The 2020 elections are still 18 months away and yet pollsters are out in force, giving us just enough information to break out our crystal balls and make wildly irresponsible predictions. This is the first post in a three-part series looking at the generic ballot and it’s utility as an election predictor
Democratic presidential primary polls have been dominating election headlines, but some congressional polls have been been released too, albeit to much less fanfare. We should probably ignore these polls — it’s too early for them to be predictive. Prognosticating off these preliminary numbers is rash, reckless and generally unwise. And so that’s exactly what we will do here in a three-piece series. This post will focus on how predictive the “generic ballot” is on, or close to, election day. The next post will look at how valuable these numbers are now, given that the election is still a year and a half away. The last will look at the relationship between popular vote and seat apportionment.
In “generic ballot” polls, respondents are asked to choose between a nameless Republican and Democrat for Congress. Gallup asks it this way: “If the elections for Congress were being held today, which party’s candidate would you vote for in your congressional district — the Democratic Party’s candidate or the Republican Party’s candidate?” It measures national support for the two parties without some of the baggage carried by their polarizing national figures (think Donald Trump, Bernie Sanders, Mitch McDonnell, Nancy Pelosi, Alexandria Ocasio-Cortez, etc.). It can also bring forward nuanced voter trends. For example, some voters might support a Democrat for President but Republican for House because they want the grinding impotency of split government. Another might vote the same way but because they support Trump but want continued House oversight. Another might choose to reward the current power dynamic because the economy is strong.
Final generic ballot polls — the ones taken within a few days of the actual election — are a better indicator of election outcomes in midterms years than in presidential years. Since 1948, the final generic ballot has missed the real midterm vote by an average of only 2%. The fallibility of the generic ballot in presidential years, though, is clear in the chart below.1Galup data from 1988 was not available.2Data is based of ‘likely voters’ for 1976 and 1996-2006 and ‘registered voters’ otherwise3The two-party generic ballot was used when available (2004-2016)4Data from Real Clear Politics was rounded to nearest integer to make it consistent with Gallup’s data
Since 1968, the generic ballot has missed the real House popular vote by an average of 4% and until 2008, it consistently overestimated Democratic support. Both of these problems have been ameliorated in recent years, resulting in a more accurate and balanced generic ballot since the late 1990s. And even though the generic ballot is less accurate in presidential years, it is still correlated with election results.7Data is based on the chart above. However, data from Real Clear Politics (2004-2016) is rounded to two decimal places rather than the nearest integer as in the chart
A party that performs better on the generic ballot will generally earn more votes in the election. The R2 in the bottom left corner indicates how much of the variance in the dependent variable (Democratic Margin in the Two-Party Vote) can be explained by variance in the independent variable (Democratic Margin in Generic Ballot). In English: R2 shows how well Variable 1 can explain or predict Variable 2. Here, R2 equals almost exactly 0.5, meaning that about half of the variance in the popular vote can be explained by the generic ballot.
Looking at the equation above the R2 value, the trend line has a slope of 0.54. This means that in general a 1% increase in a party’s margin in the generic ballot translates to about a 0.54 % bump in the House popular vote. There is still plenty of variation, though (see how the data points are not clustered too closely along the trend line), so this is not at all a perfect measure for any individual election. Even with this variation, and the knowledge that a lead in the generic ballot usually overestimates electoral success, a party is better off leading the generic ballot than trailing. And their better off leading with a larger margin than a smaller one.
Two caveats to this data. First, the makeup of the House of Representatives is not determined by the popular vote. Seat apportionment, gerrymandering, demographic & partisan sorting and other structural imbalances mean that a party’s share of the national House vote can differ substantially from the share of seats they actually win. Second, this data is from polls within a few days of the elections. We’re still a year and a half away from November 2020. The next two posts will focus on understanding these qualifications to the generic ballot.
If you don’t care about fundraising details, here’s a summary up top: The races with lots of money on both the Republican and Democratic sides are going to be closely contested and combative. So will the primaries that have multiple high fundraising candidates within one party. GA-07, GA-06, NM-02 and NY-11 are shaping up to be exciting general elections; GA-07 and GA-06 will also feature interesting primaries.
As Democrats vow to reduce the influence of money in politics, it’s notable that the top five Quarter 1 fundraisers for the 2020 House “Toss Ups”8As rated by Sabato’s Crystal Ball and The Cook Political Report are Democrats. The top slot goes to Antonio Delgado from NY-19, the newly minted Representative who won in 2018 with the help of an $8 million war chest. Both Sabato’s Crystal Ball and the Cook Political Report rate the district a Toss Up, but Delgado’s impressive cash flow and his opponent’s $2,300 fundraising totalare a good sign for the incumbent.
The other battleground incumbents brought in, as expected, plenty of money as well. The exceptions are Jared Golden from ME-02 and Kenny Marchant from TX-24. While they both vastly outraised any competition — Golden, in fact, has no competitor and Marchant’s strongest fundraising opponent9Who I interviewed! pulled only $19,000 — their numbers should raise alarm bells. As a previous post detailed, fundraising and advertising is rarely definitive in congressional races. Kenny Marchant has a massive $1.7 million stashed, so the tangible impact of his fundraising is even less consequential. Instead, the numbers matter because they can indicate voter enthusiasm.
Some non-incumbent challengers also had impressive Q1 hauls. Carolyn Bordeaux in GA-07 raised $372,000. Bordeaux was the 2018 Democratic nominee who lost to Republican incumbent Rob Woodall in the closest House election in the nation. Woodall announced his retirement earlier this year, drawing further attention to the seat on both sides of the aisle. Perhaps even more impressive on the Democratic side is first time candidate, Nabilah Islam, raising $102,000. While Bordeaux’s numbers suggest an impressive donor list from last cycle, Islam’s fundraising indicate her political aptitude as a newcomer and an appetite for a younger, diverse, more progressive candidate.
