being still being counted in Democratic-leaning states like California.

FairVote's Monopoly Politics U.S. House projection model measures nationwide partisan sentiment more accurately by comparing winning candidates' two-party margin of victory to the underlying partisan landscape of their districts, as determined by districts' presidential vote in 2012 compared to the nationwide presidential vote. This metric lets us look at how heavily the country as a whole leaned Democratic or Republican relative to a uniform baseline.

This year, the underlying preference of House election voters was 52.05% for Republicans to 47.95% for Democrats - nearly an exact mirror of the  2012 advantage for Democrats. Take a look at the swings in national two-party preference in the House in each election since 1996 (values above 50% indicate Democratic lean):

Republican's 52.05% advantage this year is well in line with historic norms and represents a meaningful but relatively modest nationwide lean. For context, in the Republican wave year of 2010, the national two-party preference was 53.8% Republican, while the 2008 Democratic wave was borne out of a national two-party preference of for Democrats of 54.0%.

Stay tuned for more on FairVote's Monopoly Politics 2016 analyses and their implications for American elections.

Wins and Losses for People of Color this Midterm Election 2014

The election results rolling in this week largely have focused on Republican control of Congress, but not much has been said about the progress or regression of candidates of color this midterm election. A record number of people of color ran for Congress and statewide elections.


African Americans Candidates
  • The Senate is 2% African American. In early 2013, there were zero African American Senators, and only five African Americans have been elected to the Senate since Reconstruction.
  • Susana Martinez (NM) and Brian Sandoval (NV) were easily re-elected as governors.
  • David Ige (D) was elected governor of Hawaii after beating the white incumbent in the primary and then winning a hotly contested general election. Nikki Haley (R) was re-elected as governor of South Carolina.
  • Of the ten Democratic incumbents that lost this past election cycle, three of them are people of color--Joe Garcia (FL), Steven Horsford (NV), and Pete Gallego (TX).
  • Out all 535 seats in Congress, only 46 members of the new 114th Congress are African American, 32 are Hispanic and Latino, nine are Asian American, and two are Native American. 
  • Appointed to office in 2013, Tim Scott (R-SC) became the first African American to be elected to the Senate in the South since Reconstruction.
  • Cory Booker (D) was re-elected to his U.S. Senate seat on Nov.4 in his state of New Jersey, after winning a special election in 2013.

Latino Candidates
  • According to the National Association of Latino Elected and Appointed Officials (NALEO),the Latino roster in Congress will be the largest yet, with an increase of five Latinos to a total of 29. Ruben Gallego (D) AZ-7, Pete Aguilar (D) CA-31, Norma Torres (D) CA-35, Carlos Curbelo (R) FL-26, and Alex  Mooney (R)-WVA-2 add to the Latino voice in Congress. Mooney is the first Latino to serve in the House from West Virginia.
  • 12 Latinos (possibly 13) will serve in statewide executive office, an increase of two Latinos. Alex Padilla (D) was elected to be the first Latino elected to be California’s Secretary of State, and George P. Bush will serve as the first Latino Land Commissioner in Texas.

Asian American Candidates
  • Stephanie Chang (D) because the first Asian American women to serve in Michigan legislature.
  • Rady Mom becomes first Cambodian American (D-MA) state legislator in the country.

Women of Color:
  • There will be a record total of 32 women of color in the House (29D, 3R), including 18 African American women (17D, 1R), 9 Latinas (7D, 2R), and 5 Asian/Pacific Islander Americans (5D).
  • The new House members include 5 women of color (four African Americans and one Latina): Alma Adams (D-NC), Brenda Lawrence (D-MI), Mia Love (R-UT), Norma Torres (D-CA), and Bonnie Watson Coleman (D-NJ). A new delegate, Stacey Plaskett (D-VI), is African American.
  • Utah’s Mia Love (R) was elected to the House, becoming the first female African American Republican in Congress. 
  • Bonnie Watson Coleman (D) beat out Dr. Alieta Eck to be elected as New Jersey’s first African American woman in Congress.
  • Evelyn Sanguinetti (R-IL) will be the first Latina lieutenant governor in any state and the only new woman of color in a state’s number two post.
  • Nellie Gorbea, Rhode Island’s new secretary of state, is the first Latina elected statewide in that state and the first Latina elected to a statewide executive post in New England.


