How big a mandate?

How big is Barack Obama’s presidential mandate? And how much is it a personal mandate for the remarkable candidate that is Mr Obama, and how much is it a mandate for his Democratic Party, and by extension, for its policy preferences?

These are highly subjective–if important–questions, and I shall not pretend to have the answer to “how big” in some absolute sense. However, we can look at some electoral data for a cross-temporal comparative perspective. How does the Obama/Democratic ‘mandate’ compare with those for other “change” presidents?

For purposes of this discussion, I am going to look at all of the newly elected presidents since 1932 whose election marked an alternation in the party controlling the presidency. We will consider the size of the president-elect’s own mandate in both votes (the real kind, i.e. the ‘popular vote’) and electoral votes.

I will then look at the extent to which the change in the presidency was reflected in the House of Representatives. Did the president’s party gain seats upon his election? If so, was the gain sufficient to alternate control of the chamber? If there was a big surge in House seats for the new president’s party, we have a classic “coattails” effect, whereby the enthusiasm for the incoming president motivates voters to cast votes for his party, too.

Then I will look at changes in the House over a two-election period. The logic for doing so is that, if control of the House alternated at the previous midterm, that election may be proven to be a harbinger of partisan change, which voters then can confirm at the subsequent presidential election. This would not be a coattails story, except perhaps one of reversed coattails: The presidency might be said to have changed partisan hands because the voters were in a two-year process of conferring a mandate on the party.

The data I am going to present provide some support for the notion that the Obama/Democratic mandate is historically significant by each of these measures. To summarize the conclusions just a bit, Obama’s personal victory, at least in the popular vote, was one of the biggest we have seen in decades. His coattails, measured by copartisan gains, were substantial, though not historic. His party’s gains over the two-election cycle are, however, historic. In fact, 2008 marks the first time since 1918-1920 that a partisan change has occurred in the House and then been confirmed at the next presidential election–and 2006-2008 was bigger than the one in 1918-1920.

Here are the data (and you may click here to open up a larger version in a new window).

2008 mandate in perspective

I would submit that this is the biggest PARTISAN MANDATE we have seen in the USA since FDR.

First, let us look at the personal support Obama received, compared to the other “change” presidents. At 52.4%, it is the largest by a Democrat since FDR (who had 57.4%), but clearly falls short of Republican Eisenhower’s 55.1%. Thus, measured by share of the popular vote, Obama received a bigger personal mandate than any change president in fifty-six years.

I also include share of two-party vote, because it has been rather common for alternations to occur in years in which there was a significant third-party or independent candidate who shared with the victorious major-party challenger an articulation of voter desire for change. Notable cases are Ross Perot in 1992 and George Wallace (if you want to call that “change”) in 1968, as well as John Anderson in 1980. The exclusion of third-party/independent votes from the denominator tends to make the newly elected president’s share of the 2-party vote greater than his share of the total vote. I assume that is because the incumbent president (or the candidate of the outgoing president’s party) loses some support from voters who can’t quite bring themselves to vote for the other major party’s candidate. Indeed, we see that Obama’s share of the 2-party vote is comparable to (and slightly behind) that of Clinton in 1992. It is also well behind that of Reagan in 1980 (as well as, of course, the figures obtained by FDR and Ike).

In looking at these cases of presidential alternation, it is striking how low the overall popular vote is for so many of them: Five failed to win even 51% of the vote, including each of the last five before this one, and two could not reach 45%. A change vote that breaks 53% of the 2-party vote is, therefore, impressive. Obama is only the third to do so since Ike.

Now let’s turn to the House, and a look at coattails and reversed coattails.

Strikingly, no newly elected president since 1952 has brought the House over to his party along with his election. And in the case of Eisenhower, it was actually a spurious majority (his party’s votes were less than that of the Democrats). That’s pretty much the definition of a personal mandate, even if the party did gain the majority. Not surprisingly, it would lose it at the very next election, in 1954–and not get it back for forty years.

