Enter your email address below to receive our latest articles

A startlingly original piece by a former student. Your comments are very welcome. Please respond to marvin.zonis@chicagobooth.edu
-- Marvin Zonis

Why Democrats Win in 2020

By Edward Hamburg
Attempting to cope with the Trump Regime, Democrats yearn for answers to the question of what will happen in November. Polls are examined, pundits are consulted, primary results are scrutinized, and complex predictive models are developed in efforts to provide them with some guidance —and even a little hope.  
While sharing these same goals, this analysis examines no polls, consults no pundits, looks at no results from primaries, and develops no complex predictive models. It instead uses historical results primarily from the 2016 and 2018 elections for the United States House of Representatives to project results for the Presidential and House elections in 2020. Like other studies[1], it finds compelling empirical support for a projected decisive Democratic victory in November, with the determining factor being the mobilization of a Democratic electorate that is more engaged and expanding than at any point since the turn of the millennium.

We begin by considering the data in Table 1 below, which compare results aggregated at the national level for elections to the US House of Representatives in 2016 and 2018. Presidential voting results in 2016 are shown as well, but only to provide context. As a guide to understanding the discussion of each table, data cited about Republican voting changes and margin advantages appear in red; these same figures for Democrats are shown in blue.

The data start with the reminder that Donald Trump won the Presidency in 2016 with -2.9 million fewer votes than Hillary Clinton, or a negative margin of -2.2%. Republican Congressional candidates, however, were still able to scratch out a 1.4 million vote, 1.1% margin, over Democratic Congressional competitors nationwide.
Table 1 goes on to show how, as expected in midterm elections, the total number of votes cast for the Congressional candidates of both parties dropped-off from the numbers posted in the immediately preceding presidential-year contest. Yet the difference in these drop-off rates is striking: Republican House candidates received -19.5% fewer votes (-12.3 million in total) than in 2016, while Democratic House candidates received only -2.0% fewer votes (-1.2 million in total) than two years earlier. This difference drove the shift from the 1.1% Republican margin in the 2016 House elections to a nationwide Democratic vote margin of 8.7% in 2018  (a gap of 9.7 million votes).  Subtracting the Republican  percentage vote drop-off rate from the same percentage change measure for Democrats produces a net change score displayed in the green shaded box of 17.5%, which in this analysis is called the “Democratic Mobilization Margin,” reflecting the loss by Democrats of 11.1 million fewer voters than Republicans between the 2016 and 2018 Congressional contests.

Table 2 below puts these observations in a limited historical context by showing the same cited data for the last five presidential-year to midterm election transitions. Note how the Democratic Mobilization Margins in midterm Congressional elections suggest how the party would perform in subsequent presidential-year national contests: the -6.8% score in 2002 presaged its lackluster results in 2004; the 15.9% measure in 2006 heralded the decisive

Obama-led victories in 2008; the -26.0%score in 2010 signaled that Democrats would lose much of their 2008 margins in 2012, and the -9.1% measure in 2014 warned that the party was poorly mobilized for 2016, in contrast to the popular notion that these elections were largely theirs to lose.

But also note how the Democratic Mobilization Margin in 2018 suggests that the party is in a strong position going into 2020: the measure of 17.5% is even higher than the 15.9% recorded in 2006, when Democrats two years later led the Presidential vote by an impressive 7.4% and the vote for House candidates by an extremely strong 11.1%.



Chart 1 above provides additional support to these observations. Shown are the number of votes received by Republican and Democratic candidates to the US House of Representatives in only midterm elections going back to 1994. This focus on midterm results is critical, as research shows how presidential-year contests are comparatively high-stimulus events more sensitive to short-term personality and other factors that often overstate the amount of fluctuation in voter behavior over time.[2]  Midterm elections, by contrast, are less sensitive to short-term political stimuli and more reflective of the underlying distribution of partisan loyalties among voters. Because midterm results are generally more stable, noticeable changes between them can be meaningful.

Note in the chart how the number of votes for Congressional candidates of both parties between 1994 and 2014 alternate within a relatively narrow 30-45 million vote band, consistent with the historical narratives describing the rise and fall of the George W. Bush administration, the more dramatic rise and fall of Barack Obama, and the continuation of Republican Congressional majorities into 2016. But this pattern explodes in 2018: the number of Republican Congressional votes jumps to an unprecedented 51 million level, an increase of 10 million votes, or 27%, from 2014, while the Democrats reach an even higher unprecedented 60.6 million vote level, an astounding 25 million vote, 70% increase from the previous midterm election. Changes of these magnitudes reflect more than the movements of swing voters and the effective mobilization by Republicans and Democrats of their respective bases; they also indicate the rapid entrance of new voters into the active electorate and the attraction by Democrats of a markedly disproportionate share of these new votes. It is hard to imagine that these dramatic changes in 2018 would not have consequences this November. 

Chart 2 below emphasizes just how remarkable the 2018 midterm election was within this historical context. The usually clear and expected down-up-down-up pattern in the number of votes cast for US House candidates between presidential-year and midterm elections is exhibited from 1992 to 2016, with midterm counts ranging from 60% to 86% of the votes received by both parties in presidential-years. The number of votes cast for Democratic Congressional candidates in 2018, however, was 98% of the amount in 2016, a figure unequalled in almost nine decades of elections to the US House of Representatives.

While much attention is focused on which candidate will become the Democratic nominee to confront Donald Trump in 2020, the Democratic rank-and-file appears to be highly engaged and potentially poised for a historic performance at the national level.

But what about electoral performances at the state level, which proved so critical in Trump’s Electoral College victory despite his -2.9 million vote loss and -2.2% vote margin nationwide? The following discussion examines the usually identified critical swing states, sorting them into those that should already be in the Democratic fold, those that are likely to become or revert to Blue, those that are genuine but tough toss-ups, and those that will become more competitive but unlikely to shift to the Democrats in 2020. It then concludes with examples of some good long-term bets Democrats could make to further strengthen their future prospects with many more years of work. 

Swing States Likely Already in the Democratic Fold

Wisconsin should be Democratic in 2020 because it actually became Democratic in 2016. As shown in Table 3 below, only 22,000 votes state-wide out of almost 2.8 million votes cast for either Trump or Clinton put its ten Electoral votes in the Republican column and wiping out the 110,000 vote, 4.2% margin achieved by Democratic Congressional candidates. These Democratic Congressional electoral margins expanded to 195,000 and 7.7% respectively in 2018 — only the state’s gerrymandered districts kept five of its eight US House seats under Republican control. The positive but low Democratic Mobilization Margin of 6.8% should be a warning to Wisconsin party operatives that more voter mobilization is required in the coming months. Yet the number of Democratic Congressional votes between the 2014 and 2018 midterms did increase by 24%, compared to a drop in the number of Republican votes of -5% in the same period. Such movements, combined with solid Democratic Congressional vote margins established in the last two elections and the continued successful efforts to combat voter suppression evident in 2018, should be sufficient to maintain Wisconsin as a majority Democratic state in Congressional voting in 2020.

To address additional concerns about the possible 2020 vote for President in the state, note how in 2016 there were -135,000 fewer votes for Republican Congressional candidates than were cast for Trump. This ballot roll-off of -9.6% compared to a mere -3,000, -0.2% roll-off between the number of Democratic Presidential and Congressional votes in the state. A sizable proportion of the Trump-but-not-Republican voters in 2016 were likely ticket-splitting Democrats, and it would not take a significant drop in the number of such voters to wash away the razor-thin margins Trump somehow eked out four years ago.

Iowa is often classified as a Red-leaning state in 2020, but the 2018 midterm election significantly altered its coloration. As shown in Table 4 below, Trump indeed enjoyed a 10.1% vote margin and Republican Congressional candidates a 9.3% vote margin over the Democrats in 2016. In 2018, however, a steep -24.7% drop in the number of Republican votes combined with a minimal -1.3% drop in Democratic participation to produce a state-wide vote margin of 4.2% for the Democrat’s Congressional candidates and enabled them to capture three of the state’s four US House seats. Iowa’s Democratic Mobilization Margin in 2018 of 23.4%, higher than the national margin of 17.5%, should help strengthen its Democratic majority in Congressional voting and even deliver its six Electoral College votes to the Democratic Presidential candidate in 2020.

Swing States Likely to Become or Revert to Blue

Table 5 below describes how Donald Trump won Pennsylvania’s twenty Electoral College votes in 2016 by a margin of less than 1% and barely 0.007% of the votes cast for either major party Presidential candidate in the state, although his Republican Congressional counterparts garnered an almost 500,000 vote, 8.2% margin over Democratic House candidates. But a lot changed in the 2018 midterm election: the total Republican US House vote in the state dropped from 2016 by -28.8%, or almost 900,000 votes, while the Democrats actually INCREASED their Congressional vote count by 3.4% — a very unusual accomplishment between presidential-year and midterm contests. These two very different trends enabled Democrats to forge vote margins of 500,000 and 10.3% over Republicans and capture nine (up from five) of the state’s gerrymandered eighteen US House seats. Moreover, Pennsylvania’s Democratic Mobilization Margin of 32.1% is almost twice the national margin, suggesting that its Democratic electorate is more than ready to do battle with Republicans in 2020, including delivery of its Electoral votes to the Democrats.

