Overcoming Bias Sometimes Makes us Change our Minds, but Sometimes Not.

Suppose our opponents Overcome their Cognitive Biases? What then?

Take Global Warming. The only good reason for Believers in Global Warming to advocate expensive actions to reduce greenhouse gases is that the Benefits of prevented harm are greater than the Costs of reducing the gases. 

But suppose the vividness of climate catastrophe scenarios activates Believers’ Availability Bias, which in turn causes Focusing effects, narrowing the Believers’ focus to exclude that:

a)    There are other concurrent global crises – Malnutrition, Communicable Diseases, Access to Clean Water, Access to Education, Poverty, etc.– which are arguably as Harmful as Warming, but
b)    not enough economic resources (moneys) are available to deal with both Warming and also all the other global crises, so
c)    therefore priorities must be set, with better cost/benefit ratios winning.

Now suppose Believers Overcome their Availability and Focusing Biases.

Then economists tell Believers that the Cost/Benefit ratios for action on Warming are 1 to 5–e.g.,Costs of  reducing greenhouse gases of $10 billion gives $50 billion in Benefits of prevented harm– as  Nordhaus, Cline do.

But suppose Confirmation/Disconfirmation Bias causes the Believers to avert their eyes from Cost/Benefit ratios that economists give us for Malnutrition of  1 to 10, Communicable Disease of 1 to 50, etc.

Now suppose Believers Overcome their Confirmation/Disconfirmation Bias too.

Wouldn’t they  have to change their minds about the relative priority of expensive actions to reduce greenhouse gases?

(It did for some Nobel Prize-winning economists, UN Ambassadors, and Youth Groups. But it seems that changing their mind also required seeking the greatest good for the greatest number of people, which is not a universal value.)

On the other hand, charges of  ad  hominem  bias against believers (e.g.,  Believers just want  government grant money) won’t change our minds, because ad hominem arguements have no weight.

What about the other charges of Cognitive Bias against the Believers:
o    Overconfidence in the models that predict Harm
o    Loss Aversion and Endowment Effects causing overestimates of Costs
o    Illusion of Control causing overestimates of avoidability of Harm.

Under what circumstances would these give good reasons to change our minds?

GD Star Rating
loading...
Tagged as:
Trackback URL:
  • Stuart Armstrong

    Dear Bruce,

    You seem to be implying that those who care about global warming would not care at all about malnutrition or communicable diseases. In my experience, this is very far from the case. Rather the opposite, in fact.

    But people should invest their efforts in the domain where they can make the greatest marginal difference. Which domain do you feel is more open to change? GW affects the rich world, so it is more likely that money will be forthcoming for it. This money is not from a zero sum game – people are probably willing to pay much higher amounts to combat GW than to combat malnutrition. And, people not being economists, they will continue to contribute to charity at the same time. Their contributions may drop slightly – but it is probable that the marginal gain to the world from dedicating your life to GW is higher than the marginal gain of dedicating your life to malnutrition.

    Also, solutions for GW may be expensive, but they are much easier to implement than solutions for malnutrition or poverty – we know what they are, for a start.

  • Stuart Armstrong

    Damn! Sorry for the double post – don’t know how that happened.

    Anyway, to answer your actual question, if GW Believers overcome their biases, then they will be closer to the truth. So what happens then depends on what exactly that truth is. But that is the whole point of the debate! So we’d have to wait until we have a (pretty unbiased) assessment of the truth, then we can ask what unbiased believers would look like.

  • Doug S.

    Well, there are “easy” fixes that are also expensive. I can’t, for the life of me, figure out why a carbon tax is anything but a great idea.

  • http://cumulativemodel.blogspot.com aaron

    Oh Bruce, in your previous post you missed a skeptic bias, though I’ve never been accused of it.

    Curiosity.

    I want to see what happens both with the climate and what comes from a more robust economy. Change is good.

