The Future of Cost Effectiveness Analysis

In this episode, Sabeel Rahman, former acting administrator of the federal Office of Information and Regulatory Affairs, discusses updates to regulatory policy that reflect a positive new approach to how climate (and other) regulations will be assessed and crafted.

(PDF transcript)

(Active transcript)

Text transcript:

David Roberts

When President Biden first took office, his administration released a series of "Day One executive actions." Among them was reforming the way federal regulations are developed and evaluated. This is not exactly something the public was clamoring for, or even aware of, but it is foundational to the administration's ability to achieve its other goals.

The agency in charge of reviewing proposed federal regulations is called the Office of Information and Regulatory Affairs, or OIRA, which sits inside the Office of Management and Budget (OMB). It is a fairly obscure corner of the federal bureaucracy that doesn't come in for much public scrutiny, but as the gateway through which all federal regulations must pass, it is immensely powerful in shaping the space of possibilities for any administration.

A few weeks ago, OIRA answered Biden's call by issuing updated versions of two crucial documents: circular A4 and circular A94. The former contains guidance for agencies on how OIRA will evaluate regulations; the latter contains guidance for how it will evaluate public investments.

These guidance documents have not been updated in more than 20 years, so this development is long overdue. The new circulars contain some fairly technical updates to the way OIRA does cost-benefit analysis — and the goals toward which it deploys cost-benefit analysis — but they are incredibly important, evidence of a generational philosophical shift.

To unpack these changes, I talked with Sabeel Rahman of Brooklyn Law School, who served as acting administrator of OIRA last year while its current leader was being confirmed by the Senate. Rahman was intimately involved in designing the updated guidance, so I was eager to talk to him about the new approach, how it was developed, how it reflects Biden's priorities, and what it means for the future of climate and other regulations.

I know this sounds wonky, but it is worth your time. I promise you will come out of it excited about cost-benefit analysis.

With no further ado, Sabeel Rahman, welcome to Volts. Thank you so much for coming.

Sabeel Rahman

Thanks so much for having me.

David Roberts

This is awesome. I'm so excited to talk about this.

Sabeel Rahman

It is wonky, but it is awesome ...

David Roberts

Wonky and awesome. I've had sort of a side obsession with these issues for a long, long time, and this is really a perfect opportunity to jump into them. But before we jump into too many wonky details, I want to do some scene setting just so people know, kind of have a sense of what we're talking about in general. So when Biden came in, he issued this sort of list of "Day One," what they call "Day One priorities." And one of those was to update regulatory policy, basically how regulations get assessed and crafted. This is not something I think the public is beating down the door demanding this.

This is something that has a behind the scenes air about it, but it's also clearly a political priority. So maybe just to start with, let's just talk about what is the Office of Management and Budget, OMB, what is OIRA, which is the Office of Information and Regulatory Affairs?

Sabeel Rahman

Yes, absolutely.

David Roberts

First try.

Sabeel Rahman

There we go.

David Roberts

What, OIRA is within it, and why what they do seems to have such political presence in the administration that it made its way to the top of this Day One priority list just to sort of set the scene.

Sabeel Rahman

Yeah, absolutely. That's great. And thanks so much, David. And you're exactly right. This is very much kind of behind the scenes type of sets of issues, but really, really important for all the day-to-day stuff that government needs to do, and especially in this moment when we're thinking about the climate crisis or we're thinking about trying to address systemic inequality. So the fact that this was part of the Day One suite of actions from the present is, I think, pretty indicative. So there was the headline stuff, the new climate regulations, the Equity Executive Order, responding to COVID right there's, all of that headline stuff.

And then this regulatory review piece was also there because that's actually part of the back-end to make all those other policies work. So we're used to thinking about the President comes in, president can make all kinds of sweeping policy decisions or kind of really important policy decisions. The day to day of how that happens involves the federal agencies. And the agencies, they can make enforcement actions, they can spend money or they can write rules. And it's that rule writing part that goes through the Office of Information and Regulatory Affairs. And so this office that you talked about, OMB, the Office of Management and Budget, builds the President's budget every year.

