Skip to main content
Loading Explore guides…
Environmental Policy Costs & Benefits · Acedra
Home
Explore
Economics
Environmental Policy Costs & Benefits
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
118 segments
Copy
Hide video
0:05
[MUSIC] >> Dan Phaneuf: Good morning and welcome. My name is Dan Phaneuf. I'm an environmental economist at University of Wisconsin Madison. And you can see the title of our session here, the costs and benefits of environmental policy. Obviously, we could spend about two days on this, but we'll take about 50 minutes. I put the paragraph that you see up there, was taken off of the agenda, that is posted on the web. And I'm gonna use that to organize my introductory remarks here.
0:33
I'll set the stage by making some brief remarks on the use, the nonuse, and the misuse of cost benefit analysis. And then Bjorn Lomborg and Caroline Secote will have a little more time to make some detailed comments. Okay, so my background is in estimating the benefits and costs primarily in the non-market space of conventional air and water pollution. And so I'm gonna focus a little bit on examples of benefit cost analyses through the lens of conventional air pollution, conventional water
1:09
pollution largely motivated by the Clean Air Act and the Clean Water Act. And the starting point for any discussion of this type is going to be executive order 12291. This was issued back in 1981 by the Reagan administration, and it was maintained and updated by all subsequent administrations since that point. And basically, what it does is it requires all rulemaking by federal agencies to conduct regulatory impact analyses, including benefit cost analyses,
1:43
to understand the economic impacts of substantial policies. So substantial is probably not substantial in today's dollars. So a lot of regulatory analyses are done under this regulation.
1:58
And I'm gonna point to some of the regulatory analyses that have been done in EPA's administration of the Clean Air Act and the Clean Water Act. Yeah, so this is one of my favorite tables. So what it does is it provides an overview and comparison of cost benefit analyses that have been done for many clean air and clean water programs. And I'd like you to focus perhaps on the air pollution column and the surface water pollution column and look first at the air pollution.
2:32
So you can see that the regulations in this space tend to have benefit cost ratios that are substantially greater than one. Nick alluded to this, and there's many reasons for this, but I'd like to highlight potentially two. So, first of all, the benefits from air policies tend to be contemporaneous human health improvements, and our methods suggest that these are highly valued, and so the benefits are potentially large from these programs.
3:00
Second, there's at least some elements in the Clean Air Act Amendment and its amendments that attempt to use cost effective policy vehicles. So cap and trade for the sulfur dioxide program, for example. And at least in some cases, these have controlled costs. And so we have this rather healthy outcome where we have, you know, big benefit cost ratios.
3:22
I personally work more in the water space and you can look and see the contrast with air. So most of the benefit cost analyses show ratios, benefits to cost smaller than one. And I think that the reasons for this contrast nicely with the air case. So one, most of the policies for surface water pollution are very much command and control. They're technology mandates. They're non-tradable effluent permits. So very high cost type policies.
3:52
It's also worth pointing out that the absence of regulation on non point source pollution precludes perhaps the use of low abatement cost sources. And so we have high cost policies. The second thing to point out is that for the most part, surface water quality provides ecosystem services. So recreation and amenity type services, rather than human health. And these probably have smaller values than human health and are also gonna be tougher to value.
4:22
And so I think these are nice contrasts about how we can learn different things about different types of environmental policy through the lens of cost benefit analysis. And this slide just tries to summarize kinda what I think of as the main message. I guess this is more of a personal list from my time working in this space. But I like to think that these requirements for cost benefit analyses encourage transparency.
4:54
They discipline our thinking and force us to think about very specific cost and benefit generating mechanisms.
5:02
They work well as organizational and viewpoint aggregators, organizational vehicles and viewpoint aggregators. And I think importantly, these requirements for cost benefit analyses help us understand the role that policy structure plays in generating costs. So costs can be very different depending on how we structure things. And perhaps a side benefit is that benefit cost analysis requires the development and maintenance of expertise in quantitative policy analysis.
5:31
And perhaps this crowds out less constructive forms of policy advocacy based more on argument and that. So I end by saying, a useful organizational vehicle for supporting decision making. If I had my dispassionate or boring hat here to compliment Terry's, that's what I would put on. But my persona usually just conveys that. >> [LAUGH] >> Dan Phaneuf: So I'll just have to let that do. So just- [INAUDIBLE] >> Dan Phaneuf: Yeah, I think we're just passionate.
6:05
There's enough passion in environment stuff. Let's bore it down a little bit. Or as Nick might say, dial back the charisma. That's my other job.
6:13
Yeah, so I'm just gonna. I've got two more slides so the first is that, commenting on the nonuse of benefit cost analysis. And I try to think of ways to categorize this. And maybe this is partially being cynical, but I think there's the philosophical nonuse, where we suggest that an environmental issue is just simply too important to be put to sort of benefit dollars and cents. Or we face an existential threat that goes beyond these types of calculations.
6:41
And I think we see a lot of that rhetoric in climate change, and I suspect that Bjorn will be making reference to that.