  • http://cumulativemodel.blogspot.com aaron

    Doug here’s a few possibilities, carbon taxes would require a shift of the tax burden to both low margins industries, businesses, and most of all the poor.

    A big increase will put pressure on people to change their behavior who have a very poor understanding of what good behavior is. Without understanting about how cars and traffic work, people will change their behavior in ways that will increase traffic and consumption. Most people falsely believe that slower acceleration saves fuel. Very much not true, faster, smooth acceleration is more efficient generally or atleast neglibly less efficient at the car level. The impact on traffic is bigger though.

    It may cause people to shift to bikes, but unless there is proper infastructure to prevent the interfenance with traffic emissions may go up (not sure how safe biking is either).

  • Bruce K Britton

    Dear Stuart:

    1. You say “You seem to be implying that those who care about global warming would not care at all about malnutrition or communicable diseases. In my experience, this is very far from the case. Rather the opposite, in fact.”

    No, I am implying that IF they are affected by the Availibility and Focusing Biases, THEN they would not address Malnutrition or Communicable Diseases, simply because those issues would not be present in their focus.

    But IF they are not affected by the Availablity and Focusing Biases, THEN they are like the people ‘in your experience,’ who have malnutrition and communicable disease along with warming in their focus.
    The next question is, are those people in your experience (the ones with Malnutrition, etc, along with Warming in their focus) are they subject to Confirmation/Disconfirmation Bias causing them to avert their eyes from comparing the relative cost/benefits ratios of Warming, Malnutrition, etc? I’m hoping you’ll get back to me on that question, since you are the one who has experience of these people, not that I don’t know some of them too.

    2. You say “GW affects the rich world, so it is more likely that money will be forthcoming for it.”

    Your arguement here — that people in the rich world will be more willing to pay to combat Global Warming because it will affect them more than Malnutrition in poor countries — is precisely in line with my point that

    “seeking the greatest good for the greatest number of people … is not a universal value”

    Instead, your hypothetical deciders hold a different value, namely the value of ‘seeking the greatest good for the people in the rich world.’

    Since values are not a matter of fact, not empirically decidable, subject to the ‘is/ought gap’, your deciders’ differing value here is not a matter that is arguable on the basis of facts, but only on the basis of ethics, which is more dicey, and altogether beyond our scope, I hope we can agree.

    3. You say “So what happens then depends on what exactly that truth is…”

    Based on your comment and my 2 above, it seems to me that what you mean is that ‘what happens then depends on’ (not ‘that truth’ in the sense of ‘truth’ as factual material) but upon the ‘values’ the deciders hold, in this case whether they value most ‘the people in the rich world,’ in which case choose GW, or whether they value most ‘the greatest number of people’ which includes those in the undeveloped world who suffer from Malnutrition and Communicable Diseases? In the latter case wouldn’t they have to decide in favor of the largest cost/benefit ratio?

  • Hopefully Anonymous

    Bruce,
    Are you a “global warming believer”? that also believes greenhouse gas emmissions should be curtailed? If you are, then I think this is a useful thought experiment that you’re engaging in. If you’re not, then the supposing here seems a little one-sided, without a good explanation for why it’s so one-sided, in my opinion.

  • Constant

    “ad hominem arguements have no weight”

    Whether they have weight depends. For example, if you tell me that I just want government to get more power, then that will probably not persuade me. But it might persuade somebody else who was listening to me.

    Why and under what circumstances might it persuade them? Ad hominem undermines the credibility of the speaker. This is irrelevant if the speaker is merely outlining an argument that is visibly sound. Such arguments stand apart from the speaker, whatever the speaker’s credibility. But the credibility of the speaker is relevant if the argument relies on it. And many speakers do. Many speakers do take a, “trust me”, approach. For example, if a computer model predicts global warming, then the credibility of that prediction depends on the credibility of the computer model. If the audience does not examine the computer model directly, then the audience must rely on the assessment of others who do examine the computer model directly, and those assessors need to have credibility, and if that credibility is successfully attacked then the argument is undermined. This is in contrast to, say, a mathematical proof, which remains exactly as sound as it ever was, no matter what terrible things are revealed about the mathematician who created it.

