Economic Expertise and Moral Mathematics

Ryan Avent, following up on Ezra’s post on the policy prestige of economists in general and benefit/cost analysis in particular, writes:

[I]f we can’t use cost-benefit, how do we make these decisions? How in hell do we figure out which trade-offs are sound ones and which are damaging to society on net? To the economist, and indeed to many policymakers, eliminating cost-benefit analysis is like depriving them of language — how can you discuss the problem without it?

In fact, we really have no other way of wrestling with these issues. One can argue by moral imperative — that we don’t have the right to impose serious costs on others — but we still must determine how much compensation or preventative action those others are owed, and from where the resources to pay or take action should be drawn. These questions involve trade-offs, which must be weighed in some fashion, and so again we find ourselves turning to economists.

Why do we turn to economists again?

I agree that it is impossible to think intelligently about policy without some minimum of economic literacy. But the economist has no competence whatsoever to tell us, say, the appropriate discount rate to apply to future costs and benefits, to take one important example. I’ve heard philosophical arguments to the effect that the discount rate for future welfare should be zero and that the discount rate should approach infinity as we consider the welfare of furture beings with whom there is no possibility of reciprocity. The funny thing is that I think people get the implications of discount rates wrong, and that both zero and infinity point to more or less maximizing growth. A zero discount rate plus a basic grasp of the relationship between technology and growth plus a reasonable projection of the current trend of technological progress implies an obligation to maximize economic growth rates with no concern whatsoever to avoid the incidence of future externalities of current activity. This is an economic argument, but it is also something rather more. Likewise, an infinite discount rate implies that we should do the best we can for our children and grandchildren, and leave it to our grandchildren to worry about their grandchildren. If we’re doing something now that might hurt people none of us will coexist with 100 years, then so what?

Of course, neither of these arguments will convince Ryan or Ezra. I’m sure we disagree both about the discount rate, which is not itself an economic question, and about the implications of the discount rate, which is only a partially economic question. And I think it gets even worse: economists have almost no competence whatsoever in telling us what counts as a cost or a benefit. That’s pretty important, isn’t it?

So why do we give so much weight to the opinions of economists?

Author: Will Wilkinson

Vice President for Research at the Niskanen Center

28 thoughts

  1. “So why do we give so much weight to the opinions of economists?”For *opinions* we shouldn't. But the value of economics is that you would not be able to have the discussion above – to realize the significance of the discount rate in the calculations, etc. – without economics.Of course economistrs shouldn't opine on the value of nominal quantities as if it is scientific, but that shouldn't stop them from telling us the implications of different assumptions about the discount rate.They have, in fact, done that – hence you arte able to have the discussion above.

  2. This is essentially asking why the Kaldor-Hicks efficiency criteria is morally compelling, is it not? I'd guess that people pay attention to it largely because it's a relatively neutral way of weighing things, and thus has broad appeal (to those with economic literacy) across the political spectrum – do you have any others in mind that wouldn't implicitly tell people that some of their desires shouldn't have positive weight?It's also far easier to impute a good measure of dollar costs/benefits than trying to weigh policies in terms of utils or hedons or whatever else.It might not be the ideal weighting scheme, but it might also be the least-worst one when trying to communicate with those who may not share your particular values.

  3. Adam Gurri – Collector of Stories. Co-founder of Sweet Talk. Follow me on <a href="">Twitter</a> or at <a href="">my personal blog</a>.
    AdamGurri says:

    I guess people just weren't able to make decisions before economic theory came along. Or maybe they just made stupid, superstitious decisions, unlike those rational, scientific ones we make today.Please. Rationalists like Avent cling to the idea that there is some value-neutral criteria to make policy decisions, but the the bottom line is that there is no such criteria. Policy matters by their nature are based on value judgments; including judgments of what constitutes good and bad effects on society.

