Americans are usually forgiving when they vote a man into the White House and he wants a second term. Of the last eight elected presidents, all but two — George H. W. Bush and Jimmy Carter — got their four more years. Which is why the conventional wisdom long held that Barack Obama would most likely weather his midpresidency slump to win another term.
Then came the debt-ceiling debates of July and August, which seemed to crystallize Obama’s vulnerabilities in a way that even the Democrats’ midterm disaster of 2010 did not. It’s probably because he handled the situation so poorly, simultaneously managing to annoy his base, frustrate swing voters, concede a major policy victory to Republicans and — through the fear imported into the market by the brinksmanship in Congress and the credit-rating downgrade that followed — further imperil the economic recovery. On Aug. 12, a week and a half after the debate ended in Congress, Obama’s stock on Intrade, a popular political betting market, dipped below 50 percent for the first time. It has hovered just below the 50 percent threshold, usually at about 48 percent, ever since.
Obama has gone from a modest favorite to win re-election to, probably, a slight underdog. Let’s not oversell this. A couple of months of solid jobs reports, or the selection of a poor Republican opponent, would suffice to make him the favorite again.
Nevertheless, this is an unusual circumstance. Roosevelt, Eisenhower, Nixon, Reagan, Clinton and both Bushes all looked like the favorite to win a year in advance of their re-election battles, either having strong approval ratings or good-enough ones accompanied by robust economic numbers. When we look at the last eight elected presidents, only Carter faced a situation worse than Obama’s: approval ratings in the low 30s rather than low 40s, the likelihood rather than the mere possibility of a recession, a primary challenge rather than a clear path to renomination and a crisis in Iran rather than a string of foreign-policy victories.
The other seven had stronger fundamentals heading into the election year. This includes the elder Bush, who lost despite an extremely high approval rating and a disarrayed Democratic field (Mario Cuomo and others skipped the race at a time when Bush appeared unbeatable). It also includes Reagan and Clinton, who had grave problems early in their terms but who saw their numbers tick upward at the very moment in their presidencies that Obama’s have continued to slide. None of this is news to the White House. Following the debt-ceiling debate, Obama’s strategists started comparing their boss with the original comeback kid, Harry Truman — an implicit concession that the president will most likely fight 2012 from behind.
The fundamental flaw with much of what passes for political analysis is the tendency to think small. I blame soccer moms for this. Not the moms themselves — or their more rough-and-tumble companions from four years later, Nascar dads — but the narrow worldview that these demographic labels represent. The sophomoric strategist thinks he can slice the American electorate into a million little pieces and make it more than the sum of its parts. The smart one recognizes our common bonds.
A recent example of the flawed type of thinking came in September, after Democrats lost a special election to fill Anthony Weiner’s old seat in a Brooklyn district with a heavy concentration of Orthodox Jewish voters. There was much commentary about Obama’s popularity among Jewish Americans and about what this could mean — most importantly for his prospects in Florida.
Obama does indeed have a “Jewish problem.” Polls find that his standing among Jews has deteriorated: only about 54 percent of them approved of his performance in the most recent Gallup survey. But this is to be expected when a president has a 40-something approval rating. He also has a Hispanic problem and a problem among the white working class. He has a problem in Ohio and a problem in Florida and a problem in New Hampshire. He even has, to a mild extent, an African-American problem: Obama’s approval ratings among black voters are still high, but down to about 80 percent from 90 percent.
All of these, however, are symptoms of Obama’s larger problems, a set of three fundamental misgivings shared by much of the American electorate.
• First, many of us understand that Barack Obama inherited a terrible predicament. We have a degree of sympathy for the man. But we have concerns, which have been growing over time, about whether he’s up to the job.
• Second, most of us are gravely concerned about the economy. We’re not certain what should be done about it, but we’re frustrated.
• Third, enough of us are prepared to vote against Obama that he could easily lose. It doesn’t mean we will, but we might if the Republican represents a credible alternative and fits within the broad political mainstream.
