Monday, March 23, 2020

Four Reasons to Be Cautious About Trump's Approval Ratings

Ever since the COVID-19 coronavirus crisis began to accelerate a couple of weeks ago, political obsessives have wondered about the likely effect that it would have on President Trump's job approval ratings—and, by extension, his chances of re-election in November. The few surveys that have been conducted over the past week or two don't show any dramatic movement—the FiveThirtyEight model currently estimates Trump's national approval at 43 percent, which is more or less where he's been since the end of the government shutdown in February 2019. But the lack of any apparent decline, combined with recent surveys that showed more respondents endorsing Trump's handling of the coronavirus outbreak than disapproving so far, have inspired some analysts to argue that the president may actually be benefiting politically from the crisis at the expense of the Democratic opposition.

Perhaps that's true. But realistically, it's far too soon to glean much about either the American public's ultimate response to Trump's management of the pandemic or its implications for the upcoming election. Here are four good reasons to exercise some patience before jumping to conclusions:

1. Political leaders' popularity often rises temporarily after the onset of a crisis. Political scientists call this pattern the "rally effect," and it's been documented many times over decades of history; most dramatically, George W. Bush's job approval shot up from about 50 percent to about 90 percent virtually overnight after the events of September 11, 2001. There are several plausible factors contributing to this phenomenon: citizens close psychological ranks around their national leaders in a moment of uncertainty and fear; they evaluate these figures on different criteria than they did before the crisis erupted; and the normally critical opposition party (sometimes) mutes its attacks on the incumbent. Both French president Emmanuel Macron and Italian prime minister Giuseppe Conte have enjoyed spikes in popularity during the current coronavirus outbreak, despite (especially in Italy's case) substantial national dislocation and tragedy.

But these popularity bumps fade with time. Either the crisis is soon resolved and citizens turn their attention to other things, or it is not, in which case they start to grow impatient with the effectiveness of their leader. The 2020 general election is still far enough away that even if Trump were to benefit from the rally effect in the short term, it wouldn't be a very reliable signal of his popularity more than seven months in the future.

2. Americans are still learning about the severity and likely duration of this crisis. National journalists are closely following each development of the coronavirus outbreak; most also live in places like New York and Washington where daily life has already been dramatically changed. But large sections of the country aren't as strongly affected so far, and the less obsessively attentive typical resident of middle America will not have experienced the same degree of disruption. Many Americans are presumably still unaware of the probable length of time before things get visibly better, much less return to normal—for example, while it seems quite apparent based on the trajectory of infection that many public school systems that are now closed are unlikely to reopen before the end of the academic year three months from now, few state authorities have yet acknowledged as much in public. Citizens who anticipate that the crisis will only last a matter of a few weeks may see little reason at this point to re-evaluate their opinions about the president, but they may start to feel differently if the inconvenience persists for longer than they first assumed.

3. Americans already have strong opinions about Trump, and most of them disliked him before the crisis. Trump has never consistently exceeded 50 percent job approval as president even during three previous years of relative peace and prosperity. His approval rating among Democrats has seldom reached double digits—it stood at 7 percent in the last Gallup survey—and his approval among independents (as measured by Gallup) has generally stayed between 35 and 40 percent for most of his presidency. The only way that his popularity could fall much further would be for elements of his remaining base—consisting almost entirely of habitual Republican voters—to become disenchanted with his performance, but these citizens are unusually resistant to changing their minds about him. Their partisan alignment means that they are already predisposed to support a Republican president, they are prone to discount criticism from Democrats and mainstream journalists, and they are disproportionately exposed to Trump-friendly messages from conservative media sources.

The good news for Trump is that he may be spared significant erosion in his popularity by the strong loyalty of his fellow partisans. But a steady approval rating could also be a problem, because he's already in a vulnerable position heading into re-election and is consistently running behind likely opponent Joe Biden in national polls. Even if the crisis were merely damaging enough to prevent Trump from boosting his public support by November, that itself might turn out to be a decisive factor in the election.

4. It's not the virus, it's the economy (stupid). The most worrying component of this crisis for the Trump administration and re-election campaign isn't the spread of COVID-19 itself and the casualties that it leaves in its wake. Instead, it's the larger impact on the economy. While the specific quantitative estimates of current forecasting models should be treated with skepticism in such an unprecedented and fast-moving environment, it seems inevitable that there will be a sudden and catastrophic economic shock that will at least temporarily push the U.S. into a recession.

Trump and his supporters will argue, with justification, that it would be unfair to blame him for the economic misery that a worldwide pandemic is poised to inflict on the nation. But voters tend to reward presidents for good economic times and punish them for bad times regardless of the incumbent's actual responsibility for either outcome. It wouldn't be surprising if Trump's popularity remained stable or even rose a bit over the next few weeks during (presumably) the strictest anti-COVID remediation measures, only to decline later in the year once the larger consequences, especially declining income and employment, became more visible to average citizens. He therefore has every reason to wish not only for a shallower economic plunge than most analysts now foresee, but also for a historically rapid national rebound once the worst is over.