  4. Will is right.My quick 0.02: I think the mathematization of policy analysis is a move towards greater impartiality and reconciliation and away from division and partisanship.Math won't solve all our problems because often we don't know how to translate a policy problem to math in a way that encompasses all the subtleties. But wherever we can translate arguments about values into arguments about math, we have made the important step of reducing the chance that we will petulantly disagree for trivial reasons.In other words, consider that one of the biggest problems in any policy discussion is that people will let ego, tribal associations (liberal/conservative/etc.), or ingrained emotional beliefs determine their judgement. This is far less likely to happen if we are haggling about numbers.Now, statistics are one of Mark Twain's three kinds of lies. And people do smuggle assumptions into the math, all the time. (“This discount rate implies we should ignore global warming? OK, try another value!”) But, the more we mathematize our decision making, the more we will become forced to be consistent. If I argue using a discount rate of X for global warming but using a discount rate Y for nuclear proliferation, that is an obvious and (crucially) quite public problem with my analysis. Inconsistency in choosing modeling parameters creates obvious and public credibility gaps, which is just as it should be.It will always be possible to smuggle politics into numbers. That's been going on for centuries. But, overall I still think that mathematization creates more light and less heat.But yes, economists can't figure it all out. And there are many tasks required in setting up such a model that economists are not trained for.

  5. Don't ask my opinion. I'll elicit an objective function from you and tell you what *your* opinion is… assuming you want to maximize that objective. Economics is a discipline not a doctrine. We're supposed to be good at writing down objective functions and then cranking the knob on the cost/benefit calculations. Not everyone can be good at those things.

  6. And I think it gets even worse: economists have almost no competence whatsoever in telling us what counts as a cost or a benefit. That’s pretty important, isn’t it?

    Economists tend to operate by a moral rule that anything that people view as a cost or benefit, is one. A strict utilitarianism. However, economists are certainly known to produce reports for political bean-counting that ignore these things.And certainly it's a moral claim that doesn't answer questions of whether non-humans (or babies or fetuses) have any sort of moral claim outside of what adults can put on it.

  7. Obviously people can lie with economics just as they can lie with statistics. Most of that lying occurs in the assumptions, though. Even if I may suspect, e.g., that Ryan Avent will invent externalities and costs and benefits to justify trains everywhere no matter what the cost, but having a framework to discuss that is still very useful.Economics is also another one of those disciplines where the media, the public, and politicians are most interested in hearing economists on issues where economists disagree, not where they agree.

  8. And I think it gets even worse: economists have almost no competence whatsoever in telling us what counts as a cost or a benefit. That’s pretty important, isn’t it?I think you're being a bit harsh on economists, here. I think what you're really complaining about is the pretty asinine nature of cost-benefit analysis. When confronted with a policy decision – “To A? Or not to A?” – one way to sound serious is to intone “We need a cost-benefit analysis!” But among the first observations one makes upon opening the pages of the Book of Political History is that such predictive analysis is accompanied by the foul, sulphurous odor of the Devil in the Details and the Demon of the Unforeseen Consequence. My favorite recent example – phasing out lead in gasoline. Not even the nuttiest environmentalist at the time would have argued about discount rates and crime statistics in advocating for the phasing out of lead. Yet Lo! Look at the effect. And of course, it works the other way as well. What does economics give us, then? Well, it gives us tools to measure, for one thing. Costs and benefits? On what scale do we compare them? All too often, the only measure is money. And from money, we get all of this other apparatus; discounting, temporal normalization, estimating total social [dis]utility by dividing total $ by the number of people in the affected population, etc, etc. Money is also enormously flexible. How much is a human life 'worth'?Hayek pointed it out. Price is information. It's data we can rely on. Economists worry about money.

  9. As much as I enjoyed the digs at economists,” If we’re doing something now that might hurt people none of us will coexist with 100 years, then so what?”If that doesn't feel wrong to you in your gut, you were probably deprives of a few socialization genes that (most of) the rest of us have … beware “individuality” arguments from genetic outliers.

  10. I'm gonna say it loud and clear, so that they'll have to say later that I was ahead of the curve – there is now a bubble in the public bashing of economists, who despite their embrace of the dangerous ideology of Keynesianism in macroeconomics, still have created a very decent formulation of the microeconomic foundations of our world. This fad created by the new Know-Nothings like Taleb has to end.People listen to economists because they have something rigorous to say about normative political decisions. This idea that economists were collectively wrong about the financial crisis, so they now are more than useless, is a fashionable yet stupid one. Eric Falkenstein has made the point, in my opinion, that even though economists have had predicted little, our political overlords have had a much worse track record. We're still talking in the broadest strokes, at the highest levels of our government, about the dangers of “deregulation” and “too much CDS”, which under closer scrutiny have revealed themselves to be terrible explanations for the crisis. Who else acts as economists do, who else isn't reasoning with undefined assumptions and most of the time explicitly political rhetoric? You can still say, “THIS IS FOR THE CHILDREN”, in this modern era, with no other caveats at all, and people will listen to you. If society has a well-defined framework to work within, society can make better decisions. Reasoning without models and mathematics gave us Marx and the Misesian School of Austrian Economics. I'll take my chances with the beleaguered pointy-headed fellows.