Each of these factors, in turn, can be quantified.
• The first factor, Americans’ performance reviews of Obama, can be measured through his approval ratings.
• The second factor, economic performance, can be measured through statistics like G.D.P.
• The third factor — essentially, the ideological positioning of the Republican candidate — is sometimes thought of as an “intangible.” But it can be measured too, and it matters a great deal.
As we get closer to the election and more data become available, we can indulge — even encourage! — greater complexity in the analysis. Figuring out how Obama is performing in individual states and how this translates to the Electoral College, for instance, requires a fair amount of attention to detail. But it is premature to do that now. Instead we should think big, and focus on the three fundamentals.
A president’s approval rating at the beginning of his third year in office has historically had very little correlation to his eventual fate. In January 1983, Reagan had an approval rating of just 37 percent, but he won in a landslide. George H. W. Bush had a 79 percent approval rating in January 1991 and was soundly defeated. But voters start to think differently about a president over the course of his third year; they view him more on the basis of his performance and less on the hopes they had for him. These perceptions are sharpened by the beginning of the opposition party’s primary campaign, which, of course, accentuates the negatives.
A president’s approval rating toward the end of his third year, therefore, has been a decent (although imperfect) predictor of his chances of victory. Reagan saw his approval rating shoot up to 51 percent in November 1983 amid the V-shaped recovery from the recession of the previous year — the first sign that he was headed for a big win. Obama’s approval rating may have rebounded by a point or two from its lows after the debt-ceiling debacle — but not by much more than that. In late October, it ranged between 40 and 46 percent in different polls and averaged about 43 percent.
There have been two presidents stuck with similarly low approval ratings a year in advance of the next election. Gerald Ford had a 44 percent approval rating a year before his loss to Carter. Johnson had a 41 percent approval rating in November 1967, and although he was eligible for another term, he opted not to run. His vice president, Hubert H. Humphrey, did, and he lost.
In fact, since 1944 (when approval ratings first became reliable), there have been five cases in which the incumbent party’s president had an approval rating below 49 percent a year ahead of the election — as Obama almost certainly will, unless he finds the cure for cancer after our issue goes to print — and each time the incumbent party lost.
It would be possible, however, to draw some improper conclusions from this. The more robust way to analyze these data is not just to consider the winners and losers, but also their margins of victory or defeat. Ford and Humphrey (serving as Johnson’s surrogate) may have lost their elections, but not by much — Ford by just 2 points, and Humphrey by 1. The correct conclusion, then, is that other factors being equal, an approval rating in the low 40s a year before the election makes a president a slight but not overwhelming underdog.
If Obama’s approval ratings were in the low 30s or worse instead of the 40s, that would be another story. The three presidents to fit this description — Carter in 1979, George W. Bush in 2007 and Harry Truman in 1951 — saw their parties take big defeats the next year. It would also be another matter if Obama’s approval rating were closer to 50 percent. Of the six presidents to have approval ratings in that range a year ahead of the election — Nixon, Clinton, Bush the younger, Truman in 1947 and Reagan in 1983 and 1987 (with Bush the elder running as his proxy in the latter case) — all six won.
As you go farther up the approval-ratings chart, you begin to see exceptions to the rule: George H. W. Bush’s approval ratings were still high in late 1991, but he lost. So were Bill Clinton’s in 1999 and Dwight Eisenhower’s in 1959, but their coattails were not strong enough to carry their vice presidents to victory. Thus, approval ratings have been a useful indicator but not an infallible one.
The direction of the economy over a president’s fourth year is also a huge factor in his chances of re-election. Here, there is some good news and bad news for Obama.
The good news is that voters have short memories. If there are hopeful signs during an election year, they may be willing to forget earlier problems. Reagan, Nixon, Eisenhower and Truman all won despite recessions earlier in their terms. Moreover, voters’ evaluations of the economy are relatively forward-looking. Even if the economy is below its full productive capacity — as it was in November 1984 when the unemployment rate was 7.2 percent, and as it certainly was in 1936, when it was still around 17 percent — voters may be willing to overlook this, provided it seems headed in the right direction.
