Divide and Conquer Strategy Failing?

Ron Brownstein argues that the Bush-Rove strategy of focusing on generating high turnout among Republicans even at the cost of alienating Democrats and moderates has had a price.

On balance, that equation worked for Bush in his first term. Bolstered by his post-9/11 glow, Bush inspired an enormous Republican turnout that spurred GOP congressional gains in 2002. In 2004, another Republican surge powered gains in Congress and Bush’s reelection over Democrat John F. Kerry. For Karl Rove and other top GOP strategists, those victories were evidence that Bush was building a narrow but stable electoral majority.

But even amid success, the limitations of the strategy were evident. Although Bush inspired passionate commitment from his supporters, he did not generate anywhere near the breadth of support that other two-term presidents, such as Dwight D. Eisenhower or Ronald Reagan, achieved at their apogee.

Eisenhower was a beloved war hero who could have had the nomination of either party. His presidency was before the television era, let alone the cynical media climate of the post-Vietnam/Watergate period. His average approval rating 65% and he ended his presidency with 59% approval.

Ronald Reagan, who was much more polarizing, had similar numbers: 64% and 57%, respectively.

Bill Clinton, who was actually impeached but rebounded to great popularity, was in that ballpark as well: 65% and 57%.

I don’t have those figures for Bush 43; the article I’m citing for the others was written right before his 2001 inauguration. If present trends continue, his final numbers will be awful as will his average. It’s noteworthy, though, that he had near-record approval ratings in the months after the 9-11 attacks and after the successful ouster of Saddam.

Bush’s margin of victory over Kerry, measured as a share of the popular vote, was the smallest ever for a reelected president. Even in the usual post-election honeymoon period, Bush’s approval rating never exceeded 55% in Gallup surveys, below the high point for every other reelected president since World War II. Bush’s support fell back beneath 50% even before his second inauguration.

Bush got a higher percentage of the popular vote than Bill Clinton did in 1996 (granted, in a three way race but one where the third party candidate mostly pulled votes away from his Republican opponent). Still, Brownstein is right about the margin of victory.

The casaulity here, though, is not established. Indeed, almost all observers would agree that the reason for Bush’s low numbers has more to do with the unpopularity of the war in Iraq than with polarizing domestic tactics.

All of this meant that even on Bush’s best days, nearly half the country opposed him and his direction. That didn’t leave him with much of a cushion for bad days, which have come in bunches during his second term.

Certainly true. Then again, he may well not have had a second term without doing all he could to maximize turnout among his natural constitutents.

While it seems increasingly less likely, a turnaround in Iraq would cause a rebound in Bush’s approval ratings. Conversely, if things continue to deteriorate in Iraq, no amount of reaching out and consensus building is going to give him much of a boost. Ultimately, results matter and he has staked his presidency on One Big Thing. So far, it appears to be a losing bet.

FILED UNDER: Congress, Uncategorized, US Constitution, , , , , , , , , , , , , ,
James Joyner
About James Joyner
James Joyner is Professor and Department Head of Security Studies at Marine Corps University's Command and Staff College and a nonresident senior fellow at the Scowcroft Center for Strategy and Security at the Atlantic Council. He's a former Army officer and Desert Storm vet. Views expressed here are his own. Follow James on Twitter @DrJJoyner.


  1. M1EK says:

    “in a three way race but one where the third party candidate mostly pulled votes away from his Republican opponent”

    This is a myth. The reliable figures from 1992 indicate that Perot voters came more from Clinton than from Bush (hard to find reliable figures for 1996); and we would have to suspend an awful lot of disbelief in order to buy the argument that in 1996, most Perot voters were Dole-second-choicers. Even so, even if every single vote for Perot had gone to Dole, Clinton still would have won.

    Perot took my vote from Clinton in both 1992 and 1996, in case you’re wondering.

  2. Derrick says:

    I wouldn’t blame the Iraq War for all of Bush’s problems. Take a look at Bush’s approval rating graph and you’ll see that after the 2004 election, the first real decline in his standing coincides with his push to reform Social Security. Also, his approval rating prior to 9/11 was pretty mediocre as well. I do think that his presidency won’t have any lasting currency precisely for the reasons Brownstein races. As a moderate Democrat who has raised the lever for more than a few Republicans, Bush and Rove’s hyper-partisanship alienates me in ways that few Republicans outside of the Strom Thurmands and Newt Gingrich’s ever have.

  3. Harry says:

    “Conversely, if things continue to deteriorate in Iraq, no amount of reaching out and consensus building is going to give him much of a boost.”

    The nature of Bush’s Presidency suggests pretty strongly that reaching out and consensus building are not things he is at all inclined to do. His White House tends to close itself off even from other Republicans, so opening up to Democrats is highly unlikely. If the Democrats do take over one or both houses, his last two years may be very frustrating for him and those who must work with him.

  4. legion says:

    This should not be surprising. The standard GOP strategy for some time now has been appealing to single-issue voters via hot-button issues… Not always terrorism, but also on religious issues (stem-cells, religion in schools, creationism, etc.), as well as the old canard of fiscal responsibility (SS reform, Medicare, deficit reduction).

    Unfortunately, those issues can only be harped on for so long… eventually the voters expect action on their issues, or they begin to feel used & betrayed and wind up just as virulently opposed to the group that just used them. The religious right is just now beginning to see this. There’s hints that the middle-class NASCAR dad/soccer mom crowd is seeing the same thing reflected in their lack of economic improvement vs. corporate America’s success, as well as the Iraq war.

  5. cian says:

    Everyone that comments here probably knows of at least one family that has been split apart by Iraq and the present administrations handling of that war. Again and again Bush and the republicans demonised the democrats and in so doing demonised millions of Americans.

    It may have consolidated their base, but most Families are tired, dispirited and fearful- not of the mad mullahs of republican imagination, but of the direction the country has taken and these families are only now beginning to find a common language through which they can start talking together about the last five years.

    Karl Rove, Cheney, the right wing talking heads and President Bush himself have failed to understand or even notice this change. Republicans in the field are hearing this new voice, and know its calling them out.

  6. Tano says:


    Minor point, but you got the presidential approval numbers mixed up a bit.

    It is true that Ike had an average rating of 65, and an end-of term rating of 59.

    But the Reagan and Clinton numbers were both higher for the end of term than for the average – i.e. Reagan and Clinton had identical averages – 57%, with Reagan leaving with a 64 and Clinton a 65.

    In other words, in your sentence:”Ronald Reagan, who was much more polarizing, had similar numbers: 64% and 57%, respectively.” – the word “respectively” is incorrect.

  7. Wayne says:

    First many of these polls still weight them with higher Dems than Reps. One I saw had Dems with 55% compare to 33% Rep. The only poll that count is election day which Bush beats Clinton on having a higher % votes compare to Clinton.

    One should remember that Dems candidates can’t do any wrong to their base while GOP expect their candidates to do everything the way they want it. Reminds me of high school sports where a terrible player get nothing but praises from their parent why the best kid on the team is constantly be ask to do better by their parent.
    I fall into the later category but I recognize who are the best players are.

  8. M1EK,

    The 1994 exit poll data shows that 67% of those who voted for Perot in 1992, voted for the republican in their congressional district in 1994. If you look at the 1992 election data and divided the Perot vote between Clinton (33.33%) and Bush (66.67%), you find that 12 states flipped (Colorado, Connecticut, Georgia, Iowa, Kentucky, Maine, Montana, Nevada, New Hampshire, New Jersey, Ohio, Wisconsin). These 12 states represent 106 electoral votes, which would have flipped the election. Now the fact that 2/3 who voted in 1994 for republican congress critters had voted for Perot in 1992 doesn’t guarantee that absent Perot Bush would have pulled all of those voters in 1992. But it does say that a better campaign by Bush Sr. would have potentially won them over as the GOP was able to win them two years latter.

    So while I don’t dispute your personal story, can you show what data are you using to say that it is a myth that Perot hurt Bush more than Clinton. Everything I have seen points that the majority of Perot’s votes came from those who otherwise would have supported the republicans.

  9. M1EK,

    I just ran a test on your statement about 1996 that Clinton would have won anyway if all Perot votes had gone for Bush. My looking at the 1996 data doesn’t support that statement. Simply adding the Perot vote totals to the Dole vote totals flips 10 states (Arizona, Florida, Kentucky, Missouri, Nevada, Ohio, Oregon, Pennsylvania, Tennessee, Wisconsin) for a total of 129 EV. This would have easily swung the election in 1996. While I think that the likelihood of Dole picking up 100% of Perot voters approaches zero, I believe that your statement is in error. It is true that Clinton would have pulled a larger percentage of the popular vote than Dole and Perot combined, but I hope you are aware that we elect our presidents based on electoral vote, not popular vote.

