David R. Henderson  

When Government Cries Wolf

Semi-Rivalry and Fiscal Extern... What Does Education Signal? Th...

Henderson's Law of Warnings: Trivial Warnings Drown Out Serious Warnings

If the government were more judicious with its warnings and tried to focus on the high-probability dangers--and this is what governments do in many other countries--lives would be saved. Human life is simply too valuable to endanger it by incessantly warning people about small risks.

This is from my article, "When Government Cries Wolf," published at the Hoover Institution's "Defining Ideas" web site.

Regular readers of my posts will recognize some of the arguments because I first made them, in shortened form and mainly about Proposition 65, here.

Another excerpt:

What had gone wrong? We had mistakenly dismissed the government's warning. But why had we done so? Because the government so often cries wolf. The government warns us about many risks, large and small, and rarely gives any idea of the size of those risks. We have a kind of "warning pollution." In economics, there's something known as Gresham's Law, which says, "Bad currency drives out good currency." With respect to warnings, there's a similar law, and since I'm identifying it, I'll name it Henderson's Law of Warnings: "Trivial warnings drown out serious warnings."

HT to Paul Crowley for the term "warning pollution."

Comments and Sharing

CATEGORIES: Regulation

COMMENTS (4 to date)
NZ writes:

It's probably worth taking into account that levels of risk vary depending on who and where you are. The same warning that may be trivial to ten people may also be very serious to one. Presuming the government can either warn everybody or warn nobody, at what ratio of triviality/seriousness should it draw the line and issue a warning?

Aaron Zierman writes:

I remember hearing somewhere recently that the number of tornado siren alarms had been reduced for this very reason. People weren't taking them seriously because they were going off all of the time.

I went searching a little and found this article about the recent deaths of the storm chasers in Oklahoma. This is from Accuweather Senior Vice President Mike Smith:

Smith said he’s worried that well-meaning people may react to Friday’s tragedy by enacting regulations that actually lead to more fatalities. The top priority in improving tornado warnings, he said, should be to reduce the number of false alarms – times when warnings are issued but no tornado develops.
Michael_M writes:

Perhaps the most important way to view warnings is to consider them like software license agreements - they exist not to protect the user but to protect the interests of the ones providing the warning/agreement.

Security warnings and procedures are probably the biggest examples. They don't reduce personal risk but they are probably considered good ways to reduce political risk in a government culture driven largely by attorneys.

Seeking indemnity through a warning is easier and less risky than pursuing potentially unpopular long-term solutions (closing a beach, more selective screening measures, etc.).

Peter writes:


Not going to happen without serious pressure from above. NWS management is extremely risk adverse and NWSEO (NWS union) is extremely hostile to metrics and accountability; both also benefit immensely from weather forecasting failures hence you have serious incentive problems. For example the latest round of tornadoes ended up getting all NWS furloughs, nationwide, cancelled while the 2004 Indian Ocean Tsunami netted the NWS millions. Accuweather has been arguing this for years and the fact weather forecasting isn't a market failure, i.e. the government unfairly prices private industry out. Most Americans would be surprised to learn most other nations either don't have a public weather service or, those which do, don't give out the data for free. Still Accuweather routinely fails in this argument as annual weather events keep the NWS in the news all the time (and the media nearly always place their failures in a positive light especially given NWS provides them all their data for free) and because NWS has trained the general public to never expect competent results "x% chance of this, x% chance of that, don't blame us, it's the weather, it's always changing, etc etc. ".

The standard line is (and what most people forget) is NWS doesn't control civil warning sirens, they simply issue warnings. Since it's the local government, not the NWS, which activates the sirens / closes schools / etc it's not THEIR (NWS) fault that the local government overreacted. Never mind they are the SME's and the local government is simply acting on the information they put out.

Comments for this entry have been closed
Return to top