They handle the budget side, but they also do a lot of really important work in terms of management. So how do agencies manage their personnel, operate strategically, have the highest impact for the resources they have. And then there's OIRA, which is the regulatory part, sort of the third pillar of OMB. OIRA works with the agencies will review under executive order going back to Clinton era and a practice actually dating even further behind that, OIRA will review major federal regulations, in part for the policy, in part for legal issues, but also, most importantly, to make sure they're consistent with the President's vision.

And that makes it a really important nexus for all of this stuff, which is also partly why it can come in the crosshairs at times.

David Roberts

Right? And a question about that, OIRA, what sort of police powers does it have? If an agency develops a regulation, sends it to OIRA for review, and OIRA finds a problem, can OIRA just say, "No, you got to go back and try again"? Or is it suggestions? What power does it have?

Sabeel Rahman

Right, so it's quite a powerful office, and I should say, obviously spent the first years of the Biden administration in OIRA and was acting head of the office until the confirmed administrator came in a few months ago. But the powers under the original Clinton Executive Order, which continues to be in effect to this day, really makes OIRA the kind of last stop in the policy making approval process. So agencies have to get OIRA clearance for significant regulations now to get OIRA clearance. That's not just what does OIRA think, what does that office think? The wire clearance process is really a kind of governmental peer review.

So the idea is that through OIRA, OIRA will get sister agencies, other parts of the West Wing, anyone who might have a point of view on the policy at hand to make sure everyone's on the same page. Right? And it's really that coordination that's the biggest kind of stick in that. If someone's got a problem, if people aren't on the same page, the rule is not going to go forward until there's at least an understanding about, okay, here's what it does, here's what it doesn't do. Everyone's comfortable with this, right?

David Roberts

So it's sort of like the last stop a regulation goes to before going out to the public and consequently has enormous gatekeeping significance, even though it's completely more or less outside of public view, unless except for a small handful of nerds paying close attention to these things. And just to mention, you mentioned this in passing, but just to clarify to listeners, you were the acting administrator of OIRA for the first two years of Biden's term while the current administrator was going through the process of getting confirmed. So these changes we're going to talk about. You were central in shepherding them through and shaping them, were you not?

Sabeel Rahman

Yeah, I think that's fair to say. There was an acting administrator in the first year, and I came in as the number two and then took over in the second year. But this was very much a big part of our day to day and a big part of the important work. Because going back to your first question, you can think about the individual regulations that are important trying to move good policy on labor issues or on COVID or on equity or on climate, but they all have to work through this existing regulatory review process. And so if we don't update that right.

This process has been in place for decades, then you're trying to shoehorn kind of cutting edge policy through very old procedures, right? And it's got to be updated. So this is a big part of the work for sure.

David Roberts

Yeah, and a final note on staffing, the administrator in question who has just confirmed is Richard Rivez. Richard Rivez, who is a law professor at NYU, one time dean of the NYU Law School, and a longtime heavyweight academic expert in exactly this stuff, cost-benefit analysis, et cetera, et cetera, which I just think is sort of an indication you watch staffing to see what administrations care about. This is, to me, appointing him indicates that Biden is taking this very seriously.

Sabeel Rahman

Yeah, I think that's right. And Ricky is fantastic. His leadership has already been tremendous in his first couple of months. But exactly right. He's certainly an expert in cost-benefit analysis in the regulatory state. He's done a lot of work on climate and energy. A lot of his academic work coming in was also about how to make sure that distributional questions don't get lost in conventional analysis. And when you look at the draft, you'll see a lot of those sensibilities woven through. Now, that's not just Ricky, right? The President, in his Day One memo calls out specifically climate, equity, distribution, future generations, human dignity, all these things that we want good policy to be able to speak to.

The charge was to go look at the review process and make sure that those values don't get squeezed out, don't get lost, that they're incorporated in a way that's rigorous and evidence based and all of that kind of stuff.

David Roberts

Right. One more background kind of philosophical note just to sort of set the table here also is debates about cost-benefit analysis go way back and are vigorous and ongoing. And there is a school of thought that says the process is sort of inherently conservative, inherently against pro social, long term action, and it should be scrapped in favor of something else. And then there's this other sort of school of thought, which I think Ricky Rivez is a good example of, which says, no, it is possible to sort of rethink and reimagine cost-benefit analysis in a way that serves pro-social, dare I even say progressive ends.

