Unhappiness After Hurricane Katrina

"Hurricane Katrina reduced the reported happiness of a nationally representative sample of Americans, and that happiness is correlated with but distinct from consumer sentiment."

The University of Michigan Monthly Survey of Consumers is widely regarded as a valuable tool for gauging Americans' sense of personal well-being. A team of researchers has now examined the data collected in recent surveys to determine if the respondents' feelings of happiness or unhappiness may be affected by external events that have no direct bearing on their lives.

In Unhappiness after Hurricane Katrina (NBER Working Paper No. 12062), researchers Miles Kimball, Helen Levy, Fumio Ohtake, and Yoshiro Tsutsui note that the Michigan Surveys showed a distinct and significant dip in reported happiness in the first week of September 2005. The researchers speculate that this occurred because it was in this week that the extent of Hurricane Katrina's devastation of the Gulf Coast became known. Indeed, the fall in national happiness was most pronounced in the South Central region of the United States -- that is, in the portion of the country closest to the destruction, and the unhappiness lingered longer there than it did elsewhere in the nation. But why, the researchers ask, should Americans who had not lost their homes or livelihoods to the hurricane exhibit a decline in their sense of personal happiness?

Even though Hurricane Katrina may not have directly affected an individual's material well-being, Kimball and his colleagues theorize that the disasters' dominance of the news media saddened Americans because it stimulated their feelings of altruistic concern or because of a more generalized emotional response to disaster images. The researchers do not discount people being concerned about the storms' impact on their material well-being as manifested by a possible rise in fuel prices or challenges to government budgets. But the altruistic factor is indeed measurable, given that charitable donations to help the hurricane victims are estimated at $2.65 billion, an outpouring of aid that surpasses that of the donations to the South Asian tsunami relief ($1.55 billion) and that almost matches the money donated for victims of the September 11 terror attacks ($2.8 billion). Indeed, by comparing the results of consumer confidence surveys, which in the post-hurricane weeks showed their lowest results in 12 years, and those of the overall happiness surveys in the same weeks, the evidence suggests concerns based on self-interest were less a factor than altruistic emotional responses to the hurricanes and their aftermath.

Kimball and his colleagues believe that those emotional responses were triggered by the media coverage of the disasters. A detailed analysis of major newspapers during the period shows the hurricanes having a near-monopoly on front pages, and of course television delivered real-time pictures of the havoc into millions of American homes. Survey respondents' estimates of their personal level of happiness, sadness, enjoyment of life, and depression correspond to a measurable degree with the intensity of graphic news coverage of the disasters. Interestingly, while women's reported happiness index over the entire period from August through October 2006 is significantly lower than men's, the movements in happiness during the hurricane period are similar for men and women. By contrast, while the South Central states do not normally show a happiness index significantly different from the rest of the United States, in the first week of September 2005 that index was much lower than that of the rest of the nation.

At the same time, feelings of unhappiness wore off even before the intense coverage of the hurricanes abated, suggesting people became inured to the graphic images of the devastation and suffering. This is evidence of "hedonic adaptation," that is, the tendency of measured happiness to revert to its previous value after responding to a shock. Indeed, by the end of September 2005, a more or less complete "hedonic adaptation" had taken place throughout the United States - although unsurprisingly, this was most marked in the regions of the country unaffected directly by the hurricanes.

A few weeks later, on October 8, a major earthquake struck Pakistan and parts of neighboring India just a few weeks after the Gulf states' hurricanes, and this disaster was widely covered by the American news media. The researchers therefore looked at day-by-day happiness surveys in the second and third weeks of October to see if Americans' subjective sense of well-being could be affected by such a far-away disaster. The survey shows a dip in national happiness, and the researchers believe that at least some of this decline was related to the earthquake. But they note that the statistical significance is not as marked in this instance as it was for the dip in happiness because of Katrina.

Kimball, Levy, Ohtake, and Tsutsui conclude that Hurricane Katrina reduced the reported happiness of a nationally representative sample of Americans, and that happiness is correlated with but distinct from consumer sentiment. This, they say, raises doubts about explaining these movements solely in terms of self-interest. Instead, the researchers say altruism or a more general emotional response to images of new disasters likely explains this response to bad news. Finally, they believe their methodology can help determine what kinds of events strike the average respondent as noteworthy good news or noteworthy bad news.

-- Matt Nesvisky