The bad news for Obama is that he has already missed his opportunity for a V-shaped recovery, and the prospects for a U-shaped recovery seem uncertain. In October, a panel of economists polled by The Wall Street Journal forecast 2.3 percent G.D.P. growth (adjusted for inflation) in 2012, somewhat below the election-year average of 3 or 4 percent and only enough to provide for modest job creation.
Growth rates during an election year are a good but imperfect indicator of electoral performance. The two times that economic activity actually shrank during an election year, 1980 and 2008, the incumbent party lost badly. The two times that it grew by more than 6 percent, 1944 and 1972, it won overwhelmingly. But Eisenhower won a landslide in 1956 despite tepid 1.8 percent growth, and George W. Bush won in 2004 with only 2.9 percent. The economy grew about 5 percent in 1968, but that wasn’t enough to save Humphrey.
Some political scientists have tried to explain these exceptions by resorting to an alphabet soup of economic indicators, conjuring obscure variables like R.D.P.I.P.C. (real disposable-personal-income per capita), which they claim can predict elections with remarkable accuracy. From the standpoint of responsible forecasting, this is a mistake. The government tracks literally 39,000 economic indicators each year. Although many (say, privately owned housing starts in Alabama) are obscure or redundant, perhaps two or three dozen of them are looked at regularly by economists.
When you have this much data to sort through but only 17 elections since 1944 to test them upon, some indicators will perform superficially better based on chance alone, the statistical equivalent of the lucky monkey from a group of millions who banged out a few Shakespearean phrases on his typewriter. Conversely, indicators like the unemployment rate have historically had almost no correlation with election results despite their self-evident importance. The advantage of looking at G.D.P. is that it represents the broadest overall evaluation of economic activity in the United States.
There is, however, another problem: economic forecasts are not very good. In fact, they are completely terrible. In November 1995, economists expected the economy to grow at 2.6 percent the next year; it actually zoomed upward by 4.4 percent. In November 2007, they expected it to grow at 2.5 percent, but it shrank by 3.3 percent, as the effects of the global financial crisis became manifest. Frighteningly enough, the margin of error on an economic forecast made a year in advance is about plus or minus 4 percent of G.D.P. Advance forecasts of election results must account for this uncertainty, either by expanding their own margins of error to accommodate it or by making their predictions conditional upon different economic situations (we will pursue the latter approach).
The other major unknown is the identity of the Republican nominee. This year it could make an unusually large amount of difference. In recent polls against Mitt Romney, Obama’s lead is just 1 percentage point — but he leads Herman Cain by 8, Rick Perry by 11 and Michele Bachmann by 14.
At some point early next summer, head-to-head polls like these will become the most important predictors of the November result. But this far out they can be misleading and inaccurate. In a survey conducted in late September 1983, Ronald Reagan actually trailed Walter Mondale by 2 points, and in October 1991, George H. W. Bush led Bill Clinton 55 to 20.
In this case, though, the relative standing of the Republican opponents against Obama aligns neatly with their perceived ideology. Obama does worse against more moderate candidates like Romney — and better against staunch conservatives like Perry and Cain.
Political scientists have long debated between two theoretical constructs of how Americans vote: one might be called the referendum model, and the other might be called the median voter model. In the referendum model, only the performance of the incumbent party’s candidate matters: the opponents could nominate Lyndon H. LaRouche or Jesus H. Christ, and it wouldn’t make much difference. In the median voter model, it is only the ideological positioning of the candidates that matters, and presidents are not evaluated on performance at all. As long as Jimmy Carter’s political positions were closer to the center of the electorate than Ronald Reagan’s (they probably were in 1980), who cared about an inflation rate of 12 percent or the hostage crisis in Iran?
