Last month, at the Annual Conference of the Society for Benefit-Cost Analysis, I had a fascinating conversation with an economist from North Dakota. He was struggling with a concern he had in his work about the implications of his cost-benefit analysis methods. Essentially, he was trying to calculate what the impact would be if a city stormwater system got overwhelmed.
From an engineering perspective, the answer was quite clear. The excess water would move downhill, and flood two adjacent neighborhoods: one with moderate-income single family houses, one with low-income trailer homes. Barring a complete overhaul of the entire stormwater system, it appeared as if the city had to choose which way to direct the water in the event of a severe storm.
Using a basic cost-benefit analysis framework, it seemed that allowing the trailer homes to flood would be a much less costly outcome. This is due to the fact that property values are much lower in this neighborhood and the monetary value of damages and the needed repairs would be much smaller.
However, this doesn’t take into account the structural disadvantages that residents of this neighborhood face. Although it might cause a higher amount of property damage should the single-family homes flood, people living in them would be more resilient to that event.
This type of equity concern is extremely difficult to account for in cost-benefit analysis, and it has become one of the most pressing current issues in the field.
Beyond this lunchtime conversation, multiple paper sessions and panel discussions focused on this same question. How can we take this analysis tool that is so narrowly focused on measuring efficiency, and use it to ensure we achieve equitable outcomes as well?
This is an interesting element to try and add into a cost-benefit analysis framework, because often we assume that there is some tradeoff between equity and efficiency. In his 1975 book on the topic, Arthur Okun describes this tradeoff using the metaphor of a leaky bucket.
He posits that pursuing an equitable outcome is like moving water between wells with a leaky bucket. The resources we use to improve outcomes for less welloff individuals comes from people who are more well off, and to get them where they are more needed there will be some economic loss along the way.
While it is not impossible for some policy decisions to improve both equity and efficiency, it is often acknowledged that we need to balance these two policy goals.
The most commonly proposed method to incorporate these equity considerations into cost-benefit analyses is distributional weights. In practice, distributional weights suggest that costs and benefits matter more or less depending on who accrues them.
From a theoretical perspective, we can justify this by looking at the marginal utility gained or lost by different groups of individuals. For example, we might expect very wealthy individuals to lose less utility from some income-reducing policy, because they care less about that marginal amount of income.
In practice, this really isn’t an efficiency concern. If we weight outcomes for different groups, then we are no longer looking for the most economically efficient outcome. Instead, we are trying to find the “socially optimal” outcome.
This blend of efficiency and equity analyses doesn’t neatly fit into the rigid lines of economic analysis. When we combine these two into a single algorithm, we are trying to pursue two goals at once, and distributional weights are an attempt to make them comparable.
If it sounds like I don't think distributional analysis is a good addition to cost-benefit analysis, that is certainly not true. Theoretical questions aside, the current state of economic analysis does tend to favor currently advantaged groups, like the owners of those single family homes. Economists should be considering equity in their analyses, they just maybe should be willing to say that equity is important out loud.