About Pew Research Center Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of The Pew Charitable Trusts. Newsletters Donate My Account. All four simulations suggest a future electoral landscape that is more favorable to the Democratic Party.
Under those assumptions, demographic change, especially by generation, creates a consistent and large headwind for the Republican party—shrinking the size of the groups that support them and increasing the size of those that oppose them. It does not mean that Republicans are doomed to lose but it does suggest that in the absence of substantial changes in the voting behavior of American demographic groups, the political terrain will continue to change in ways that make their success more difficult.
Figure 5 below summarizes the results of these simulations, presenting both the national population vote margin and the electoral college vote margin. In the sections below, we walk through those simulations in detail. In , Hilary Clinton won Going out further, the Democratic margin increases every election cycle by a little over 1 point, resulting in a 7. The primary reason for the projected rise in the Democratic popular vote margin is that the underlying racial and education composition of the electorate has undergone significant change between and Specifically, there has been a decline in the share of eligible voters who are white noncollege and a rise in the share of eligible voters who are nonwhite and white college graduates.
These changes are projected to continue throughout the s and into the s. Since white noncollege voters are strongly Republican while nonwhites are even more strongly Democratic and white college voters now lean Democratic, these demographic shifts create a more favorable electoral environment for a Democratic candidate.
However, elections are not determined by the popular vote. Using the support and turnout rates from the presidential election, we find that a Republican candidate would lose Michigan and Pennsylvania in but just barely take Wisconsin, resulting in a dead heat of in the Electoral College allocating congressional district-specific votes in Maine and Nebraska to the respective winners of those states. But going out further, Florida and Wisconsin are added to the Democratic column by ; Georgia and North Carolina by ; and Arizona by Thus, even in a scenario where generational preferences do not count, the electoral college quickly becomes more difficult terrain for a Republican candidate.
The U. In this simulation, we assume that the turnout rates of different race, education, gender, and age groups within states hold steady at the levels observed in the presidential election. We make a similar assumption about political preferences, but they are held steady among different race, education, gender, and generation groups within states. Only the underlying composition of the eligible electorate changes, with different groups growing and shrinking in size over time.
The result is striking. While the no generational effects simulation anticipates that the Democratic win margin in the national popular vote would shift from 2.
Going out further, the Democratic margin increases every election cycle by about 3 points, resulting in an These results illustrate the incredibly powerful effects that generational change could have under the assumption that generations hold their political preferences as they age. As these younger generations age—making up a larger share of voters as their turnout increases and older generations decline in size—they work to reshape the American political landscape. This contrasts with the first scenario where the political preferences of different age groups were assumed to be relatively stable and only subject to change because of shifts in their racial, educational, and gender compositions over time.
Given these dramatic effects of incorporating stable generational preferences, it is not surprising that the simulation projects a very substantial reordering of the Electoral College.
Under these assumptions, we find that a Democratic candidate in would win in the Electoral College by flipping not only Michigan, Pennsylvania, and Wisconsin—the states that President Donald Trump won so narrowly in —but also Florida and Arizona. Going out further, Georgia and North Carolina are added to the Democratic column by making the electoral vote split ; Ohio and Texas by ; Iowa in ; and the previously red state of Alaska in Our third simulation modifies the full generational effects scenario to allow for the possibility that generations may become more conservative over the course of the life cycle.
In this scenario, generations start in different places in terms of their political preferences but they undergo a similar conservatizing process as they age. Specifically, we assume all generations grow more conservative at a rate of a 1. This scenario produces results intermediate between the first two scenarios, forecasting a 4. Going out further, the Democratic margin increases every election cycle by about 2 points, resulting in a Even with a conservatizing trend built in, generational change still exerts a powerful force on electoral results going forward.
And, as in the full generational effects scenario, we can see generational effects at play in how age groups move Democratic by substantial margins over time. Under these assumptions, we find that a Democratic candidate would win in the Electoral College in by flipping Michigan, Pennsylvania, Wisconsin, and Florida. In later elections, we continue to see states added to the Democratic column, including Arizona and North Carolina by making the electoral vote split ; Georgia and Texas by ; Ohio in ; and Iowa in Our fourth simulation considers the possibility that newer generations not yet fully in the electorate may not be as pro-Democratic as the Millennial generation.
Specially, we assume that Gen Z members and the as-yet unnamed generation behind them will exhibit greater Republican preferences than the Millennials. This scenario also produces results intermediate between the first two simulations but quite close to the third simulation. This generational simulation would forecast a 4. Here, too, we see age groups moving Democratic over time through generational change but with an interesting difference.
