The good versus the bad

I was wanting to touch on something that I’ve tried to discuss before. I didn’t do a great job last time, and a I may not do a great job this time, but I’ll give it a go. A regular criticism of the coverage of climate change is that there is too much focus on the bad aspects, and not enough – if any – on potentially good aspects. Personally, I think this criticism is wrong, and I will try to explain why.

Imagine we were actively trying to change our climate; that we felt that we could change it to something better than we have now. Before doing so, we would want to know how our climate would change in ways that we would regard as good, ways we would regard as bad, how – overall – this compared to our current climate, and the costs associated with making the change. We could then decide if changing our climate were worth doing.

However, this isn’t what we’re doing. Climate change is inadvertant; it’s a side effect of energy generation. In a simple sense, the correct comparison is between the benefits of the energy we generate and the risks associated with doing it as we do. Consequently, we really need to consider the possibility of severe outcomes due to climate change, the benefits of the energy that we generate, and the costs associated with generating it in a way that minimises these risks. That some of the impacts of climate change might be good, doesn’t trump that some might be very bad.

We might decide that the benefits outweight the risks and that we should leave things as they are. We might decide that the risks could be severe and that it would be worth generating energy differently, or that it’s worth changing our lifestyles so that we use energy more efficiently (or both). We may even find that it would be quite straightforward to generate energy in a way that would minimise the risks without severely impacting our lifestyles. However, whatever we decide, I don’t think that the possibility that there might be some good aspects to climate change is all that relevant.

Therefore, until such time as we actually want to change our climate, that there could be some positives doesn’t – I think – really come into the analysis. Of course, if we had perfect knowledge, it might, but we don’t, and that’s the key point. We’re dealing with possibilities, and so the possibility of really severe outcomes, trumps the possibility of some good outcomes. To be clear, though, this is just my current thinking on this, and I don’t think I’ve explained this as clearly as I would have liked. I’m quite interested in what others might think, so if anyone thinks differently, or would like to expand on this, feel free to do so through the comments.

Advertisements
This entry was posted in Climate change, Global warming, Personal, Science and tagged , , , . Bookmark the permalink.

92 Responses to The good versus the bad

  1. tallbloke says:

    On the current long term (120myr) trend, CO2 levels will drop below 170ppm in about 1.7 million years and all life will end on Earth as vegetation and plankton die at the base of the food chain.

    Humans are doing a great job extending this remarkably short timespan. We’ve bought another 5m years for the planets so far if we can keep up the good work.

  2. tall,
    I’ll post your comment, but won’t bother responding. It doesn’t really deserve one.

  3. tallbloke says:

    Cool, thanks. Just a further observation if I may. We’ve been stuck in an ice-age for several million years, but if it ends, the average temperature is likely to rise 8C or so, as it has in the deep past. Our intervention could therefore be very important, as extending the conditions in which plants and plankton can survive by 3m years could make all the difference. We don’t know if the ice age will end in 1m years, 4m years or 10m years. But where there’s life there’s hope. And we’ve increased the chances of survival quite a bit.

  4. I’ve no idea why you’re thanking me, I wasn’t being complimentary. Your second comment is no better than your first. In fact, it may be even more ridiculous.

  5. Marlowe Johnson says:

    small nitpick. it’s not just energy generation as any casual look at the inventory of global emissions will show. land use change, process emissions, and cow farts all play a big role as well.

  6. Marlowe,
    Indeed, there are other things that are also relevant. The same basic idea should apply though, it should be some kind of cost benefit analysis and that some of the inadvertant impacts might be good doesn’t really come into it.

  7. > In a simple sense, the correct comparison is between the benefits of the energy we generate and the risks associated with doing it as we do

    This looks a mistake. You don’t compare benefits with risks; you compare them to costs.

    > we really need to consider the possibility of severe outcomes due to climate change, the benefits of the energy that we generate, and the costs associated with generating it in a way that minimises these risks

    Ah, see, you knew that really. You’ve now shifted to what you really meant to say: you’ve decided that our priority is to minimise risks, and you’re going to balance energy-benefit against costs-assoc-with-gen-in-risk-minim-way. That’s possible, but its a choice you need to make explicitly.

    > We might decide that the benefits outweight the risks

    Oh. Now you’re back to comparing benefits to risks. I think you’re confused (well, at the least inconsistent) in the language you use.

    I think its a cost-benefit calculation. Both costs and benefits are weighted according to the probability of their occurring. Things that are unlikely but severe costs are called “risks”; but in the calculation, they’re just a cost weighted by probability.

    > the possibility of really severe outcomes, trumps the possibility of some good outcomes

    Why you think that isn’t at all clear. If you were trying to say the probability-weighted cost of “really severe outcome” was greater than (the probability-weighted benefit of “some good outcomes” + “the benefits of the energy that we generate”) and so we shouldn’t do it then I’d kinda understand, but ask you for your numbers; but use of the word “trumps” implies, as in cards, that you’re not doing that balance.

  8. John Davis says:

    Of course with hindsight, the benefits of both fossil fuel and of the climate change we have already caused since the 1850’s have been immense.

  9. RickA says:

    ATTP:

    In order to do a proper cost benefit analysis we need to know what happens if we do nothing (inaction), as well as what happens for every proposed action.

    We need to know what bad things will happen and how much they will cost.

    We need to know what good things will happen so we can net them against the bad.

    So a proper cost benefit analysis does need to know the “good” things which will happen, for every proposal – as well as for inaction.

    I am not a policy expert – but I thought that in a cost/benefit analysis the proposal (whatever it may be) was compared to doing nothing (inaction) – and the costs and benefits of the proposal were weighed against the doing nothing case (which may have its own costs and benefits).

  10. If global warming were a net benefit, I would still be against a geo-engineering policy to purposefully increasing CO2. I would not do so because of the risks (and because of the regional variations in costs and benefits). In this sense, you can compare benefits to risks even if it is an apples to oranges comparison.

    When you compute how big the (negative) impacts of climate change are, however, you naturally take all effects into account. When you look at the impact on plant growth, you take into account that CO2 improves plant mass growth under ideal conditions and that plants die due to heat waves, being badly adjusted to the new seasonal cycle and drought. When you look at the impact on costs of air-conditioning, you take both less heating in winter and more cooling in summer into account. When assessing the impacts, scientists do and should also take the positives into account to get an unbiased estimate of the impact.

  11. WMC,
    I knew this would be a post that would get you going 🙂 . Once again, it may well be me using terminology that isn’t strictly correct, or standard. I’m just a lowly physicist.

    This looks a mistake. You don’t compare benefits with risks; you compare them to costs.

    Okay, in a strict sense this is true, but – presumably – assumes that one can associate a cost with every possible outcome. But, how do we estimate the cost of ecosystem collapse, for example?

    Now you’re back to comparing benefits to risks. I think you’re confused (well, at the least inconsistent) in the language you use.

    My language may be confused. I really am just talking about possible outcomes that have a negative effect. So, when I say “risk” I’m really just talking about the risk of something negative that presumable carries some cost.

    Both costs and benefits are weighted according to the probability of their occurring. Things that are unlikely but severe costs are called “risks”; but in the calculation, they’re just a cost weighted by probability.