  • Hopefully Anonymous

    Great point, Constant.

  • Stuart Armstrong

    Dear Bruce,

    To answer some of your points, and to provide as full a picture as I can of these people:
    are they subject to Confirmation/Disconfirmation Bias causing them to avert their eyes from comparing the relative cost/benefits ratios of Warming, Malnutrition, etc?
    No, they’re just not economists. They don’t see why they can’t solve GW, and malnutrition, and poverty, etc… Their answer to any attempt to bring up cost/benefits is that “there is enough money out there to solve all these problems” (which is of course true). When told that this money isn’t available, they respond “it should be. And I’ll work to make sure it is.” I think they lack the concept of “opportunity costs” more than anything. Seems more ignorance than bias.

    For the other bit, you have misunderstood my emphasis on the rich world. They are not interested in solving GW for the rich world (or not much, anyway). The point is that since the rich world is affected, then vastly more sums of money will be available to combat the problem (compare the funding, during the 90’s, for HIV prevention and for malaria prevention). So here the cost/benefit problem does not apply, since the cost side is not fixed. This money can then be used to solve the issue globally, which is what these people really want.

    Their ethics, for which I have some sympathy, is “the greatest good for the greatest number, but with special emphasis on those who are at the bottom of the poverty and misery scales”. Their major bias is some sort of affect bias, which can be summarised as “people animated by good intentions will help more than those without” – i.e. aid is better than trade. It’s not an overly strong bias, and they will accept counter-examples and arguments, but it’s clearly in the background of their minds.

    what you mean is that ‘what happens then depends on’ (not ‘that truth’ in the sense of ‘truth’ as factual material) but upon the ‘values’ the deciders hold
    No. The biases you treated in detail may depend on believers’ values, and consistency with those values. But look at the biases you listed at the end:
    o Overconfidence in the models that predict Harm
    o Loss Aversion and Endowment Effects causing overestimates of Costs
    o Illusion of Control causing overestimates of avoidability of Harm.

    All these are biases causing factual errors, not value issues. Before even thinking of addressing these biases, we have to have a good idea of the reliability of the models, the true costs, and how truly hard it is to avoid harm.

  • http://yorkshire-ranter.blogspot.com Alex

    Flaw: this all assumes that malnutrition and communicable diseases are independent of climate change. Which is an incredibly stupid assumption; perhaps the biggest single climate worry is the effect on agriculture (I shouldn’t need to spell out that food has something to do with hunger), and water supply (i.e. a major determining variable for the prevalence of communicable disease). You think it’ll be easier to keep shit out of the drinking water when half the population of Bangladesh is in refugee camps in West Bengal?

    This is a very serious factual issue which the Lomborg fan club just don’t engage with and have never done.

  • Constant

    Alex writes: “this all assumes that malnutrition and communicable diseases are independent of climate change. Which is an incredibly stupid assumption […] This is a very serious factual issue which the Lomborg fan club just don’t engage with and have never done.”

    That was easy to check and, as it turns out, disprove.

    Googled article: “Some scientists question such concerns. Danish statistician Bjorn Lomborg has become a spokesman for the view that trying to repair global warming will cost more money than just working to adapt the world to it. He suggests, for instance, that it would be cheaper to cure and eradicate malaria than to attack the rising temperatures that could expose millions more people to the disease.”

    Source:

    http://www.lcv.org/newsroom/in-the-news/page.jsp?itemID=32635952

    Article by Lomborg and Rose: “Mr. Gore says that global warming will increase malaria and highlights Nairobi as his key case. According to him, Nairobi was founded right where it was too cold for malaria to occur. However, with global warming advancing, he tells us that malaria is now appearing in the city. Yet this is quite contrary to the World Health Organization’s finding. Today Nairobi is considered free of malaria, but in the 1920s and ’30s, when temperatures were lower than today, malaria epidemics occurred regularly. Mr. Gore’s is a convenient story, but isn’t it against the facts?”

