Funding for Disaster Recovery: Increased Taxes or Charitable Donations to Nonprofits?

Abstract

Governments oftentimes fail to satisfy high demanders of public services, leaving room for the nonprofit sector. How can the government deal with this government failure? Must it resort to tax increases or should it engage the nonprofit sector? Using the public polls conducted in the wake of Hurricane Katrina, this study aims to explore an empirical answer to this question. In particular, we contrast the determinants of the willingness of an individual to pay more tax with those of the willingness to make more voluntary donations for the disaster recovery purposes. The results show that attitudes on governments' roles in disaster recovery are not consistent with charitable giving behavior and are affected more by the survey respondents' political orientation than voluntary motivation. We discuss this finding's implication for designing a more effective disaster response system.

Keywords:

• disaster recovery
• tax increase
• nonprofit
• public-nonprofit cooperation

Notes

¹However, it may be misleading to contrast more taxes with more donations. First of all, willingness to pay more taxes does not necessarily mean that the individual would be less willing to donate. Second, the government-nonprofit relation can be understood in terms of complementary as well as supplementary one (Young, 1999; Salamon, 1987). Due to the possibility of the voluntary failure (Salamon, 1987), partnerships with government may enable nonprofits to overcome its narrow parochial focus. We will pick up this point in the discussion.

²It might be problematic to use public opinion surveys conducted right after Hurricane Katrina to which the government egregiously failed to respond effectively. It may raise the issue that the survey respondents might be framed against more government roles in the disaster relief efforts. In spite of this possibility, we decided to use the afore-mentioned surveys. First, we cannot find relatively recent surveys that both provide comparable information sufficient to build regression models and were conducted after 9/11 or after the major hurricanes with which government has successfully dealt. Second, while the respondents of the surveys we are using might be biased against expanded government roles, as described below, two of our models include the variable of changes in the level of confidence in governmental ability in disaster recovery efforts. We believe that this variable may in part control the framing effect of governmental failure in dealing with Katrina.

⁶ Unequal variance t-test. Only the significant difference is reported. * = <0.1; ** = <0.05; and *** = <0.01.

⁷ The actual statement for this variable in each survey is slightly different from the other. The CBS survey asks “Has the government's response to Hurricane Katrina made you more confident in the government's ability to respond to a terrorist attack or natural disaster, less confident in the government's ability to respond to a terrorist attack or natural disaster, or hasn't it affected your opinion?” In the PEW survey it reads “Has the government's response to this disaster affected your confidence in the government's ability to handle a major terrorist attack, or not?” The former survey asks the respondents to assess the government's ability in responding to a natural disaster as well as handling a terrorist attack.

³The average correlation coefficients are .277 between Democrat and Liberal (.259 in CBS and .285 in PEW) and .304 between Republican and Conservative (.377 in CBS and .319 in PEW).

⁴This deserves further consideration in that a lowered level of confidence is not necessarily contrasted with impressive activities of the nonprofit organizations. If it is, it might lead to a higher inclination to make nonprofit donations.

⁵Western states include Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming.