Progressives Control the Future
By Sean McElwee and Colin McAuliffe
While much has been said about the demise of progressives in the wake of the Biden victory, the reality is that the future of the country will be more Democratic and more progressive than the past. Over the next decade, as millennials reach middle age and begin turning out to vote at a higher rate, they will create a unique opportunity for progressive policymaking.
Policy Views And Partisanship Are Not Set In Stone, But They Are Very Sticky
Today’s young voters are strikingly liberal and Democratic, to a degree that is not comparable to any past generation. While the current state of age polarization is sometimes imagined to always have been a feature of modern politics, it is relatively new. Ronald Reagan actually performed incredibly well among young people. In 2000, young and old split roughly evenly for Bush and Gore. Popular narratives sometimes paint baby boomers as a generation of young hippies who settled down and turned rightward, but in reality boomers were notably more Republican than other generations when they were young.
Cohorts of voters tend to get more conservative and Republican as they age. This is a complex phenomenon that we will unpack in detail, but first, note that younger cohorts are drifting Republican and conservative at a much slower rate as they age. In the chart below, we show that the average year on year shift to Republicans in party identification and conservative ideological identification has been declining significantly. Presidential vote tends to be a bit more stable over time, and also exhibits some cyclical patterns as some voters jump back and forth between parties in each election. This can make it a bit more difficult to interpret the average rate of change for presidential voting. But what we observe is that since the base level of Democratic voting preference for millennials is so strong, unless something happens to cause a rightward shift in their politics at a rate that is substantially faster than what has occurred for any past generation, they will remain solidly Democratic.
These results are based on a quantitative analysis of several decades of survey data, but there are some limitations for attempting to generalize these results to make statements about what might happen in the future. The intervals for our estimates of the rates of change for younger generations are wider because we have fewer years of data, but qualitative insights from political science research can help us fill in the gaps in our quantitative analysis and reason about how the Millennial generation’s politics might change over time.
There are many different reasons that political views and age are correlated, and it’s important to recognize that many of these are not intrinsically connected to age or aging per se. Different age groups have different priorities at any given time, and since certain experiences--getting married, having children, buying a home, accumulating wealth--can change people’s political views and interests, age is essentially serving as a proxy for groups with a set of similar experiences. No age cohort walks through life in exactly the same way, and we therefore should not expect the political leanings of each generation to evolve in the same way as the generation ages. We cannot directly generalize the political trajectory of any generation to make statements about the likely political trajectory of any other. Instead, we have to take a close look at the structural factors which shape the politics of each generation individually.
With these caveats in mind, there are several important results from the political science literature that we can be reasonably confident in applying across generations. A recent study by Jonathan Peterson, Kevin Smith and John Hibbing analyzing the Michigan Youth-Parent Socialization Panel Study finds that voters do become modestly more conservative over time, but that attitudes are remarkably stable. To the extent that people change partisanship, there is evidence that it tends to be connected to major life events. Research by Yair Ghitza, Andrew Gelman and Jonathan Auerbach finds that a predictive model based on the formative events of an individual’s life can strongly predict lifetime voting preferences, with millennials being the most liberal cohort of white voters. Work in sociology has come to similar conclusions.
The Millennial Experience: Iraq, Katrina, the Crash, Obama, BLM, Trump, Coronavirus
These findings suggest that millennials will probably end up somewhat more conservative in the future than they are now, but their politically formative years have largely already passed. The model of the formative years by Ghitza, Gelman, and Auerbach suggests that the popularity of the president is ultimately very influential. For millennials, this formative period was defined by the failures of two Republican Presidents (W. Bush and Trump) and the success of a popular Democrat (Obama). Research has shown that millennials’ views on racial justice have been positively shaped by the Black Lives Matter movement. Other research suggests that recessions during an individual's formative political years durably increases support for more redistribution. Millennials have already experienced two incredibly destructive recessions, punctuating an economy characterized by increasing income volatility, increasing underemployment, and meager wage growth which has never managed to reach its high in the late 90’s before any millennial had entered the labor market as an adult. As far as formative experiences go, these are pretty well optimized to produce a left-leaning cohort.