Table 6 below describes how Donald Trump won Michigan’s sixteen Electoral College votes in 2016 by a margin that was only a quarter of his microscopic margin in Pennsylvania and barely 0.002% of the votes cast for both major party Presidential nominees in the state, although his Republican Congressional counterparts did scrape out a 49,000 vote, 1.1% margin, over Democratic House candidates. But in the 2018 midterm election the total Republican US House vote in the state dropped from 2016 by -17.7%, or almost 400,000 votes, while the drop-off in Democratic votes was under 30,000, or -1.3%. These two different results enabled the Democrats to build 2018 Congressional vote margins of 319,000 and 7.9% over the Republicans and capture seven (from five) of the states’ fourteen US House seats. Michigan’s Democratic Mobilization Margin of 16.4% is just under the national margin, but probably still high enough to maintain its US Congressional vote majority and deliver its Electoral votes to the Democrats in 2020.

Arizona is still very much a swing-state, but probably one with a strong enough Democratic trajectory to make it reliably Blue in 2020. As described in Table 7 below, Trump won the state’s eleven Electoral votes by a margin of 3.8% over Clinton in 2016, and Republican Congressional candidates posted a solid 8.0% margin over the Democrats. But the same unusual vote drop-off pattern observed in Pennsylvania occurred in Arizona in 2018: the number of Republican Congressional votes fell by -10.0% from 2016 as is customary in the next midterm contest, while the Democratic Congressional vote total actually INCREASED by 9.3%. These very different results produce a Democratic Mobilization Margin for Arizona of 19.3%, above the national margin and probably set to increase given the state’s dramatically changing demographic characteristics.

If the predictions made to this point are correct in 2020, then these Electoral counts would be added to those for the states of Washington, Oregon, California, Nevada, Colorado, New Mexico, Minnesota, Illinois, Virginia, Maryland, DC, Delaware, New Jersey, New York, Connecticut, Rhode island, Massachusetts, Vermont, New Hampshire, and Maine (2 of 4) to provide the Democrats with 295 Electoral College votes in 2020, 25 more than the 270 required to win.

Genuine But Tough Toss-Up States

Tables 8, 9, and 10 below show that Texas, Georgia, and Ohio, respectively, were decisively won by Donald Trump and Republican Congressional candidates in 2016, but they also show dramatic gains made by the Democrats in teh 2018 midterm elections. Texas and Georgia exhibited the same unusual differences in Congressional vote drop-off rates as in Pennsylvania and Arizona, with Republican vote counts falling and Democratic vote counts INCREASING between the presidential-year and midterm contests. In Ohio the drop-off for Democrats was only -3.3%, compared to drop for Republicans of -23.5%. Ohio’s Democratic Mobilization Margin of 20.2% is above the 17.5% national margin, while that for Georgia is an impressive 33.6% and Texas an even more impressive 37.1%. These figures helped drive down the Republican Congressional statewide margins of victory in Texas from 21.4% in 2016 to 3.5% in 2018, from 20.5% to 4.6% in Georgia, and from 16.3% to 4.8% in Ohio.

The holes in these states from which the Democrats must climb to turn them Blue in 2020 may simply be too deep, but seismic demographic changes in Texas and Georgia in particular could continue to accelerate the force of the Democratic mobilizations exhibited in 2018 and produce some compelling electoral surprises.

Competitive But Unlikely to Shift Swing States

While Democrats made progress in both Florida and North Carolina between 2016 and 2018, Tables 11 and 12 below clearly show that without improvements in the party’s abilities to mobilize significantly more voters, these states will likely remain under Republican control.

While Trump’s margin of victory in Florida was just 113,00 votes, or 1.2%, out of the over 9 million cast for the major party Presidential candidates in 2016, Republican Congressional candidates posted a solid state-wide margin of three-quarters of a million votes, or 8.6%. The number of Republican Congressional votes in 2018 dropped by -22.4%, but the number of votes received by Democratic Congressional candidates also fell significantly by -17.0%, yielding a low Democratic Mobilization Margin of 5.4%, less than a third of the national margin. The 5.3% vote margin still enjoyed by Republicans in Florida in 2018 will be difficult to overcome in 2020 at this low level of Democratic voter mobilization, although the latter might be just high enough to tip the state’s twenty-nine Electoral votes into the Democratic column for President.