While we must congratulate all of the candidates of color for their win across the nation, it is essential to remember that on all levels of government, we must look to the losses to create a real depiction of the political playing field for candidates of color. Here are a few unsettling facts:

  • Latina Republican Marilinda Garcia of New Hampshire lost her chance at Congress during the midterms, with many claiming she was too young and did not have enough experience to lead at 31 years old. While it is difficult to make a comparison, it is notable that white Republican of New York, Elise Stafanik, 30, was able to take the position as the youngest woman of Congress in history this election.
  • Anthony Brown (D), could have been the third African American to be elected as governor since Reconstruction but lost against Larry Hogan (R), despite the advantage Brown had in residing in a heavily Democratic state. Analysis of the defeat has largely ignored the proverbial elephant in the room: Brown did much more poorly than white Democrats running statewide in areas of the states with large white majorities. Instead of noting the loss of a possible history-making election for African American representation, the post-election coverage has focused on Brown for not having a strong enough platform  and that, as the Washington Post reports, “polls suggested that nearly a third of all voters did not have a clear impression of Brown.” But in various media outlets, it was widely reported that the majority of the races on the 2014 midterm elections rested on weak platforms or riding on attacks for the other opponent. So why was Brown given more resistance than his white counterparts?
  • A record number of 83 African American politicians, a majority of them Democrats, made the congressional ballot this year, an increase of 11 from 2012. But while it may be tempting to celebrate this number, the catch is that 32 of them ran in the South, an area where Democrats most likely lose. David Bositis, a senior research associate at the Joint Center for Political and Economic Studies, noted

I wish I could write with confidence that these increases in black major party nominees was a positive development, but the fact is that many of the increases are occurring in states (especially in the South) where most whites are withdrawing from Democratic party politics — leaving black candidates the nominations by default.

  • With the changes that were made to Congress, stagnation is also crucial to spotlight.
    1) Next year there will be only three governors who are men of color -- Asian Americans Bobby Jindal (R) of Louisiana and David Ige (D) of Hawaii and Latino Brian Sandoval (R) of Nevada -- and two governors who are woman of color, Susana Martinez (R) of New Mexico and Nikki Haley (R) of South Carolina. There will be no African American governors.

    2) The Senate still has the most African Americans it has ever had-- a whopping two of them -- and there are still only three Latinos in the Senate, with no Latinos making it on the ballot in any Senate races this election.

This begs the question: Are we more critical, whether it is consciously or subconsciously, of candidates and elected officials when they are from a minority group? Do people of color face more barriers when running and once elected than white candidates in the U.S.? FairVote will be delving into the election results of candidates of color, the coverage of those results, and electoral reforms that might change opportunities for people of color in the year ahead.

Incumbents' Advantage in U.S. House Elections Is in Free-fall

2014's midterms saw a continuation of a trend in declining advantages to incumbents in the U.S. House, a likely indication of the increasing dominance of party affiliation over candidate-specific traits as the key driver of votes in federal elections.

Incumbents enjoy inherent electoral advantages over challengers, such as greater campaign spending ability, more press coverage, more experienced campaign operations, and ongoing delivery of constituent services for their district. FairVote's Monopoly Politics U.S. House projection model quantifies the advantages of incumbency by comparing the electoral performance of incumbents to their districts' underlying partisan lean. This "incumbency bump" roughly measures the degree to which incumbents would outperform hypothetical generic candidates of the same party running in their district. By this metric, the advantages of incumbency are in rapid decline. In the 2000 election, the nationwide incumbency bump was almost 8%; by last Tuesday, however, the incumbency bump had plummeted to just 2.55%. Take a look at the dramatic fall in the electoral benefits to incumbency in the House over the past decade:

Democrats bore the brunt of Tuesday's relatively weak pro-incumbency sentiment: while Republican incumbents enjoyed a boost of 4.64%, the bump for Democrats was just 0.46%. Here's a look at the split between the parties' incumbency bumps since 1996: 

This decline in incumbency's advantages likely reflects an increasingly efficient sorting of House representation by party: with districts increasingly likely to elect a representative of the same party as the overall partisan affiliation of their voters, cross-party incumbents are becoming increasingly unable to compete.