    Fine, this is essentially what I’m getting at. We could use the term “cost” but I’m still not convinced that we can necessarily assign an actual value to every outcome; well in a way that doesn’t require a massive amount of judgement. So, if you think that “cost” is a better term than “risk”, that’s fine by me.

    Why you think that isn’t at all clear. If you were trying to say the probability-weighted cost of “really severe outcome” was greater than (the probability-weighted benefit of “some good outcomes” + “the benefits of the energy that we generate”) and so we shouldn’t do it then I’d kinda understand, but ask you for your numbers; but use of the word “trumps” implies, as in cards, that you’re not doing that balance.

    I’m trying to get at something here that I may not have done very well. Climate change is inadvertant, not planned. We aren’t trying to do it; it’s a consequence of how we generate energy (and food, …) Therefore, it’s not clear to me that the possibility of some good outcomes really comes into the calculation because we don’t know if they will actually materialise. Ultimately we would normally want to avoid the bad outcomes and compare how we might do so with the cost associated with doing so.

    For example, consider a scenario where the probability-weighted benefit of some good outcome + the benefits of the energy we generated exactly balanced the probability weighted cost of really severe outcomes. Let’s also imagine that we could easily change our energy generation so that we were no longer changing our climate as a consequence of energy generation. Wouldn’t we go ahead and do so, given that doing so would allow us to avoid the potentially bad outcomes? That there may now be some good outcomes that will no longer materialise seems irrelevant, given that we aren’t doing this in order to achieve those outcomes.

  12. Rick A,

    We need to know what good things will happen so we can net them against the bad.

    So a proper cost benefit analysis does need to know the “good” things which will happen, for every proposal – as well as for inaction.

    I’m disagreeing with this. In a sense, the “good” is the energy that we generate, the “bad” is that this is inadvertantly changing our climate. My argument is that the possibility of some good outcomes doesn’t mean really come into it because this isn’t what we’re trying to do; we’re simply trying to generate energy and we should really be considering the risks (or costs) associated with changing our climate and the costs associated with generating energy in a way that doesn’t (or not as much) change our climate.

  13. Of course with hindsight, the benefits of both fossil fuel and of the climate change we have already caused since the 1850’s have been immense.

    Hmmm, but that doesn’t mean that we should continue as we are.

  14. Victor,

    When assessing the impacts, scientists do and should also take the positives into account to get an unbiased estimate of the impact.

    Yes, but isn’t this slightly different? Trying to asses the impacts you should consider everything. However, when doing a cost benefit analysis, it still seems that the possible negative costs are more important that possible positive benefits. Lets imagine that we actually had some evidence that the impact of climate change might be positive. Wouldn’t we then want to be pretty certain that this would actually be the case if we were to decide to simply carry on as we are?

  15. ATTP: “ Wouldn’t we then want to be pretty certain that this would actually be the case if we were to decide to simply carry on as we are?

    Yes, that was my first paragraph. Plus that it is impossible to be “pretty certain” when changing a complex system like the climate system. There will always be unknown unknowns. Call me a conservative, but you simply do not mess with a complicated system our civilization is build on.

  16. Willard says:

    > Things that are unlikely but severe costs are called “risks”; but in the calculation, they’re just a cost weighted by probability.

    Other things that are more likely and less severe can too be called “risks.” There are risks this thread will lead to more ClimateBall ™, for instance. There could be benefits too, of course.

    ***

    > I think its a cost-benefit calculation. Both costs and benefits are weighted according to the probability of their occurring.

    Perhaps one day we’ll see such calculation regarding AGW. I’m not sure we’ve ever did so for something that is remotely the size of the Earth. I expect to find such calculation right next to an engineer-level formal derivation of the doubling of CO2:

    http://neverendingaudit.tumblr.com/tagged/engineerilyderiving

    In the meantime, I think it’s safer not to take such parlance too seriously.

  17. Victor,

    There will always be unknown unknowns.

    Ahh, yes. I think what I’m suggesting is essentially a form of uncertainty isn’t your friend.

    Willard,

    There are risks this thread will lead to more ClimateBall ™, for instance. There could be benefits too, of course.

    I wrote this partly for you 🙂

  18. Willard says:

    Political risks might be tough to calculate:

  19. And Then There’s Physics says: “I think what I’m suggesting is essentially a form of uncertainty isn’t your friend.

    Surely you are not referring to this wonderful post: Fans of Judith Curry: the uncertainty monster is not your friend

  20. Victor,
    Yes, that would probably be the one 🙂

  21. paulski0 says:

    I don’t see why the intentionality of the climate change matters in this context? The question is whether we should intentionally change our course of action because of this unintentional climate change. That means analysing the costs and benefits of comparative scenarios whereby we continue as we are or cut fossil fuel emissions, trying to understand all impacts of each course, positive and negative.

    As a close analogy, the cooling effects of aerosols are unintentional but have acted to slow warming, which has been arguably beneficial. If there were no health implications of aerosols would we ignore this benefit and actively try to cut emissions regardless?

  22. Willard says:

    Some more intangible risks:

    Do central banks set their interest rates based on any formal cost-benefit analysis?

  23. Paul,
    As far as intentionality goes, I was trying to distinguish between what we would do if we were actively trying to change our climate and what we would do if – as it is – it were unintentionally. If it were intentional, we would presumably want much greater certainty before deciding whether to go ahead, or not. Since it is unintentional, we can make decisions even if the outcomes are not that certain. Also, the argument I’m trying to make (and which may be wrong) is that potentially good outcomes that are not what we are trying to do, have less value than potentially bad outcomes. If the good outcomes that we don’t specifically want, don’t happen, that presumably has less impact on us than if the potentially bad outcomes do happen.

  24. I think I am with Willard here. Apply a kind of cost-benefit analysis designed to assess the relative merits of some new, say, transport system, is orders of magnitude simpler than trying to do such a thing for the planetary system (of which our human economics is a part).

    We can do a complex Probabilistic Risk Assessment for a nuclear power plant but our failure of imagination and simplistic methodology will ensure we don’t spot a design flaw in an indicator light when we are looking for the risk of operational failure. If we fail with TMI, what chance the planet?

    Species loss is an ‘issue’ (something that is happening, rather than something that might happen: a ‘risk’). It is happening and with global warming it gets worse. Who puts a price on that? Not economists because it is unquantifiable.

    The other concern I have with the discussion here is that the ‘cost/benefit’, even if it could be reformulated into something more useful than a NPV for Earth ( 🙂 ) is a changing value. If we look at 30, 60, 90, … years … what do we see? Short term value (the 3rd world taken out of poverty with the help of King Coal) at the expense of a trashed future?

    Maybe instead of trying to do a cost/benefit analysis we did something more like a balanced scorecard for the planetary system, with an emphasis on qualitative boundaries not numerical (chalk and cheese) comparisons … hmmmm … hasn’t someone done that already?

    http://www.stockholmresilience.org/21/research/research-programmes/planetary-boundaries/planetary-boundaries-data.html

  25. A cost benefit analysis also just looks at the mean. When you are betting our common house, one should add a few times the uncertainty to that, at least in my view as a conservative.

  26. > Fine, this is essentially what I’m getting at

    OK, good. I do think its a good idea to make it clear what your words mean, though.