And what we're looking at here is that school of thought showing what it can do. Overall, what we're seeing here is an effort to make cost-benefit analysis more, let's say, pro social and far seeing and less of the sort of conservative process that it has traditionally been. Is that fair?

Sabeel Rahman

I think that's fair. And I would say it's an effort to make the cost-benefit really the impact analysis because it's not just costs and benefits, right? There's other stuff that don't fit into those buckets. So it's really about a more holistic impact analysis and to make that sort of as strong and robust and cutting edge as it can be. And this was part of the President's Day One charge as well, because if you look at that original memo, that memo sort of reaffirmed the President's commitment to the enterprise of impact analysis, to the conventional sort of role that OIRA plays, but set this task that the role needs to be exercised in this more modern, cutting edge way.

And that's what for me, reading it in context of the other sort of substantive Day One commitments was really important because that's the substance that the process has to be in sync with.

David Roberts

Right.

Sabeel Rahman

Because we got big things that we're trying to do on climate, on equity, on all the crises that were swirling at that time. So I think you're exactly right about the broader philosophical debate. I should say this is so important because my own entrey into this, I'm a political theorist by training I spent before the administration I was President racial equity advocacy organization Demo.

So we did a lot of work on these kinds of issues. And it's a very real debate that I imagine will continue and should continue about what is the right way to review and analyze public policy. But what I think is true, sort of regardless, is that we need something much more multifaceted holistic flexible than what we had before, even under previous Democratic administrations.

David Roberts

Yeah, what we've got now clearly doesn't work. So let's talk about then specifically what's happening. So there are two documents guidances being updated here. The first one is called Circular A4, which is basically guidance on how OIRA assesses regulations. And then there's Circular A94, which has to do with how OIRA assesses public investment. And my understanding is that the issues and updates in A94 regarding public investment more or less track what's going on in A4. So I'm going to mostly focus on A4 we can touch on. If you have specific things to say about A94 later, we can get back to it.

But I think if we cover A4, we'll more or less be hitting the big issues in A94 too. So A4 is a guidance for what OIRA is doing when it does this regulatory review. So maybe let's just start with what is this circular A4? Where did it come from? And how long has the sort of existing guidance been around? Like, who wrote the one we're using now?

Sabeel Rahman

Yeah, absolutely. So there's a whole world of government documents that is not beyond the executive orders, and we're entering into it now. So Circular A4 has never been updated. It was first issued in 2003 during the Bush years and has stayed in place since then. So that, right off the bat, I think tells you a lot, right, about just kind of how important this update is. The fact that it's a circular that's really guidance not just for OMB and OIRA, but it's meant to be guidance for the agencies. The idea is that the agencies are using it as their sort of touchstone of here's what we should be striving towards.

This is also the kind of stuff that OIRA, when your rule comes in to OIRA for final review, this is the stuff that OIRA is going to kick the tires on. And so it really kind of sets up those conversations. The other thing that's really important to know that it's a technical document. I mean, it's technical. If you read it, it's technical to read. But even its status is as a technical document, less as a political document because it actually goes through and the version that's released now is about to go through public comment and peer review.

And so the idea is that it's supposed to represent a sort of expert state of the field that is not meant to be kind of changing every time the White House changes hands. It's really meant to be. This is what public policy and social science across the board, people agree that this is a kind of best practice. Right?

David Roberts

Got it.

Sabeel Rahman

And so the ambition here is to update that old document but really update it in a way that has that seal of approval from, you know, the evidence and the research and, and, and the field yeah. So that it can have some lasting, lasting staying power.

David Roberts

Right. So this is not something that was ever intended to be sort of updated administration to administration. It's supposed to be sort of a stable guiding document over time, but maybe like having it be 20 years old is maybe a little bit longer between ...

20 years is probably too long. And I'm sure we'll get into the weeds of you can definitely see the drift that has happened right. As the world has moved on in the last 20 years.

Yeah. So several changes to A4 substantive changes that I want to get into, but one just sort of kind of technical change right off the top that I kind of thought was interesting is because I've discussed it on Volts before, is what triggers OIRA review. And so currently the threshold was if the regulation has $100 million or more of impact, that triggers full OIRA review, which is a pretty exhaustive process. It's time consuming and staff consuming and that threshold has been raised to $200 million, as far as I can tell, just to sort of reflect inflation, et cetera.