Stay tuned for more on FairVote's Monopoly Politics 2016 analyses and their implications for American elections.

The Need for Better Recount Laws in Virginia

The Virginia Senate race between incumbent Mark Warner (D) and Ed Gillespie (R) could become Virginia’s third statewide recount in the past ten years. However, if a recount occurs, it will have the same outcome as the previous two – it will not change who wins. According to FairVote’s research of all statewide recounts occurring from 2000-2009 and updated review of recounts from 2010-2012, full statewide recounts are rarely necessary, as the average change in victory margin is less than 0.03%. Based on the original tally in the Virginia Senate race, the margin between the two candidates is about 0.79%.

Despite this large margin, current Virginia recount laws does allow a candidate to apply for a recount if the difference between the apparent winning candidate (Warner) and the apparent defeated candidate (Gillespie) falls within one percent or less of the total vote cast for those two candidates. In this case, the margin of difference in votes between those two candidates is less than one percent. If Gillespie does request a recount, he will be required to pay $10 per precinct. If the recount alters the election outcome in his favor, then he is would be refunded the money. There are 2,557 precincts in Virginia – which would require Gillespie to pay $25,570 to recount all precincts in the state. 

There are three key problems with Virginia’s recount laws that can be changed. 

First, FairVote recommends that Virginia fund risk-limiting audits that can catch potential fraud or technological error in any race before the outcome is final. The number of ballots to review should increase as the victory margin decreases. 

Second, states should create a sensible minimum threshold for triggering an automatic, state-financed recount in statewide elections. The victory margin should be no larger than the greater of 0.15% or 2,000 votes. Our research shows that any larger margin is highly unlikely to change outcomes. Candidates could still request a recount in elections with larger margins, but be willing to pay for it unless the recounts changes the outcome. The end result is that recounts that should happen will be paid for by the taxpayers.

Third, candidates and voters should be allowed to petition and pay to accelerate the initial vote counting in order to allow time for a recount prior to any deadlines. The amount would be based on a state or counties public estimates of acceleration costs.

The need for better recount laws in Virginia has come to light again in this election. Without the recommended changes, recounts will most often be an expensive and lengthy process unlikely to result in a new outcome.

Wednesday, November 5, 2014

National Trends in 2014 Highlight Accuracy of FairVote's Monopoly Politics Model

FairVote’s Monopoly Politics projection model is designed to predict the results of as many U.S. House elections as possible using the most simple of inputs. Using only data about the incumbent’s past margins of victory and the district’s vote in the most recent presidential election, the 2014 model predicted 368 House races in 2014 with 99.4% accuracy. New data from Tuesday’s election, however, allows us to evaluate the model more completely. When we incorporate two basic pieces of new information into the model’s default assumptions, it correctly projects the makeup of the entire House in 2014.

The Monopoly Politics model’s outputs are influenced by two kinds of national sentiment, both of which vary in magnitude considerably across elections: the electorate’s preference for one party over another and the electorate’s preference for incumbents over non-incumbents. The model accounts for these trends with two numerical inputs. First, the national two-party preference, which describes the electorate’s overall partisan lean and is expressed as the percentage of major-party voters at the polls who, all else being equal, preferred Democrats over Republicans; a national partisan swing of 46%, for example, would indicate a strong Republican wave year. Second, the national incumbency bump, which measures the degree to which incumbents benefited at the polls relative to non-incumbents, all else being equal; it’s expressed as a percentage value indicating how much better incumbents performed compared to a hypothetical generic candidate of the same party in an open-seat race in the same district. A national incumbency bump of 3% would indicate that incumbents tended to receive an extra boost of about 3 percentage points solely as a result of their incumbency.