Saturday, March 21, 2020

Nelson W. Polsby's Analysis of Presidential Nominations Still Applies After 37 Years

Today at the Washington Post's Monkey Cage blog, I reflect on the continued relevance of Nelson W. Polsby's 1983 book Consequences of Party Reform for the events of the last several weeks. The media-driven ups and downs of the Democratic presidential candidates in the 2020 election illustrate the enduring insights of Polsby's research, as does his newly-resonant concern that excluding party leaders from playing a central role in selecting nominees increases the risk of electing a president who is unable to meet the expectations and challenges of the office.

Wednesday, March 11, 2020

March 10 Primary Review: From "Contested Convention!" to "Over Already?!?"

It wasn't all that long ago that the prospect of a contested Democratic convention was every savvy Washingtonian's favorite topic of conversation. Traditional political journalists, who are habitually bored and irritated by the smoothly stage-managed and substantively anticlimactic nature of modern conventions, seem to envy the days when their forerunners like H. L. Mencken witnessed the excitement of a political party actually resolving its nomination contest in a blur of stem-winding speeches, repeated roll-call votes, and dark-horse bandwagons. But their predictable musings, usually accompanied by ill-disguised hope, at the beginning of every nomination season that the imminent primary race could easily produce such a thrilling grand finale received reinforcement this year from the ranks of the cutting-edge quantitative analysts. According to the FiveThirtyEight forecasting model, the probability of no single Democratic candidate receiving a majority of delegates spiked after this year's Iowa caucus, briefly rising well over the 50 percent mark prior to Super Tuesday.

Of course, that all seems rather quaint now. Joe Biden's remarkable resurgence, beginning with his modest second-place finish in the Nevada caucus and picking up rocket fuel the following week in South Carolina, extended through another week's worth of elections this Tuesday night. Barely a month after his damaging losses in Iowa and New Hampshire seemed to bring his campaign to the edge of a cliff, Biden is poised to assume the mantle of presumptive Democratic nominee.

What's more, the procedural mechanism that previously made a contested Democratic convention seem so very possible—the party's requirement that delegates must be allocated in proportion to the popular vote in individual states and congressional districts—is the same thing that will now help Biden's campaign make the case that the race is effectively over. At the beginning of the primary sequence, with so many active candidates dividing voter support, it's only natural to suppose that no single contender may be able to attract an outright national majority of delegates. Republican candidates can quickly amass a strong advantage even without a popular majority by placing first in multiple states that award delegates in a winner-take-all fashion, but national party rules close off that path for Democrats.

At the current stage of the nomination sequence, however, proportional allocation becomes a Democratic front-runner's best friend. Biden's lead in the delegate count, though it may not appear numerically lopsided at first glance, has in fact become sufficiently large that only a series of crushing defeats in state after state would seriously endanger it. A Republican candidate in Sanders's position could conceivably ride victories in multiple winner-take-all states to make up ground quickly in the back half of the primary calendar, but the Democratic Party's proportionality requirement dramatically blunts the impact of upsets late in the process. Both Hillary Clinton in 2008 and Bernie Sanders in 2016 notched some notable wins in populous states during the final phases of the primary season, but they had both already fallen far enough behind their opponents that anything short of repeated landslides wasn't going to throw the final outcome into doubt.

And so, like in many other years, the 2020 presidential nomination system has quickly transformed apparent chaos and uncertainty into order and predictability. There are still more states to vote, more questions to resolve, and more strategic choices ahead for both Biden and Sanders. But unless a truly extraordinary disruption occurs in the race, the process has now foreclosed any other outcome than the one that now stands before us.

Wednesday, March 04, 2020

Super Tuesday Review: Biden Back in Front

The vote-counting from Tuesday's elections will continue not only through the night but also, in California anyway, for a couple of weeks to come. Yet the overall picture is relatively clear. Joe Biden appears to be the winner in ten states (Alabama, Arkansas, Maine, Massachusetts, Minnesota, North Carolina, Oklahoma, Tennessee, Texas, and Virginia). Bernie Sanders has placed first in four states (California, Colorado, Utah, and Vermont). Mike Bloomberg won the caucus in American Samoa, and Elizabeth Warren finished no better than third in every state or territory—including her home state of Massachusetts.

Here are some of the most important implications of the Super Tuesday results:

1. All of a sudden, Biden is once again at the front of the Democratic race, in what might be the most dramatic apparent comeback in the modern history of nominations (no candidate has previously survived finishing fourth in both Iowa and New Hampshire to vault back into the lead). The Biden surge of the past week was spread across the entire nation, and only the prevalence of early and mail voting in California, Texas, and Colorado kept him from amassing a near-decisive lead in delegates.