But the effect will be fewer regulation, like the number of regulations that needed OIRA review is sort of piling up and getting unwieldy and staffing shortages. So among other things, this will free up OIRA staff a little bit to focus a little bit more on the truly significant regulations. And I always like administrative capacity is one of my passions.

Sabeel Rahman

Absolutely.

David Roberts

This is sort of a way of freeing up some administrative capacities.

Sabeel Rahman

That's exactly right. And the new executive order also puts a provision that that now $200 million dollar threshold will be updated automatically every three years indexed to GDP. Which is ...

David Roberts

Makes a lot of sense.

Sabeel Rahman

... technical and wonky, but just removes this problem of like a number that may or may not have made sense 20 years ago, definitely doesn't make sense now. Just like make it an automatic thing. And it is really a capacity management tool to your point. Like our civil service, I think is incredibly important. They're a crown jewel of our democracy as far as I'm concerned, and it takes staff time and attention and resources to make good policy in service of the public.

You got to focus the efforts right on the most important stuff.

David Roberts

Alright. Okay. So let's talk about we're going to go through three big changes in A4. The first one is an update of discount rates. And discount rates are not something I think that are widely understood or widely discussed in the public. I once did a long piece on it. It was one of the first sort of super long wonky things I ever wrote. Just the wildly positive reaction sort of set me off on my career path. So I have a sort of fondness for discount rates, but let's just explain briefly, if that is even humanly possible, what do we mean by a discount rate and what is its significance?

Sabeel Rahman

So the basic idea is how do we trade-off or weigh impacts that might happen today versus impacts that might happen a long time in the future? And in general, if you have a higher discount rate, you're really favoring impacts that happen right now because you're discounting impacts that might happen, say, 100 years, 200 years in the future for a lot of day-to-day stuff that doesn't really matter all that much. But anytime you're talking about policy, the obvious big one is climate crisis policy. But anything that is going to have a longer term multigenerational tale of impact.

If you don't have the discount rate right, you're going to be systematically off. You're either going to be overweighting to the present and undercounting the future right in terms of costs or benefits. And we talked about how long it's been since A4 is updated. A4 has written into it a 3% discount rate that was written in in 2003. And so that's been the rate that agencies have been using for a long time. That's not the rate now. By all the best science out there.

David Roberts

The way sometimes I try to explain it to people is like, what if I offered you I made you choose between I could give you $100 today or $105 in a year.

Sabeel Rahman

Totally.

David Roberts

Or how about $106 in a year, $107? How much would it take for you to delay getting your money? And if you would take $200 a year from now to compensate for $100 now, you have like 100% discount rate. Or conversely, if you're like $100 a year from now, $100 now, either one is exactly the same to me that's a 0% discount rate, like future benefits are worth exactly as much to me as present benefits. And it's just I think a good heuristic sort of indicates how much do we value future benefits. So a couple of things about this.

One thing I want to ask about is the 3% discount rate that's in the previous A4, the unupdated A4 was developed via a procedure. And the new I saw Ricky give a presentation on this. The new discount rate, which is now 1.7%, was developed basically by using the same procedure, just updated numbers. So what is that procedure? How do you come to this number?

Sabeel Rahman

It's super wonky. There's a model that OMB and a lot of folks in the field have been using to basically try to take into account all the different complicated factors that might weigh into the kinds of policy impacts that might happen over the long run. You're trying to factor in changes in human behavior, changes in market conditions, all the stuff like that. So what the proposal? The new A4 proposal has actually done is two things. One is it's done a straight, just keeping the old formula, but updating the data with the latest data that we have right up to 2022, or at least through 2021 as far as data was available, and running those numbers, that gets you on the same model, a much lower number because it's got more recent data baked into it of 1.7%.

But the new version is also put out for peer review and public comment. And one of the questions that is being asked of expert reviewers is, are there variations to the model that should be considered now? That's a heavier lift, right? Because then you have to construct a new model. It has to be sort of something that meets field wide approval in terms of peer review and all of that. But the advantage of putting both of these out is say, okay, if we take a kind of small C conservative approach and keep the old model but just update the data, that already gets you a much more up to date number.