Given the Democratic “lock” on the House between Eisenhower’s first midterm and Clinton’s first midterm, the presidential changes to Republican in the interim both failed to bring about alternation in the House. By the same token, that meant that the changes of the presidency back to Democratic control in the same period meant no alternation: they were more “restorations” (of unified government) than alternations.

Of course, the 1980 election merits special note. The Republican Party gained what is still the highest number of seats in a presidential-alternation year since 1932. Nonetheless, their 34 seats, propelled by a 3.1 percentage-point increase in the party’s House vote compared to 1978, was insufficient to bring Reagan’s party the House majority. (The party did, of course, take over the majority of seats in the Senate.)

The other alternations from 1960 through 1992 are remarkable only for the absence of coattails. The parties of Kennedy and Clinton actually lost seats (22 in the case of JFK!), Carter’s managed only a one-seat pickup, and Nixon’s only six. Three of these presidents saw their parties lose House votes (as a percentage) as they were being elected (Kennedy, Carter, and Nixon). I already mentioned the spurious majority for Eisenhower’s party, though the party did manage a minuscule gain in votes percentage.

Obama’s party’s gain of 19 (pending final counts that could push it to 20 or higher) is the second highest since 1952 (and the higher one, in 1980, still left the new president with divided government).

As for the possibility that a partisan mandate may begin at the previous midterm and be confirmed by the presidential alternation, Obama’s and the Democrats’ example is the first in a very long time. In 2008, with a net change of 50 seats (and it could go higher) we see the highest two-election gain in seats for the newly elected president’s party since the remarkable 147-seat gain in 1930-32.

Thus there is some evidence of a reversed-coattails effect, whereby Obama rode a strong pro-Democratic wave that began in 2006. ((Gains in a concurrent election really could be either coattails or reversed coattails.)) While the Republican gains over two elections in 1968 were a bit greater than those in 2006-08 and the gains in 1974-76 and 1990-92 were similar in magnitude, none of these produced an alternation.

Now, about that 1918-1920 partisan mandate. It is the most recent case, before 2006-08, that I could find of alternation in House control followed by presidential alternation. And it deserves an asterisk. ((Speaking of asterisks, I am leaving out of this discussion the presidential alternation of 2000, as it was not driven by a popular vote plurality. I may add the rest of the numbers in a comment later, just for comparative purposes.)) Republicans had a 215 – 214 plurality in the House after 1916 (after having been in the minority at the 1914 election), and took the majority (240 – 192) in 1918, then presidency in 1920. (The 1916 House had Prohib 1, Socialist 1, others 4.) So it is not as if the Democrats held a clear majority at the time this alternation began–unlike the Republicans when the current one began in 2006. So, this current case appears to be the only clean example of a two-election partisan alternation in the House and presidency (with the Senate, too!) in over a century.

In short, Barack Obama and the Democratic Party have just won a mandate of historic proportions.


6 thoughts on “How big a mandate?

  1. Although I put the electoral-vote percentages in the table, I forgot to say anything about them. Obama’s percentage stands at the 67.7% shown in the table, pending the final call of Missouri (and the one electoral vote in Nebraska that he may win). It looks like Missouri is likely to go to McCain, but if it were to go to Obama, then our 2008 winner would pass Clinton in the electoral votes.

    At the current count, Obama’s percentage is clearly non-historic. It ranks fifth among the eight “change” elections, beating only Kennedy, Nixon, and Carter. Of course, this narrower (albeit still decisive) electoral-college win reflects the growing ideological distance among state majorities, compared to earlier periods. What really matters for judging the scope of a mandate is the extent of the national popular-vote swing in both the presidential and congressional elections. And, as noted above, the swing was substantial in this case.

  2. I should also note that the House national vote total will not be available for a while, but it clearly will show a swing. It was already +7 from 2004 to 2006! It was probably another 2 or so in 2008.

    If it is indeed around +9 over the two elections, it would be by far the greatest for a partisan-change period since 1930-32.