2. We won't know the final results in California for a while, and it's possible that Sanders did well enough there to keep the total Super Tuesday delegate margin between him and Biden from becoming too lopsided. But the most damaging result for Sanders on Tuesday wasn't the delegate count—it was his unexpected losses in Massachusetts and Minnesota. Beginning with next week's primaries in Michigan and Missouri, many of the key states in the post-Super Tuesday phase of the nomination calendar are urban states in the Northeast and Midwest, including New York, Pennsylvania, Ohio, and Illinois. Had Sanders been able to confine Biden's victories on Tuesday to the string of states from Virginia to Texas, he could have argued that Biden's strength was mostly regional and unlikely to endure once the contest moved northward. But Sanders's losses in two medium-sized, mostly white urban states where Biden didn't even campaign or spend money are much more troubling omens for his candidacy.

3. These results show why a truly contested convention—despite dominating media speculation in the early stages of every election season—is unlikely to occur except under very unusual circumstances. Nearly always, nomination contests naturally narrow down fairly quickly to one or two viable candidates; it's very hard for three candidacies to sustain themselves through the entire gauntlet of primaries. With only two contenders (at most) left standing by the end of the schedule, one or the other can count on a first-ballot majority at the convention, even if the arithmetic technically requires a bloc of delegates previously pledged to withdrawn candidates to pitch in enough votes to put the leading delegate-winner over the top. (Before a recent round of rules changes barred their participation on the first ballot, superdelegates could also perform this service, as they did for Barack Obama in 2008.) Democratic voters in the first 18 states have reduced what was once a large field of candidates to two plausible remaining options—Biden and Sanders—and the role of Democrats in the remaining 32 states is to determine which of these two will be the nominee.

4. Biden's now the favorite once again, but Sanders is by no means out of the running. More twists and turns are still quite possible, if not likely. But this is usually the kind of defeat that compels a candidate to make adjustments: tweak the campaign message, revise the strategy, target a new constituency. A key question hanging over the rest of this race is whether Sanders, who prides himself on his consistency, will rethink his approach or simply plow ahead on his current path.

5. A lot of people seem to have drawn the conclusion from the last few days that campaign ads and field organizing have become fairly meaningless in modern elections, since Biden engineered his historic surge while being massively outspent and out-organized by Sanders and Bloomberg. The reality is probably more complicated. It's certainly true that national media and social media are more important factors in the nomination horse race, and local organizations less important factors, than they once were. But Biden also has a unique advantage: everybody already knows who he is, and Democrats already have positive views of him, so television ads and campaign mailers are much less necessary to boost his name recognition or get his message out than would be true for other candidates.

In fact, it's very possible that Biden's lack of money and organizational capacity severely damaged and almost ended his candidacy in Iowa and New Hampshire—especially in Iowa, where the caucus system rewards candidates who have the infrastructure to identify supporters, drag them to the caucus meetings, and keep them there until the voting is complete. Similarly, while Bloomberg's money wasn't sufficient to deliver him the nomination, one glance at the Super Tuesday results is enough to confirm that he was able to buy himself a significant, though ultimately insufficient, amount of popular support simply by spending at unprecedented rates.

Rather than decisively declaring one factor "the real story" and other factors "worthless," we analysts should acknowledge the extraordinary complexity of multi-candidate nomination contests. It can be tempting to declare Biden's comeback inevitable now that it's happened, but nominations are much less predictable and more contingent than that. All of us are students rather than masters of this subject, and the unusual events of the past few days have shown how much there always is to learn.

Sunday, March 01, 2020

South Carolina Primary Review: Biden's Back in the Game, But He's Not Winning Yet

The one silver lining in Joe Biden's weaker-than-expected performances in Iowa and New Hampshire was that they gave him the chance for a "comeback" in South Carolina, where he retained a potential reservoir of support among the black and moderate white voters who dominate the state's Democratic electorate. Biden's advantage was temporarily shaken—several polls after New Hampshire showed his lead in South Carolina narrowing to single digits—but a combination of his second place finish in Nevada last Saturday, sharper-than-usual debate performance on Tuesday, and key endorsement from veteran congressman Jim Clyburn, the House Majority Whip, on Wednesday helped propel him to a victory of nearly 30 points.

It's likely that Biden will benefit from a few days of very positive media coverage heading into Super Tuesday, and that a number of elected Democrats will rally around him as the most viable remaining alternative to Bernie Sanders. But Biden still has a ways to go before he reclaims his position as front-runner in the race. Sanders will probably win a decisive victory in California on Tuesday, where much of the vote has already been cast by mail and is thus insensitive to a Biden surge, that may alone provide him with a significant lead in the national delegate count. It's also unclear what proportion of any last-minute decline in support for Warren, Buttigieg, or Bloomberg will migrate to Biden, and how much will be transferred to Sanders instead.

It does seem as if the Democratic contest is quickly heading toward a showdown between Sanders and Biden, though polls suggest that Bloomberg is poised to accumulate a chunk of delegates on Super Tuesday that could conceivably matter to the final outcome. (A two-candidate race would provide some clarity to the question of whether there will be a contested convention, since one or the other will have a majority of delegates.) But can Biden, who suffers from a much smaller war chest and weaker campaign organization than one would expect a two-term vice president to have, actually keep up with Sanders once the election calendar accelerates from Tuesday onward? The answer to that question may well hold the key to the nomination.