Next door, in GA-06, money is flowing both to Democratic incumbent Lucy McBath and her highest profile challengers, Karen Handel and Brandon Beach. National attention focused on this district back in 2017 for what became the most expensive House race of all time, with spending on the race totaling $55 million. And that well is not dry. McBath raised $482,000, the 6th highest total among candidates in Toss Up district. Handel, who won the 2017 special election but lost in 2018 to McBath, collected $260,000 and State Senator Brandon Beach totaled $124,000.
The last two notable races are NY-11 and NM-02. In the former, incumbent Max Rose raised 603,000 and his challenger, Republican State Assemblymember Nicole Malliotakis, raised $301,000. In the latter, incumbent Xochitl Torres Small raised $453,000 and Yvette Herrell, the 2018 Republican nominee seeking a rematch, raised $211,000.
These races — GA-07, GA-06, NY-11 and NM-02 — where both incumbent and challenger(s) raised impressive amounts of money will get the most national media attention, featuring highly engaged voters and inter-party contention. Before the general though, candidates need to clear the primaries. The races with multiple candidates from the same party with impressive fundraising — so far GA-07 and GA-06 — are the primaries to watch. They’ll likely expose intra-party division and clashes between different wings of the parties.
Congressional candidates have started to shake their collection tins. Most candidates see fundraising as a top priority, as an unfortunate necessity so they can pay staff, run polls, travel and, most of all, run advertisements. The media fixates similarly on money — analyzing fundraising hauls, ranking candidates and pontificating on what it all means.
It’s all overblown. When you ask, “Is anybody actually persuaded by these political ads?” The answer is: not really. There’s plenty of research showing that money has little effect on the outcome of congressional races. Voters hold their political affiliation tightly and their grip is unlikely to be broken by a congressional campaign, no matter how much money it has raised. Split ticket voting is increasingly rare — meaning that the national environment and upper-ballot races are far more important than any congressional candidate’s political messaging.
While political ads might not sway the outcome of a general election, they are useful:
When a candidate is unknown. Political ads are good at telling voters that a candidate exists. They can build name recognition and give voters a first impression. Political ads are bad at persuading voters to change their mind about a candidate they already know. These dynamics make fundraising more important for lesser known candidates . This means that advertisements are more important for challengers than for better known incumbents. Once candidates have raised enough money to build name recognition, their efforts have diminishing returns. Similarly, in races heavily covered by the media where candidates are well known, political ads don’t have much effect.
In primaries. Unlike in the general election, primary voters are choosing between candidates only from their team10Unless it’s an open primary, in which case there will be some non-party members.. They don’t have their minds made up along partisan lines and are more willing to swing between candidates. Additionally, primary candidates are usually less well known than general election candidates and, as discussed in #1, ads are more effective for political unknowns.
As an election indicator. While fundraising may not be that important in actually persuading voters or determining the election outcome, it is a sign of grassroots support. It also indicates who donors think will win. Donors want to tie themselves to the winning horse, so the strongest candidates are also more likely to tally big fundraising numbers.
In building a media narrative. The media loves to cover political advertisements — just look at the coverage of the 2018 Georgia Gubernatorial election. Brian Kemp’s ads got nationaland local press — likely reaching more voters than the advertisements themselves. Candidates can put out ads in order to establish a media narrative — and it works. Mark McKinnon, a media advisor to several Republican presidential candidates, said in an interview with NPR “sometimes, we just put out an ad, and it’ll only be up for a day. And we knew that it wouldn’t get seen by voters, but it would get coverage by reporters.”
So, what does this mean for first quarter fundraising totals (all of which became available April 15)? The big picture: look for totals outside the normal range as these could be signals that an incumbent is vulnerable or that a challenger is particularly formidable. For battleground incumbents, a haul of about $250,000 to $500,000 looks normal. For challengers (non-incumbents), than number is about $50,000 to $100,000. For most congressional races, though, their hours and hours of ‘phone time’ are trivial. Come the general, candidates from both parties will be flush with cash and have no problem establishing name recognition.
Later this week we’ll dig into these Q1 fundraising numbers for the 2020 battleground
districts11Those rated “Toss Ups” by either Sabato’s Crystal Ball or The Cook Political Report and what it means for the candidates and races there. Sneak peak: The biggest fundraiser in any of these races was Antonio Delgado from NY-19 with a whopping $755,000.
RECAP FROM THE LAST TWO WEEKS Twenty-six percent of Americans consider themselves Republican, 31% percent Democratic, and 38% Independent, according to a recent Pew study. Because independents who lean towards one party vote for that party at almost the same rate as party members, they will be considered party members for this analysis. When these independent leaners are included in party totals, 39% of the public is Republican, 48% is Democratic and only 7% is Independent.
The charts below show who Independents, Democrats and Republicans are. This is different than showing how demographic groups align. For example, the first chart shows that 45% of Independents are women, but it does not show what percentage of women are Independents.
THE INDEPENDENT VOTING BLOC Gender
Race AgeEducation Data: Pew, More Pew
*Used 2016 presidential vote as a stand in for party
Independents differ demographically from both Democrats and Republicans. The share of men among Independents, 55%, is higher than the general voting population. It’s even higher than that of the Republican Party, typically considered a male-dominated voting bloc. Independents are less white relative to the overall voting population (50% vs 70%) and are much more Hispanic (23% to 9%). Independents are also younger than partisans — 59% are under 50 years old while the same is true for 52% of Democrats and 42% of Republicans. Independents are also much less educated than both Democrats and Republicans. 46% of Independents have only ‘high school or less’ education, while 32% of Democrats and 34% of Republicans say the same. In some ways (age, racial diversity) the Independent voting bloc looks more like the Democratic Party) while in others (gender, education levels) it aligns more closely with the Republican Party.
Independents fall in-between Democrats and Republican on most political issues. These numbers do not measure intensity of individual voters’ feelings but how the voting bloc is split. Independents sit in-between Democrats and Republicans for five of the six policies in the chart. The exception is marijuana legalization, which Independents support in greater number than both Republicans and even Democrats. In addition to marijuana legalization, Independents are closer to Democrats than Republicans on support for same sex marriage, immigration, and a belief that the economy is generally unfair. The one issue on which Independents more closely align with Republicans is wanting a smaller government. This ideological combination — socially liberal and wanting a smaller government — signals a libertarian streak in the small portion of the country that considers themselves true non-partisan independents
Independents are Less Engaged Independent voters are less engaged than party members. ‘No lean’-ers are almost half as likely to vote in 2018 as Republican party members — 33% compared to 61%. Relative to Republicans, Democrats vote at a slightly lower rate, 59%, but still vastly outperform Independents.
Their disengagement is as a result of their independence and a cause of it. A lack of interest in politics causes independents to identify as such, simply because they are disengaged from political life. Other Independents do not alight ideologically with Democrats or Republicans, and, without a team to support, are less likely to engage in political activity. The directionality runs both ways.
Independents Don’t Like the Parties The partisan antipathy dominating contemporary politics is not isolated to Democrats’ and Republicans’ mutual disdain for each other. Independents partake as well. But while most voters view their own party favorably and the other unfavorably, Independents are more even handed. 37% of Independents have an unfavorable view of both parties, 22% have a favorable view of both parties, and about 20% view one of the two parties favorably.
Independents who do lean towards a party say that “other party’s policies [are] bad for the country” is the number one reason for their partisan lean. These partisan leaners even view members of their own party unfavorably more than half of the time.
Independents Aren’t as Important as We Think Independents are seen as the crucial tie-breakers in a nation evenly divided by two political parties — if independents swing towards one party, so goes the nation. This is not true. Independents (the only data available here is for leaners and non-leaners alike, so that what is used here) voted for Trump in 2016, but the popular vote went to Clinton. A similar split happened in 2000, 2004 and 2012. Elections can turn on how well each party turns out its base — a task that is often more challenging for the larger, but less engaged Democratic Party. The parties and their candidates cannot expect to win by only targeting the 7% of truly independent voters if it means sacrificing a portion of their base.
When choosing between energizing the party base versus swinging moderate/independent voters, the question parties should ask themselves is ‘why not both?’ In earlierposts, I suggested that the parties might do well by highlighting issues that 1) are seen as priorities by the general public and 2) the public supports their policies over the opposing party. But maybe they instead should focus on issues that 1) excite their base and 2) are supported by independents. If they go with route #2, Democrats should emphasize health care, the ‘unfair’ economy and socially liberal issues like marijuana legalization and same-sex marriage. Republicans have less to work with, but could highlight limited government and the economy (especially if it holds strong through 2020).
RECAP FROM LAST WEEK Twenty-six percent of Americans consider themselves Republican, 31 percent Democratic, and 38 percent Independent, according to a recent Pew study. Because independents who lean towards one party vote for that party at almost the same rate as party members, they will be considered party members for this analysis. When these independent leaners are included in party totals, 39% of the public is Republican, 48% is Democratic and only 7% is independent.
The charts below show the makeup of the Democratic and Republican parties. This is different than showing how demographic groups align. For example, the first chart shows that 61% of Democrats are women, but it does not show what percentage of women are Democrats.12I don’t know how, but I feel like the ‘all squares are rectangles but not all rectangles are squares’ logic applies. The ‘Democrat minus Republican’ rows show the difference in how much each bloc is composed of a given demographic.
Race & Education
Urban, Suburban and Rural
Data: Pew, More Pew
*Used 2016 presidential vote as a stand in for party
** Used Data from CNN Exit Polls
THE DEMOCRATIC PARTY The stereotypes about the Democratic party are true: the party has more women, minorities, college educated voters and religiously unaffiliated voters than the country overall. And when compared to the Republican party these gaps are even more pronounced. But the size of some of these coalitions — like minorities and college educated voters — are often overstated. The media rarely makes the distinction between a demographic group being overrepresented in a party versus a demographic group making up the vast majority of the party. So which numbers in the charts above can help give a more nuanced understanding of the Democratic Party’s demographics?
First, The Democratic Party has a huge gender gap — much bigger than that of the Republican Party — of 61% women and 39% men. Women, though, only account for 38% of Democratic Representatives and 37% of Democratic Senators. So while 2018 may have been “The Year of the Woman”, even the Democratic Party, being 60% women, dramatically under-elects them.
White people make up a majority of the Democratic party. But all the talk about the Democratic Party being the party of minorities has skewed perceptions here. It might be surprising that 57% of Democrats are white, but it is less so when you know that 70% of the general electorate is. Uneducated white voters make up 32% of the Democratic Party. While this is low — especially because they makeup 45% of the electorate — it is still a big coalition within the party. The Democratic Party, known for repelling uneducated whites, is one third…uneducated whites.
Similarly, Democratic youthfulness and education levels are often overstated. Only 20% of Democrats are 18-29 years old, almost exactly the same percent that is 65+ (19%). The majority of Democratic voters, 61%, are in-between 30 and 64 — precisely the same percent that are in the same age range for the country overall. And even though the Democratic Party is thought of as the party of college graduates, it’s not really true. A majority of the party, 62%, does not have a bachelor’s degree. And while the percent of Democrats with a college education is greater than that of the general public, the 4% divide is not as stark as media and stereotypes would have us believe.
Democrats are religiously diverse. Seventy one percent of the party is religiously affiliate. The largest religious coalition in the party, Protestants, make up 34% of the party. And while the party under-represents white religious voters, this is because the party under-represents white people generally. The exception is white evangelical voters, who make up 35% of the Republican party and 20% of the general electorate, but only 8% of Democrats.
Lastly, the Democratic urban/suburban/rural split matches the country well. Urban voters are underrepresented in both parties. How is this possible? Suburban and rural voters vote at much higher rates than their urban counterparts, so while Urban voters may lean more heavily towards democrats, they still make up a smaller portion of the party.