  11. This post is just weird. Economists are useless… once you've learned enough economics to use and apply analytical tools that were, by and large, invented and developed by economists?

  12. I don't want to impose a particular treatment decision on any of my future patients. But I'd like to do as good a job as possible presenting the probability of certain outcomes. I can always say “I don't know, because, how can we predict anything in life, after all?” But that doesn't help people considering aggressive chemotherapy or radiation, who want to know these interventions would barely improve their chance of survival (as in most types of pancreatic cancer) or bring their survival rate to more like 90% (as in certain types of leukemia). Of course, I would make the caveat that not everyone responds in the same way, and that my calculations are not based on complete information. But my (or some health wonk's) best calculations are much better than no information. Then, of course, I allow the patient to make his or her own decision. How do I know how many months of life is worth the experience of nausea hell for a particular person?Saying that economists can and should attempt particular calculations is not the same thing as saying economists are subsequently uniquely qualified to prioritize certain costs and benefits. No one could do that, unless we all had the same preferences. But making our best possible predictions is better than going in blind, so long as those who rely on economists' work realize that human predictions are not prophecies.

  13. “A zero discount rate plus a basic grasp of the relationship between technology and growth plus a reasonable projection of the current trend of technological progress implies an obligation to maximize economic growth rates with no concern whatsoever to avoid the incidence of future externalities of current activity.” This is – blandly – assuming that Julian Simon was completely right and that Herman Daly was completely wrong, with no uncertainty or qualification whatsoever. If it wasn't for what you follow this with, I would only be suggesting that you consider Daly's arguments against this picture and such reliability. The question is a bet that is at least somewhat more difficult.But as it is, I'm appalled and disappointed to see how philosophically easy you are making this picture and conclusion for yourself. With “the discount rate should approach infinity as we consider the welfare of future beings with whom there is no possibility of reciprocity”, and “if we’re doing something now that might hurt people none of us will coexist with 100 years, then so what?”, you simply shrug at the whole question of whether an outcome might possibly be bad. “Future beings” seem to be like people who are in another other dimension, which we can affect but which cannot ever affect us – where you are saying it would be morally meaningless, and only advantageous to us, if by deriving economically useful particles from their world we removed the oxygen from its atmosphere or afflicted its people with a disease. A difference is that, unless we go extinct by the time the people you care about die, the “future beings” are not hypothetical – they are absolutely certain to be real and be there. And you shrug at the idea that we might perhaps affect them badly or reduce their options or… whatever. If we did – if history should go badly – it's nothing to you. (After all, you've read philosophical arguments.)You are arguing here from lack of interest, taking it as a significant premise and challenging anyone to stir you from it. I have never taken libertarianism, or good economic thinking or good philosophical thinking, to require such a thing or to stem from such a thing. You should revisit this picture and see how it looks if you don't lean so comfortably on easily absolute assumptions. If you have a reason. I expected better. But my previous readings of you didn't include this sort of emotional okayness with lack of concern.

  14. Will, have you seen this? It's a paper claiming that, in addition to the well-known correlations between openness and conscientiousness and political attitudes, the agreeable are likely to be leftist in economics (but not necessarily on social issues) and the emotionally stable are likelty to be libertarian in economics (without any correlation on social issues). To shoehorn into the topic of the thread, public discourse is bound to become more quantitatively-oriented as we find stuff out, but that doesn't necessarily imply permanent domination by economists as other social sciences learn to use numbers.

  15. I can't imagine what you are thinking in saying economists have no competence to say what is a cost or benefit. That seems to me to be one of the things we know best. And the market interest rate clearly gives the opportunity cost of resources spend in the future. You complain about cost-benefit analysis, but seem to have nothing to offer in its place.

  16. I have not slept a lot so i might have missed something, but he seems to ignore Talebs assertion that past data is useless in predicting asset prices because the economy, altough it is a deterministic system in theory, is so complex that in practice we should treat it as random… but he does not quite because he uses fractals… wich I don't get…anyway the fact that he does not disclose the results of all his funds are interestingcheers

Comments are closed.