Weighing The World

Recent graduate of Cambridge University, Markus Anderljung wrote his masters dissertation on the Post-2015 Consensus. The entire dissertation is available for download following the brief summary. 

A Review of Some Ethical Critiques of the CCC

The Copenhagen Consensus Centre (CCC) seeks to inform public policy by conducting Cost-Benefit Analyses (CBAs). The rationale behind this is the thought that public policies should be chosen largely based on their effects. One should choose the policies for which the benefits most greatly outweigh the costs. Below are some ethical critiques that could undermine this rationale. None of them are decisive.

Firstly, the effects of policies may not matter all that much. Critics of CBA often attack it for ignoring what are not effects of policies, such as violations of rights (e.g. Nussbaum 2001, 169). Some have tried to respond to these worries by trying to put a value on rights (Lowry & Peterson 2012, 260), but the CCC’s response is better: a CBA need not be a complete decision procedure (Lomborg 2014, 4). That is, the CBA only informs the policy-decision, such that policy-makers can take considerations such as potential rights-violations into account. If anything, this critique may be getting things backwards, as we should likely worry more about policy-makers overemphasising rights over consequences than the other way around.

Secondly, it may be that the effects of policies hold a lot of weight, but that the CBAs fail to capture the effects that matter. These are the kinds of issues I am particularly worried about. One such concern relates to the social discount rate. This is the rate at which the value of a benefit diminishes with time. In the CCC, this rate is 3% and 5% a year. The social discount rate means that ridding someone of malaria today is more valuable than doing so in a year’s time. There are a number of good reasons to think this, for example that there will be economic growth and technological progress. A bad reason sometimes given for the discount rate is a pure time preference, where one simply asserts “the sooner, the better”. Nearly all ethicists, who seldom agree on anything, agree that one ought not have a pure time preference: a life lived now is just as valuable as a life lived a hundred years from now. This is not to say that we should sacrifice all of our current wellbeing for the sake of future wellbeing, as doing so may for example violate the rights of current people, but it does suggests that some sacrifices may be appropriate. And so, if the CCC’s discount rate relies in part on a pure time preference, I would argue along with many ethicists, that they are undervaluing future benefits. 

Another set of worries suggest that the CCC’s methodology does not give enough weight to poor individuals. A first way to express this worry is that the distribution of welfare, not just the aggregate welfare, matters – that a more equal distribution of welfare is better. One response to this worry is to incorporate concerns for inequality into one’s CBA: for example by valuing welfare-benefits to well-off people less highly than the same absolute benefit to less well-off people (Adler 2012, 307; Johansson-Stenman 2005, 340). A crucial difficulty with this approach is deciding the shape of the function from how well off someone is to the value of a welfare-benefit. That decision seems to be a matter of contentious political and ethical issues. Instead of wading into these issues, it may be better to follow the CCC’s line, holding that though inequality may matter, it is best for policy-makers to take this into account when considering their CBAs instead of incorporating the concern into the analyses (Lomborg 2014, 4). Additionally, the CCC takes wealth inequality into account indirectly, in its mainly conducting CBAs on Low- and Middle-Income Countries.

A second way to express the worry about not giving enough weight to poor individuals concerns wealth distortion. I am particularly worried that the CBAs of the CCC fall prey to wealth distortion, and argue in the paper that this distortion ought to be removed. CBAs often measure how much a good, such as being more healthy, improves someone’s welfare by looking at how much they are willing to pay for that good. For example, to measure the value of saving someone’s life, one extrapolates from how much people claim to be willing to pay or in fact pay, to avert some risk of death. One problem with this method is that how much someone is willing to pay for a good does not only depend on how much benefit they derive from that good, but also on how much money they have. There is a diminishing marginal utility to consumption. Someone who lives on $1 a day receives a lot more welfare from spending that money than I would by spending $1 on a pack of gum.

The CCC’s CBAs mitigate this concern somewhat by mainly concerning Low- and Middle-Income Countries. However, these countries still vary substantially in terms of GDP per capita. This means that analyses measuring, for example, benefits from education in terms of wage-increases (such as in Psacharopoulos 2014) undervalue the benefits of those wages to people living in Low-Income Countries. However, this is only a problem if the benefits vary greatly between countries of different income levels. A conservative fix to the wealth distortion could be for the CCC to more diligently follow its current practice of having its analyses concern Low- and Middle-Income Countries – CBAs that include High-Income Countries (such as Anderson 2014) may not be comparable to CBAs that do not, due to wealth distortion – and flagging when benefits vary greatly between countries of different income-levels.

I hold that a better fix to this distortion may be similar to the one discussed above regarding inequality. One finds the relationship between how well off someone is and the benefit to welfare they get given some absolute increase in consumption. This relationship can be used to translate how much an individual (or an average individual in a country) is willing to pay for a good into how much benefit they derive from it. A main problem with this suggestion is finding the relationship. However, unlike with the adjustment for inequality, the shape of the diminishing marginal utility curve is a scientific question. It can be answered using economics and psychology research – for example by looking at the relationship between consumption and happiness as people are better off. Though finding this relationship is still an ongoing research project, the UK Treasury has introduced an adjustment where as an individual’s income is doubled, the marginal value of consumption to individuals is halved (HM Treasury 2011, 93). This may be a good enough approximation.

Sources Cited

• Adler, Matthew D. 2012. Well-Being and Fair Distribution: Beyond Cost-Benefit Analysis. USA: Oxford University Press.
• Anderson, Kym. 2014. “Trade Assessment Paper.” Working Paper. Post-2015 Consensus. Copenhagen Consensus Center. http://www.copenhagenconsensus.com/sites/default/files/trade_assessment….
• HM Treasury. 2011. The Green Book – Appraisal and Evaluation in Central Government. London: TSO. https://www.gov.uk/government/uploads/system/uploads/attachment_data/fi….
• Johansson-Stenman, Olof. 2005. “Distributional Weights in Cost-Benefits Analysis – Should We Forget About Them.” Land Economics 81 (3): 337–52.
• Lomborg, Bjorn. 2014. “Prioritizing the World: Introduction.” In Prioritizing the World. Co-penhagen Consensus Center.
• Lowry, Rosemary and Martin Peterson, 2012. “Cost-Benefit Analysis and Non-Utilitarian Ethics.” Politics, Philosophy and Economics 11(3): 258-279.
• Psacharopoulos, George. 2014. “Education Assessment Paper.” Working Paper. Post-2015 Consensus. Copenhagen Consensus Center. http://www.copenhagenconsensus.com/sites/default/files/education_assess….

You can download a full version of Markus Anderljung's dissertation here.