However, these numbers aren’t set in stone. They vary unpredictably between elections, and their values for a given year can only be determined after the fact by examining actual election results. Thus a key feature of the Monopoly Politics model is that these values can be tweaked by users to simulate different national trends, such as a strong wave year for one party or a strong anti-incumbent sentiment. The 2014 model is archived here; the values in the top-left hand corner can be edited by viewers to simulate these effects.

The model’s default projections for 2014 assumed that voters as a whole would prefer Democrats and Republicans to the same degree, i.e. a national two-party preference of 50%. As well, they assumed that the incumbency bump would remain at its 2012 value of 4.5%. The model predicted 368 of 2014’s races with 99.4% accuracy using these assumptions. But now that the 2014 election is over, we can plug the actual 2014 national two-party preference and incumbency bump values back into the 2014 model and see how it performs. Let’s take a look at the model’s projections using the actual 2014 national two-party preference of 48.11% (see methodology notes below) and the actual national incumbency bump of 2.36%:

Strikingly, the model predicts the final tallies with complete accuracy – even including those races which it categorized as too close to make an official projection.

These results speak to the strength of the model, but also to the impact of 2014’s strong climate for Republicans and relatively weak climate for incumbents. In a nationally even partisan year in which incumbents were favored at the same rate as in 2012, the model predicts Republicans would have won 13 fewer seats, adding just two to their existing majority. Instead, with 51.89% of major-party voters preferring Republicans to Democrats, 2014 was a substantial if not overwhelming Republican wave year in the House. Meanwhile, 2014’s 2.36% incumbency bump was unusually low by recent standards: the national incumbency bump from 1998 to 2010 averaged 6.34%. However, peering below the aggregate number, it was Democratic incumbents who bore the brunt of low pro-incumbent support: while the median Republican incumbent enjoyed a reasonably healthy boost of 4.26%, the median Democratic incumbent’s advantage was just 0.46%.

A more detailed explanation of our methodology for calculating national two-party preference and national incumbency bump follows. A Google Sheets document containing the below calculations and relevant data is available here.

National Two-Party Preference. To calculate the national two-party preference, we first calculate each district’s partisanship value (expressed as a percentage of voters favoring Democrats), which measures the underlying partisan preference of the district’s voters independent of national trends. Partisanship is calculated by taking Barack Obama’s two-party vote share in that district in 2012 and subtracting half of his national margin of victory. Next, we take the two-party vote percentage for each House winner and subtract his or her district’s Democratic partisanship percentage to calculate the degree to which the vote in that district deviated from the district’s baseline partisanship. We then take the median result of these comparisons for each of three categories: open seats, seats held by an incumbent Republican, and seats held by an incumbent Democrat. We take the mean of these three medians to arrive at the national partisan swing, a percentage value indicating the deviance from an even 50-50 nationwide partisan split; negative values indicate a Republican lean and positive values indicate a Democratic lean. In 2014, the national partisan swing was -1.89%. We add this number to 50% to calculate the 2014 national two-party preference of 48.11%, which indicates that 48.11% of major-party voters preferred Democrats over Republicans.

Incumbency Bump. We use a similar process to calculate the national incumbency bump. We compare each incumbent’s performance to the partisanship of their district to calculate the degree to which incumbents over- or under-performed their district’s baseline partisan preference. We take separate medians of these values for Democratic incumbents and Republican incumbents. We then take the mean of these two medians to arrive at the 2014 national incumbency bump of 2.36%, which indicates incumbents could expect to perform 2.36 percentage points better than a generic candidate of the same party would have performed in an open-seat race. 

Monopoly Politics Projections for 2014 US House Election 99%+ Accurate

Republicans had a strong showing in congressional elections on Tuesday, backed by a clear majority of voters at the polls. Over a year before this November’s election—and the campaigning that led up to it—FairVote made its 2014 US House projections as part the biennial Monopoly Politics report. FairVote projected the outcome of 368 of the 435 House races:  205 to Republican Party candidates and 163 to Democratic Party candidates (see here for a more detailed analysis).

