A Burning Situation

A fire broke out at a baker's in the City of London on September 2nd, 1666. This lead to The Great Fire of London which burned for 4 days consuming 436 acres, including 80% of the City. Out of the ashes of this ruin came a new business; the competing for-profit fire insurance companies, each one of which also owned a fire brigade. For the next two centuries London held this form of protection.

History tells us that in the earliest days of insurance, companies gave plaques to policy holders who would display them on their homes. This showed them to be covered. The brigades would refuse assistance to those homes not holding a plaque. Further, an insurance company would extinguish a fire only if it was in a home insured by that exact company, so a house with the plaque of a different company would receive no assistance.

In present day Obion County, homeowners must pay $75 annually for fire protection services from the nearby city of South Fulton. Residents in the rural area of Obion had no fire protection until the county established the $75 fee in 1990. If they did not pay the fee and their home caught fire, they do not get fire protection services under any circumstances. Someone must always learn the hard way.

An article found here, describes how a man’s home caught fire last week and he called 911. Having not paid the $75 fee, he was offering to pay all expenses related to the Fire Department’s response. He was refused service. Firefighters did however come out when the man’s neighbour (who had paid the fee) called 911, worried that the fire might spread to his property. Once there, members of the fire department stood by and watched, acting only when the fire reached the neighbour's property. What did the man have to say for himself?

"I thought they'd come out and put it out, even if you hadn't paid your $75, but I was wrong."

This quote captures the “free rider” problem perfectly. Many economics textbooks discuss fire protection with respect to public goods. Economists define public goods in terms of non-rivalry and non-excludability. Non-rivalry is when a good can be used by one person without affecting the use of the good for another person. Specifically, the marginal cost of an additional person enjoying the good is zero. Non-excludability occurs when it is impossible to exclude anyone from the benefits of a particular good. A pure public good is one that has both absolute non-rivalry and absolute non-excludability.

The supply of pure public goods requires government involvement in order to be efficient. A free-rider problem often arises due to the non-excludability, meaning no one person has any incentive to provide it. Thus private provision of these goods leads to production of an insufficient level of consumption or supply. Many goods may have only a degree of non-rivalry and/or non-excludability. These goods are sometimes called “impure” public goods. Fire protection falls in this category.

According to Joseph Stiglitz, in his Economics of the Public Sector, fire protection is similar to a pure public good in that the marginal cost of use is very low. The article from Obion, states that the firefighters were at the scene and had all the equipment available to put the fire out.

However, the ease of exclusion for fire protection is similar to that of the pure private good. If fire companies cannot discriminate between the insured and uninsured, a significant free-rider problem will exist. Without discrimination, no incentive would exist for people to pay fees like the one for $75. With fire companies only interested in protecting those who paid the fees, they had no incentive to help burning uninsured houses. So there you have it, fire protection possesses attributes of both a public and a private good.

From Economics of the Public Sector:

'Most of the time, fire fighters are not engaged in fighting fires but are waiting for calls. Protecting an additional individual has little extra cost. Only in that rare event when two fires break out simultaneously will there be a significant cost to extending fire protection to an additional person. But even here, matters are more complicated: if we want to protect the building next door which has paid for fire protection, it may be necessary to put out the fire in the building which has not paid for protection – exclusion may not really be feasible (Stiglitz, 2000, p. 134).'

In another article by Levy, which hits closer to home for this blog, the idea that buildings would be allowed to burn in a fee-based system was suggested. We would encourage those interested to read it.

Stiglitz, J. (2000) Economics of the Public Sector, 3rd edn., New York: W. W. Norton & Company.

Levy, J. (1995) Essential Microeconomics for Public Policy Analysis, Westport, CT: Greenwood Press.