The Shortcomings of Cost-Benefit Analysis

In the comments section of a previous post, Ron H and Steven make good points on Cost-Benefit Analysis (CBA).  I regret not addressing them sooner, but I have been sick with a nasty cold.  Allow me to address them now:

Steven writes:

Well, this is the problem, you can’t force anyone to accept a Kaldor-Hicks criteria. They use a measure which substantially discounts those costs, and there is not a sense in which they’re wrong in terms of economic logic.

Steven is 100% correct.  Kaldor-Hicks (the technical name for cost-benefit analysis) is a bit subjective in how these costs and benefits are measured; the assigned values of the like.  He and I, both trained economists, could both use CBA on the same problem and come out to different answers.

Steven continues:

Also,can’t this same distributional argument you make against instating the minimum wage can be used against removing it? There are losers in both situations. Pareto improvements are elusive.

This is also a good point.  Pareto improvements (that is, improvements that occur so that no one is made worse off) are elusive (or, one might say, damn-near impossible).  A CBA argument can just as easily be made to keeping minimum wage as eliminating it.

Ron H writes:

There is also a moral question here. Is it OK for third parties to decide that one person’s well being should be advanced if it means another person will lose their job?

There is a moral question, one which CBA cannot, or does attempt to, answer.  And thus why we must rely on our own facilities to answer them.  In describing one of the shortcomings of CBA, one of my professors likes to point out that, given certain conditions, the Holocaust could be a net benefit if there were more Nazis (this professor is not advocating the Holocaust but pointing out that, under a strict Kaldor-Hicks analysis, if the benefits to the Nazis were higher than the costs to the Jews, then it would be seen as a Kaldor-Hicks improvement).

These points raised by the commentators I would have brought up in this post by myself, but they beat me to it.  These shortcomings are not to be ignored, and why I stress CBA is a starting point in the conversation.  It is what we economists can contribute; it should not be the whole conversation.  We would need our ethical considerations to play into it, too.  Sometimes CBA can help inform this process (“the current FDA procedure kills 2 people for every 1 saved” for example) or may not.  But regardless, it is a good starting point.