What A Democrat Wants The charts below show what percentage of Republicans and Democrats think a given policy “should be a priority for Trump and Congress” relative to the general public. It also shows which party voters think handles those policies better. Data: Pew, Pew, Gallup, PollingReport
Democrats are in a good position for campaigning. For each of the Democrat’s top seven priorities, the public believes they are better with the issue than Republicans. The most important issue to Democrats, “Health Care Costs”, was a winning issue for Democrats in the 2018 midterms. And its easy to see why — the general electorate prefers Democrat’s approach by a seven point margin. It’s also an the second most important policy issue for the public, trailing “The Economy” by just one percent. The general electorate most strongly prefers Democratic leadership in “Climate Change” and “The Environment”, although only 44% and 56% of all voters think they should be a priority.
The two policy areas that could give Democrats trouble are “The Economy” and “Terrorism”. They are both relatively far down among Democratic priorities — ranking eighth and tenth — and voters prefer Republican leadership on the issues. So, while Democrats can probably avoid these issues in the primary, they have two options for the general.
1) Avoid these issues on the campaign trail. This, however, could make them seem disinterested in matters that are top priorities for much of the country. Not usually a winning campaign strategy.
2) Emphasize and debate these issues on the campaign trail. If they hold the line on their unpopular policies without persuading voters, though, it will cost them votes. They will have to communicate their policies in a way that voters like. The three ways to do this are: win over public opinion with strong arguments, frame the issues in bland, inoffensive platitudes or adopt the more popular Republican stance on the issues.
Another possibility, one that makes this exercise seem a bit futile, is that voters don’t care about policy. Some research shows that voters — rather than forming opinions on their own and voting for politicians with similar ideas — take their queue on policy from political leaders. The idea that voters don’t care much about policy is not too wild. Trump transformed the party of free trade into a tariff-loving community in just two years. When Democratic primary voters break down their first and second choice presidential candidates, the plurality of Biden supporters choose Sanders, perhaps the most dissimilar candidate in the field, as their second choice. And vice versa.
This policy agnosticism is probably due to the increasing sports-like nature of party politics. Voters support their team rather than strict ideological beliefs. There is however, a small number of voters who are without a team. Next week on ESY…a look at independent voters.
Nobody Knows the Parties Perceptions of Republican and Democratic voters are wildly inaccurate. Take a guess at a) What percent of Republicans make over $250,000 per year and b) what percent of Democrats are gay, lesbian or bisexual. The real numbers is in Footnote 1. —>13A) 2.2% B) 6.3%
If you overestimated both, you’re not alone. The chart below compares the share of the Republican and Democratic parties that belong to a demographic group versus public perception. People overestimate, by 1736% (!!!), the percent of Republicans who make over $250,000 per year.
To understand the real makeup of the U.S. electorate, the next three posts will break down the country’s main voting blocs: The Republican Party, The Democratic Party and Independents/Unaffiliated Voters.
Twenty six percent of Americans consider themselves Republican, 31 percent Democratic, and 38 percent Independent, according to a recent Pew study. Because independents who lean towards one party vote for that party at almost the same rate as party members, they will be considered party members for this analysis. When these independent leaners are included in party totals, 39% of the public is Republican, 48% is Democratic and only 7% is independent.
The charts below show the makeup of the three coalitions. This is different than showing how demographic groups align. For example, the first chart shows that 53% of Republicans are men, but it does not show what percentage of men are Republicans. The ‘Democrat minus Republican’ rows show the difference in how much each bloc is composed of a given demographic.
Race & Education
Religion …Religion Continued
Urban, Suburban and Rural Data: Pew, More Pew *Used 2016 presidential vote as a stand in for party ** Used Data from CNN Exit Polls
THE REPUBLICAN PARTY The charts mostly speak for themselves: The Republican party is overwhelmingly white (86%), religious (only 12% are unaffiliated) and skews older (58% over 50). But with a bit more context, some less obvious numbers jump out.
While there is some truth to stereotypes of the Republican Party — old, male-dominated and white — they obscure demographic nuance. The party is surprisingly diverse when it comes to gender, age, education levels and where voters live. Forty severn percent of Republicans are women; Sixty five percent live in urban or suburban areas; Sixty six percent have some college experience; Only 25% of the party is 65 and older and only 35% live in rural areas. And while this just reflects that there are fewer older and rural people in the country overall, it still means the party is more diverse than stereotypes suggest.
This more-diverse-than-one-might-expect coalition could be trouble for the Republican Party. It means they will have to champion policies that can appeal beyond old rural men. The red-meat culture fights are not going to appeal to all these voters and could make holding the coalition together a challenge.
What A Republican Wants The charts below show what percentage of Republicans, Democrats and Independents think a given policy “should be a priority for Trump and Congress.” It also shows which party voters think handles those policies better.
Republicans care most about terrorism and the economy. The next crop of policies — Immigration, Social Security, Medicare and the military — are about 10% down.
Immigration could give Republicans trouble. It’s a priority for their party’s voters (68%), meaning that it will be foregrounded in Republican primaries. Primary candidates will be pressured to align with the president and take a hard stance on the issue, which will scare off general election voters — who prefer the Democratic Party’s approach to immigration by 14%.
Social Security and Medicare are similarly problematic for Republicans. They are priority for a majority of the party (68% and 60% respectively). This makes sense as the party skews older. Republicans risk scaring off this core constituency if they continue to emphasize unpopular policies around Social Security and Medicare. The general public supports the Democratic approach by 9 and 19 point margins. Republicans scrambling to defend pre-existing conditions and cast Democrats as a threat to Medicare means that they may be pivoting away from their entitlement-cutting agenda due to its unpopularity.
Republicans do have some popular policies. Voters approve of their handling of the economy, military and terrorism over that of Democrats. Electorally the party would likely be better off emphasizing these issues over their unpopular and divisive social policies. But candidates need to win primaries. Meaning they need to win over the Republican base and restraint is unlikely. Republican candidates will probably continue to sprint rightward in primaries and tiptoe back towards the center for the general.
This is the second post in the series “Mercurial Nation”, which will look at the elements that make up the political climate (the ‘mood’ of the country) and how they will affect the 2020 House Race.
