Homeland Security Watch

News and analysis of critical issues in homeland security

June 19, 2006

Becker and Posner examine the DHS grant issue

Filed under: Risk Assessment — by Christian Beckner on June 19, 2006

Worth reading: a fascinating exchange – in three posts – between Nobel-winning economist Gary Becker and Judge Richard Posner (who has written a couple of books on homeland security-related topics) on their joint blog, on the topic of the rationale and methodologies for homeland security grants, in light of the recent controversy over the funding cuts to DC & NYC. There are some interesting thoughts in the posts, including a proposal by Becker to require states & cities to match some of the federal grant funds:

The Department of Homeland Security should offer to give cities a certain number of dollars for each dollar they spend on antiterrorist activities. For example, if a city spent $30 million, they might get an additional $60 million from the federal government. In this example, a city would get to spend $3 dollars for each dollar they used from their own funds to fight terrorism.

Federal matching of this type discourages cities from cutting back on their spending to fight terrorism since they lose say $3 dollars for each dollar they cut back. Matching grants also induce cities to give de facto recognition to the fact that each dollar they spend helps residents of other cities as well by improving the overall American fight against terrorism. The ratio of federal spending to city spending should be a measure of the ratio of the benefits to other cities compared to the benefits to the city spending their own money to fight terrorism.

…and a solid analytical construct by Judge Posner about how to prioritize and consider the cost & benefits of homeland security grants:

Ideally, one would like the grant moneys to be allocated in such a way as to maximize the excess of benefits over costs. The costs are relatively straightforward, but the benefits are not. The benefits of an antiterrorism measure, for each potential target, depend on (1) the value of the target (not just in terms of financial loss, of course) to the United States, (2) the likelihood of its being attacked, (3) the likely damage to the target if it is attacked (which requires consideration of the range of possible attacks), and (4) the efficacy of a given measure to prevent the attack or reduce the damage caused by it. (2) and (3) are probably the most difficult to estimate accurately, because to do so would require extensive knowledge of the plans, resources, number, location, and motivations of potential terrorists. But (4) is very difficult too, because the effectiveness of increasing the number of policemen, or of installing surveillance cameras on every block, or of increasing the number of SWAT teams, or of taking other measures of prevention or response, is extremely difficult to assess in advance.

Overall, an interesting exchange by two very smart guys.

Share and Enjoy:
  • Digg
  • Reddit
  • Facebook
  • Yahoo! Buzz
  • Google Bookmarks
  • email
  • Print
  • LinkedIn

No Comments »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>