    > Climate change is inadvertant, not planned. We aren’t trying to do it; it’s a consequence of how we generate energy (and food, …) Therefore, it’s not clear to me that the possibility of some good outcomes really comes into the calculation because we don’t know if they will actually materialise.

    You said the bit about not-planned in your original post. I don’t understand why that makes any difference at all. You do a cost-benefit analysis either way, whether you’re planning to make the change or not. Suppose some denialist were to say “I like the idea of a warmer Earth. As far as I’m concerned, this change is planned”. Does the fact that he considers the change to be planned make the slightest difference to the costs and benefits?

    > For example…

    Ah, now you touch on something completely different, which I think might be what you actually meant in the first place. Roughly, that if we have a situation where C_1 = (A_1, B_1, …) and C_2 similarly; with C_n being the overall benefit (or cost if negative) of scenario n; and A_n, B_n etc are the components of the balance; then we should prefer a scenario which also minimises (|A_n| + |B_n| +…).

  27. There have been other climatic periods before the current era which a completely different set of flora and fauna found to their liking, and thrived. Human development really took off 10,000 years ago, after the last ice age, so we must reasonably assume that the climate that has generally prevailed was the climate which is ideal for our species to thrive. Over that period of fairly stable weather—no sudden and major climatic change—a rich and varied selection of flora and fauna has lived alongside us and, we must assume, also mostly thrived in the prevailing climate.

    Since 1850, when compared with past global changes in pre-history, we unwittingly began to change our climate very rapidly as a by-product of technological advance. In hindsight this was very dangerous, as we had no understanding of possible side-effects—desirable or not—and it’s only recently that we’ve become aware of the negative impacts of the new climate we’re heading into.

    When we eventually understand all the nuances of climatic change on our civilisation and the other species that share our world with us—and we’ve carried out a thorough, scientifically-comprehensive, risk-benefit analysis—we might decide that a certain type of change has benefits. However any change needs to be carried out slowly if we are to allow some of the flora and fauna we know and love to adapt and evolve. To launch into a dramatic change by accident and then continue on regardless even before we understand all the impacts, is the totally reckless behaviour we usually associate with drug addiction.

  28. Sam taylor says:

    John,

    I think it’s more the stability than anything else that’s made advanced civilsation possible over the last 10k or so years. Things have been remarkably stable. Whether or not the current conditions are “optimal” seems less important to me than the fact that we’ve had 10,000 years to optimise things like agriculture and so on to the prevailing conditions.

  29. WMC,

    I do think its a good idea to make it clear what your words mean, though.

    Indeed, but – I find at least – that in this context it come sometimes be harder than it might otherwise be.

    I don’t understand why that makes any difference at all. You do a cost-benefit analysis either way, whether you’re planning to make the change or not. Suppose some denialist were to say “I like the idea of a warmer Earth. As far as I’m concerned, this change is planned”. Does the fact that he considers the change to be planned make the slightest difference to the costs and benefits?

    Well, because if we were planning to do so, I think we’d want much better climate models than we currently have. So, it still seems to me that how we might assess this is different if we were planning to change our climate to if we are inadvertantly changing it.

    Ah, now you touch on something completely different, which I think might be what you actually meant in the first place. Roughly, that if we have a situation where C_1 = (A_1, B_1, …) and C_2 similarly; with C_n being the overall benefit (or cost if negative) of scenario n; and A_n, B_n etc are the components of the balance; then we should prefer a scenario which also minimises (|A_n| + |B_n| +…).

    This may have been what I meant, but you might need to explain this to me a little more before I can say for sure.

  30. Willard says:

    Let’s start the year by quoting Nordhaus on net benefits:

    Suppose we were thinking about two policies. Policy A has a small investment in abatement of CO2 emissions. It costs relatively little (say $1 billion) but has substantial benefits (say $10 billion), for a net benefit of $9 billion. Now compare this with a very effective and larger investment, Policy B. This second investment costs more (say $10 billion) but has substantial benefits (say $50 billion), for a net benefit of $40 billion. B is preferable because it has higher net benefits ($40 billion for B as compared with $9 for A), but A has a higher benefit-cost ratio (a ratio of 10 for A as compared with 5 for B). This example shows why we should, in designing the most effective policies, look at benefits minus costs, not benefits divided by costs.

    Let’s also skip to the punch line:

    [T]here are indeed substantial net benefits from acting now rather than waiting fifty years. A look at Table 5-1 in my study A Question of Balance (2008) shows that the cost of waiting fifty years to begin reducing CO2 emissions is $2.3 trillion in 2005 prices. If we bring that number to today’s economy and prices, the loss from waiting is $4.1 trillion. Wars have been started over smaller sums.10

    http://www.nybooks.com/articles/2012/03/22/why-global-warming-skeptics-are-wrong/

    It would be interesting to adapt Nordhaus’ DICE model to analyze the net benefit of creating a food crisis:

    It took the brilliant minds of Goldman Sachs to realize the simple truth that nothing is more valuable than our daily bread. And where there’s value, there’s money to be made. In 1991, Goldman bankers, led by their prescient president Gary Cohn, came up with a new kind of investment product, a derivative that tracked 24 raw materials, from precious metals and energy to coffee, cocoa, cattle, corn, hogs, soy, and wheat. They weighted the investment value of each element, blended and commingled the parts into sums, then reduced what had been a complicated collection of real things into a mathematical formula that could be expressed as a single manifestation, to be known henceforth as the Goldman Sachs Commodity Index (GSCI).

    For just under a decade, the GSCI remained a relatively static investment vehicle, as bankers remained more interested in risk and collateralized debt than in anything that could be literally sowed or reaped. Then, in 1999, the Commodities Futures Trading Commission deregulated futures markets. All of a sudden, bankers could take as large a position in grains as they liked, an opportunity that had, since the Great Depression, only been available to those who actually had something to do with the production of our food.

    http://foreignpolicy.com/2011/04/27/how-goldman-sachs-created-the-food-crisis/

  31. Sam taylor says:

    Willard,

    The Chinese in no sense “own” America. All of the US debt that is owned by China is denominated in US dollars, which can only be spent in the USA, on American goods. An analogy to this is when someone buys you a gift voucher for somewhere like homebase, and you find yourself wondering around homebase with a fifty quid voucher wondering what to buy there. The same is true of the Chinese. They only have those vouchers in the first place because they decided to send goods to the USA, the goods were purchased with dollars, and the Chinese then used these dollars to buy bonds, which basically means they opened a savings account. The long and the short of it is that American citizens got real goods and services, and the Chinese got bits of paper with old dead white men on them. Now they have a dollar denominated savings account, and need to figure out what to buy with it, however whatever they buy it’ll have to be denominated in dollars and thus will be made/sold in the USA. The money need never leave. That’s why having a reserve currency is so powerful.

  32. Actually, what Victor says may be something like what I was trying to get at here

    A cost benefit analysis also just looks at the mean. When you are betting our common house, one should add a few times the uncertainty to that, at least in my view as a conservative.

    We could go and assess the impact of climate change and estimate some kind of cost (I guess this is what IAMs do). However, there will clearly be uncertainty in this estimate with some probability of there being a net benefit and some chance of there being a very high net negative cost. My argument is essentially that even if the mean cost were small, and the distribution were symmetric (i.e., as likely to have net positive as net negative outcomes), you’d still want to give extra weight to the negatives than the positives. It’s my view that a positive impact that you’re not specifically wanting has less value than a negative impact that you might really want to avoid.