    Source:

    http://www.opinionjournal.com/editorial/feature.html?id=110009552

    Evidently, and unsurprisingly, Lomborg specifically recognizes and addresses a potential dependency of disease on climactic change.

    Keep in mind that I am not either agreeing or disagreeing with Lomborg’s claims. I am only addressing the claim made by Alex that “the Lomborg fan club” (a label presumably intended to include Lomborg himself) “just don’t engage with” the potential dependency of disease on global warming.

  • http://cumulativemodel.blogspot.com aaron

    Alex, regugee problem of bangledesh is as likely with cooling as warming. Either way, it will continue to sink and erode.

  • http://cumulativemodel.blogspot.com aaron

    Unless of course they, you know, develope reservoirs, sanitation, dykes, levies, etc.

  • http://yorkshire-ranter.blogspot.com Alex

    He hasn’t engaged with it; he’s made a claim regarding one particular disease and one particular city, which does not invalidate the general case. Very specifically, he chose not to deal with the question but to dish Gore on the specific instance. Here are WHO documents regarding malaria epidemics in Kenya. As it happens, there certainly were areas of Kenya that had no malaria in the 20s – it is theorised that it was introduced by humans after a railway was built.

    Curiously I can’t seem to find an actual first order reference to Gore actually saying this, but it is heavily quoted by folk like Iain Murray of the CEI, whose bona fides I dispute for reasons too frequently stated to need rehearsal.

  • http://yorkshire-ranter.blogspot.com/ Alex

    Unless of course they, you know, develope reservoirs, sanitation, dykes, levies, etc.

    And a pony.

  • http://yorkshire-ranter.blogspot.com/ Alex

    More broadly, I think this is an example of a problem with the OB project. Classifying people who disagree with you as the victims of “bias” is not actually all that far from classifying them as “mad”. It’s quite easy to slide into using “bias” as a delegitimiser of dissent.

  • michael vassar

    Unfortunately, if bias, madness, or whatever is ubiquitous, it is quite likely that some (most?) dissent from some (most?) positions is illegitimate. Pretending that illegitimate dissent is legitimate, say, by engaging with creationists, just wastes time. If the insane outnumber the sane, such “titration” can potentially waste *all* of the efforts of reasonable people.

    OTOH, the delegitimization of dissent is Powerful stuff, very dangerous, and the historical precedent for its use has been very poor. I have long believed that Popper’s basically bogus definition of science was intended primarily as a political weapon against all supposed sciences that aimed to delegitimze dissent and only secondarily as a crude epistemological tool for those who had previously possessed only even cruder epistemological tools. Historically, qualitative improvements in research program quality seem to have resulted from epistemological tools for delegitimizing arguments of specific and fairly easily verifiable forms, as in the categorization of logical fallacies. This works better if one is attacking specific arguments made by specific people than if one is attacking positions. However, it is ultimately necessary that arguers as well as arguments be vulnerable to attack for dishonesty, as otherwise dishonest arguers are prone to wasting people’s time by repeatedly making dishonest arguments for the same position. Academic credentials are one extremely crude way of doing this. Reputation markets could probably do the job MUCH better, as, for that matter, ordinary reputations already do within small circles.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Michael, I agree that “arguers as well as arguments must be vulnerable to attack for dishonesty,” though I have less hope for “reputation markets” than you.

  • Bruce K Britton

    Dear Stuart:

    I think we are making progress. We now have 2 classes of believers

    Class 1 is my guys, who have Availability, focusing, and confirmation/disconfirmation biases, so do not consider malnutrition, etc, but only warming.

    Class 2 is your guys, who don’t have those biases, are not economists, think we can solve all the crises mentioned, if told there is not enough money they pledge to work to make sure there is, lack the idea of opportunity costs, give special emphasis to the poor in their values, and have the bias that ‘people with good intentions will help more than those without’.