Looking beyond the formative years, there are several cultural, economic, and demographic factors that make it unlikely that we will see a large rightward shift among millennials any time soon. Millennials are getting married and having kids later than past generations, thereby delaying key life events which tend to cause voters to become more conservative. Millennials lag behind past generations in accumulating wealth, and their prospects for future wealth accumulation do not look great, removing yet another factor that can drive voters to become more conservative over time.
Millennials also lag behind past generations in home ownership and face soaring housing costs relative to their income. For the bottom of the income distribution, this is largely explained by stagnant wage growth. But for the higher end of the income distribution, whose wages have grown significantly, this is explained entirely by rising housing prices. Policies to favor housing wealth have been implemented over the past several decades, and the housing share of national income has nearly doubled over that time period. These policies are designed to inflate house prices, and the fact that now many well heeled young professionals still can not afford to buy a home is proof that they work as intended. From a political perspective, this means that a demographic that would have been likely Republicans in generations past is now largely shut out of homeownership and therefore are less likely to move rightward.
Increasing home prices, high returns on stocks, free flowing credit, and mass incarceration acted as substitutes for the large social safety nets that are typical in rich countries. But this approach relies on maintaining a base of middle class voters who are dependent on perpetuating this system. The policies that created a bloc of older voters who use the political system to fiercely defend the value of the assets they hold also created a bloc of asset-less younger voters who have no stake in an economic system that suppresses their wages and piles them with educational and consumer debt in order to fuel growth in asset prices. In fact, the current moment in politics gives us a window into an alternate version of history. Without rising home and stock prices, it is easy to imagine that middle class voters would never have tolerated the tepid economic growth and abysmal wage growth that have typified the neoliberal period.
Politics is now dominated by a new social class of petty capitalists, which includes a large number of older blue collar workers who depend on seeing returns on their property for financial security. At the opposite end of the political spectrum, a new class of grand proletarians is emerging; well educated young professionals who do not own property and whose fears of downward mobility are driving their demands for the US to adopt some semblance of social democracy. Combined with the fact that millennials are the most diverse and best educated generation at a time where racial justice movements are gaining momentum and partisanship is sorting along the lines of racism and xenophobia; we have a perfect storm for producing the stark and completely unprecedented age polarization that exists in politics today.
Setting aside the rapid changes in political views that tend to occur in the formative years, as well as the changes associated with life events, it is still possible for older generations to experience political shifts. In fact, the ANES data shows that the silent generation is still getting more conservative as they age, and is even doing so at an accelerated rate.
Differential mortality is probably playing a role here, but we can confirm this result using panel data from Democracy Fund Voter Study Group, where the same respondents have been reinterviewed in 2011, 2016, and 2019. Here we should emphasize that while the literature shows that political views are most malleable when people are younger and more stable afterwards, this does not mean that they are immutable. It is likely that the increased salience of immigration and demographic change are causing certain segments of older voters to cast their ballots for Republicans. In this case, the rightward movement of the silent generation is the mirror image of the leftward movement of young generations, driven by the long term trends in politics which we have discussed earlier. The current state of age polarization is therefore not the result of anything intrinsic to age itself, but is a manifestation of other structural trends which are correlated with age.
The future is uncertain, but based on political, economic, cultural, and demographic trends combined with the best available knowledge we have from the political science literature, we should not expect to see millennials make a large rightward shift as they age. The trends that have created today’s politics have deep structural roots, and can be observed in several other rich countries to varying degrees. Large segments of white voters who formed the traditional center left voting base are flocking to anti-immigrant parties, while center left parties find their ranks being filled by well educated professionals. Despite this drastic realignment along educational lines, lack of wealth has remained a very consistent predictor of center left voting preference.
It will be difficult to comprehensively test our theory that structural factors in politics will ensure that the millennial generation will remain quite liberal without waiting to see what happens in the coming years. But according to the political science literature, if millennials were to become significantly more conservative, we should be able to already see evidence for it in voting patterns and survey data from recent years.