  33. Andrew dodds says:

    Sam – indeed. As we found out with Iceland, you can’t reposess a country.

    It’s also true that the amount of money is not a conserved quantity, which tends to trip up a lot of people. It only works if we think of it as more or less conserved, the moment people stop believing that it ceases to be money and becomes toilet paper.

    To me, the real problem with using abstracted-to-money cost benefit analysis for global warming is that you have thrown so much information out of the window in the abstraction that the result becomes meaningless. As in ‘in XXX years, the economy will be 4 times bigger so we can lose a third of it and still end up ahead of the game’ – which is both technically true and blatantly insane, as the loss of a third of the economy is utterly catastrophic no matter what the starting point.

  34. Willard says:

    > The Chinese in no sense “own” America. All of the US debt that is owned by China is denominated in US dollars, which can only be spent in the USA, on American goods.

    I’m not sure I need to defend that the Chinese owns America, Sam. I’d settle for Mao asking Abe “Can I haz that trillion back?” while he tremedously (H/T the Donald) sits on his gigantic lap. Besides, it’s not really one trillion:

    On June 30, 2015, debt held by the public was $13.08 trillion or about 74% of the previous 12 months of GDP. Intragovernmental holdings stood at $5.07 trillion, giving a combined total gross national debt of $18.15 trillion or about 102% of the previous 12 months of GDP. $6.2 trillion or approximately 47% of the debt held by the public was owned by foreign investors, the largest of which were the People’s Republic of China and Japan at about $1.3 trillion and $1.2 trillion respectively for the two countries.

    https://en.wikipedia.org/wiki/National_debt_of_the_United_States

    To give a sense of perspective, here’s the budget for Occupy Irak:

    And here’s the size of the Libor scandal:

    The price of Mr. T’s cost on sensitivity issues has been lukewarmingly estimated at 10 trillions.

  35. Andrew dodds says:

    Willard – its like this:

    Abe prints off a trillion dollar note.

    Mao has no choice but to accept it, because if Abe says it is a trillion dollar note then it is.

    Nb. Only works if your debts are all in your own currency. Excessive use may destroy your economy. Thinking about this too much may end up with you going to live in a shack in the woods.

  36. verytallguy says:

    Any quantitative cost/benefit or risk analysis is doomed.  The uncertainties of not only emissions and climate sensitivity but also impacts, together with the extreme effect of parameters such as discount rate makes these useless.   See for instance

    http://www.nber.org/papers/w19244

    The assumptions in these calculations are  then,  in effect, proxies for values.   Values such as

    – how much is our natural world worth? 

    – what are future generations worth vs our own?

    – how much risk are we comfortable with? 

    – how important is personal freedom vs societal  good? 

    Climate change is a moral and political issue,  not an economic or scientific one. 

  37. Sam taylor says:

    Andrew,

    I largely agree. I read dale Jamieson’s book over the holidays which makes many of the same points and goes to great lengths to illuminate the limits to how useful things likw cost benefit analyses are in the context of climate change, especially analyses like Nordhaus’s, which is basically an exercise in consumption smoothing.

    Furthermore there’s the issue that pricing these things in terms of money is sticking a linear marker onto what is a fundamentally nonlinear system. It’s basically using an extensive value which can be infinitely subdivided without changing its quality (money) to try and describe the state of something which, if you eventually chop it into a small enough piece ( eg a fishery) eventually undergoes a rapid state shift. Trying to measure systems in this way can lead to undesirable results as you’re simply not getting the information out that you need. It’s also why I think the whole ‘natural capital’ thing is hooey, to be honest.

  38. Isn’t a large part of this discussion related to Martin Weitzman’s conclusions (c.f. “Fat-Tailed Uncertainty in the Economics of Catastrophic Climate Change” or search “the dismal theorem”).

    In essence, he argues that at the tail-end of potential damages, the costs get extremely high and rise faster than the probability of such events can fall, so the risk – in WMC’s calculus being cost(i)*probability(i) – becomes unbounded. It is a deeply frustrating (and dismal and disturbing) result, but ultimately the risk proves so large – even at low probabilities that it completely dominates the results, and traditional CBA (cost-benefit assessment) modelling breaks down. According to the “math”, we should be willing to pay *any* amount to avoid that type of risk.

    Which is essentially a useless result because what do you do with it? Nonetheless, the “benefit” side of the equation fades into inconsequence.

    Weitzman himself is pretty conservative and is himself frustrated with this conclusion, as I understand it.

  39. rust,
    It may be similar to that kind of idea, but what I was getting at what was – I think – somewhat simpler than that. I’m still trying to decide if I’ve written quite what I intended (or that what I’ve said has merit), but the general idea is essentially that something unexpected, but good, has less value than something unexpected, but bad – even if their costs were formally calculated to be the same. Weitzman’s idea is probably more formal in that he seems to be doing a cost benefit analysis but is arguing that the fat tail means that – as you say – the risk/cost of the outlier events becomes so high that we should do all we can to avoid them, and the benefits become inconsequential.

  40. verytallguy says:

    ATTP,

    what I was getting at is essentially that something unexpected, but good, has less value than something unexpected, but bad – even if there costs were calculated to be the same

    I think you are describing “loss aversion” ,  wherein:

    Most studies suggest that losses are twice as powerful, psychologically, as gains

    https://en.m.wikipedia.org/wiki/Loss_aversion

    I’m not sure I would agree that formalising this in economic decision making  is a good idea. 

  41. vtg,

    I’m not sure I would agree that formalising this in economic decision making is a good idea.

    I’m not sure I was suggesting that 🙂 . Loss aversion may be one way of describing this, but I think I’m trying to suggest something subtler. We are not intending to change our climate. Hence any possible positive benefits are not things that we’re actively trying to achieve. Hence, if they don’t materialise (because we choose to generate energy in a way that does not change our climate) we haven’t really lost anything.

  42. Mal Adapted says:

    There is also the question of who “we” are. AGW is a result of privatizing the benefits of fossil energy production while socializing much of the costs. We residents of the developed world have enjoyed benefits in proportion to the prices we’ve been paying for energy, but many residents of the developing world, who’ve enjoyed little if any benefit from fossil energy, will pay a large portion of our externalized costs. Think Tacloban.

  43. Mal,
    Indeed, that is another issue. A cost benefit analysis may conclude that we should do little (and some certainly argue that this is the case) but we can’t ignore that those who benefit may not pay the costs. Hence, vtg is probably right that climate change is ultimately a political and moral issue. We can certainly use cost benefit analyses to inform decision making, but the decisions will clearly be based on judgements, not simply on costs and benefits.

  44. JCH says:

    The Chinese debt would not allow them to own America. We could swap it for Hawaii. Maybe even just part of it.

  45. verytallguy says:

    ATTP,

    I think what you are suggesting is that we should only weigh costs, but not benefits, of the unintended consequences of decisions.

    I have no idea why you think this!

    An analogy:

    I decide to consider going to to Rome on holiday in October because I love ancient remains and want to see the Colloseum before I die.

    An unintended benefit is that it’s lovely and warm when it’s cold and orrible in the UK
    An unintended cost is that I get hassled by people selling crappy souvenirs

    Why shouldn’t I weigh in both the weather and the souvenir sellers when making my decision?