  3. My election friend, I’m back with another count.

    I went to the CNN website and tallied up all the House district votes. There were 58 districts where one party or the other did not run a candidate and CNN just left those blank. I think in 44 of those districts there was no Republican candidate and in 14, no Democratic candidate. Also, CNN doesn’t include any third party candidate.

    Of course, the returns aren’t final or official anywhere and there are probably still several million votes to be counted, so this is strictly preliminary.

    So, anyway, these are the totals for the 377 districts.

    Democrats 54,598,746 53.6%
    Republican 47,296,887 46.4%

    If we adjust for the candidate unopposed, who in every state but Florida is still on the ballot and still gets votes, by say giving them 150,000 votes, we can readjust the preliminary figures for a more accurate total.

    Democrats: 44 unopposed canidates * 150,000 = 6,600,000

    Republicans: 14 unopposed candidates * 150,000 = 2,100,000

    Democrats: 54,598,746 + 6,600,000 = 61,498,746 55.3%
    Republicans 47,296,887 + 2,100,000 = 49,396,887 44.7%

    I’d say that’s approaching a landslide win.

  4. A longer term look at the trend to the Democrats

    It’s interesting to compare the results of this election to the results of the 1988 election.

    1988 was the beginning of the Democratic Party changeover from when it was ‘captured’ by the liberal “Great Society” Democrats that included Walter Mondale to the modern Democrats of today. Michael Dukakis in many ways was the beginning of the modern Democrats despite his being successfully labeled ‘a liberal’, largely because he came from the Northeast. He was actually a ‘technocratic’ Democrat.

    We saw in this election that the Republicans tried to label Obama as a ‘liberal’ or ‘socialist’ Democrat and it made no difference to voters except to the already persuaded Republican hard core.

    I know the 2008 figures aren’t final, but they’re close enough.

    008 totals, 1988 totals, D then R
    Thanks to Dave Leip’s web site

    39-61, 39.9-59.2 — Alabama,
    36-62, 36.3-59.6 — Alaska,
    45-54, 38.7-60.0 — Arizona
    39-59, 42.4-56.4 — Arkansas

    61-37, 47.6-51.1 — California,
    54-45, 45.3-53.1 — Colorado,
    61-38, 46.9-52.0 — Connecticut,
    62-37, 43.5-55.9 — Delaware,

    51-49, 38.5-60.9 — Florida,
    47-52, 39.5-59.8 — Georgia,
    72-27, 54.3-44.8 — Hawaii,

    36-61, 36.0-62.1 — Idaho,
    62-37, 48.6-50.7 — Illinois,
    50-49, 39.7-59.8 — Indiana,
    54-45, 54.7-44.5 — Iowa,

    41-57, 42.6-55.8 — Kansas,
    41-58, 43.9-55.5 — Kentucky,
    40-59, 44.1-54.3 — Louisiana,

    58-40, 43.9-55.4 — Maine,
    62-37, 48.2-51.1 — Maryland,
    62-36, 53.2-45.4 — Massachusetts,
    57-41, 45.7-53.6 — Michigan,
    54-44, 52.9-45.9 — Minnesota,
    43-57, 39.1-59.9 — Mississippi,
    49-50, 47.9-51.8 — Missouri,
    47-50, 46.2-52.1 — Montana,

    41-57, 39.2-60.2 — Nebraska,
    55-43, 37.9-58.9 — Nevada,
    54-45, 36.3-62.5 — New Hampshire,
    57-42, 42.3-56.2 — New Jersey,
    57-42, 46.9-51.9 — New Mexico,
    62-37, 51.6-47.5 — New York,
    50-49, 41.7-58.0 — North Carolina,
    45-53, 43.0-56.0 — North Dakota,

    51-47, 44.2-55.0 — Ohio,
    34-66, 41.3-57.9 — Oklahoma,
    57-41, 51.3-46.6 — Oregon,