While provisional and mail ballots are still being counted, FairVote can confidently say that we were  99.7% accurate in our projections of 2014 US House races. This accuracy was achieved despite the fact that we made our projections in mid-2013 based only on the 2010 and 2012 election results in each district and the presence of an incumbent.

Using the same methodology, FairVote will project outcomes for the November 2016 election on November 6, 2014, just two days after this year’s election

The single error in our projections was in New York’s 24th district. FairVote projected the Democratic Party candidate, Dan Maffei, would win with over 58% of the vote.  Few figured NY-24 was in play until the last week of the election campaign.  The Republican Party actively targeted NY-24, especially Syracuse, as part of its strategy to pick up extra seats in New York. As the results rolled in on Tuesday night, it quickly became apparent the Maffei had been defeated by Republican challenger, John Katko. Indeed, Katko picked up a remarakable 60% of the vote. NY-24 has now changed hands in each and every of the last four elections.

Still, this unanticipated House result is an outlier. We already have a good sense of who will win in 2016. Stay tuned for FairVote’s 2016 US House projections, which will be available tomorrow on our Monopoly Politics page.

Women Win Large Majority of Ranked Choice Voting Races in Bay Area, Including Two Mayoral Races

Women have a history of doing well in ranked choice voting (RCV) elections. The incentives it creates to find areas of common ground with backers of opponents may be easier for many women candidates. The less polarized, negative style of campaigning may invite more women to run.

We'll need to see what happens with men and women candidates over time, but certainly women did well in the four ranked choice voting elections held yesterday in cities in the Bay Area. Consider:

Oakland: Women may sweep all eight RCV elections. Libby Schaaf won the mayor’s race, defeating ncumbent Jean Quan, and women were the top three finishers in first choices. Women won the citywide auditor race, at least two of three Oakland city council seats being elected (including newcomer Annie Campbell Washington, while Dana King narrowly trails in another open seat race), and at least two of three school board  seats (with Nina Senn ahead in the one still-undecided  race).

San Leandro: Pauline Cutter won the open seat race for mayor, defeating another woman in the final instant runoff. Women also won two of three city council seats, all of which were open seats.

San Francisco: All incumbents won easily in the five RCV elections for the Board of Supervisors and  two citywide offices , including women in three of five Board of Supervisors seats and one of two citywide offices.

Berkeley: Berkeley was used RCV to elect four city council seats and a citywide race. One female and two male incumbents retained their seats, while the open seat race is too close to call in the final instant runoff between a man and woman. A female incumbent was re-elected as city auditor.

Overall, that's 24 seats elected by RCV yesterday. At least 15 women are winners, with a chance to win three more, including women who are African American, Latina, Asian American and white. You can see an overview of the races at the East Bay Express.

RCV to the rescue: non-majority winners in gubernatorial races

In Tuesday’s midterm elections, 36 states held gubernatorial elections. Of those races for governor, at least eight resulted in non-majority winners--meaning that no candidate received support from more than 50% of voters. This scenario played out in Alaska, Colorado, Florida, Maine, Massachusetts, Oregon, Rhode Island, and Vermont. That means that in almost 1 in 4 states that elected a governor yesterday, a majority of voters will be governed for the next four years by a candidate they did not vote for.

Rhode Island’s Gina Raimondo (D) won the governorship with a startlingly low 40.2% of the vote, while third-place finisher and Moderate candidate Bob Healy garnered 22% of support; her predecessor won the 2010 race for governor with an even lower share of the vote. Maine re-elected incumbent Paul LePage with only 48.3% of the vote, while challengers Mike Michaud (D) and Eliot Cutler (I) split the remainder of the vote with 43.3% and 8.3% respectively. Of Maine's last 11 elections for governor, nine were won with less than 50%.

Florida, one of the most hotly contested gubernatorial races of the year, saw incumbent Rick Scott (R) win narrowly (1.2 percentage points) with 48.2% of the vote. Had that race tightened ever so slightly, the 3.8% of votes garnered by Libertarian candidate Adrian Wyllie would have loomed even larger.