    (There are also risks to consider such as the likelihood of suffering death in an Italian taxi driven by a lunatic simultaneously gesticulating out of the window, speaking on his phone and driving at speed)

  46. vtg,

    An unintended benefit is that it’s lovely and warm when it’s cold and orrible in the UK
    An unintended cost is that I get hassled by people selling crappy souvenirs

    But didn’t you go to Rome because it is likely to be lovely and warm? It seems to me that most of what you’ve described are intended benefits and costs, not unintended. I might be wrong, but it does seem to me that unintended benefits have less value than unintended costs. Although, part of why I’m thinking this is because we don’t have perfect knowledge about how our climate will change, so really I’m thinking of possible unintended benefits. If we could determine the costs and benefits of climate change reasonably accurately, then it may well be different since we would essentially know what will happen.

  47. > I have no idea why you think this!

    Me neither. And yet it seems to be so “obvious” to ATTP that he can hardly even understand the question 😦

  48. Sam taylor says:

    Vtg

    Pindyck has a more recent paper on the use and abuse of models, not sure if you’ve seen it:

    http://web.mit.edu/rpindyck/www/Papers/PindyckClimateModels2015.pdf

  49. WMC,

    Me neither. And yet it seems to be so “obvious” to ATTP that he can hardly even understand the question 😦

    Well, no, it doesn’t necessarily seem that obvious to me, but it’s not obvious to me that it’s wrong either. It’s possible that I’m explaining myself poorly, but it’s also possible that I’m wrong. Again, I’m really referring to completely unintended benefits that are only possible (because of our lack of knowledge) rather than unintended benefits that we may be able to definitively determine. If we aren’t specifically aiming for a particular benefit, then if it doesn’t happen we haven’t really lost anything (or, maybe, we haven’t lost anything that we were trying to achieve). On the other hand, an unintended cost that does materialise may be something we really would rather avoid. So, an unintended benefit that doesn’t materialise might (in some sense) have less value than an unintended cost (of the same magnitude) that does.

    Maybe you could also explain your C_n example again, because that may be along the lines of what I was getting at, but I wasn’t quite sure what you meant.

  50. verytallguy says:

    Sam, thanks. I am a blundering amateur in these matters, not even familiar with the basic literature.

    For me, even pindyck in that paper relies on a discount rate. It’s worth noting that a reasonable range of discount values has an effect of orders of magnitude on the outcome of a simple npv calculation over a century. But his suggestion still seems much better than IAMS.

  51. > C_n example again

    All I meant by that was that instead of maximising sum(benefits – costs) you should want, in some sense, not to be doing a difference of large numbers; both because such a sum is usually ill-determined; but also (in this case, because the sum is global) because there are winners and losers in each term. So you might wish to maximise your benefit, subject to some constraint resembling each of the terms being as small as possible; thus minimising in some way sum(abs(terms)).

  52. WMC,
    Okay, thanks. That may be similar to what I’m suggesting, but I’ll have to think about it some more. Again, I do think there is a difference to how we would approach this if we were actively trying to change our climate, to how we would approach this if climate change were inadvertant (as it is). It seems as though you’re suggesting that the approach should be the same in both cases, but maybe I’m misunderstanding what you’re saying.

  53. I do think there is a difference to how we would approach this if we were actively trying to change our climate, to how we would approach this if climate change were inadvertant (as it is)

    There seems to be a difference in practise. I do not think that there should be one, but the way people talk about climate change sure suggests that it makes a difference whether global warming is intended.

    The same mitigation sceptics that cheer at inadvertent geo-engineering by CO2 are terribly afraid of any other form of intentional geo-engineering.

    The private sector making money from inadvertently changing our climate is naturally also a lot better than a government program to intentionally increase the global temperature in the published opinion of these strange times we currently live in.

  54. > Abe prints off a trillion dollar note. Mao has no choice but to accept it, because if Abe says it is a trillion dollar note then it is.

    Indeed:

    In World War II, much as it had during World War I, the US benefitted from its late entry into the fray. It spent much of the early part of the war shipping exports to the Allies, and collecting much of their gold as payment. After the war, it was impossible to reconstitute the gold standard mostly because the US owned almost all the world’s gold.

    So instead of basing the world monetary system on gold, the advanced economies based their currencies on the US dollar. The dollar was still linked to gold, meaning foreign government central banks could, in theory, redeem exchanged dollars for US gold whenever they wanted. Under this system—the post-war arrangement known as Bretton Woods—the exchange rates to the dollar were fixed at a certain price, but there was some leeway for devaluation if necessary.

    There was, however, an inherent tension in this system. Under the Bretton Woods system, most countries—besides the US—operated with trade surpluses: They exported more than they imported. As a result, they ended up with growing piles of dollars. (The Chinese have a similar situation today.) They needed a place to put these dollars. The only financial market big enough to absorb them was the market for US Treasury securities. In other words, these countries were all lending money to the US.

    That encouraged the US to run large deficits and build up its foreign debt. Eventually foreign lenders began to doubt, rightly, whether it could cough up all the gold they were theoretically entitled to. That’s why president Richard Nixon snipped the last strand of a linkage between gold and the US dollar in the early 1970s.

    http://qz.com/294809/why-the-us-has-the-most-powerful-currency-on-the-planet/

    Mao may not be sitting on Abe’s lap for that note alone.

  55. To return to our program, here’s the abstract from a paper that offers a distinction which may solve some of the conceptual problems we are facing:

    Nearly all discussions about the appropriate consumption discount rate for climate-change policy evaluation assume that a single discount rate concept applies. We argue that two distinct concepts and associated rates apply. We distinguish a social-welfare-equivalent discount rate (rSW) appropriate for determining whether a given policy would augment social welfare (according to a postulated social welfare function) and a finance-equivalent discount rate (rF) suitable for determining whether the policy would offer a potential Pareto improvement.

    Distinguishing the two rates helps resolve arguments as to whether the choice of discount rate should be based on ethical considerations or empirical information (such as market interest rates), and about whether the discount rate should serve a prescriptive or descriptive role. Separating out the two rates also helps clarify disputes about the appropriate stringency of climate change policy.

    We find that the structure of leading numerical optimization models used for climate policy analysis may have helped contribute to the blurring of the differences between rSW and rF. In addition, we indicate that uncertainty about underlying ethical parameters or market conditions implies that both rSW and rF should decline as the time-horizon increases.

    I rediscovered that paper because of franktoo’s shopping list at Judy’s:

    What large part of the consensus do you accept? The consensus advocates for an 80% reduction in emission of GHGs by 2050. It doesn’t take much of a bias in the pillars of this policy to change from prudent to foolhardy: 1) Overestimating ECS, particularly on the high end of the pdf. 2) Underestimating the GMST at which warming ceases to be beneficial (today’s temp or +1 degC). 3) Overestimating rate at which damages grow with temperature. 4) Applying an artificially low discount rate for government borrowing, when investment in low-carbon energy is made mostly by businesses that charge their customers a much higher cost of capital. 5) Counting world-wide benefits against national costs of mitigation. 6) Underestimating adaptation. Given the distortions you acknowledge, how can you be sure the social cost of carbon dioxide emission is high enough to justify mitigation?

    http://judithcurry.com/2015/12/20/what-is-there-a-97-consensus-about/#comment-754298

  56. izen says:

    @-“I think you are describing “loss aversion” , wherein:
    Most studies suggest that losses are twice as powerful, psychologically, as gains”

    That is because vicious circles are far more probable than virtuous ones.