Each of these models is ridiculous when taken to its logical extreme. But the evidence points on balance toward the referendum model (Carter lost in 1980, for instance). That is why our view of the fundamentals focuses on two performance-based indicators for the incumbent president — approval ratings and economic growth — but only one variable related to the opposition candidate.
Still, that one variable is an important one. In their book “The Party Decides,” the political scientist Marty Cohen and his colleagues estimated the ideological positioning of past opposition-party nominees based on a combination of objective indicators like Congressional voting records and surveys of presidential historians. I have translated Cohen’s estimates to a scale that runs from 0 for an extremely moderate nominee to 100 for an extremely liberal or conservative one, with 50 representing the average.
Historically, this indicator has had a good deal of predictive power. When the incumbent party faced an opposition candidate with an extremism rating of 50 or higher, it won re-election in six out of eight cases. When it faced one with a rating of 50 or lower, meaning a more moderate nominee, it won just three times out of nine.
As with our other metrics, this one can be outweighed when the other fundamentals are not aligned with it. The moderate Thomas E. Dewey (who scored an 18) lost not once but twice. The very conservative Reagan (who came in at 85) won in 1980 because voters could find too few positives in Jimmy Carter’s record. Electability exists along a spectrum and not in absolutes.
Still, if the performance-based factors are unkind to Obama, this one is potentially more favorable. I have estimated extremism scores for this year’s Republican candidates by combining data from the three principal objective methods that are used to estimate ideology, one based on Congressional voting, one based on fund-raising contributions and the other based on voters’ assessments of the candidates’ ideology in polls. I then placed the Republicans onto the scale based on how their figures compared with past candidates. Here’s how they stacked up:
Jon Huntsman 40
Mitt Romney 49
Herman Cain 60
Gary Johnson 63
Rick Santorum 64
Rick Perry 67
Newt Gingrich 68
Michele Bachmann 83
Ron Paul 96
As you can see, Romney’s score of 49 is to the left of every Republican candidate except for Huntsman. But the G.O.P. as a whole has moved to the right, so Romney is about average in the broader scope of history.
The bigger problem for Republicans might come if they nominate a candidate further to Romney’s right, like Perry, who has an extremism score of 67. The difference between Romney and Perry amounts to about 4 percentage points at the ballot booth. If conditions were otherwise very favorable (or very unfavorable) to Republicans, this wouldn’t be enough to matter: an election that Romney would win by 10 points, Perry might win by 6. But an election that Romney might win by just 2 or 3 points, Perry could easily lose.
A Three-Factor Model
By combining these factors — approval ratings in the year before the election, G.D.P. growth during the election year itself and the ideology score of the opposition candidate — we can come up with a forecast of next year’s election. By design, it is not an exceptionally precise forecast. There are all types of factors that the model does not explicitly consider, among them the possibility of third-party candidates or differences between the popular vote and the Electoral College. Moreover, voter perceptions about the economy, or the ideological positioning of the candidates, may differ in practice from what the objective data say about them. Rather than pretending to have all the answers, the model knows how much it doesn’t know and allows for a reasonably wide range of possible outcomes.
The model’s error can be reduced, however, if we are willing to consider certain possibilities that fill in some of the blanks. We will evaluate four such scenarios here. They represent cases in which Obama is matched up against either a relatively moderate nominee like Romney or a more conservative one like Perry (who will be used as our guinea pig despite Herman Cain’s recent rise in the polls). Further, they represent cases in which the economy either slips into recession or instead beats expectations and grows by 4 percent next year (each has about a one-in-three chance of happening). A multitude of other situations are possible — but these are the paradigmatic cases. And they produce vastly different results.
Case Study No. 1: Romney and Stagnant Economy
Obama approval rating in November 2011: 43%
G.D.P. growth in 2012: 0%
Probability of winning the popular vote: Romney: 83%, Obama: 17%
We begin with the worst of these situations for Obama: Mitt Romney is the Republican nominee, and economic growth, rather than continuing along sluggishly, comes to a halt (perhaps the debt dominoes have fallen in Europe). Under these assumptions, Obama would only have a 17 percent chance — about one in six — of winning a majority of the popular vote.
His chances are slim enough in this case that if I woke up next November to discover that we would have four more years of Obama, I might ask whether there was some sort of October surprise: “Mitt in Torrid Affair With Filipina Housekeeper.” Subhead: “Illegal Immigrant Got Free Romneycare.” Then I might ask if Sarah Palin had run on the Tea Party ballot line and taken 6 percent of the vote.
In a more normal election, an Obama victory would have resulted from some combination of three factors. The first two are arguments that we might expect to see as part of Obama’s campaign, and which are viable even under this worst-case scenario. The third is simply an observation.
First, Obama may have some untapped potential in highlighting his foreign-policy wins. The killing of Osama bin Laden did not produce a lasting bounce in the polls. But — sorry to be so gauche — killing the world’s most wanted terrorist will make for a nice sound bite. Meanwhile, the NATO intervention in Libya reached a successful if bloody outcome, and the troops are coming home from Iraq. The White House has not done a good job of turning the discussion to foreign policy, and it has to be careful that in doing so it does not appear to be ignoring the economy. But foreign policy matters to voters, and Obama will have a marketing budget of hundreds of millions of dollars to get the message right.
Second, the overall unpopularity of the Republican Party could rub off on Romney, who will have to endorse most of its platform. Only 35 or 40 percent of voters take a favorable view of the G.O.P., which has not improved at all from 2006 or 2008. The Republicans in Congress are less popular still and have an agenda with some prickly bits — like Paul Ryan’s plans to privatize Social Security — that Romney could have a hard time distancing himself from.
All of this, probably, would not be enough to save Obama. Romney is a skilled debater, a cautious candidate and a smart man — exactly what you want when the fundamentals otherwise favor you. Under these conditions, in fact, Obama would be as likely to lose by double digits as to win the popular vote.
Still, the White House would be able to make some credible arguments and Obama might have an outside chance at an upset. But how does it get around the problem of the economy?
The observation: even if the economy goes into recession again, voters might not care much about the difference between slow growth and no growth. Some 80 percent of voters already think we’re in a recession. They are not happy about it, but expectations are so low that a sort of fatigue may have set in about further bad news.
In practice, voters may think about the economy as falling into one of three basic categories — Good, Bad and Getting Better — rather than along a continuum. Obama would benefit if he could make a credible case for Getting Better, something he would not be able to do in this situation. But since he’s already unable to make that case now — remember “Recovery Summer”? — it’s plausible that a deterioration in the numbers would not hurt him as much as an acceleration of growth might help him. Beating Romney with 0 percent growth would not be easy, but it might not be that much more difficult than beating him with 2 percent growth (also no piece of cake, of course).
Case Study No. 2: Romney and Improving Economy
Obama approval rating in November 2011: 43%
G.D.P. growth in 2012: 4%
Probability of winning the popular vote: Romney: 40%, Obama: 60%
Obama would be far better off if he could make the Getting Better case. Imagine, as before, that Romney is the nominee. But rather than going into recession, the economy grows by 4 percent next year, enough to make a real dent in the unemployment rate. This would be enough to make Obama the favorite.
But not by all that much: he’d have only about a 60/40 edge.
Why not larger? The key to understanding this one is that Obama has a lot of gravity to overcome. Voters usually put their earlier concerns aside if there is an improvement in the economic fundamentals in the election year. But there have been exceptions: growth was quite strong in 1992, but voters were still punishing George H. W. Bush for the 1990-91 recession and the jobless recovery it produced.
Romney is much different stylistically from Bush’s opponent, Bill Clinton, but both are skilled at driving an economic message. Romney would bring out his PowerPoint and seek to persuade voters that the growth had been too little and too late. After all, if killing bin Laden couldn’t lift Obama’s approval rating much above 50 percent, who knows whether one year of good-but-not-great growth would?
