The Politics Of Disasters

Throughout the first 200-plus days of Donald Trump’s presidency, it’s been common for analysts to say he is struggling through sub-40% approval ratings despite not having to reckon with a major non-scandal crisis. Whether that was true before last weekend is debatable -- do North Korea’s provocations count? -- but it’s almost certainly not true now after Hurricane Harvey struck Houston and southeast Texas.

The immediate physical and emotional impact of Harvey is clear: catastrophic flooding, billions of dollars in damage, thousands of displaced persons, and multiple deaths. Some areas of Texas received more than 50 inches of rain, a figure that is difficult to fathom.

But what is uncertain is the political fallout -- or lack thereof. Congress will likely face decisions regarding emergency funding appropriations, which could create political conflict. Trump attracted criticism for his tweeting during the storm, which included a random book plug, and for pardoning controversial former Maricopa County (Phoenix) Sheriff Joe Arpaio (R) just as the hurricane was hitting Texas. The pardon upset not only Democrats but also many Republicans, such as Arizona’s two senators and Speaker of the House Paul Ryan of Wisconsin. Additionally, Trump’s decision to go ahead with a ban on transgender military service was revealed as Harvey struck on Friday. In the past few days, Trump has talked about unity in the face of the storm, but whatever one thinks of the transgender ban and the Arpaio pardon, the president made two highly partisan, divisive announcements in the midst of the storm, actions that make it hard for him to credibly call for harmony.

How the public judges Trump over the government response to Harvey, and whether it will be a key moment in his presidency, remains to be seen. And with the storm still raging, the jury’s out on the quality of the response to a horrific event, including whether Congress can provide sufficient funding to deal with its aftermath in a bipartisan way.

The president has actively tried to make it appear as though he is on top of the situation. The White House press team has been putting out a series of updates including photos of Trump holding conference calls with top officials. Trump also visited Texas on Tuesday, again giving off the impression that he cares and is in control. We’re emphasizing what public relations professionals might call “optics” here because this is really all any White House can reasonably control in a situation like this. The administration wants to prevent the public from feeling like Trump doesn’t sufficiently care or is out of touch -- the latter danger exemplified in moments like then-President George W. Bush surveying the wreckage of Hurricane Katrina from Air Force One in 2005, which as NPR’s Domenico Montanaro recently recalled, “made him look aloof -- even though, as his White House pointed out, Bush had good reason not to land Air Force One, and the president wanted an aerial view to capture the full magnitude of the devastation.”

For better or for worse, the obvious point of comparison is the catastrophic aftermath of Katrina in 2005, a disaster whose political impact merits further examination. That storm, the costliest natural disaster in U.S. history, made its second landfall in Louisiana on Aug. 29, 2005, which caused numerous breaches in New Orleans’ levee system and flooded most of the city. Bush and his administration eventually came under fire for their response to the crisis, as did state and local officials in Louisiana.

But how much did the Katrina fiasco hurt Bush in the eyes of Americans? Charles Franklin, who runs the Marquette Law School Poll, recently recapped an analysis he undertook in 2006 to gauge how the storm affected Bush’s approval rating. Overall, Franklin found that “the immediate impact of Katrina was a loss of 1.4 percentage points in President Bush’s job approval rating.” While 1.4 points sounds insignificant, Franklin noted that Bush’s approval had been falling at a rate of about one point per month prior to Katrina, meaning that a 1.4-point drop was the equivalent of about one-and-a-half months of approval decline. So Bush’s approval took a small but notable hit. Still, the rate of approval deterioration did not increase in the long run. With an eye on Harvey, Franklin concludes his recap by pointing out that “the lesson of Katrina is the impact may be smaller than might be expected.”