6:49
A second one I might say is more cynical. It's more political non use. So if a policy is not really about a sober assessment of trade offs for society's wellbeing, but rather about serving constituents or achieving political power, then we might wanna strategically avoid doing this kind of thing. I think the consequences in either case can be large. So climate policy is sort of a favorite of mine to think in these terms. So think about the Inflation Reduction act.
7:16
We don't have a good sense of the relative costs and benefits of a lot of these programs. They sound good on paper, but we don't really know what they pencil out to be. I for one, would be really interested in knowing how much more expensive our policies under the Inflation Reduction act than, say, a standard carbon tax or a cap and trade program for the same carbon reductions. Or my other favorite point, what if we emphasize conventional air pollution, which we know has immediate and
7:46
large human health benefits rather than future climate change? Then we may be able to think about policies that are efficient in the current term and have long term co benefits in the climate space. So a couple of things to think about there that evolve from the nonuse. Okay, promise, last slide. So I'll just close by giving three examples of what I think of as non-use. So three categories I've given here, so execution, limitations, and undue influence.
8:15
Execution of this can be tough, so mistakes can get made. And I think one thing that is going to come up in the next session is the importance, but the difficulty of thinking about adaptation. So adaptation can reduce the benefits of environmental improvements, but adaptation can be costly, so it should enter into both sides of the ledger.
8:35
The second thing that I think is worth emphasizing is recognizing the limitations of cost benefit analysis. So this is very much about efficiency, right. There may be other criterion that society cares about, distribution, justice, and so on. And I, for one, am not a fan of trying to cram these other types of criterion into cost benefit analysis. And that relates to my final point, which is failures of inference. I try to teach my students that we should think of cost benefit analysis as not
9:07
giving the answer in quotes, but rather informing us about an important dimension, efficiency. We should consider the other dimensions, equity, justice, and so on, on their own merits. And I think we do damage to both the efficiency understanding and the distributional understanding if we try to cram them into the same tool. So hopefully, some of these ideas will come up again and be complemented by Bjorn and Caroline's talks. With that, I'll stop, and I'll invite Bjorn to come to the podium.
9:35
>> [APPLAUSE] >> Bjorn Lomborg: Thank you, and so I'm gonna really take some of the points that Dan just mentioned, especially if I have my slides. Yes great. So it's very clear that cost and benefit analysis can't just be the answer. So I work in an organization, the Copenhagen Consensus, where we go out and try to get all kinds of governments around the world. We just work with Tonga, the last thing, the Pacific Island nation of Tonga.
10:03
And obviously, governments have a lot of other interests apart from costs and benefits. And so you're right, this is in some way just a way to convince people, this is also an important thing. So I think of it. I often use the metaphor to say, imagine if you went into a restaurant and you got this wonderful menu with all these great things. Sounds really great. But there were no prices and no sizes on the menu. Unless you're on a really good expense account, you'd worry about ordering
10:35
from that menu because you have no idea what you're going to get. Are you gonna get a tiny pizza or a huge pizza? Is it gonna cost a dollar or $1,000? You'd like to know. And so what cost benefit really does is it puts prices and sizes on the social menu. But we are also the kind of guys who would end up saying, you know what, we've looked at all of this and we found out that spinach is really cheap and it's really good for you. But if you don't like spinach, [LAUGH] you're not gonna buy that.
11:05
We are just putting prices and sizes on the social menu and then it's up to society to make those decisions. So I think it's incredibly important to say prices and sizes on a menu matter. It's certainly something we should know, but it's not the only thing we should know. So I took this because I spend most of my time when I'm not talking about things that have nothing to do with what we're talking about here. I spend a lot of my time on climate, so I was gonna do the climate bit.
11:34
And I think what I'd like to just spend my twelve minutes talking about is really talk about how I try to get people to realize that you need to think about cost benefit in a climate perspective. I think the fundamental point is to get people to understand, sure, climate means costly damages, but climate policy also means costly damages. Both of these have costs, and since we have to pay both of them, we should actually try to find the place that minimize the sum of those two costs.
12:08
It's not quite what you can reformulate it into what Terry just told us about, but you shouldn't equate the two. You should just simply find where it is the cheapest of these outcomes. And so I'd like to share with you, so you may know that Nordhaus, the guy who won the Nobel Prize in Climate Economics, has a new paper out where he's actually redone his cost benefit. It just came out this month and I just could not make the numbers work. So I'm going to show you the old one.
12:36
He does have a new paper out where he increases the cost and he just basically decides, I'm not going to look at any of the likely increase in cost of policy. I find it somewhat unconvincing, but there is another paper out here. But I'm gonna show you the work that he won the Nobel Prize for. And this is also what Richard Tull and many others have done. If you look out here in the temperature increase and then see what is the impact on global GDP, now,
13:07
not all of them do the same thing, and there's a lot of uncertainty in this. These are all the measurements that Nordhaus used in 2017. Tull has updated this to 2024, and he gets very much the same outcome. Nordhaus has actually included this, he now, as I'll use, the Burke estimate, which you may know, it is the one that gives you a 23% cost, which is very far outside, I'm sure we could talk about that if you want to. That's one of the reasons why he find a higher cost.
