Oren Bar-Gill, an economics and law professor at Harvard, recently wrote a paper critiquing the use of willingness-to-pay (WTP) as a proxy for utility because WTP is affected by wealth (some economists refer to WTP as “willingness-and-ability-to-pay” for this reason). Specifically, all else equal, wealthy people often have a higher WTP for goods. This means that standard, microeconomic welfare estimates using WTP—i.e. consumer surplus maximization—implicitly add greater weight to the utility of wealthy individuals relative to poor individuals.
We’re wondering if you all (a.) think this is a problem and (b.) have come across/can think of solutions for dealing with this concern.
Extra info: We think this issue may have real-world consequences. Economists regularly use consumer surplus to make policy decisions. For example, the FTC uses consumer surplus as a chief consideration when making decisions about antitrust regulation. Additionally, well-respected economists regularly use consumer surplus maximization as an approximation of welfare for policy papers such as in this paper about price ceilings in natural gas markets.
Given that this concern seems to have real-world consequences, we find it strange that professors at our university (which has a well-regarded economics department) didn’t address it. Most undergraduate microeconomic courses used surplus maximization as the core tool for welfare maximization. Yet, in our experience, no professor brought up the fact that using WTP may add greater weight to the utility of wealthy people. This made us think we were missing something, but professors seemed to agree that this was a concern when we asked in office hours.
If you also think this is a problem, we’re very interested in hearing how you think it can be (at least partially) resolved. Here are a few examples of “solutions” (more like band-aids) we thought about:
Split up groups by income and separately analyze the WTP (and the derived consumer surplus) of each group. Then apply different weightings to each group’s consumer surplus and sum across. We took a stab at this in this spreadsheet and explain our process in this doc.
Control for poverty when conducting a regression analysis using survey data. Feel free to link examples.
We’d love comments on these proposed solutions or other suggestions. Thanks!!
I agree that this is an important topic. Economic welfare analysis is a widely used method for cause prioritization, and a lot of it looks very different from the kind of cause prioritization work that 80k or Open Philanthropy would do. So it’s useful to look into why that is.
That said, a lot of the economics literature goes beyond the simple “maximize surplus” approach that you highlight. The main subfield to look at is public economics (or a narrower subfield called public finance). That’s where you find professors who understand welfare economics the most. Since at least the 1970s, mainstream public economists have studied optimal policy using social welfare functions which capture the idea of diminishing marginal utility from wealth (James Mirlees’ Nobel prize-winning work on optimal redistribution is a good example. Atkinson and Stiglitz’s textbook, first published in 1980, is a great resource for learning this kind of stuff. And here are a few examples of papers published in the last few years). Roughly, those papers adjust WTP by a factor to account for differing marginal utilities, as you suggest in your google doc.
Many economists who most directly influence public policy are public economics or public finance professors (as an example, all 5 professors from the University of Chicago who served on the Council of Economic Advisors in the past 12 years were from public economics), so narrow “maximize willingness to pay” thinking shouldn’t be a problem for them.
However, many economists outside of public economics often do engage in the simple “maximize WTP” thinking that you’re talking about. Unfortunately, more sophisticated welfare analysis often isn’t a standard part of graduate school curricula (let alone undergrad courses), so a lot of economists don’t know about it. These economists could be having negative impacts in a number of ways:
They directly influence policy. The main channel for this, I would expect, is macroeconomics policy. It may also impact development policy, but from what I’ve heard development economists rarely do any sort of formal welfare analysis (they don’t even consider consumer surplus), and instead just look at ad-hoc objectives like increasing health or output.
They teach welfare analysis poorly to students. This leads to more economists who don’t know any welfare analysis beyond willingness to pay.
I’m not sure how big this negative impact is, but if it turned out to be big, I think the solution would be on the education side. Basic concepts in welfare economics, including optimal redistribution, should be a part of standard graduate and undergraduate economics curricula. Interested students should be informed that if they want to learn more they should take a public finance elective. From what I hear, this is fairly common at European universities, but not at American ones.
Thank you for this informative answer!
I vaguely recall hearing an economist say that welfare economics ceased to be part of the undergraduate curricula in American universities at some point in the past. I wonder if it might be worth tracing down the history of this development and examine it as a potentially instructive case study. Quick googling uncovers an interview with Amartya Sen in which the Indian economist recommends Tony Atkinson’s The strange disappearance of welfare economics as the “best article on that sad neglect”.