Millennials Are Really Liberal, And They’re Staying Liberal
The millennial generation is on the whole far more liberal than any generation before it, not just because it is more diverse, but because millennial whites are also more liberal than their parents. The Voter Study Group panel survey now allows us to analyze attitudes among millennials as they have become older.
The Democracy Fund Voter Study Group (VSG) has surveyed the same respondents in multiple waves starting in 2011, with the most recent wave in 2019. This allows us to analyze the partisan identification of the same millennials over a nearly decade-long period (we defined millennials as anyone born in the year 1981 or later; there were no Generation Z participants in the survey). We find remarkable stability among millennials, and if anything, a slightly liberalizing effect. In 2011, 51 percent of millennials in the Voter Study Group survey identified as Democrats (29 percent identified as Republicans), while in 2019 that number grew to 53 percent (32 percent identified as Republicans). This finding holds up if we analyze only white millennials. In 2011, 47 percent of white millennials identified as Democrats (compared to 35 percent as Republican), which is dramatically different from non-millennial whites (38 percent Democrat, 48 percent Republican). Eight years later, 48 percent of white millennials identified as Democrats, and 38 percent as Republican. Non-millennial whites as a whole were somewhat but not distinctly more conservative in the 2019 survey than they were before (36 percent Democrats, 51 percent Republican).
In 2008, 59 percent of millennials voted for Obama, in 2012, 57 percent did, in 2016 56 percent of millennials voted for Clinton and in 2020 59 percent of millennials plan to vote for a Democrat. This suggests that over 12 years, the percent of millennials prepared to vote for Democrats has not changed much (while the percentage voting for the Republican candidate dropped seven points).
Turning to policy, the Voter Study Group has consistently asked a question about whether it is the government’s role to ensure everyone has health insurance. In 2011, 46 percent of millennials said yes and 39 percent said no (42 percent of non-millennials said yes and 38 percent said yet). In 2019, 52 percent of millennials agreed the government should guarantee healthcare and 38 percent said no. If anything, millennials have become more liberal on healthcare policy. They also became slightly more supportive of increasing taxes on the wealthy (56 percent support to 59 percent).
Most jarringly, millennials have become dramatically more liberal on issues of race. In 2011, 44 percent of millennials agreed that slavery and discrimination held back black Americans and 42 percent disagreed (this compares to 32 percent agree and 57 percent disagree among non-millennials), in 2016, 50 percent agreed and 39 percent disagreed (the question was not asked in 2019). Among whites, net disagreement was -9 in 2011 (compared with -40 percent among non-millennial whites) and in 2016 support for the statement was even for white millennials, compared with -37 percent among non-millennial whites.
In 2011, 31 percent of millennials believed that immigration should be easier and 31 percent thought it should be harder (compared with 25 percent and 44 percent of millennials). In 2019, 41 percent of millennials believed immigration should be easier and 25 percent believed it should be harder (compared with 27 percent and 42 percent of non-millennials). White millennials moved from +0 to +17 easier, while white non-millennials moved from -22 to -18. This is consistent with the academic literature on policy change, which suggests that ideological shifts tend to be driven by generational replacement more so than individual level attitudinal change.
Conclusion
According to Pew, millennials will make up a quarter of the electorate in 27 percent and Zoomers 10 percent (together millennials and Zoomers will be the same share of the electorate as The Silent and Boomer generations). But it will likely not be until 2024 or 2028 that millennials make up a larger share of the electorate than boomers, because of lower turnout. However, millennials are starting to reach the middle-ages, where individuals become regular voters. It is probably a safe bet to say that millennials will still be quite liberal although somewhat more conservative than they are now when this moment comes.
Demographics are not destiny, but structural factors in politics are. Unless nearly every major structural trend in politics manages to reverse itself and produce a degree of rightward movement among millennials of a size and pace that has never happened to any cohort in the past, millennials will have the opportunity to dramatically reshape American politics in a more progressive direction. We can perhaps already see the beginnings of this new kind of politics with the arrival of new millennial legislators, but whether or not progressives will be prepared to capitalize on their moment remains to be seen.
Colin McAuliffe (@colinjmcauliffe) is a co-founder of Data for Progress.
Sean McElwee (@SeanMcElwee) is Executive Director of Data for Progress.