13:39
But this is the estimate that he makes of what is likely the cost of different temperature increases. Now, the point here is not to get ourselves to think about, this is the absolute truth, but it's simply the starting point to recognizing that climate change, in any realistic setup, is not the end of the world. It's a problem, it is something that has definite costs, but not the end of the world. If we look at the maybe 7 degree Fahrenheit, if we do nothing,
14:10
we're talking about 2 to 3, maybe 4% of global GDP by the end of the century. One way that I like to try to emphasize this is if you look at the sort of standard estimate, both from, actually from Nordhaus and from OECD, but also from the IPCC, the SSP2, the middle of the road scenario. We estimate the average person, by the end of the century, on the planet will be about 450% as rich as he or she is today. That's a tremendous achievement.
14:44
Of course, we should recognize this is just a much, much better world. But because of global warming, this is going to feel like not 450%, but 434% as rich. Now, the important point here is to recognize this is a real loss in income. But it's also important to recognize this is not the end of the world. It is a world that's much better, but slightly less much better than it would have been without climate change. This puts the conversation and climate change in a very different term,
15:15
and I'd love to go into this more. But in my understanding, this is probably the most important issue, to get people to recognize that global warming is not something that leads us to the end of the world, but it's an impeditor to the world moving into a much better face. So instead of things getting much better, they get much better slightly slower. That's a problem, but again, not the end of the world. So what Nordhaus did, and this is famously his estimate, he looks at what's
15:43
the cost if we do nothing, it's $140 trillion in his estimates in 2016 version. And then if we reduce temperatures and again, reduce temperatures smartly, the cost is lower and lower. This, I would argue, is what everyone has understood from all of the climate. Conversations that we've had over the last 20 years. The more you do to regulate temperatures, the more you do to cut carbon emissions, the lower the damages. This is absolutely true.
16:11
And this, of course, if that was the only thing you had to think about, that obviously means we should go for the 3.9, which was back then, the only thing he could make his model get to. Remember, we've actually promised, in some ways, maybe 2.7%, sorry, 2.7 degrees Fahrenheit, temperature increase, 1.5 degrees, which in his model is almost impossible. But of course, we need to recognize that there's also a cost to the climate policy. And so, just looking at one side, but not the other one,
16:44
you get a wrong understanding. So, what's the cost of no policy? Not surprisingly, the cost of no policy. The policy cost is nothing. Then the cost increases as we go up here, and it becomes much larger down towards the end. So, when we reduce temperatures a lot, there's a huge climate policy cost. And the point, and this is really just Terry's main point, is, since we have to pay both, we really should put both of these together.
17:12
So, these are the climate cost, and then we add the climate policy cost, which is nil up there in top, and then a little more, and then a lot more here. This is just basic economic understanding. You need to add both of these. And that was why Nordhaus found that this was the optimal outcome in his particular model. He got a lot of flack for that. But the fundamental point is to recognize if you do too little to climate,
17:36
you end up with too many climate damages that don't outweigh the fact that you have fewer climate policy damages. But if you do too much, you end up with much lower climate damages, but much, much higher climate policy damages. Since you have to pay both, you should be cognizant of both. And this is really just the basic point. It's not rocket science. And I think in some ways, we need to make sure that everyone gets the general point.
18:03
I love Julian Simon was the guy who got me started off on this. And his metaphor, which I also quoted in one of my books, was, if you think about doing the dishes, everyone knows that. It's not about cleaning the dishes down to the last molecule. If you've lived in a dorm, you certainly know. But the main point is, it's about diluting the dirt to an acceptable level. That's what we do with air pollution, that's what we do with the water pollution.
18:33
That's also what we need to do with climate policy, we find the place where it's acceptable to live with. We never solve problems, we just improve them. And so really what we need to recognize, and this is, I think, one of the underappreciated graphs of how you could really visualize Nordhaus. And again, it's changed a little bit. So now he finds the 2.6 degrees Centigrade is actually the optimal outcome. I'm sorry, I haven't translated it into Fahrenheit, but it's much lower.
19:08
It's like 4.1 or thereabouts. But the main point here is that if you look at the wealth that we lose to climate and climate policy, it is very small, unless we do a lot. If we do a lot of climate policy, we actually end up losing. More. Interestingly, in his new paper, and again, I also wanna show that, but I didn't want to just show a little part of it. He also looks at what's the cost and benefit of 1.5 degrees, so the thing that all politicians are now talking about. And interestingly,
19:40
he finds that that has a cost of beyond $1,000 trillion in discounted benefits. So that's about a fifth of the global welfare for the rest of humanity. So this could actually cost us 20%. So no matter what we do with climate, it's not really going to matter all that much. But if we do the 1.5 degree policy, that will actually make us substantially poor. I think that's an important point, and we don't hear that enough. And I just wanted to show you, this is the outcomes from the climate
20:12
change economics that Robert Mendelsohn edited last year. And I'm a little surprised by the fact that we've now, at least for what, since 2015, had a global policy of basically going net zero around 2050. Yet we've had no academic estimates of what that cost. As McKinsey has done a pretty good study, I think, of what would be the cost. But we have had no estimates of this, and we should, because this is by far the biggest and most costly policy that mankind has ever committed to.