Thanks! Both the Atkinson article and the Sen interview are very interesting. I would like to see some actual data on the teaching of welfare economics/public economics. People seem to agree that it’s declined, but I’m not sure I would agree that it has “disappeared” (anecdotally, I know many people who were exposed to models of optimal redistribution during undergrad. Some of these people were exposed through a required course, and others chose to take a public finance elective). My impression that welfare economics teaching is more common at European universities is also just based on anecdotal evidence. Some actual data would be helpful.
I agree that this is a problem and had previously raised the question in a post on the Forum, (though it is my lowest scoring post ever so evidently lots of people disagree with my argument!)
This issue became especially clear in early attempts by economists to put a value on the life of people across countries. Since people in poor countries took on greater risk for less money, their lives were valued at a fraction of those in rich countries.
Another example is tickets. Suppose that we are selling tickets to the final of Euro 2020 and that Warren Buffet buys all the tickets for the game because he likes to watch games in empty stadia. Economists often say that because willingness to pay tracks utility across persons and the tickets went to the highest bigger, this outcome enhances social welfare compared to a world in which the ticket prices were kept artificially low. But obviously the fact that Buffet is willing to pay so much for the tickets is more a reflection of his massive wealth and peculiar tastes and not the fact that his welfare would actually be enhanced more than everyone else.
Economists then try to solve this by having independent fairness or equity constraints. But the market outcome is bad on utilitarian grounds.
For governments who have the option to tax, WTP has obvious relevance as a way of comparing a policy to a benchmark of taxation+redistribution. I tentatively think that an idealized state (representing any kind of combination of its constituents’ interests) ought to use a WTP analysis for almost all of its policy decisions. I wrote some opinionated thoughts here.
It’s less clear if this is relevant for a realistic, state and the discussion becomes more complex. I think it depends on a question like “what is the role of cost-effectiveness analysis in contexts where it is a relatively minor input into decision-making?” I think realistically there will be different kinds of cost-benefit analyses for different purposes. Sometimes WTP will be appropriate but probably not most of the time. When those other analyses depend on welfare, I expect they can often be productively framed as “WTP x (utility/$)” with some reasonable estimate for utility/$. But even that abstraction will often break down in cases where WTP is hard-to-observe or beneficiaries are irrational or whatever.
I think for a philanthropist WTP isn’t compelling as a metric, and should usually be combined with an explicit estimate of (utility/$). I don’t think I’ve seen philanthropists using WTP in this way and certainly wouldn’t expect to see someone suggesting that handing money to rich people is more effective since it can be done with lower overhead.
I would like someone with a background in both economics and EA to offer an articulation of the best defense of using willingness-to-pay in cost-benefit analysis. My experience is that when people raise this objection, many economists (e.g. Robin Hanson) respond by saying that the critics haven’t really understood the methods of economics. But I have never seen a clear explanation of why the objection is mistaken.
I think it is also worth noting that the economists themselves do not appear to apply willingness-to-pay consistently. John Broome (an economist by training) explains (Climate Matters, pp. 144–145):
I think the “main” (i.e. econ 101) case for time discounting (for all policy decisions other than determining savings rates) is roughly the one given by Robin here.
I don’t think there is a big incongruity here. Questions about diminishing returns to wealth become relevant when trying to determine what savings rate might be socially optimal. Analogously, questions about diminishing returns to wealth become relevant when we ask about what level of redistribution might be socially optimal, even if most economists would prefer to bracket them for most other policy discussions.
Ideally we’d move onto measures of subjective well-being, like life satisfaction and just use them directly, but I expect data to be much harder to obtain (at least I’d guess there’s much less data now, and I expect trying to estimate the effects of various goods on life satisfaction would require large samples of subjects or experiments to detect effects). Your solution 1, using weights based on subjective well-being like you describe, seems like a good approach.
It is a problem and there are already solutions for that in the field (Public Economics/Environment Economics). This is called the Distributional aspects of CBA, if you want you can take a look at the course Cost-Benefit Analysis from the University of Helsinki which provides related material. The link to the course: https://studies.helsinki.fi/courses/cur/otm-e45f15a4-2322-4e44-93ba-e3bbb2e95d63
The 2 research papers that they used in the course, you can take a look here:
https://www.econstor.eu/bitstream/10419/102065/1/781242428.pdf
https://www.sciencedirect.com/science/article/pii/S0921800917318074
Hope that helps!