    @-“I think what you are suggesting is that we should only weigh costs, but not benefits, of the unintended consequences of decisions.”

    Its a second order effect resulting from inherent instabilities. In climate, ecologies and economies.
    benefits tend to shrink with large or rapid change.
    Risks accumulate.

  57. There’s a problem here.

    Look, I’m all for doing something about AGW. But look, we have CAGW activists in our country who are looking to do tremendous harm. Look at Obama’s tweet. Look at C13. I mean, these people, they’re not lukewarm honest brokers, okay? Look at what happened to Judy. Look at what happened to Junior, with, you know, audits. Other people going to be audited, they’re so badly injured. We have a real problem. There is a tremendous hatred about GRRRROWTH out there. You know, you can’t solve a problem until you find out what’s the root cause. And I wanna find out what is the problem, what’s going on.

    I’ve had so many people and email me and say thank you. Now, if you remember, when I audited CAGW a week ago it was like bedlam. All of a sudden – and you watch at Judy’s, and you see people commenting. They said, “well, Willard has a point. We have to get down to the problem.” The people that are friends of mine called and said, “Willard, you have done us and a tremendous service.” Because we do have a GRRROWTH problem. And we have to find out what is the root cause.

  58. Willard, too many 3-silable words.

  59. Sam – actually the dollar being a reserve currency is beside the point. As long as you borrow in your own currency – reserve currency or not — you run no chance of becoming Greece. The problem comes with borrowing above the means to meet interest payments to such an extent that high inflation becomes endemic because the treasury’s printing presses are never allowed to stop. The USA is not historically close to that point.

    There is an old saying, often attributed to John Paul Getty: If you owe the bank $100 that’s your problem. If you owe the bank $100 million, that’s the bank’s problem.

    As you rightly point out, China can’t really do much with our currency except spend it. And because they hold a large amount of US currency they do *not* want to see it’s value fall significantly – since that would be *their* problem as much as ours.

    The past 8 years have revealed that conservative and libertarian leaning economists simply do not understand the basics of macroeconomics (if you needed further proof). How many choruses did we hear of skyrocketing interests rates, inflation and the debasement of the dollar that was sure to follow the stimulus program that the US embarked on following the economic crisis of 2008? The post-crisis statistics are pretty clear – the larger the stimulus, the quicker the recovery. Supply-side and it’s close relatives simply failed at the basics. Quadrupling the money supply simply did not have the effect those of that ilk expected. Meanwhile, basic New Keynesian, IS-LM models turned out to be spot on the money.

    But much like ClimateBall, EconomicBall means never having to admit you were wrong. Facts and coherent theories be damned

  60. phlambo says:

    The problem is existential. The ultimate cost is destruction which has been witnessed many times and the ultimate benefit is immortality which has never been witnessed. Thus costs are worth avoiding more than benefits are worth achieving.

  61. Joshua says:

    ==> “How many choruses did we hear of skyrocketing interests rates, inflation and the debasement of the dollar that was sure to follow the stimulus program that the US embarked on following the economic crisis of 2008? “

    Thank you very much, Mr. Chairman, for allowing me to appear before your subcommittee this morning to discuss the feasibility of establishing a gold standard.

    As you know, I have introduced, and other members have cosponsored, H.R. 7874, which is a comprehensive bill to place the United States on a full gold coin standard within two years of the date of its passage.

    I believe such a standard to be not only desirable and feasible, but absolutely necessary if we aim to avoid the very real possibility of hyperinflation in the near future, and economic collapse.

    Ron Paul…..February 23….

  62. Joshua says:

    Oh wait….

    I forgot….

    February 23, 1981

  63. Andy Skuce says:

    Part of the problem with cost-benefit analysis is that there’s an assumption that a net global benefit is a good outcome. CBA is all very well performed at the level of an individual, a family, a firm or even, in some cases, a country. But done at a global level, it’s problematic because, in practice, the winners don’t compensate the losers. If you get flooded or your harvests fail year after year, that’s just planetary tough luck. The winners who might benefit, say, from longer growing seasons or CO2 fertilization will attribute their improved fortunes to their own hard work and, figuratively, if not literally, their industry.

    For example, Alberta’s wheat farmers don’t share their bumper harvests with their drought-stricken Syrian counterparts.

    There’s a good moral reason for focussing on the harms rather than the benefits of climate change. It’s a more a matter of distributional equity rather than some net sum of all the goods and bads.

  64. mwgrant says:

    Preceding comments of risk and cost/benefit:

    Hmmm, Willard…a lack of CLARITY™. :O) Happy New Year.

  65. Not sure what you mean, mwg.

    Wait. I see what you did there.

    Happy New Year!

  66. rustneversleeps says:

    The assymetry of costs and benefits, or loss aversion as another poster alluded to, is pretty straightforward I would think.

    if the global GDP is say $100 trillion, and there is a 5% probability of climate benefit somehow doubling it to $200 trillion, and there is also a separate 5% probability of climate cost sending it to $0 trillion Ior $1 trillion, etc.), we are not going to perceive those outcomes as a wash, a trade-off, a netting out of potential costs and benefits.

    We would almost certainly be williing to pay quite a bit to try to avoid the latter, and certainly much, much less to encourage the former (remember, there would be a 95% chance of the benefit not occurring).

    And there is also the issue of “”what” you would be paying for. In the latter case (minimize system change to avoid costs), you would be opting to keep the climate/atmosphere close to what we already have and are pretty sure how it has behaved under those conditions. In the former case (accept/encourage change and weigh benefits against costs). presumably you would be deliberately* forcing the system to a state where we are far more uncertain how it will respond.

    * of course, globally, we have been/are not-deliberately doing just this, and seemingly relatively unconcerned about the outcome to-date…

  67. RNS: that’s a decent point, but not I think perfectly expressed. Firstly, I’d say its fairer to think of multiples, so a better comparison would be between GDP doubling, and halving. That’s no longer quite as fatal, but it does show the problem. Which is related to what I think is called marginal utility or somesuch by economists: as you get more $, they become less useful to you (oh yes: https://en.wikipedia.org/wiki/Marginal_utility). So, indeed, given a 50-50 change of doubling my income or halving it, I would see that as a bad bet[*].

    However, that’s not quite right, as its not really the “bargain” being offered. The “standard” offer is something like a 5x increase in GDP over a century (from BAU) and a pick-your-number-say-10% drop in GDP from climate damage. Versus, say, a 4x increase in GDP and no damage from a use-less-fossil-fuels-scenario. So either way round in those std scenarios you’re looking at increases in GDP (/wealth, /whatever you’re choosing to call it).

    No scenarios offer a 2x improvement from “climate benefit” and none end up with 0 GDP, or GDP/2.

    [*] Or rather, I think most people would. As it happens, due to my current circumstances, I’d see it as a *good* bet; but that I think is unusual.