20:47
Just to give you a sense of proportion, this is probably in the order of hundreds, possibly even thousands of trillions of dollars. The next costliest policy that mankind committed to was the Versailles treaty. Had we actually forced Germany to live up to it, it would have cost about half a trillion dollars. So we're talking about something that is hundreds or possibly thousands of times more expensive than what we've ever promised before. And yet, we've done no cost benefit analysis on that.
21:15
That's probably not a good idea. Now, I want to show you why we haven't done that, because Mendelssohn actually asked them, all of the academics, to do cost and benefit. Very few of them did. So there's only one cost, sorry, one benefit estimate and three cost estimates. But I'm gonna show you. So this is Richard Tol's benefit estimate. So as you can see, by 2100, if we do nothing, this is in a world where we basically just let go, the cost is going to be a little more than 3% of global GDP.
21:47
So this clearly indicates there is a good argument to do something about climate change. I'm going to show you now the three cost estimates. So this is Richard Tol's own cost estimate. As you can see, it vastly outweighed, there's a lot of conversation about what is net zero by 20, 51.5 degrees. Roughly the same. That's why they are using it here in the same way. As you can see, the cost outweigh the benefit all the way through the 21st century. Now, eventually, of course,
22:15
this seems to indicate that it will cross in the 22nd century. If you look at the other two from MIT, this is the realistic 1.5 degree cost. It is much higher. It's above 10% most of the century. And if you look at the net zero cost, because we just don't have good technology to achieve net zero around 2050. So it's a huge spike, up to about 18% of global GDP by 2050. And then it cuts down. What this basically shows you if you just do an average across the century.
22:45
And again, I'm not saying this is a good way to do it, but it gives you a good sort of rough sense of this, the benefit of climate policy in the 21st century. And, of course, that leaves out what happens in the coming centuries, is about 1% of global GDP. The average of these three is about 7% of GDP. Or to put it differently, if you just do it in terms of actual dollars, the total benefit per year undiscounted across the 21st century is about $4.5 trillion, which, of course, is a lot of money.
23:15
And that means we should definitely sit up and pay attention. But the cost of doing this is almost $27 trillion. So about $0.17 back on the dollar. Again, I've done no discounting. I find that most people, when you talk about discounting, they think that you're just cheating, which I understand. And there's a lot of ways that you could do this. But I think this sort of gets people's attention that currently, we're making arguments that are fundamentally incredibly expensive and
23:45
not actually delivering the benefits that we would hope they do. So what we should do, and this is what we've been spending a lot of our time on thinking about, just to give you a sense of what works and what doesn't. So the EU 2020 climate policy we know doesn't work. Possibly gives you about $0.03 back in the dollar, which is a bad idea. The net zero, 1.5 degrees, as I just showed you, about $0.17 back on the dollar. Again, a bad spending. If we do a smart global carbon tax,
24:14
we could actually do $2 of social good for every dollar spent. That's a good idea. It's not a fantastic idea. Remember, I have a new book out on how we can do about $52 per dollar spent for especially the world's poorest. So there are much, much better things to be done in malaria, and tuberculosis, and immunization, and education, and land tenure security, and all these things, but $2 definitely worth taking. And what we found was by far the best investment is to invest in green energy.
24:44
R and D fundamentally make future green energy so cheap that everyone will want to buy, that'll deliver $11 back in the dollar. Again, there's no way these numbers are true, maybe possibly even the first digit. They're roughly true, but they give us an understanding of what it is that we should be focused on and where we can spend our money best first. Thank you. >> [APPLAUSE] >> Caroline Cecot: Hi, everyone. Thank you, Dan and Bjorn, that was terrific.
25:11
So you've heard some great things about CBA's potential to improve government interventions, whether generally or in particular in climate policy. At the very least, it provides a transparent list of all the expected effects of an intervention. And at its best, it can help government agencies better balance trade-offs, anticipate unintended consequences, follow the scientific evidence, or even consider approaches that might be more effective, whether it includes market based approaches, etc.
25:47
But what if highlighting important trade-offs, proposing a creative or innovative approach, or even acknowledging the full range of costs that need to be justified in order to adopt this policy, put the agency's action at risk of being invalidated by a court. So I'm a law professor, so that's why I worry about these kinds of things when it comes to the future of CBA. And I love that this panel, we're all kind of picking up on different things that keep us up at night.