Hurricane Katrina also has to be placed in the context of other events that were going on in 2005-06. Most notably, Bush had invested a significant amount of political capital in an unpopular and unsuccessful attempt to change Social Security, and the Iraq war was becoming even more divisive. Around the time of Katrina, the public was about evenly divided over whether the U.S. had made the “right decision” to use military force in Iraq and regarding how well the conflict was going. And the polling trends were generally moving toward more unfavorable views of the war. Richard Eichenberg of Tufts University, Richard Stoll of Rice University, and Matthew Lebo of Stony Brook University authored a 2006 paper that attempted to model Bush’s job approval from 2001 to 2006 while accounting for a number of variables, including economic conditions, American battle deaths in Iraq, and major events such as Hurricane Katrina. Modeling for the entire period and the “war weeks” (from early 2003 to early 2006, the end of analysis for the article), in only the latter did they find a significant, negative coefficient for the Katrina variable as it related to Bush’s approval. In general, this finding could be said to echo Franklin’s in that it suggests a negative effect, but one that indicates Katrina did not hugely influence Bush’s approval rating on its own.

Based on their work in a 2008 article, Neil Malhotra and Alexander Kuo of Stanford University (Kuo is now at Cornell University) suggest that partisanship can affect the public’s view of who is to blame for a poor governmental response to a crisis. Using a set of seven actors from the Katrina disaster -- including Bush, Louisiana Gov. Kathleen Blanco (D), New Orleans Mayor Ray Nagin (D), and Bush-appointed FEMA director Michael Brown -- the authors found that partisanship affected who people targeted for blame, with Democrats and Republicans differing significantly on the question of who was “most to blame.” Their study found two-thirds of Democrats blamed Bush the most while a plurality of Republicans (35%) blamed Mayor Nagin the most. But the authors noted that “a surprisingly large 21.6% of Republicans found President Bush most to blame. This may help to explain the deterioration of Bush’s approval among conservatives and moderates post-Katrina.” On the one hand, this suggests Trump could lose some GOP support if he gets broadly negative reviews of his performance regarding Harvey. On the other hand, Trump’s approval is already low and very polarized by partisanship, so it’s tough to see the hurricane greatly altering how the public perceives the president.

Ultimately, people are going to judge the response to Harvey on subsequent reporting and the grisly details of the loss of life and property damage, and whether the combined local, state, and federal response was adequate. But just appearing to take the disaster seriously can positively impact voter perceptions.

In a 2010 paper, Andrew Healy of Loyola Marymount University and the aforementioned Neil Malhotra studied gubernatorial responses to tornados and found that “voters punish incumbents for tornado-induced economic damages…in both the county itself and in nearby counties. Moreover, voters seem to only punish incumbents in the absence of a response to the negative event.” Basically, if a governor was seen to have taken the event seriously (by issuing a disaster declaration), he or she did not pay a penalty.

Anecdotally, the perception or reality of a strong response to a natural disaster can help an incumbent governor in his or her reelection bid. In 2011, Joplin, MO was struck by a horrible tornado, and then-Gov. Jay Nixon (D) was widely credited with leading an effective response. It seems possible that the response helped Nixon at the polls when he ran for a second term in 2012. While Nixon’s statewide margin of victory fell from 19 in 2008 to 12 points in 2012 -- a drop mirroring Barack Obama’s marginal decline in the state after he contested but narrowly lost the Show Me State in 2008 and then abandoned it four years later -- Nixon actually improved on his 2008 margins in Jasper (by nine points) and Newton (seven points) counties, where Joplin is located. It’s probably not a coincidence that those are the only two counties in southwest Missouri where Nixon improved on his 2008 showing.
Granted, this study was about governors, but it speaks to a larger point about the importance of leaders projecting the image (or fact) that they are taking a natural disaster seriously. Gov. Greg Abbott (R-TX) is up for reelection next year and as of yet has no real Democratic opposition; it will be interesting to see if we see changes in his level of support in Southeast Texas depending on his response to this disaster, which is yet to be fully determined as of this writing.