Speaking of Libertarians, a key U.S. Senate race deserves honorary inclusion in this review: Virginia's incumbent Senator Mark Warner (D) has survived a challenge from Republican Ed Gillespie, with Libertarian nominee Robert Sarvis garnering far more votes than the margin, just as he did in the 2013 election for governor in Virginia.

In each of these states, minor party and independent candidates garnered support from voters in already relatively closely contested elections, resulting in a non-majority winner for governor. In each case, ranked choice voting (RCV) would have avoided such a result, and allowed for a consensus winner. Mainers, who have now experienced this phenomenon in back-to-back gubernatorial elections, collected signatures on Election Day to move forward with implementing RCV in all state and federal races. Other states would do well to follow suit, and give themselves an opportunity to vote their conscience on Election Day, without fear of electing leaders that do not have a majority of the electorate’s support.

Democracy on the ballot in election 2014

Huffington Post has a useful site to track statewide ballots. Voters sent mixed messages. Notable statewide ballot measure results include:

  • Election Day registration maintained in Montana, rejecting repeal placed on ballot by legislature
  • Top Two primary loses two-to-one and in every county in Oregon despite more than five million dollars spent in its advocacy
  • Redistricting commission wins in New York despite divisions in the reform community and opposition from most editorial board
  • Connecticut and Missouri won't allow early voting
  • Voters want retirement ages for judges. (Take note, Supreme Court!)

Below is a full report.


[Put on ballot in party-line vote in legislature.

Referendum LR-126 - Change Late Voter Registration Deadline  Closes late voter registration the Friday before election day.
100% reporting

Against 202,150 56.8%
For 153,563 43.2%


Proposal 1 - Redistricting
Creates a redistricting commission to draw district lines every ten years.
99% reporting

Yes 1,576,395 57.3%
No 1,172,717 42.7%


Measure 90 - Open Primaries
Creates an open, top-two primary election system.
84% reporting

No 810,868 68.0%
Yes 382,445 32.0%


Amendment 1 - Absentee Ballots
Allows the legislature to expand early voting.
83% reporting

No 410,142 52.8%
Yes 367,238 47.2%

Amendment 6 - Allow Early Vote
Establishes a six-day-long early voting period.
99% reporting

No 983,504 70.3%
Yes 414,911 29.7%


Question 3 - Constitutional Convention
Dictates whether or not Rhode Island holds a constitutional convention.
99% reporting

Reject 157,015 55.3%
Approve 126,721 44.7%


Limited measures, but Utah removes requirement for Lt. Governor appointee to have special election, while Maryland voters allow counties to have special elections for county executives. As reported below, Florida voters retain them for judicial vacancies. 

Utah - Constitutional Amendment B - Lieutenant Governor 
Removes the requirement that an appointed lieutenant governor stand for election in the next regular general election following his or her appointment.
99% reporting

For 256,923 55.6%
Against 204,824 44.4%

Maryland -  Amendment - County Special Elections
Permits a county charter to provide for filling vacancies in county executive offices through special elections.
99% reporting

For 1,176,720 80.1%
Against 291,920 19.9%


Florida rejects appointments for judicial vacancies

Amendment 3 - Judicial Vacancies

Allows the governor to fill judicial vacancies by appointing a justice or judge from a slate of nominees.
100% reporting

No 2,792,619 52.1%
Yes 2,567,630 47.9%

Hawaii overwhelmingly opposes raising mandatory retirement age for judges

Amendment State Justices Retirement
Increases the mandatory age of retirement for judges and justices to 80.
100% reporting

No 247,765 76.7%
Yes 75,205 23.3%

Louisiana rejects raising mandatory retirement age for judges

Proposition 5 - Repeal Retirement Age
Removes the retirement age requirement from judicial offices.
100% reporting

No 782,393 58.2%
Yes 562,790 41.8%

Tennessee moves away from judicial elections

Amendment 2 - Judicial Selection
Empowers the governor to appoint judges subject to confirmation by the general assembly.
99% reporting

Yes 830,418 60.9%
No 532,779 39.1%