26:22
So to back up a little bit, Nick and Terry introduce the theme of this conference, markets versus mandates, and our focus in this panel and others, how best to ensure that the health and environmental impacts of air and water pollution, often a result of productive economic activity, are valued. So the major federal statutes governing the space, in particular the Clean Air Act and the Clean Water Act, arose at a time of discontent about how the free and
26:55
imperfect market, transaction costs, asymmetric information, discontent about how it was striking the balance. These statutes then were written very heavy-handed with mandates, often silent about cost-benefit analysis or even cost consideration in general. In this context, CBA still emerged, albeit slowly, as a way for mandates to not stray so far from socially beneficial outcomes. Dan mentioned already the Reagan administration's executive order first
27:33
requiring the practice for major regulations. But in the early days, it seemed much easier to proxy for social costs, would expect a compliance cost, and much harder to value social benefits, and still, in many areas, still hard. And obviously a one-sided cost benefit analysis is not very informative or useful for sensible government policy. But over time, we've had great advances in valuing benefits, especially in the airspace, and
28:05
CBA started to provide some real information driving the stringency of standards as much as relaxing the stringency of standards. Agencies were also more willing to rely on cost benefit analysis when their statutory authorizations gave them a lot of discretion about how to think about these problems. And courts appeared to respect that reliance when this cost-benefit analysis at least appeared relatively complete. But an agency is only authorized to do what Congress allows it to do.
28:39
So the executive orders, while useful for promoting cost benefit analysis in the background, those can't be used as an agency's justification for actually relying on a cost benefit analysis. That has to come from Congress. And this constraint has sometimes worked against CBA. So, notably, when interpreting the determination of the National Ambient Air Quality Standards under the Clean Air Act, the Supreme Court held that EPA is prohibited from considering costs.
29:13
That it must set the max at a level requisite to protect the public health with an adequate margin of safety, no matter the cost, because Congress didn't explicitly allow cost consideration. Now, there are many, many, many standards in the Clean Air act and the Clean Water act. Some mention costs, some do not. The Supreme Court has since pulled back on a default assumption that silence on cost consideration is a no on cost consideration. And even as recent as 2015 in Michigan v EPA,
29:48
the Supreme Court has recognized the important role that cost consideration and benefit consideration play in encouraging rational, appropriate government action. Okay, so why my worries? In 2022, the Supreme Court decided West Virginia v EPA, holding that with respect to questions of vast economic and political significance, the agency can't act absent clear congressional authorization. So this essentially changes the default inference from silence, from an inference
30:25
that there is authorization to an inference that there is not authorization. Is an agency finding an innovative way of deploying, say, a market based solution or a cost effective solution to tackle a new problem, a major question that requires a new statute? And if so, does it mean the agency doesn't do anything under its current older statutes and authorities? Under many statutes, doing nothing might not even be an option. Instead, this might mean a return to more traditional command and
31:01
control regulation that is not influenced by cost benefit analysis. We instead might get less bang for our buck. And CBA, in the way that it identifies, say, more efficient ways of approaching problems, becomes less relevant and useful. Second, and relatedly, the Supreme Court this term is considering getting rid of a doctrine called the Chevron doctrine that encouraged courts to defer to agencies. The Chevron doctrine emerged in a case allowing the Reagan EPA to
31:38
allow a more flexible approach to defining when a source triggers new source review in non-attainment areas under the Clean Air Act. If Chevron is overruled, this too could have ramifications for agencies to consider innovative and flexible approaches under discretion in their mandates. Finally, even in the 2015 case that I mentioned, Michigan v EPA, which sanctioned cost consideration, which is great because that's what makes cost benefit analysis possible.
32:16
The Supreme Court seems skeptical about allowing an agency to consider the full costs and the full benefits. Why I say this, the court seemed, appeared to be, and again, this was dicta, which means this isn't part of the holding. This wasn't on, this wasn't up for the court's review. But the court seems sympathetic to an approach where the consideration of benefits might be limited by Congress, at least when they're silent about it,
32:45
to whatever was the explicit goal that Congress had in enacting its statute. Whether it's some specific solution or pollutant, etc. Well, if the court goes that route, then that could be just as bad as prohibiting cost consideration, because a cost benefit analysis is only useful and informative if it presents as close a picture of costs and benefits as possible. And if costs or benefits are arbitrarily truncated, then the CBA ceases to provide meaningful information.
33:23
It honestly would be irrational to rely on something where one side is not the full picture. So I'm hopeful that CBA survives these potential threats, especially now when its promise was really seeming to get off the ground as data and scientific gaps were more likely to be closed. As we were getting more scientific information incorporated, as we're filling in these costs and benefits. But whether it survives, whether it's because we'll have less judicial
33:59
intervention, or maybe it's because we'll have more congressional explicit authorization for these things, that remains to be seen. So I was gonna stop here, but after Dan and Bjorn's remarks, I wanted to make two additional comments, and if there's any interest, we can talk more about these in the Q&A. But one comment, so on climate change now, and this is a continuation for start in the Obama administration and continuing now to the Biden administration.