Still, the most likely eventuality in this case — enough economic growth that the White House gets to make the Getting Better case while maintaining a straight face — is a narrow win for Obama. Perhaps it would be somewhat like the one that George W. Bush secured in 2004: it would keep the network anchors up late, but it wouldn’t be close enough to put us in Recount Land.
Case Study No. 3: Perry and Improving Economy
Obama approval rating in November 2011: 43%
G.D.P. growth in 2012: 4.0%
Probability of winning the popular vote: Perry: 17%, Obama: 83%
Imagine Rick Perry as the Republican nominee instead, and Obama has considerably more margin for error. Against Perry, should economic growth improve to 4 percent next year, Obama would be about an 83 percent favorite. This is the most favorable arrangement for Obama — good enough that you might even hear talk about Obama expanding the map again (can he win Georgia? Arizona?).
The usual danger to the favorite in these cases is complacency, something that befell Gore in 2000 and Dewey in 1948. Perry, if he wouldn’t go over well in high-income and socially liberal states like Colorado or Virginia, could make a credible-enough case about his working-class roots and his jobs record in Texas to appeal to voters in hardscrabble states like Ohio and Pennsylvania. With an improving economy, it would be a narrow path to victory, especially given that Florida — where Perry’s position on Social Security would not play well — might be a difficult sale for him.
But Obama could still go down if he appeared out of touch. This particularly holds if he were too quick to declare “mission accomplished” on the economy, or if the Democratic base remained as disinterested in the election as it was in 2010. No one should mistake the Occupy protests, on Wall Street and elsewhere, for a movement of the Democratic base (one poll found that most of the Wall Street protesters identify as independent). But you sometimes get the sense that the left has already gotten over Obama and is focusing on longer-term problems. This may be good for them, but it isn’t necessarily good for Obama.
And then there is the October surprise potential: “Obama Sent Crotch Pics to Congressional Staffer.” Subhead: “Who Used to Be Lobbyist for Solyndra.”
Case Study No. 4: Perry and Stagnant Economy
Obama approval rating in November 2011: 43%
G.D.P. growth in 2012: 0%
Probability of winning the popular vote: Perry: 59%, Obama: 41%
Perhaps the most interesting case is the one in which Perry (or Cain) is the nominee but the economy gets worse and falls back into recession. This would be a true test of the two political-science perspectives on the election: the referendum paradigm, which would favor Perry, against the median voter paradigm, which would favor Obama.
Our model suggests that the referendum paradigm is slightly more likely to win out, as it did under somewhat similar circumstances in 1980. Specifically, it suggests there is about a 60 percent chance that Perry would beat Obama under these conditions, as Reagan beat Carter.
True, Rick Perry is no Ronald Reagan. While they have some biographical similarities (multiterm governors of major states; movie-star good looks; former Democrats) and some ideological similarities, Perry is much more likely to make the mistake of turning the race into a popularity contest rather than asking voters, as Reagan did, whether they were better off than they had been four years earlier. But Reagan won by 10 points, which means that even if Perry has disadvantages relative to Reagan (and Obama has advantages relative to Carter), they might not be enough to overcome the weight of a sagging economy.
So Perry would be a favorite under these conditions — but only a modest one. Republican strategists like Karl Rove are sensible in their desire not to nominate him; he might provide the clearest case in American history of a party blowing an election it by all rights should have won.
The Bottom Line
Average these four scenarios together and the probabilities come out to almost exactly 50-50. A month or two ago, when Perry and Romney appeared about equally likely to be the Republican nominee, it would therefore have been proper to think of the election as a toss-up.
With Perry having slumped in the polls, however, and Romney the more likely nominee, the odds tilt slightly toward Obama joining the list of one-termers. It is early, and almost no matter what, the election will be a losable one for Republicans. But Obama’s position is tenuous enough that it might not be a winnable one for him.