    Class 1 I claim will, when they overcome their availability focusing, and conf/disconf biases, see that there are other crises, see the others have higher cost/benefit ratios than warming and opt to do the other crises instead of warming.

    Class 2 will (you tell me but I’ll guess now) work to make sure there is enough money to solve all the crises. If they succeed, then they will set off to solve all the crises. If they fail to get the money to solve all the crises, then they are back to having to prioritize. How do they prioritize? One way would be by cost/benefit ratios. Another way I guess you are suggesting is that they would ignore the cost/benefit ratios and prioitize by wealth of the recipient, i.e., by helping the poor more than the rich countries.

    Is this the scenario you have in mind? Lets get this straight and then go on.

    On another matter, I see from your last paragraph in your last comment that you were referring in your second comment to the next to last paragraph in my original post, which is good, and I would like to discuss that issue, of the other biases– overconfidence, loss aversion and illusion of control– with you separately from the above issue, from which it can be separated.

  • Stuart Armstrong

    I see from your last paragraph in your last comment that you were referring in your second comment to the next to last paragraph in my original post,

    One of the greatest sentences ever written ^_^

    I’ll respond substantively to your comments tomorrow. Though I do need to ask – the views you state about cost/benefits, are they the general consensus?

  • Constant

    “He hasn’t engaged with it; he’s made a claim regarding one particular disease and one particular city”

    That’s engaging with it. What you should have said, if you are right, was that, “apart from XXX, he has not engaged with it.” However, having falsified your claim once, I am not particularly motivated to check your new claim – that he has done no more than make a claim regarding one particular disease and one particular city. I’m not your butler. I cleaned up after you once, but no more.

  • Stuart Armstrong

    Is this the scenario you have in mind? Lets get this straight and then go on.

    Your description of class 2 seems close enough to real people to be usable.
    I doubt they would ever overcome their biases – the most likely is that they get heavily involved in one type of project and focus narrowly on it (in fact, once they are heavily invested in one type of project, then the marginal difference they can do is probably higher by staying in that project rather than shifting).

    But let’s imagine they overcome their biases (but not their values). They will be more open to private sector involvement in their projects – nothing earth shattering there. But now let’s assume they are open to cost benefit analysis, and overcome any confirmation bias. What will happen then?

    Well, if your cost-benefit analysis is correct, they will shift their efforts to malnutrition and those other things. If my analysis is correct (that the effect of GW on the rich world means that much more money will be available to combat GW that malnutrition) then what happens gets more complicated. If they do not benefit from this extra funding (for instance, if they manage a fixed endowment), they will shift to malnutrition. If they benefit from the extra funding (they manage a GW charity or a research institute into alternative energy, for ex.) they will remain in GW.

    But there are many other moral issues that cloud the whole picture – GW was caused by the rich nations, for example. So depending on what moral value people put on concepts like ‘historical responsibility’ (I put none at all, by the way – I refuse to feel ashamed or proud of my ancestors in any way) they may or may not change. There are other moral values, arguments and biases (some of which I share) that also advantage GW over other issues.

  • http://profile.typekey.com/brucekbritton/ Bruce Britton

    Hi Stuart:

    I think we have made excellent progress.We agree that if the cost/benefit analysis is correct, and they do not benefit from the extra funding you expect, then they will go for the options with the highest cost/benefit ratios.

    The whole exchange with you has been helpful in clarifying several matters for me. Thanks very much.

  • StefanoC

    What happened to mixed strategies and diminishing return ?

    I don’t think that the donors/investors should put all their resource on the climate change problem, or all on the malaria problem, or all on the malnutrition one.

    A balanced “portfolio” is always best. On this I think all GWB (the Global Warming Believers, I mean) that have overcome their bias would agree, as would those of the anti-GWB that have overcome _their_ bias.