  10. Tano says:


    I read two rather exhaustive analyses of the ’92 vote, both of which reached the conclusion that the Perot vote (were he not in the race) would have broken ever so slightly for Clinton, rather than Bush (50+%, and 52+% – sorry I forget the fractions). So I think the notion that without Perot in the race, x number of states would have flipped for Bush is just fantasy.

    Looking at the 94 data is not justifiable – obviously there was movement away from the Dems between 92 and 94. If Perot voters went 67% to the GOP in 94, you can be certain that they went to the GOP far less so in 92.

    Clinton did appear to have won back more of the Perot vote than did the GOP in ’96 – Clinton’s vote when up 6 points, Dole up 3 over Bush, so maybe if Perot had not run in 96 his vote would have helped Dole – but it wouldnt have been enough.

  11. M1EK says:


    Tano’s addressed your points fairly well – there is, in fact, polling data available from 1992 referred to here:


    Again, 1996 data is harder to find, but here’s just one example:


  12. James Joyner says:

    Tano/Mike: I’m not asserting that Perot was the reason Clinton won in 1992 or 1996 (although there’s a very roundabout case to be made for 1992) My remarks above were entirely about 1996 and simply about the margin of victory.

    My sense of the Perot voter was that they were mostly anti-incumbent.

    In 1992, a large chunk of the electorate was dissatisfied with Bush but few outside the hard core Democrats trusted Clinton. Perot was an alternative and, once he dropped out and all-but-endorsed Clinton, a gateway.

    In 1996, Clinton was again quite polarizing but Dole was incredibly uninspiring as an alternative. Perot was a protest vote.

    The CNN polling is interesting. I’m not sure I buy it, though.

  13. M1EK,

    I can show you polls in June and October of 2004 that says Kerry would win in 2004. That can hardly be taken as proof that Kerry won. So I am less than impressed if that is all your data. Also, the 1996 poll is nothing compared to looking at the actual results of the election. That shows that going state by state and adding Perot’s vote total to Dole would flip 10 states and the election.

    I would be interested in seeing the “exhaustive analysis” you talk about but don’t link to. As I said in my comment, the 1994 break down is far from conclusive as to what happened in 1992. But I stand by the idea that it shows those voters were persuadable absent Perot. I wasn’t claiming that the be definitive, but just one data point that might suggest that Perot pulled more from Bush than Clinton.

    Of course looking at the broader picture, no democrat has gotten a majority of the presidential popular votes for the last 40 years except Carter (who got 50.08% in 1976). So for at least the last 40 years, the over whelming verdict of the voters at the presidential level has been “not the democrat” as expressed by votes cast for candidates other than the democrat. The republicans problem is that they have won 7 of the 10 presidential elections in the last 40 years, they have only been able to get a popular vote majority 5 of those 10 times (so you can say the result was “not the republican” 5 out of 10 times).

  14. M1EK says:


    When you combine the polls from 1992 (which, I suppose, now, you don’t argue with) with the fact that essentially all of the remaining Perot voters would have had to go for Dole to make a difference (not 75%, but 98%), it’s nonsensical to continue to argue that Perot cost Dole the ’96 election.

  15. Tano says:


    I guess there are all manner of ways of looking at things.

    Since you are talking about levels of popular sentiment, rather than electoral technicalities, then I think you have to acknowledge that the Dems won the 2000 election, in terms of getting more people in America to vote for their candidate. So it is not 7, but 6 out of 10 for the GOP. Or 4 out of the last 8. Or 1 out of the last 4.

    Since the relevance of these comparison fades the longer back you go, I prefer that last statistic.
    If you add up the Dem and GOP presidential vote for the last 4 elections, you come up with roughly 201 million for Dems, 190 for GOP. In other words, over the past 14 years, at the presidential level, the Dems have won an average of 51.5% of the vote, GOP about 48.5% (not counting minor parties).

    I guess the question really is this – was the reversal of that average number in the 04 election just a fluke, or a sign that maybe the GOP might be reemerging as a credible presidential party.

    (For the sake of your health, please do not look at any approval rate polls before attempting to answer that question).