34:31
Federal agencies use the social cost of carbon in their cost benefit analysis to sort of track some of the benefits of carbon in their analyses. This hasn't been not controversial, but it's been defensible, it's been transparent. And so at the end of the day, I think this offers a transparent way of doing something while we build consensus for bigger moves and the forms that they could take. And then the second comment I wanted to make, you know,
35:08
the difference between, and Terry and Nick alluded to this, the difference between a mandate and a market, to me, comes down to some sense of coercion, though for some, with perhaps lower means, the market might feel close to the kind of coercion that a mandate might for others. When we regulate, there's always winners and losers. I am in support of government analyses, doing a distributional analysis, and being transparent about who benefits and who is burdened by regulation.
35:49
And that I see as a separate distributional analysis. And I happen to think that adhering to cost benefit analysis will often produce results that promote equity as well. And I think this because a lot of the problems and the folks that it affects happen to be folks that have lower means, can't escape from those on their own, are more limited in markets. So I think that a cost benefit analysis would provide benefits to those groups more often than not, than raw politics, which might funnel
36:26
benefits more to those who are more advantaged in the political system. So that's my own point of view about how those would intersect. It's an empirical question that we could see if we had the data, but I also am not in support necessarily of incorporating them into one thing, because that just adds that reduces some of the transparency that I think is valuable in making this kind of move. And from what I've seen so far with the distributional analyses that agencies have
36:55
made, I mean, they're nowhere ready. There's no benefits and costs by subgroup where we're at the point where we can apply even some kind of a multiplier that might make sense. I look forward to questions. Thank you very much. >> [APPLAUSE] >> Dan Phaneuf: Yeah, so we have about ten minutes for questions, and I guess I will try to moderate. The first hand I saw is here. Yeah.
37:30
>> Speaker 4: Thank you. Great discussion of the views of the issues. Now, I'm a numbers guy, and so when.
37:42
Okay, all my degrees are from the University of Wisconsin. So, yes, go Badger. Now, one of the numbers that I've written about is the 2010 IWG social cost of carbon. And in a paper I published, and I think we'll talk 2012, I showed that the Nordhaus code for the social cost of carbon had a mistake in it. That basically, he assumed that CO2 emitted in the future [COUGH] would start warming the atmosphere today. And that kinda bumped up the social cost of carbon. A detail, yes, I know, a detail.
38:32
And as one macroeconomist said, Ken, it's all garbage anyway, all these numbers. I don't like that attitude. And Nordhaus had no objection, he said he's gonna have to change his code. But the EPA guys who used the same code in their IWG, they wrote a comment that just indicated that people at the EPA have no idea how to do these numbers. I mean, they just don't know elementary differential equations, they're just incoherent, their defense.
39:07
Also, the other thing is a lot of the EPA stuff is closed, and numbers come out, but you can't get the documentation for it. I've tried getting it for EPA for a couple different climate models. Absolute refusal to get the code. There was an MIT energy project that had code, it published it in Science magazine. They initially refused.
39:37
However, I looked up the chain of command. And at the top of the chain of command was George Shultz in their external advisory board. He told them to release the code, and they did. But this is the kind of institutional resistance to this. Now, the Society for Benefit Cost Analysis had a two-day workshop in February. And it was abysmal hearing their excuses. Like one thing, they said, well, we're limited in computing power.
40:11
Now, okay, if you insist on doing things on a laptop, yes, that's a limitation.
40:19
However, I reminded them of a paper I wrote that appeared in JPE, where we used several million core hours on a supercomputer. It can be done. It does affect the results.
40:35
And also, I think I reminded them that the DOE spends hundreds of millions of dollars each year on computing resources to make sure that if the president wakes up one day and wants to incinerate hundreds of millions of people, he can push a button, and everything will work. It's called the NNSA part of DoOE. That money is available for that kind of work. However, DoE will not allow their machines to be used for this kind of work or economics in general. I have personal experience with that.
41:10
I've dealt with DOE, and that's their priorities. So when the government starts giving me numbers, I wonder. I mean, I hear the people generate the numbers, and it's closed science. And the other thing that's missing in all these numbers is the uncertainty. The expected wealth in 2100 is four and a half times today.
41:36
But when you look at macroeconomic risk and fluctuations, if I may use the term, there's nearly a unit root. Okay, those of you who understand that know that that's important. We really don't know what 2100 GDP is gonna be. It could be an enormous range, which means enormous difference in temperatures, et cetera. That uncertainty is not represented. But as I argue, we should be thinking in terms of insurance in these policies, not just simple deterministic kind of model. Thank you.
42:10
>> Dan Phaneuf: Just one follow up comment. You refer mostly to climate examples.
42:15
My experience in the conventional water and air pollution examples is somewhat different. I mean, I've reviewed and seen many of the analyses that have come out, and I think it supports one of my points to point out that we've been doing those for decades, and the expertise has been developed, and those are much higher quality. I think the requirement to do these in the climate space is newer. I think that there will be learning and
42:38
hopefully improvements along the dimensions that you've referred to.
42:43
>> Caroline Cecot: If I can add, as the vice president for society for benefit cost analysis, apologize. You didn't enjoy that experience. But I welcome everyone to a lot of workshops on cost benefit analysis that are usually well received. And just one small comment that I had. Well, thank you to your marks. And just to sort of recenter it, because this is really important stuff, bringing up errors that are in the analysis cuz it matters.