Another group of political scientists, Fordham University’s Boris Heersink, University of Virginia PhD candidate Brenton Peterson, and the University of Southern California’s Jeffery Jenkins (all three have UVA connections), argue that natural disasters can harden partisan feelings. They studied disasters from 1972-2004 and found: “In disaster-affected counties that were safely in the incumbent party’s column, candidates of that party were rewarded in the wake of a natural disaster. In disaster-affected counties that were safely in the opposition party’s column, incumbent party candidates were punished severely -- even if they provided relief efforts.” They also found a similar effect on then-President Obama’s 2012 electoral performance in a more recent natural disaster, Hurricane Sandy: “Obama did better in safely Democratic counties that were hit by Sandy, compared to similar counties that were not affected by the hurricane. In contrast, Obama did noticeably worse in safely Republican counties.”

At the time, Obama’s active response to Sandy (and the embrace of GOP Gov. Chris Christie) appeared to be a factor in widening Obama’s narrow lead over Mitt Romney. At the very least, Romney’s campaign believed that Sandy took the focus off their candidate at a crucial moment: “Every race that I’ve ever been involved in where we beat an incumbent, we always controlled the dialogue at the end of the race,” said Romney strategist Stuart Stevens at a post-election conference. “I think the impact of the storm was we lost control of the race and the ability to control the race.” It’s impossible to know how things may have turned out had Sandy not occurred, but George Washington University’s John Sides and FiveThirtyEight’s Harry Enten did not find that Sandy impacted the polls all that much.

One other devastating natural disaster that may have had a presidential impact was Hurricane Andrew, which struck Florida in late August 1992 while beleaguered President George H.W. Bush was seeking a second term. As the New York Times reported at the time, “local officials and disaster relief experts said that federal efforts had been inadequate and often confused and that four days after Hurricane Andrew struck South Florida, uncounted thousands of residents still lacked food, water and shelter.”

Did Andrew hurt Bush? Maybe a little bit, but it’s hard to disentangle the hurricane from Bush’s other challenges during his rocky reelection bid. When Andrew happened, independent candidate Ross Perot was not yet back in the race, and Clinton had a solid, 10-point lead over Bush. Perot reentered the race later in the campaign, and Bush consistently trailed in the fall. Andrew may have stepped on Bush’s post-convention polling bounce but his chances of winning were probably minimal at that point anyway. Bush actually won Florida in 1992, the last time the Sunshine State failed to vote for the presidential winner, and 1992 was the last time that Florida had a large GOP lean compared to the nation (it has had a smaller Republican lean since 1996, when Bill Clinton won the state by about six points when he was winning nationally by nine). If Andrew could have hurt Bush anywhere, it should have been in Florida, yet the state’s voting was generally in line with how it had voted relative to the nation in the 1980s.

Ultimately, it seems like the presidential effect of natural disasters is minimal, and it’s hard to find an instance where a disaster was decisive. Without Katrina, George W. Bush’s popularity probably still would have declined, as it did throughout the tail end of his presidency. And without Sandy or Andrew, Obama probably still would have won a second term and Bush-41 probably still would have been denied one.
The immediate presidential political consequences of Harvey are scant: Trump doesn’t face the electorate for another three years.

Additionally, impressions of Trump are already largely baked in. At the time of Katrina, Bush’s approval rating was a weak 44% approve/52% disapprove in the RealClearPolitics average. Trump’s approval is already significantly worse than that -- 39%/56% when Harvey hit -- and it does not appear to have moved much over the past several days as the hurricane has dominated news coverage.
Very little appears to affect Trump’s standing significantly, and both the history of Trump’s presidency so far as well as the recent history of natural disasters should make us cautious about expecting major change on account of Harvey.

Kyle Kondik is a Political Analyst at the Center for Politics at the University of Virginia.

Geoffrey Skelley is  the Associate Editor  at the Center for Politics at the University of Virginia.

See Other Political Commentary by Kyle Kondik.

See Other Political Commentary by Geoffrey Skelley.

See Other  Political Commentary.