This reflects the assumption about more conservative post-Millennial generations. Under these assumptions, we find that a Democratic candidate would win in the Electoral College in by flipping Michigan, Pennsylvania, Wisconsin and Florida.
In later elections, we continue to see states added to the Democratic column, including Arizona and North Carolina in ; Georgia in ; and Texas and Ohio in In our earlier report from , we found that demographic change—even without generational effects taken into account—had the potential to reshape the political landscape in favor of Democratic candidates.
We also found that strong movement toward Republicans among white noncollege voters—their strongest, albeit declining, demographic group—could potentially counter these underlying changes. In this report, we have calculated the potentially game-changing effects of generational change in the electorate. Generational change—working in tandem with other underlying changes by race and education—has the potential to create much stronger headwinds for future GOP candidates than we have found previously.
This is not to say that Republican candidates cannot win in the future. As we have repeatedly emphasized, the estimates presented here are simulations that explore very specific ideas about how the electorate might evolve. We generate them not to predict the future but rather to estimate the potential impact of underlying demographic changes in shaping election outcomes.
But, while none of the estimates presented here are predictions in the conventional sense, they are suggestive of the kinds of advantages and disadvantages parties might face in the not too distant future. Even under scenarios where these generations grow more conservative as they age, the current political disposition of these groups is so lopsided toward the Democratic party that the party would not be wise to rely solely on the conservatizing effects of age.
Thus, while demographics are not destiny, it is fair to say that demographic, especially generational, change is likely to have profound effects on the competition between the parties in this decade and beyond.
We cannot say which party will best handle these changes, only that they will have to do so. Robert Griffin is the research director of the Democracy Fund Voter Study Group, focusing on demographic change, public opinion, and American political behavior. He has taught courses on research methodology, statistics, public opinion, and political advocacy for The George Washington University, Pennsylvania State University, and Loyola University Chicago.
Ruy Teixeira specializes in the study of demographic change, voting behavior, and public opinion. He is a senior fellow at the Center for American Progress and co-director of its Politics and Elections program.
He is also a co-director of the States of Change project. William H. Frey is an internationally regarded demographer, known for his research on urban populations, migration, immigration, race, aging, and political demographics, as well as his expertise on the U. He is a senior fellow at the Brookings Institution, a research professor in population studies at the University of Michigan, and a co-director of the States of Change project.
The eligibility rates for the different populations presented in this report are projections. The first step in this process was taking data from multiple years of the American Community Survey ACS and dividing up the American population into groups based on state, race, age, and gender —for example, Hispanic men ages 30 to 34 in Colorado.
Multilevel statistical models were then used to estimate the unique eligibility rates—the rate of citizenship among a given group—and naturalization rates—the rate at which these groups gained citizenship over time—for each state, race, and age group.
These groups were then tracked forward in time and had those unique naturalization rates applied to them as they moved into older age groups. The rates of eligibility for those migrant and immigrant group was based on a similar multilevel modeling process. The end result is a procedure that is sensitive to the different rates of naturalization experienced by each of these groups, as well as the migration and immigration rates each state is predicted to experience in the future.
From there, the educational attainment of these eligible voter groups is projected in a similar fashion. Data was taken from multiple years of the ACS— to —and then the U. We then used multilevel statistical models to estimate the unique education rates the rate of college education among a given group and attainment rates the rate at which these groups gained education over time for each state, race, and age group.
Why doesn't this data include agricultural jobs? Other measures Other states United States. Voting in United States. Presidential election vote Total Votes. Percentage of Votes. Turnout showed distinct increases from among Asian American, Latino or Hispanic, and non-college white voters.
Each displayed to turnout increases that exceeded 6 points or greater. The non-college white rise in turnout is especially significant, as this group is linked to voting Republican in presidential elections, including in This rise occurred among both non-college white men and women Download Table A.
Yet increased turnout was also evident for Asian American voters a point increase and Latino and Hispanic voters a 6-point increase.
Black voters also showed a turnout spurt from their dip in the election. Consistent with much of the recent past, older adults voted at higher rates in than their younger counterparts. However, younger age groups showed the greatest turnout increase in that election, rising by 8 points for those aged 18 to In total, over half of these young adults turned out to vote. The impact of this voting bloc—which leaned heavily Democratic in the most recent election—was influential in many close swing states.
The voter population for presidential elections continues to change in its demographic makeup. Thus, for the first time in a presidential election, white voters without college degrees comprised less than two-fifths of the voter population.
These changes look quite different from , when non-college white voters comprised more than half of the voter population and nonwhite minorities comprised only one-fifth. The shift in the race-ethnic makeup of the population—especially the younger population—is evident when looking at voters in the past five presidential elections.
During this period, younger generations of voting-age citizens have become more racially diverse.
0コメント