43:08
But I thought of the perfect as the enemy of the good in the sense that the actual counterfactual to a lot of this is the agency not doing this analysis. They're not required to in most cases, they're not. So the actual counterfactual is just making a decision and justifying it broadly by pointing to some expected benefits without looking at any of these details. And the courts would uphold that under most of these current statutes.
43:37
What I see as a benefit of cost benefit analysis is that allows for this kind of conversation. It makes the agency reveal that it's relying on some kind of estimate and someone else could come and say, well, you know what? That's not the right way to think about this. And that could emerge. But of course, the agencies are always behind the science in these kinds of areas. So I completely agree with Dan was saying, thank you for that comment.
43:59
>> Bjorn Lomborg: And just very briefly, I think you're absolutely right. I mean, we need to also talk about the distribution. We should also recognize, and I think that's something the adaptation people will talk about in the next panel, that if we're much poorer, were much worse off in so many other ways. But I just struggle if you're going to tell people a story, try to make it so simple that it actually makes sense. And I think a lot of us are very good at making it so
44:26
complicated that it just simply whoosh for everyone and then there is no point. So there really is a trade off between what's sort of presentable in six or 800 words and what is true.
44:41
>> Dan Phaneuf: Why don't we collect a couple of comments and then maybe we can respond to them. And I've totally lost the string, so I'll rely on you guys. >> Bob Benson: Hi, I'm Bob Benson. Question on making cost benefit analysis again. I've done a few, I think I'm a truth seeker. But the levelized cost of energy, I've looked at that a little bit, and I'd like to hear your comments on. So I have solar, I'm a big fan of it, but when I look at the numbers of solar and
45:10
wind, it seems like they're truncated and they don't take into account that you have to have some backup energy to deal with that, yet everybody's bought into it. I'd like to hear comments on that. >> Dan Phaneuf: Thank you.
45:28
>> Roger Pielke: Roger Pielke, University of Colorado, Jordan, I'm going to put you on the spot and invite everybody else to answer also. On current policy trajectories the world is undershooting the Nordhaus optimal temperature figure. So do you think cost benefit analysis allows us to come to the conclusion that we should be actively trying to increase temperatures to that level above where we're currently headed?
46:01
>> Josh Rauh: Josh Rauh, thanks for a very interesting panel. As an economist, I mean, there is no alternative. You have to have cost benefit analysis. And I think, as Bjorn has shown us in the discussion of CO2 and net zero, you just have to introduce some basic CBA to the discussion of climate policies to find out what works and doesn't. You can make a lot of progress with that. I worry a lot about two things regarding the way that CBA is implemented by
46:23
regulatory agencies, and it goes to much simpler things than whether they can do differential equations. So, first of all, the guidelines for the assumptions that regulatory agencies make, they set the playing field in such an important way. So Biden administration has just totally redone the rules. It's called the revision of circular a four. Now, there are distributional assumptions that can be baked in at the discretion of the agency doing the RIA.
46:50
Now, the discount rate has been lowered to 1.7% real, which basically makes it so that policies that have short term costs and long term, very, very uncertain gains now all of a sudden are much more justifiable by the agencies. So I think that moving target can make it kind of challenging. And related to that, I worry about the incentives of those conducting cost benefit analyses. And I always encourage observers, as a finance professor, whether it's companies
47:21
that are doing discounted cash flow is their version of cost benefit analysis, or whether it's about agencies. And I ask the question, who is doing the cost benefit analysis and what are their incentives? And when I think of a company that does its version of cost benefit analysis, which is discounted cash flow, a company, if they make the wrong decision, they're going to suffer. The executives are going to lose their job. Think about some of the great failures of corporate decisions.
47:46
Kodak didn't move into digital photography. BlackBerry decided to create the BlackBerry Storm. These are business decisions that tanked companies. The EPA does an RIA, and the assumptions are not good. Well, there isn't really any Any check on them? So I guess my question maybe this is, sorry for being long winded, but the question is, what actually are the checks and balances that exist on agencies to get it right? Do we have a framework in place?
48:15
Isn't it problematic that any administration could just change the rules at will? I think it's something different when Bjorn Lomborg is doing cost benefit analyses or some of the people in others in this room are doing it. Versus when an agency is doing it, where the bureaucrats have their own incentives.
48:32
>> Dan Phaneuf: Thank you. So, Nick is kind of looking over at me. Yeah, we're two minutes over. I guess there's two options. We can try to respond to these, or we can use them for fodder and. >> Speaker 8: 2635. >> Dan Phaneuf: Okay, levelized cost. That's not my area.