  68. rustneversleeps says:

    “The “standard” offer is something like a 5x increase in GDP over a century (from BAU) and a pick-your-number-say-10% drop in GDP from climate damage. Versus, say, a 4x increase in GDP and no damage from a use-less-fossil-fuels-scenario. So either way round in those std scenarios you’re looking at increases in GDP (/wealth, /whatever you’re choosing to call it).”

    This is a separate point, but that “standard” outcome is basically due to an assumed growth rate that is literally chosen by the modeller and input into the standard IAMs.

    More recent work is relaxing that “baked-in-the-cake” growth constraint, and there the numbers are more sobering. Especially when you consider what Andy S mentioned above about how differently winners and losers will perceive things. For instance, as opposed to global GDP being off by say 10% from BAU wealth at the turn of the century, they foresee situations where say Russia is 400+% ahead of where they otherwise would have been, while India is 92% below (i.e. Russia’s real growth goes from a business-as-usual, say, 2% to 3.9%, while India’s would go from 2% to 0.35%.) Granted, they both “grew”, but it is hard to see how we weight/compensate the foregone growth for India with the windfall for Russia.

    Just another note on the onward-we-grow inevitability of the standard IAMs… Do you know what temperature increase you need* to input into Nordhaus’ DICE model to get a 50% reduction in BAU wealth over the century (i.e. still a major increase in wealth, just 50% less growth than otherwise)?

    17 degrees C. That’s right, seventeen, not one-point-seven. If the planet were to warm 17C this century, the damage function in DICE would have us just a little less better off than we otherwise might have been.

    So take these things with a big grain of salt.

    * at least that was the case for DICE in c.2009… although I think Robert Pindyck has shown that little has changed in the IAMs since….

  69. rustneversleeps says:

    oops, here is a link to the Stanford/Berkeley paper last October… or at least the Bloomberg write-up that I have bookmarked… http://www.bloomberg.com/news/articles/2015-10-21/climate-change-slams-global-economy-in-new-study-from-stanford-and-berkeley

  70. rust,
    I think related to what you’re saying is that most IAMs are simply linear perturbation models. Hence, by definition, they can’t consider large, non-linear perturbations. In a sense, they’re assuming up front that the damages will be small (at least, that’s my understanding). So, as I understand it, if you want to know what carbon tax to set today, then using IAMs is probably fine – maybe it’s all we can do. However, if you’re wanting to know what will happen in 2100 if we choose to follow a high-emission pathway, than IAMs probably can tell you little, partly because their underlying assumption is that the perturbations will be small.

  71. Andrew Dodds says:

    I was thinking of Syria as an example, random googling brought up this (compiled just before TSHTF, to use the diplomatic term)

    http://ageconsearch.umn.edu/bitstream/155471/2/2_Al-Riffai.pdf

    Figure 4 on page 38 is interesting. It basically says that GDP loss due to drought is small – perhaps 1% max. Which is reasonable, if agriculture is c. 19% of GDP and only a portion is affected.

    This is all nice, reasonable stuff, with suggestions about how increasing drought could be mitigated. It also completely missed the bit where increasing drought lead to a chain of events which ended up practically destroying the economy and country – hard to get numbers, but perhaps -50% of GDP. Once your GDP change starts causing political change.. modelling becomes useless.

  72. -1=e^iπ says:

    Or expected social welfare maximization http://judithcurry.com/2015/11/29/decision-making-under-uncertainty-maximize-expected-social-welfare/

    I think it’s amazing how you just try to dismiss all benefits. Perhaps you simply want to avoid the complexity of the issue of climate change and the complexity of decision making.

  73. -1=e^iπ says:

    @ verytallguy

    “I’m not sure I would agree that formalising this in economic decision making is a good idea.”

    I agree. Without expected utility theory, it is basically impossible to use expected social welfare maximization to rank policy options.

    “It’s worth noting that a reasonable range of discount values has an effect of orders of magnitude on the outcome of a simple npv calculation over a century.”

    I think the reasonable range of discount values is more constrained than most people think, especially if you take the position of Nordhaus, Arrow, Weitzman, Tol and others that the discount rate should satisfy the Ramsey equation. Since real riskless interest rates and real GDP per capita growth are easily observable, ultimately the range of reasonable discount rates depends on the coefficient of relative risk aversion of humans. The coefficient of relative risk aversion is observable and some of the better estimates, particularly Layard et al. (see above link to expected social welfare maximization), are quite well constrained at slightly above 1.

    @Willard –
    “We argue that two distinct concepts and associated rates apply. We distinguish a social-welfare-equivalent discount rate (rSW) appropriate for determining whether a given policy would augment social welfare (according to a postulated social welfare function) and a finance-equivalent discount rate (rF) suitable for determining whether the policy would offer a potential Pareto improvement.”

    Yeah, that’s just the Ramsey equation.

    “In addition, we indicate that uncertainty about underlying ethical parameters or market conditions implies that both rSW and rF should decline as the time-horizon increases.”

    That’s for certainty equivalent discount rates. Alternatively, you can just do expected social welfare maximization with a constant rate to yield the same effect.

    @ William Conolley

    “Firstly, I’d say its fairer to think of multiples, so a better comparison would be between GDP doubling, and halving.”

    This corresponds to a coefficient of relative risk aversion of 1, which is arguably favoured by natural selection so is expected for humans and similar animals under evolutionary theory.

    @ Rustneversleeps

    “17 degrees C. That’s right, seventeen, not one-point-seven. If the planet were to warm 17C this century, the damage function in DICE would have us just a little less better off than we otherwise might have been.”

    Nordhaus does a quadratic approximation to damages, that’s basically a taylor approximation assuming that we are somewhat close to the optimal temperature. As long as temperature changes are moderate (say just a few degrees) it’s probably good enough. Showing that a taylor approximation doesn’t apply outside of its useful rate does not mean that taylor approximations are useless. That’s like saying we should never use Newtonian physics because it fails when things get close to the speed of light.

    @ATTP

    “IAMs probably can tell you little, partly because their underlying assumption is that the perturbations will be small.”

    Or… simply get better estimates of the damage function.

  74. Brandon Gates says:

    Joshua,

    February 23, 1981

    Didn’t some guy named Hansen tell the Congriscritters back in 1988 that the oceans were about to boil?

    [ducks]

  75. -1,

    I think it’s amazing how you just try to dismiss all benefits. Perhaps you simply want to avoid the complexity of the issue of climate change and the complexity of decision making.

    No, that isn’t really what I’m trying to suggest, but I’m either explaining it badly or it’s a bad idea. However, given our last interaction something I certainly want to avoid is wasting my time discussing this with you.

  76. -1=e^iπ says:

    “I certainly want to avoid is wasting my time”

    Isn’t that sort of my point? You want a simple answer even if no such answer exists. Because it’s easy and you don’t have to spend time addressing the complexity of the issue? Knowledge begins only after people admit what they do not know.

    “True wisdom comes to each of us when we realize how little we understand about life, ourselves, and the world around us.” -Socrates

  77. -1,

    Isn’t that sort of my point?

    No, you left out the bit where I said “discussing this with you”. Seriously, I really do have better things to do today than repeat our last interactive which I regarded as an utter waste of time. So, if your comments don’t appear, don’t hold your breath.

    “True wisdom comes to each of us when we realize how little we understand about life, ourselves, and the world around us.” -Socrates

    I always find it odd that those who quote these thoughtful bits of advice rarely seem to think that it applies to themselves.

  78. Andrew dodds says:

    Eyepie

    There is little reason to assume that evolution favours a relative risk aversion of 1. Even within a species, the sexes will typically use different values.

  79. Willard says:

    > Or… simply get better estimates of the damage function.

    That’s an interesting usage of “simply.”

  80. Willard says:

    > “True wisdom comes to each of us when we realize how little we understand about life, ourselves, and the world around us.” -Socrates

    True quotes come to otters when specific sources are being cited.

    My favorite one is:

    Source: http://quotationsbook.com/quote/44998/

    If you’re to ignore AT’s request, Minus One, please stick to formal armwaving.

  81. verytallguy says:

    -1,

    I think the reasonable range of discount values is more constrained than most people think,

    Perhaps you could put figures on your view of this constraint?

  82. -1=e^iπ says:

    @ Andrew Dodds –
    Sinn, Hans-Werner (2003), “Weber’s Law and the Biological Evolution of Risk Preferences: The Selective Dominance of the Logarithmic Utility Function, 2002 Geneva Risk Lecture.” The GENEVA Papers on Risk and Insurance Theory 28 (2), 87-100

    Zhang, Ruixun; Brenna, Thomas J. and Lo, Andrew W. (2014), “The origin of risk aversion.” Proceedings of the National Academy of Sciences of the United States of America 111 (50), 17777-17782

    @verytallguy –
    Well in https://curryja.files.wordpress.com/2015/11/expected-social-welfare-maximization-2.pdf I do give it a confidence interval, but that is based on an unnecessary generous confidence interval for the coefficient of relative risk aversion.

    Layard, R.; Nickell, S. And Mayraz, G. (2008), “The marginal Utility of Income.” Journal of Public Economics 92, 1846-1857

    Layard (2008) alone constrains the coefficient of relative risk aversion to be somewhere between 1.14 and 1.35 (95% confidence interval). Given that the average real riskless after-tax interest rate over the past 20 years for the US was 3.14% and that the average rate of real GDP per capita growth over this period was 1.475%, this suggests by the Ramsey equation that the social rate of time preference is somewhere between 1.15% and 1.46%. Thus it would make sense to use a discount rate between 1.15% and 1.46% when doing expected social welfare maximization (provided you choose the appropriate coefficient of relative risk aversion). If one is doing traditional cost benefit analysis, then one should just use the real riskless after-tax interest rate (~3.14%), though traditional cost benefit analysis would be inappropriate for the issue of climate change.

  83. verytallguy says:

    -1,

    you are suggesting that discount rates should be constrained between 1.15 to 1.46%?

    Interesting as economists have actually used rates above 4% or even higher.

    A 4% discount rate values economic activity 100 years hence 20 times lower than a 1% rate.

    http://www.rff.org/files/sharepoint/WorkImages/Download/RFF-DP-12-43.pdf

    Why are your constraints not widely accepted?

  84. anoilman says:

    I’d be careful about referring to past rates as an example of the future.

    We had the baby boomers wrecking interest rates for decades. We had cheap and plentiful fossil fuels powering our economies. We outsourced jobs to poorer nations like China. None of this will hold for the future.

    Also, all projections for future growth seem to avoid the delicate issue of economic adjustments, and failures. The longer the time line, the riskier the chances of smooth sailing appear.

    I think rates will be low, but the risks against stability are high. But that’s just my opinion.

    http://grist.org/article/discount-rates-a-boring-thing-you-should-know-about-with-otters/

    In any case, a low interest rate is in line with applying a monster carbon tax right now. I’m good with that.

  85. -1=e^iπ says:

    @ verytallguy

    “Interesting as economists have actually used rates above 4% or even higher. ”

    It depends on application. In a traditional cost benefit analysis, you aren’t taking into account intergenerational inequality so you need to discount the future more (so 3-4% might be more appropriate). And if the application involves deciding whether or not to invest in a risky projection, you may want to add a risk premium to the discount rate.

    “http://www.rff.org/files/sharepoint/WorkImages/Download/RFF-DP-12-43.pdf”

    Your link is in complete agreement with my position.
    My position is also in agreement with Arrow, Kenneth et al. (2012), “How Should Benefits and Costs Be Discounted in an Intergerational Context?” Resources for the Future Discussion Paper No. 12-53.

    “Why are your constraints not widely accepted?”

    It is somewhat accepted. Nordhaus uses a coefficient of relative risk aversion of 1.45 and a social rate of time preference of 1.5% in DICE and Tol uses a coefficient of relative risk aversion of 1.47 and a social rate of time preference of 1.07% for FUND. My constraints above are based on the results of Layard et al., which is basically one of the most well constrained estimates of the social rate of time preference in the literature. To be fair, there is likely some additional uncertainty in Layard et al. due to assumptions made about how utility and happiness are related. There are of course many other estimates in the literature.

    @ Anoilman

    The Ramsey equation allows interest rates to vary even if the social rate of time preference does not vary because real GDP per capita growth rates vary. Not to mention that in the short run Central Banks can cause divergences from what is expected by the Ramsey equation; but you can avoid that by looking at a long term average. In any case, the social rate of time preference is likely related to the inverse of life expectancy and one gets rough agreement between the two values using the best estimates of the coefficient of relative risk aversion.

  86. nobodysknowledge says:

    Perhaps so far so good. Sahara greening, better harvests etc.

  87. nobody,
    I guess you don’t get the whole idea of risk then?

  88. Willard says:

    > Your link is in complete agreement with my position.

    I’m not sure how a link can agree with anything.

    It might be more useful that you commit to one of the fork (descriptive/prescriptive) the authors offer, and recognize that your position can’t speak for the other, Minus One.

    Furthermore, if you wish to convey with your “better constrained” that your side of the fork is simply better than the other, then I duly submit that your position is not in agreement with the authors.’

  89. verytallguy says:

    -1,

    the link sets out that radically different discount rates can be used, and advocates not for one or the other, but that any usage of a discount rate should be clearly stated and justified.

    As I said upgrade, that results in differing assumptions leading to orders of magnitude difference in value of current vs future costs and benefits.

  90. -1=e^iπ says:

    @ Willard
    “I’m not sure how a link can agree with anything.”

    Page 6 has the Ramsey equation. The authors primarily discuss that different discount rates make sense for different types of analysis, which I agree with.

    “(descriptive/prescriptive) the authors offer”

    Descriptive since I use empirical evidence to obtain moral parameters like the social rate of time preference and the coefficient of relative risk aversion. Also, I have no idea how you will ever be able to obtain agreement with a prescriptive approach.

    @verytallguy
    “the link sets out that radically different discount rates can be used, and advocates not for one or the other”

    Of course, since different analyses can be used. However, when it comes to expected social welfare maximization, the range of acceptable discount rates is far more limited. And other types of analyses such as traditional cost benefit analysis or social welfare maximization under certainty are merely approximations of expected social welfare maximization (and I would argue inadequate approximations at dealing with an issue as complex as climate change).

    “that results in differing assumptions leading to orders of magnitude difference”

    Of course. That doesn’t mean those different discount rates chosen are justified.

  91. verytallguy says:

    That doesn’t mean those different discount rates chosen are justified.

    So you say. Yet justify economists do.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s