The first element, Presidential Coattails, is split into two parts. Part 1 looked into what coattails are, what causes them and why they are losing power. Part 2, below, focuses on the 2020 presidential race and what different candidate strategies and outcomes could mean for the House.
Presidents usually win reelection15Only two presidents (Carter & Bush I) since World War II who have ran for re-election have lost. Both lost partly due to a recession. and the blue wave in 2018 shouldn’t be any comfort. Midterms don’t indicate much about the following presidential election. Other presidents with bumpy midterms — Reagan in 1982, Clinton in 1992, Obama in 2010 — won two years later. There’s a good chance Trump will be around through 2024.
Still, 20 months is a long time for a president and political environment as volatile as ours. Trump’s unpopularity, which already could cost him re-election, still has room to fall. Things outside his control — the economy, the Mueller report, the Democratic nominee — could determine the election.
All this to say…anything could happen before election day. For now, we’ll assume 2020 will feature a closely contested presidential race. If, however, one party wins the presidency in a landslide — or breaks into the mid-50 percentage range — it will probably carry the House too. So, if it’s November 2020 and Kamala Harris has a 10 point lead over Trump, you can stop reading.
How tight does the presidential race have to be for the House to be competitive? If House districts moved perfectly in line with the national environment,16Districts obviously don’t move in total synchrony, but their movement does track closelywith the national environment17The following calculation gives the projected margin (+ for Democrat and – for Republican) of a district in a neutral national environment: 2018 Democratic House Candidate % – 2018 Republican House Candidate % – 8.6%(to adjust for the D+8.6 national environment) +(for Democrats) or – (for Republicans) new incumbency advantage (0% for holding a seat, 2.7 % for winning an open seat, 5.4% for beating an incumbent). Democrats would lose 17 House seats, holding their majority by just one.18Unless they win the election in NC-09 (the 2018 results were invalidated due to election fraud), scheduled for September or November 2019. If Democrats win, they will have a two seat advantage. The tipping point seat, FL-27, would go to Democrats by just 0.1%.
And while a 3% presidential margin in favor of either party would likely be enough for them to win House comfortably (In my calculations, both parties would take a 16 seat lead), to be safe, any margin less than 5% should be considered a competitive environment for the House.
Getting to 270 The Electoral College (EC) distorts presidential elections, giving one party a boost relative to their percentage of the national popular vote. That advantage, though, goes back and forth between Democrats and Republicans due to changing demographic coalitions. In 2000 and 2016 it benefited Republicans — both years their nominee lost the popular vote but won the presidency — and in 2004, 2008 and 2012 it favored Democrats. Read this FiveThirtyEight article if you’re interested how this is measured.
Because EC votes19Besides in Maine and Nebraska. are awarded all or nothing by states, candidates don’t bother to campaign in states whose fates are predetermined. California’s 55 votes will go to the Democrat; South Dakota’s three will go to the Republican. No amount of campaigning will change that. Of the 538 EC votes, Cook Political Report and Sabato’s Crystal Ball assume that between 308 and 31320The discrepancy comes from Cook rating New Mexico as “Safely Democratic” and Sabato rating it “Likely Democratic”. votes are baked in. The states that could go to either party, dubbed ‘swing states’, are the ones that matter.
The 2020 “Toss Up” EC Votes — according to either Sabato or Cook or both — will likely be Florida (29 EC Votes), Pennsylvania (20), Michigan (16), Arizona (11), Wisconsin (10), New Hampshire (4) and Nebraska’s 2nd District (1).21Shout out to Larry J. Sabato for responding on Twitter and letting me use his map
The toss up states can be divided into two groups:
Sun Belt States22States stretching across the bottom of the U.S. from Arizona to to South Carolina. that have increasingly diverse populations and big cities with lots of white, educated suburban voters. These states have been trending away from Republicans as their non-white and college educated populations grow and they become more socially liberal. Trump, with his racialized politics and combative politics, has accelerated the trend.
Rust Belt States23Mostly in the Midwest and Great Lakes Region. that have a lot of white, uneducated, “working class” voters. These voters tend to be culturally conservative and more economically populist than the Republican establishment. They notoriously swung towards Trump in 2016 after voting for Obama in 2008 and 2018, helping to break Democrats’ “Blue Wall” of Pennsylvania, Wisconsin, and Michigan and give Trump his Electoral College victory.
Of this year’s “Toss Ups”, Florida and Arizona are Sun Belt States; Pennsylvania, Michigan, and Wisconsin are Rust Belt States. New Hampshire and Nebraska-02 — both heavily white and well educated — have characteristics of both.
If these are the competitive states, Democrats will likely have 228 EC votes locked down and Republicans will have 219. The magic number is 270 — Democrats need to win 42 and Republicans 51.24There are several scenarios that could give an EV vote total of *gasp* 269-269 (if Republican won the Rust Belt States and NH but lost lost NE-02). This scenario that wouldlikely end in a Republican president.
In this scenario, the Republican can win by:
Sweeping the Rust Belt Toss Ups (46 Electoral Votes ), NH (4) and NE-02 (1)
Winning a patchwork of Rust Belt & Sun Belt states.
The Democrat can win by:
Sweeping the Rust Belt Toss Ups (46).
Winning the Sun Belt Toss Ups (40) and NH (4).
Winning a patchwork of Rust Belt & Sun Belt states
How Presidential Strategies Could Tip the House The tactics and voter appeals that each nominee uses will largely depend on how they plan to make it to 270. If the House is closely contested, the messages and strategies deployed by presidential nominees could be decisive for control of the chamber.
The charts below include the 22 House districts rated “Toss Ups” by either Cook or Sabato. The “Diversity and Education Index” takes into account the what percentage of the district is non-white and what portion of the white population has a college degree. Negative numbers indicate a higher proportion of uneducated whites; Positive numbers indicate a higher proportion of educated whites and non-whites.
The big takeaway is that the 2020 Toss Up districts include an almost perfectly even distribution of districts dominated by white, uneducated voters and districts dominated by non-whites and college educated whites.
So, how could the presidential campaigns tip these districts? Lets start with Democrats.
Democratic Path 1: Sweeping the Rust Belt Toss Ups. If Democrats aim to sweep the Rust Belt and win over white uneducated voters, their best shot is to emphasize their populist agenda. This means focusing on the economy and revitalizing rural America, judiciously criticizing Republicans for their unpopular tax cuts, attempts to repeal the ACA and ‘swampiness’. It would also require laying off unpopular culture war issues, something Democrats seem reluctant to do. While this might be the Democrat’s best chance to win the presidency — Clinton only lost these states by about 1% in 2016 and they have historically Democratic roots — it could hurt them in diverse, educated House districts.
Democratic Path 2: Winning the Sun Belt Toss Ups and NH. To win Florida, Arizona and New Hampshire, the Democrat would need to drive up minority turnout and continue to pull educated whites away from Republicans. This would require more focus on divisive social issue — “The Wall”, Charlottesville,Trump’s “Family Separation Policy”. Like Path 1, this would mean sacrificing voters — this time in the white, uneducated House districts.
Democratic Path 3: Winning a patchwork of Rust Belt & Sun Belt states. This is probably the most challenging path for Democrats. But, if successful, it would help them win the most House seats. To win over voters in both the Sun Belt and Rust Belt, the Democratic nominee would need to appeal to white uneducated voters and non-white and college educated voters. A candidate who has a flexible image could thread this needle. Someone like Obama — who had a popular economic message among rural voters and was also a symbol of social progress to educated and non-white voters — could win over the three categories of voters. A Democrat who allows voters to see the candidate they want — a populist, a moderate, a progressive or a symbol of social change — would run up numbers across the board, helping Democrats in all of the toss-up districts. This path, however, risks failing in the same way that Clinton did in 2016 — falling just short enough among white, uneducated votes to lose the Rust Belt, but not picking up enough non-white and college-educated voters to win diversifying Sun Belt states like Arizona. It is a high risk tactic that could help Democrats run up their seat margin in the House.
Republican Path 1: Sweeping the Rust Belt Toss Ups, NH and NE-02. If Trump tries to sweep the Rust Belt and win New Hampshire and NE-02, he will try to appeal to the socially conservative, economically populist Obama-Trump voters. He will invoke ‘culture war’ issues and racialized politics that will boost his (and Republican House candidates’) popularity in white, uneducated districts, and hurt them with diverse and college educated voters elsewhere.
Republican Path 2: Winning a patchwork of Rust Belt & Sun Belt states. If Trump wants to win a patchwork of Rust Belt & Sun Belt states he will take a more moderate tack. This would mean toning down his racialized rhetoric, dropping unpopular fights (like shutting down the government to get money for The Wall) and putting down Twitter while emphasizing the strong economy and making a deal or two with congressional Democrats. This would require Trump to transform into a candidate with a flexible image typified in Democratic Path 3. If he could successfully make this pivot — though nothing in his history suggests he will — towards a broad appeal, he would help pull House Republicans into office across the board.
Remember, these tactical variations will only matter if the presidency is decided by a close margin — less than about five percent. Otherwise, the winning presidential candidate’s party is sure to take the house. In a tight presidential election, both parties would do better in the House if their presidential nominee could win the election through a patchwork of Sun Belt and Rust Belt states rather than by dominating in one of the two regions.
This mixed state coalition, however, will probably not be the easiest path to victory. Presidential candidates will probably angle towards either 1) white, uneducated voters in the Rust Belt or 2) non-white and college educated voters in the Sun Belt. In doing so, they will win over additional voters in about half of the “Toss Up” House districts and sacrifice voters in the other half — helping and hurting their party’s House candidates in roughly equal numbers.
This is the first post in the series “Mercurial Nation”, which will look at the elements that make up the political climate (the ‘mood’ of the country) and how they will affect the 2020 House Race.
The first element, Presidential Coattails, is split into two parts. Part 1, below, looks into what coattails are, what causes them and why they are losing power. Part 2 will focus on the 2020 presidential race and what different candidate strategies and outcomes could mean for the House.
Excluding local race dynamics, a House candidate’s baseline support in their district — measured by popular vote percentage— would match that of their party’s presidential candidate. A popular House candidate and favorable local dynamics will lift this percentage up; an unpopular House candidate and unfavorable local dynamics will drag it down.
House candidates that win, but with a smaller percentage of the district vote than their party’s presidential candidate — and who therefore benefited from the president’s higher level of support — are riding the president’s coattails.
To walk through an example of the candidate selection process, meet Sandra in the footnote->27Sandra lives in Moorhead, Minnesota (where they’re a bit overly excitedabout being the birthplace of the DQ “Dilly Bar”) and pays peripheral attention to the news. She is a strong supporter of the Republican presidential candidate because:
1) She’s a registered Republican.
2) She just got a pay raise.
3) She is pro-life.
So, Sandra goes and votes for the Republican presidential candidate. But what about those downballot races?
Sandra came to support the president and doesn’t know much about these other candidates. She decides to vote the Republican ticket all the way down to support her presidential pick and the Republican Party.
This is a presidential coattail vote — a downballot vote that comes from support for a presidential candidate.
In another world, Sandra still lives in Moorhead (and is just as proud of the ice-cream-on-a-stick heritage) but has seen ads saying that the Democratic House candidate supports the President’s economic agenda and is pro-life (Moorhead, Minnesota actually is in one of four districts represented by a pro-life Democrat) — both things that Sandra likes! Sandra also likes that the Democrat, Colin Peterson, used to be — and this is true— in a congressional bipartisan conservative rock/country/country rock band named “The Second Amendments”.
In this world, Sandra splits her vote: Republican for President, Democrat for House. Thus (I’m allowed a pretentious word here and there) no presidential coattails.
THE POWER OF COATTAILS TODAY How powerful are presidential coattails28From here on, referred to as just ‘coattails’. and are they getting stronger or weaker? Well, it’s complicated.
Coattails are measured by tallying the number of districts where the president’s vote total was greater than that of their party’s House candidate. By this measure, coattails are weak.
WHAT ARE COATTAILS MADE OF? To understand why they appear to be weakening, we’ll look at the two pieces that determine coattail length: Presidential Popular Vote Percentage (PPVP) and Ticket Splitting Rate.
Presidential Popular Vote Percentage (PPVP) The downward trend in PPVP is clear. Looking at the chart below, no presidential candidate has cracked 55% since Reagan in 1984. It is also clear that the candidates who won in landslides — Eisenhower (1956), Johnson (1964), Nixon (1972) and Reagan (1984) — had the longest coattails.
So why are landslide elections and high PPVPs becoming less common?
Historically unpopular candidates…
along with a stubbornly divided electorate with intense party loyalty.30There are plenty of analyses on the other factors that influence presidential elections. But what matters here is identifying the causes of the downward trend in PPVP. Everything else that is happening under the surface can stay there.
This environment limits a candidate’s ability to win over enough voters to reach into the mid-fifties, let alone garner enough votes for a landslide. And even recent elections with excited voters and high turnout— like Obama’s 2008 election— still feature a polarized electorate, and PPVPs around 50%.
Recent candidates, with their record unpopularity, have also boosted support for third parties. Third party votes have been on the uptick since 2004 and jumped in 2016. According to the 2016 exit polls, five percent of third party presidential voters31Voters who said they did not vote for Clinton or Trump or gave no answer. supported a Republican or Democrat House candidate. When seats are decided by one or two percent, the direction that these voters swing downballot can be decisive.
Ticket Splitting A voter who supports candidates from different political parties on the same ballot is called a ticket-splitter. Their ballot is dubbed a split-ticket.
A popular House candidate with favorable local dynamics may pull enough votes to win a race in a district that their party’s presidential candidate lost. Likewise, an unpopular candidate with unfavorable local dynamics may lose in a district that their party’s presidential candidate won. These split-ticket districts can be one-offs, but if a national party does an exceptionally good or bad job with candidate recruitment33*cough* Democrats in 2010*cough* or fundraising, it could have an aggregate effect.
It might seem that weakened coattails would mean a rise in ticket splitting. That’s not the case; split-ticket voting is at a record low. Take a look at this Washington Post graphic. It shows that ticket splitting on a district level has plummeted, particularly since 2012. Here is another picture showing the same trend.34Since I know some of you lazy boobs didn’t click the link.
The dive in ticket splitting has continued even as states do away with Straight Ticket Voting — the option to check one box and vote for all candidates of a single party.35Only 8 states will have the option in 2020. Given the antipathy between Republicans and Democrats, they’ll surely take the time to check those boxes D or R top to bottom.
The party loyalty trickles down from presidential politics into Senate, House, Gubernatorial, and even state assemblyraces. When it comes to elections of all levels, people take their cue from national party. This might be due to the nationalization of media or because national parties are a good and easy stand-ins for voters’ values. It’s a rabbit hole that goes down deep enough for books to be written on the subject.
BRINGING IT ALL TOGETHER Reading material that dense might be throwing spaghetti on the wall. If you need a summary, click the footnote –>36Presidential coattails are a factor of 1) Presidential Popular Vote Percentage (PPVP) 2) The percentage of their voters who split tickets downballot.
PPVPs have steadily decreased in recent years — a result of unpopular parties, unpopular candidates and high party loyalty — lowering the number of districts in which presidential candidates run ahead of the House candidate, which is one way to measure coattails.
Ticket splitting has fallen in recent years due to high partisan loyalty and the nationalization of politics and media.
So…are presidential coattails getting stronger or weaker?
It depends how they are measured. While presidential votes have become more and more indicative of House votes37Measured by falling ticket-splitting rates., low PPVPs are rarely enough to pull House candidates over the finish line.
Remember, however, that these trends are not destiny — they could intensify, reverse or hold steady. We live in a volatile and Mercurial Nation. Voters happy with the economy could view incumbents charitably, keeping Trump and a Democratic House. Voters could go to the ballot with hopes of electing a split government as a check on the parties, a theory known as balancing. Or voters could be hungry for change and elect a Democratic President and Republican House.
Next week we’ll go deeper into how the 2020 presidential election could play out and what these scenarios would mean for the House.
Let me start off by saying that I have no idea who will win the House of Representatives in 2020. Phew…it feels good to get that out of the way.
Now, if you’re still here38Hi mom., let’s take a look at what happened in 2018 and the current makeup of the House. Then we’ll turn to 2020. Keep in mind that the House, Senate and Presidency are different centers of power with different constituencies and voting trends. Be wary of translating any lesson or theory from one chamber or branch of government to another.
Republicans managed to flip only three districts (if we’re being generous40They only flipped PA-14 because of Pennsylvania’s court mandated redistricting.) out of 195 possible targets — a flippage success rate of 1.5%. Democrats flipped 46 out of 240 — a much more impressive flippage rate of 19%. Democrats were successful in nearly every type of district: Romney-Trump, Obama-Trump, Obama-Clinton, Romney-Clinton, suburban, and urban (see chart below). Conventional wisdom is that they struggled in rural districts, but even that is debatable as Democrats saw some of their largest gains percentage-wise in rural districts, even if these gains didn’t turn into outright wins. In fact, Democratic gains were largest in Republican strongholds that voted heavily for a Republican House candidate in 2016.
Democrats picked up seats in nearly every type of district: 41Sorry about the ugly charts. I’ll work on it. Chart Formatting/Idea from FiveThirtyEight; Density Data from CityLab
Right now, Democrats hold 235 seats, Republicans hold 199 and NC-09 is still undecided. This means that Republicans need a net flip of at least 18 seats to win back the chamber. What are their chances?
The TLDR 42Too Long Didn’t Read (a summary). I’m throwing in some internet slang to help the older folks keep up to date. is this: history looks bad for Team Red’s chances to take back the house, but the only honest prediction out there is ¯\_(ツ)_/¯.
Looking back, it’s clear that flipping the House in presidential years is very rare. So rare in fact that it has not happened in the post 1972 timeframe. The last presidential flip was way back in 1952, 66 years ago.
And in presidential re-election years (which 2020 will be with Trump43most likely at the top of the ticket), the president’s party averages a net of +2 seats. This number includes years where the presidents both won and lost re-election. Even in successful re-election years, the president’s party nets only +8 seats on average. Not enough for 2020 Republicans to take back the House.
The best year for any incumbent president’s party in this sample was 1984, Reagan’s re-election. Republicans picked up 16 seats — close to the 18 Team Red needs in 2020, but not quite.
The President’s party doesn’t usually win many House seats in re-election years:
Data from history.house.gov44The net +/- reflect overall change from the previous national congressional election.
Incumbent President Win/Loss
Incumbent President Party’s House Net +/-
Still, history is just history and politics has changed a lot since 1972. It would be foolish to think that this sample size of 5 elections could tell us everything we need to know about the upcoming election. In politics, past is not always prologue.
To get a better picture of what’s coming up in 21 months, we need to look at today’s political trends and individual House districts. I’m going to use Sabato’s Crystal Ball and The Cook Political Report — two of the most respected and historically accurateprojection websites — to get a quick feeling of how things stand for 2020.
Both site’s rating scale include “Toss Up”45Cook Definition: “These are the most competitive races; either party has a good chance of winning.”, “Lean D” or “Lean R” 46“Cook Definition: “These are considered competitive races but one party has an advantage., “Likely D” or “Likely R”47Cook Definition: “These seats are not considered competitive at this point but have the potential to become engaged.” or “Solid D” or “Solid R”48Cook doesn’t define, but I hope you can extrapolate. districts. According to FiveThirtyEight, these districts have an average margin of victory of 0%, 7%, 12% and 34% respectively.
The biggest takeaway from their 2020 House ratings, (take a look at them here and here) is that they both give Democrats the upper hand to keep the House while also placing more Democratically held seats as Toss Ups or Leans, the two most vulnerable categories. Particularly worrisome for Democrats are the 16 (according to Cook) and 10 (according to Sabato) seats rated as “Toss Ups”. Sixteen is alarmingly close to the 18 seats that Team Red needs to flip control of the House. Many of these were extremely tight races in 2018 or voted heavily for Trump in 2016, and could easily flip back to Republicans.
Side note — Nearly 80% of districts are so heavily Team D or Team R that there is no question which party they will vote for! A divided nation indeed.
A future post will go deeper into details about these ratings, but this is getting long so it’s time to wrap up.
So, to help understand the broad but all-important concept of the political climate, ESY will start off with a series called Mercurial Nation. Each post in the series will look at one major element of the political climate and how it could affect the 2020 House Race.
Next Sunday493/3/2019: Presidential Coattails: What Are They?
Election nerds, family, friends, interweb surfers of good fortune — welcome to Every Second Year.
This is a blog about power and how voters in the United States decide who gets to hold that power.
Specifically, it’s about the race for the 2020 United States House ofRepresentatives — the 435 members who, according to the Constitution “shall be composed of Members chosen every second Year by the People of the several States”. In need of a refresher on the three branches of government? Here’s a classic. It doesn’t quite replicate the genius of I’m Just a Bill, but sequels never do. The blog will have three parts:
1. (Almost) weekly posts, published on Sundays, using electoral history and trends to analyze the 2020 House Election broadly. These are labeled “The Big Picture”.
2. Bi-weekly (every other week) posts, without a set publishing day, looking deep into one race. These are labeled “2020 Battlegrounds”.
For each “Battleground” post, a formula picks out the district that would have the smallest percentage margin between the Republican and Democratic candidate if 1. the election were held now 50“Now” meaning whenever I start researching and writing, but no more than two weeks before a post is live. 2. polls were perfectly accurate and 3. every district swung perfectly in line with the national mood.
The formula isn’t perfect, but it will highlight a variety of competitive districts and give some spontaneity. It may point to a different district every week or stay on one for a few in a row; regardless, compelling stories will ensue.
Big picture readers, feel free to skip the next paragraph — it goes through the details of the selection calculation.
Here is the formula: Average the 2020 congressional generic ballot margin 51Using FiveThirtyEight’s aggregator.52While congressional generic ballot polls are not available, I will use President Trump’s approval rating from FiveThirtyEight’s aggregator. with every congressional district’s two party margin in its 2018 House election.53Adjusted up 8.6% to account for the 2018 popular vote margin Factor in any incumbency advantage (2.7% as estimated by FiveThirtyEight).54Seats where an incumbent was beaten will be adjusted 5.2% to account for the previous incumbent’s 2.7% advantage The district which has the smallest margin will be the district de jour. If there is a tie, the district with the closer 2016 House result will be chosen.
In addition to these weekly and biweekly posts, I occasionally write about other political topics: Presidential primaries, scandals, Cardi B’s views of the government shutdown. You know … the important stuff. These come without a schedule, when I have something to say and time to write.
Be warned: this is a policy-light blog. It was almost named “Horse Race House” to really lean into the non-policy angle. But that seemed a bit heavy handed. And how we choose our leaders is important! It matters for our country’s short-term policies and long-term survival. Democracy is not predetermined. Societies can — and usually do — break down.
And on that light note, it’s time to wrap up.
In my second post, I quickly lay out the results of the 2018 fight for control of the House and look at the GOP’s chances to take back power in 2020. It’s up now! Go read it 🙂
I also want to give credit to the journalists and news outlets upon which I am building my writing style and analysis.
FiveThirtyEight— This is where I started and continue to learn about electoral politics. Much of my writing/analysis will emulate their style. My writing format— using lots of internal footnotes and plenty of links — will also come from them.
Nathaniel Rakich — His blog Baseballotgave me the idea to start one of my own. He showed that it’s okay to put your work out there before you’re an expert and to learn along the way.
It’s a bit scary posting online these days. Once you put something out there … it’s really out there. There’s no turning back. You can pretty much count on someone having a screenshot of your most embarrassing online moment. So, Future Seth, if something on this blog comes back to haunt you, take this as Past Seth’s apology.