48:53
>> Bjorn Lomborg: Just very briefly. Yes, of course. Levelized cost of electricity. Just tell you that solar is incredibly cheap when the sun is shining. But what are you gonna do if it's not? There's a wonderful study of the US that shows if you actually want to run the US entirely on solar or entirely on wind, you need three months of backup, which, of course, at current levels, would cost about a third of us GDP every year. So, yes, you have left something out of that equation. Now again
49:27
it's not reasonable to say that we would only run it by a 100% solar or 100% win. And it's apparently very, very difficult to do these calculations, but it's very obvious that something is left out and a very substantial amount. >> Dan Phaneuf: Bjorn, do you wanna briefly speak to if we should be trying to get hotter? And then Caroline can maybe end by the incentives. >> Bjorn Lomborg: So, Roger, you're absolutely right. Nordhaus model is a sketch model, if you will, and it also is a model
49:55
that's based in a world where we thought it was gonna get a lot hotter. So it seemed very unlikely that we should try to make the world hotter. If you do that, and it was mentioned, you mentioned that with the social cost of carbon. Clearly Nordhaus estimate it's about $6 on average today, and it should probably be at least double, possibly quadruple, possibly eight. I'm not sure what that's called, but so, clearly, we should do more. But again, that was not the point that I was trying to make.
50:31
I was trying to make the point that you need to compare the costs with climate, which everyone knows about, to the cost of climate policy, which seemingly no one seems to care about. And then you get this sort of curve where it's very costly if you don't do anything. It's very costly if you do a lot, and we need to get that understanding, it is very likely that that has shifted downwards as you also point out. If I could just answer the last one from Josh on discount rate,
50:57
I just think, yes, it's ridiculous. You could just turn down discount rates and then you basically get everything to work out. One of the things that I try to use that as an argument for is that when we do all the other things, you could also do like tuberculosis and education and stuff like that. They go through the roof. If you think about education, education is not a good investment. Well, sorry, education is a tough investment because all of the benefit
51:25
come 2030, 50, 80 years down the line. So that's why you need discounting. If you almost don't discount, everything you do in education becomes a great idea.
51:37
If you remember back in 2005, when Nicholas Stern did the first cost benefit of climate change with a very, very clear estimate, he had to find that this was a good idea. He set the discount rate at ridiculously low, like 0.1% plus some extra things. So very, very low. And one of the other economists who really, really liked what he did, he said that's the equivalent of saying that we should save 97.5% of all our income for future generations.
52:08
What that really means is if you set the discount rate at a very low level, because the future becomes so important, we should all just eat porridge now and leave everything for the future. But of course, the problem is that's true for all future generations as well. So it's just gonna be one infinite life of porridge eating. And then there'll be tons of money for a future that'll never actually use it, that's probably not a true outcome. And of course, it's not how people act.
52:37
>> Dan Phaneuf: Let's give Caroline one minute and the last word on incentives, and then we'll have a break. >> Caroline Cecot: So I'll talk about a different aspect of your question, Josh. So assumptions do matter, and we want them to be transparent. And I also worry about shifting assumptions, especially as times have become more polarized from administrations. But I will comment that Circular A-4 has not been revised for 21 years. So we had the prior circular during the Bush administration,
53:04
Bush two administration, 2003 unchanged. So some of what happened in this revision is just updates to some of this stuff. But then, of course, there's some changes. It remains to be seen what will happen in another administration, whether what becomes lasting and what doesn't, and how that sets the tone for what should happen going forward, because obviously, I think shifting in this way is not ideal. And in terms of incentives for those conducting CBA, this is a great question.
53:34
And of course, it's very different when it comes to government. But I'll tell you a little bit of what the framework is, and then you can decide for yourself whether you think this is enough or if you have thoughts about how this could be improved. So today agencies do have some check, and it's the judicial review process. Courts, of course, are made up. Judges don't have any expertise the way you have and others in the room about these estimates. So they often will defer.
54:01
But one thing that we do have in administrative law is the doctrine that agencies have to explain their decisions when these issues arise. So if someone, for example, writes a comment and says, this is an error and you should not be relying on this, you literally wrote the wrong number here, and the agency doesn't respond to that. And that's brought up, this could be influential. And same thing I've documented between administrations. If one administration has a CBA that shows one outcome,
54:29
another administration reverses the whole thing, shows the other thing. That's not gonna fly unless they can explain what made that difference and why that is based on some kind of thing, whether it means that they change the discount rate, but it's based on some kind of study, that might not be enough, but it's something that works to constrain some moves. And we've seen some of these constraints, at least during the Trump administration's experience.
54:53
The other constraint is political, which means that if the agency is doing something, its cbas are showing that this is great. It's obviously not true on the ground, there's a lot of pushback. There's a political response to this. And then it happens. When we get administration change under some statutes, agencies refer to advisory panels, scientific boards, and get some peer review of at least some of what they're considering to do. Maybe we could utilize more of this.
55:21
Maybe there could be some peer review of cost benefit analyses, too. Those are all ideas that could be put forward that are not systematically used at this time. Thank you. >> Dan Phaneuf: So we've gone ten minutes over. Apologies for that, but the brilliance of my colleagues, I think, was worth it. >> Bjorn Lomborg: Do you think we'll have to pay? >> Dan Phaneuf: Possibly, although I don't remember how that enforcement works. Thank you, Nick. Do you want to let people know how we'll do it?
55:45
>> [APPLAUSE] [MUSIC]
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat