Some people seem to think that a good use of economists is to estimate “the cost of” things. Particularly in the realm of environmental impacts or natural resource degradation, we often see very large numbers bandied around as estimates of “the cost of” this or that. Most recently, large estimates of “the cost of” weeds in Australia (around $4 billion per year) have been released, and compared publicly with the smaller “cost of” salinity, with implications that relative levels of public funding are inappropriate.

This is unfortunate. There are a number of problems with “the cost of …” concept that make it not very relevant to policy making . The three main problems are:

1. “The cost of” X looks at only one side of the ledger. It provides hints about the benefits of taking action to remove the problem, but no information about the costs of taking action.
2. No matter what action is taken, the problem, typically, will not be eliminated entirely. How far you can reduce the problem through practical management or policy actions is a key consideration in deciding what to do, and the degree of mitigation will vary widely between different issues and different actions. “The cost of” X can be quite misleading as it relates to an unattainable, mythical world where the degradation problem costlessly disappears.
3. From the point of view of guiding government policy, focusing on “the cost of” ignores the question of whether there is any market failure to justify governments taking action. Is it a fully private issue, where benefits and costs are borne by the same people, or an issue with “public good” dimensions; that is, where an action by one person affects others through externalities or where there are free-rider problems? Economists argue that it is usually not appropriate for governments to get involved in fully private issues. A large estimate for “the cost of” X would not affect this argument. It would be up to private individuals to judge whether the benefits to them of reducing X outweighed the costs.

For all these reasons, “the cost of” X relative to “the cost of” Y, provides no evidence whatsoever about whether X is more or less deserving of government attention or funding. A big “cost of” provides no evidence that there is any issue worth responding to.

In my view, economists should resist calculating “the cost of” anything, unless it is done as part of a broader analysis considering the costs and benefits of taking specific actions, and considers whether there is market failure to justify a government response.

In the case of both weeds and salinity, there are some aspects that are fully private, and some that are partly or fully public. In both cases, there are plenty of examples where “the cost of” a particular weed or salinity impact is very high, but the cost of repair or eradication is even greater. There are, of course, also cases where benefits of taking action outweigh the costs. A sensible decision on relative levels of public funding would require an unpacking of these issues for each case. It is not obvious to me which would “win”.

David Pannell, The University of Western Australia