    55-44, 48.4-50.7 — Pennsylvania,
    63-35, 55.6-43.9 — Rhode Island,
    45-54, 37.6-61.5 — South Carolina,
    45-53, 46.5-52.9 — South Dakota,

    42-57, 41.6-57.9 — Tennessee,
    44-55, 43.3-56.0 — Texas,
    34-63, 32.0-66.2 — Utah,
    67-32, 47.6-51.1 — Vermont,
    53-47, 39.2-59.7 — Virginia,

    58-41, 50.1-48.5 — Washington,
    43-56, 52.2-47.5 — West Virginia,
    56-43, 51.4-47.8 — Wisconsin,
    33-65, 38.0-60.5 — Wyoming

    93-7, 82.7-14.3 — District of Columbia

  5. States where Obama did better than Dukakis in 1988

    1.45-54, 38.7-60.0 — Arizona
    2.61-37, 47.6-51.1 — California,
    3.54-45, 45.3-53.1 — Colorado,
    4.61-38, 46.9-52.0 — Connecticut,
    5.62-37, 43.5-55.9 — Delaware,
    6.51-49, 38.5-60.9 — Florida,
    7.47-52, 39.5-59.8 — Georgia,
    8.72-27, 54.3-44.8 — Hawaii,
    9.62-37, 48.6-50.7 — Illinois,
    10.50-49, 39.7-59.8 — Indiana,

    11.58-40, 43.9-55.4 — Maine,
    12.62-37, 48.2-51.1 — Maryland,
    13.62-36, 53.2-45.4 — Massachusetts,
    14.57-41, 45.7-53.6 — Michigan,
    15.54-44, 52.9-45.9 — Minnesota,
    16.43-57, 39.1-59.9 — Mississippi,
    17.49-50, 47.9-51.8 — Missouri,
    18.47-50, 46.2-52.1 — Montana,
    19.41-57, 39.2-60.2 — Nebraska,
    20.55-43, 37.9-58.9 — Nevada,

    21.54-45, 36.3-62.5 — New Hampshire,
    22.57-42, 42.3-56.2 — New Jersey,
    23.57-42, 46.9-51.9 — New Mexico,
    24.62-37, 51.6-47.5 — New York,
    25.50-49, 41.7-58.0 — North Carolina,
    26.45-53, 43.0-56.0 — North Dakota,
    27.51-47, 44.2-55.0 — Ohio,
    28.57-41, 51.3-46.6 — Oregon,
    29.55-44, 48.4-50.7 — Pennsylvania,
    30.63-35, 55.6-43.9 — Rhode Island,

    31.45-54, 37.6-61.5 — South Carolina,
    32.42-57, 41.6-57.9 — Tennessee,
    33.44-55, 43.3-56.0 — Texas,
    34.34-63, 32.0-66.2 — Utah,
    35.67-32, 47.6-51.1 — Vermont,
    36.53-47, 39.2-59.7 — Virginia,
    37.58-41, 50.1-48.5 — Washington,
    38.56-43, 51.4-47.8 — Wisconsin,
    39.93-_7, 82.7-14.3 — District of Columbia

    States that voted more Republican

    1.39-61, 39.9-59.2 — Alabama,
    2.36-62, 36.3-59.6 — Alaska,
    3.39-59, 42.4-56.4 — Arkansas
    4.36-61, 36.0-62.1 — Idaho,
    5.41-57, 42.6-55.8 — Kansas,
    6.41-58, 43.9-55.5 — Kentucky,
    7.54-45, 54.7-44.5 — Iowa,
    8.40-59, 44.1-54.3 — Louisiana,
    9.34-66, 41.3-57.9 — Oklahoma,
    10.45-53, 46.5-52.9 — South Dakota,
    11.43-56, 52.2-47.5 — West Virginia,
    12.33-65, 38.0-60.5 — Wyoming

  6. Obama’s likely final vote tally would represent about 22.6% of the entire US population. That may not sound impressive, but it is. It would beat the existing record for a non-incumbent, set in 1952.

    (Nate presents the data.)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.