Trump defeated Clinton in North Carolina by a reasonably comfortable margin in 2016, but Democrats significantly reduced the Republican Congressional margin of victory between 2016 and 2018. As in Florida, this margin reduction was achieved more through the disproportionate demobilization of Republican voters than the mobilization of Democrats to the polls in 2018. And like Florida, North Carolina has a single-digit Democratic Mobilization Margin that’s less than half the margin nationwide, probably too low to overcome in 2020 the 2.8% margin still enjoyed by Republicans in 2018, as well as flip the state’s fifteen Electoral votes to the Democratic Presidential candidate.

Potential Long-Term Democratic Bets

Few might imagine Kansas, Nebraska, or Kentucky as possible Blue states in the future, particularly because Trump won each state’s Presidential vote by margins of 22.2%, 27.0%, and 31.3% respectively, and Republican Congressional candidates enjoyed even larger state-wide margins of 37.2%, 43.3%, and 41.4% respectively. Yet as shown in Tables 13, 14, and 15 below, these sizable Congressional electoral margins in 2018 were cut to a third in Kansas as well as in half in Nebraska and Kentucky. These reductions resulted from significant drops in the number of Republican Congressional votes in all three states, along with sizable INCREASES in the number of Democratic Congressional votes so unusual from presidential-year to midterm elections. All three of these states exhibit strikingly high Democratic Mobilization Margins in 2018, more than twice the national margin in Nebraska and Kentucky and over three times the national margin in Kansas. None of these observations should mean much for Democratic candidates in 2020, but these states may become very different politically over the next decade.


In their opinion article in the Washington Post (23 February 2020), Robert Costa and Philip Rucker discuss how Bernie Sanders’s victory in the Nevada primary “forces a reckoning” for the Democratic Party as it plods towards selecting a Presidential nominee in 2020. This reckoning involves the party’s eventual decision to nominate a candidate considered most “electable” to the broadest possible range of citizens willing to vote against Trump, or a “movement candidate” believing that Trump’s defeat is a necessary but not sufficient condition to leading a long-overdue transformation of American politics.

The examination of electoral data and the projected outcomes presented in this analysis suggest that either one of these choices would be effective in defeating Trump and his Republican acolytes in November. This conclusion is likely frustrating to many, particularly those who think that candidates are the primary driver of electoral outcomes. Candidates, of course, are already an important factor in 2020, but in reverse from the perspective of Democrats — the repellant nature of Republican candidates (from Trump to McConnell to Graham to Nunes) may be more powerful in motivating voters than the positive characteristics of any particular Democratic nominee. Moreover, there could also be an additional “Brownback Effect” that reinforces the negative images of Republican candidates among many voters. As the elected governor of Kansas in 2010 and 2014, Sam Brownback forcefully implemented a set of such reprehensible policies from the Republican playbook that his own legislature was forced to roll many of them back, and he left office with a 66% disapproval rating. Such reprehensible policy over-reach is, of course, characteristic of the Trump Regime, potentially unleashing the same negative electoral consequences.

Yet Democrats will still insist upon wrestling with a difficult calculus: how to weigh their customary penchant for centrist candidates and understandable aversion to risking anything in the face of a possibly extended Trump presidency, against the opportunity to establish a new electoral coalition dedicated to launching a wide ranging agenda that forcefully addresses such fundamental concerns of the American polity as income inequality, health care, and climate change. Truly transformational opportunities come very infrequently in American electoral history; at what pace it happens or whether it occurs at all will be the ultimate questions answered as Democrats make decisions in the months ahead.


Edward Hamburg serves on the boards of directors of high technology companies after many years as a senior executive in the software industry. He received a Ph.D. in political science from the University of Chicago.

[1] See, for example, the work of Rachel Bitecofer, reviewed by David Freedlander, Politico, 6 February 2020.

[2] The initial research in this area was presented by Angus Campbell, “Voters and Elections: Past and Present,” Journal of Politics 26 (1964): 749-752. A summary of other such studies can be found in Edward Hamburg, “Anatomy of a Realignment: The Dynamics of Electoral Change in America, 1874-1906” (Unpublished Ph.D. dissertation: The University of Chicago, 1982): 14-16.