Do realignments exist?

[Heimdal]

One theory in American political science is realignment theory.  I understand a “realigning election” as an election where a chunk of the electorate wanders from one coalition to the other. One party is dominant, and another is weak. In modern American politics it goes something like this: Franklin Roosevelt wins the 1932 election and ushers in a period where the progressive New Deal coalition dominates. The New Deal coalition is broken by 1968, and the Republican coalition of Nixon/Reagan/Bush rules the political landscape for the next three decades.  According to this theory, we are due for a new realignment now.

My problem with this theory is that there are far too many holes in it to make sense. The Democrats controlled both houses of Congress for most of the time Reagan and Nixon were in office. Nixon won the 1968 Election by a whisker, and the Democrats were back in the White House in 1976. Even the 1932 Election doesn’t look that much like realignment when you look at it closely. As a progressive force the New Deal coalition had broken down by the mid-1940s, and during the 1950s a Republican President was elected and re-elected in landslide victories.

So what do you guys think? Are realignments real?

[Sol]

Ehh, I don't think so, at least in the American context.

[Antonio the Sixth]

Yes, they most definitely do. Just not the way most people think of them.

[King]

Hispanics definitely realigned in 2008.

[muon2]

Whether they happen immediately or evolve over time is the subject of debate, but it's hard to argue against the idea that political coalitions realign between the parties over time. Consider these maps, each separated by 44 years and representing relatively close elections. There's a transition such that almost all states have flipped parties by the end of the sequence with the middle map showing some intermediate shifts.

1916


1960


2004

[Bandit3 the Worker]

I think the period from 1994 to 2000 was a realigning one. The Democrats collapsed in 1994. But in 2000 (or maybe 1998), the Republicans collapsed in urban areas. The GOP never really collapsed in rural areas.

In other words, the urban areas recovered from the Republican Revolution of the '90s, while the rural areas did not.

[Heimdal]

I also think it is clear that voting groups move between the different coalitions. The best examples of this may be African-Americans and White Southerners. The African-Americans started voting Democratic (at least in those states where they weren't disenfranchised) during the 1930s, and Southern Whites began leaving the Democratic coalition in the 1950s.

I think my beef is with realignment theory at its most primitive: that one of the two parties dominates the political landscape for cycles of 32 or 36 years. That isn’t to say that there aren't periods where one party has the upper hand on the other. The Republicans were obviously in a stronger position than the Democrats from 2000 to 2006, and the Democrats from 2008 to 2010.

Muon2: Thank you for the interesting maps. I think it is interesting how Maryland have stayed in the Democratic coalition the whole time.

[angus]

So what do you guys think? Are realignments real?

Definitely.

The most simplistic view is trying to superimpose Daniel Elazar's model of political culture upon American politics.  Three cultures, two parties.  Whichever party can gain two of those cultures wins elections.  This allows one party to dominate for a while.  With mobility and changing demographics, Elazar's model has become less relevant, but it still offers some insight.

You also can just look at state-by-state party control.  Two states in particular, Texas and West Virginia, have shown marked movement.  In just one generation one political party completely supplanted another.  This is a realignment. 

Or you can view it by voting blocs.  This requires that you objectify people, but if you don't mind doing that then it is easier to see the realignment.  In the US, however, the political parties re-align as much as the population re-aligns. 

I think it is interesting how Maryland have stayed in the Democratic coalition the whole time.

That is also evidence of a realignment.  To use Elazar's terms, Maryland was culturally Traditionalist in 1916.  Very much so, and it aligned with Virginia, Carolina, Georgia, etc.  At that time the Democrats were very successful at courting the Traditionalists.  By 2004, Maryland's population had changed so much that Moralists and Individualists outnumber Traditionalists.  It now aligns more with New Jersey, New York, Philadelphia, etc.  Also, by 2004 it is the Republican party that successfully courts the Traditionalists.  Maryland stays the same color because it and the parties change simultaneously (Maryland by migration and the resulting demographic realignment, the parties by strategy and the resulting ideological realignment).

[TNF]

Obviously.

They coincide with party systems. The First Party System (1790s-1830s) saw the Republican Party dominant over the Federalists, the Second Party System (1830s-1860s) saw the Democratic Party dominant over the Whigs, the Third Party System (1860s-1890s) saw the Republicans dominant over the Democrats, a trend which continued into the Fourth Party System (1890s-1930s), until the Democrats once again regained supremacy in the Fifth Party System (1930s-1960s). The Sixth Party system, I would argue, was the period of Republican dominance between the 1960s and the 2000s, and now we've probably entered into a Seventh Party System in which the Democrats will be dominant for the foreseeable future (likely the next 30-40 years, as is the usual life of these party systems).

[Clarko95 📚💰📈]

Yes, they exist in that coalitions rise and fall throughout times. But I don't believe in theories that say they run in 20/40/50/60/80 year cycles.

[Sol]

Oh, I certainly do believe that areas change hugely over time- but I don't believe in the set-year cycles, nor that it happens all in one election always.

Of course, this only applies to the US. Canada, for instance, seems to actually have proper realignments.

[pbrower2a]

1952 Eisenhower (R)/  2008 McCain (R)

1952 Eisenhower (R)/ 2008 Obama (D)

1952 Stevenson (D)/ 2008 McCain (R)

1952 Stevenson (D) / 2008 Obama (R)

Gray -- did not vote in 1952

Note that electoral votes in Maine and Nebraska went the way the states voted in 1952 (no split among districts) and except for NE-02 went as the states went in 2008.

Curricula vitae of newly-elected Dwight Eisenhower and Barack Obama could hardly have been more different.

1956 Eisenhower (R)/  2012 Romney (R)

1956 Eisenhower (R)/ 2012 Obama (D)

1956 Stevenson (D)/ 2008 McCain (R)

1956 Stevenson (D) / 2008 Obama (R)

Gray -- did not vote in 1956

All electoral votes for Maine and Nebraska went as the states went on the whole.

[DC Al Fine]

Canada, for instance, seems to actually have proper realignments.

Once again, proof that Canada has the most exciting elections of all the FPTP countries.

[badgate]

1956 Eisenhower (R)/  2012 Romney (R)

1956 Eisenhower (R)/ 2012 Obama (D)

1956 Stevenson (D)/ 2008 McCain (R)

1956 Stevenson (D) / 2008 Obama (R)

Gray -- did not vote in 1956

All electoral votes for Maine and Nebraska went as the states went on the whole.

I was curious how many electoral votes these blocs of state had in 2012 vs 1956.

1956: 332  2012: 325
1956: 125  2012: 132
1956: 74  2012: 71

[muon2]

1956 Eisenhower (R)/  2012 Romney (R)

1956 Eisenhower (R)/ 2012 Obama (D)

1956 Stevenson (D)/ 2008 McCain (R)

1956 Stevenson (D) / 2008 Obama (R)

Gray -- did not vote in 1956

All electoral votes for Maine and Nebraska went as the states went on the whole.

I was curious how many electoral votes these blocs of state had in 2012 vs 1956.

1956: 332  2012: 325
1956: 125  2012: 132
1956: 74  2012: 71

1956 was a blowout. That's why 1960 seemed like a better comparison. Close elections tell more about the intrinsic lean of a state.

[pbrower2a]

1956 was a blowout. That's why 1960 seemed like a better comparison. Close elections tell more about the intrinsic lean of a state.

Intrinsic lean of one state is one thing. In 2008, Barack Obama exaggerated those, winning every state that Gore or Kerry carried by at least 9%. 

Really, Barack Obama generally won by blowout wins in those states that he won in 2008 or did not win at all. It's telling that his tipping-point state in 2008 was Iowa, a state that he won by 9%.

In 1952 and 1956, Dwight Eisenhower did extremely well in states that we generally consider very liberal.  Ike won Massachusetts (the only state that Nixon never won) and Minnesota (the only state that Reagan  never won). I'm going to figure that kissing up to segregationist pols in the South was a poor way to win Up North in the 1950s.

He won the only two Northern states (Massachusetts and Rhode Island) that didn't go to Herbert Hoover in 1928 -- if you want to go that way.

2008 was a near-blowout election. All that prevented an Obama landslide was his race. Move the electorate 5% in the states from 2000 toward Obama, and Obama wins Missouri, Arkansas, Louisiana, Tennessee, and West Virginia (I concede Arizona due to the favorite-son effect), and Barack Obama wins 406 electoral votes.

Another interesting contrast is Carter 1976 to Obama 2008.

[badgate]

^Dont forget how close Montana was IRL in 2008.

[Miles]

Yes; otherwise Landrieu, Hagan and Pryor wouldn't be in tough races.

[pbrower2a]

Carter 1976 Obama 2012  147
Ford 1976 Obama 2012  179
Carter 1976 Romney 2012  141
Ford 1976 Romney 2012 65

(Maine districts are shown in yellow because orange does not show for districts).

[Clarko95 📚💰📈]

Carter 1976 Obama 2012  147
Ford 1976 Obama 2012  179
Carter 1976 Romney 2012  141
Ford 1976 Romney 2012 65

(Maine districts are shown in yellow because orange does not show for districts).

This is fascinating!

[Filuwaúrdjan]

Of course, but American 'realignment theory' is insane.

[pbrower2a]

Indeed I call the 1976 Presidential election "ancient and irrelevant history" to the present. You can see why.

[pbrower2a]

Demographic change from population shifts and ethnic change can turn a State from Republican to Democratic over ten years or so (California, Colorado, Nevada, and New Mexico since 1980 due to the rapid growth of the Mexican-American vote). On the other side, Democrats won West Virginia fairly reliably when many of the voters were were active members of the politically-powerful United Mine Workers Union (UMW)except in Republican landslides -- until the coal-mining industry shrank and with it the UMW, and the state has drifted unambiguously toward the Republican Party. Such does not explain states that have not undergone demographic change. Vermont and Maine have gone from two of the solidest-Republican states in Presidential elections,  as a political ally of FDR quipped during the reporting of his 46-state electoral blowout:

You must be logged in to read this quote.

https://en.wikipedia.org/wiki/United...election,_1936


It is highly unlikely that either Maine or Vermont, neither of which seems to have undergone great demographic change since the 1930s,  has changed in its political culture since the 1930s. What has happened is that the Democratic norm in Presidential politics has come to include the political cultures of Maine and Vermont that existed in the 1930s and as late as the 1980s. Maine and Vermont (also Oregon, Washington, Michigan, Illinois, and Pennsylvania) used to elect numerous liberal Republicans. That is over. There are few remaining liberal Republicans.

Politicians can rarely force demographic changes, but they can exploit them. As significantly, they can also exploit the failure of those pols who think certain constituencies or states 'sure' to vote for them. California, Michigan, Illinois, Vermont, and Maine used to be almost reliably R in Presidential elections.  They are the opposite now.   

[All Along The Watchtower]

^not sure about Maine, but Vermont has definitely undergone a lot of demographic change within the last 100 years (or even the last 50).

[TDAS04]

They do happen, even if it takes a few election cycles for them to become significant.