David Roberts

But the old formula is drawing on it's sort of indexing on market interest rates, right? Mainly, is that the main sort of indicator that the discount rate is being derived from?

Sabeel Rahman

There are a lot of inputs. That's one of the biggest ones. And in fact, one of the debates is basically in terms of the methodology, how much should one just sort of look to market interest rates full stop? And that's one of the modeling questions.

David Roberts

We're calling it a modeling question, but really it's a philosophical it's a philosophical question because if you're looking at market rates, you're looking at sort of intra generational, like how much do individual investors care about their individual benefits in the future versus their individual benefits today? But when you're talking about something like climate change, you're talking intergenerational sort of how much do I value benefits for my children versus costs for me? Which might not be the same thing. Market rates might not be a good indicator of how much we value subsequent generations, right?

Sabeel Rahman

Totally. And it's also not clear how much are market rates, in fact pricing in the real catastrophic risks of climate or other types of existential threats. And in the new A4, there's some nice language sort of framing that the point of discounting is to really try to also take into account some of those kind of really hard to quantify really catastrophic dangers that might come down the line. So it is a broader kind of conceptual and as well as analytic question.

David Roberts

Yeah, this was sort of the point of my piece that I wrote originally which is really like moral these are moral and philosophical debates sort of being waged undercover of math or undercover of models. So I wonder running the same formula gets you to a 1.7% discount rate, but then you also put the model out for comment, and I wonder, is there any sort of room in the guidance if an agency decides, well, the regulations we're developing have extraordinarily long time horizons and intergenerational effects, and so we would like to use a smaller discount rate? Like is there room for agencies to have sort of their own initiative here?

Sabeel Rahman

So there is room for that. And arguably the old A4 had some room for this too, although you can imagine this was rarely took up on that offer because it's really hard to calculate this stuff, right. Agencies often don't have the bandwidth or the kind of person power to build their own model from scratch. So that's why the default number is really important. But the new version does include a discussion about or the new proposal anyway. It does include a discussion about there might be instance circumstances where the particular nature of the policy or the issue might have different dynamics and in those cases agencies can and should come up with variations and probably just as a best practice like run it with the 1.7 and then run it with the variation.

So you kind of can show like have some an informed decision. But that's absolutely in there. And I think a general theme I would say of this new A4 is creating much more informed flexibility, right? That where things don't fit. Here's a good default, we've updated it, but where it doesn't fit, let's talk about how to make it work because the policy goal should be front and center and then you should build an analysis that can inform that rather than trying to shoehorn everything into a straitjacket.

David Roberts

One other question about discount rates is one of the places where discount rates come into climate policy is the effort to determine a social cost of carbon. Which is another thing we've talked a lot about on this pod and another thing that I think Biden is updating. So just maybe talk a little bit about just even if you just change from 3% to 1.7% how that might sort of affect how highly you price carbon, right?

Sabeel Rahman

So it is very big direct effect. Now we can talk in a minute about sort of the mechanics around the social cost of carbon update because that's happening in a different process. But basically this point about discounting future impacts has a big implication for how we might price the economic costs, the social costs of a ton of carbon pollution in the air. Basically, the higher the discount rate, the lower that social cost of carbon is going to be. Because many of those costs that might arise from too much carbon are long term costs, right? They're going to really manifest in the future.

And so for discounting that then the cost can look really small. Now, this is important because you know this better than me, David, but in the social cost universe, one of the things that the Trump administration did, they put out their estimate of the social cost of carbon as extremely low. And part of how they got to that low number was to say, well, we're going to have a really high discount rate, among other things, right? And so if you do the math, then you get this really low number. Well, okay, but if that discount rate is not rooted in reality, then of course that number is kind of meaningless.

David Roberts

And also to get back to the intergenerational thing, you can derive some pretty absurd results from a high discount rate. All of humanity goes extinct in whatever 2100 under like a 7, 8, whatever, percent discount rate. We would hardly care today that's going to happen. So you can get absurdities on both ends with discount rates. Okay. So the sort of default 3% discount rate now has been lowered to a default 1.7% discount rate. And that is, all things being equal, going to make more regulations look cost justified as a rule of thumb. It's going to be easier to justify regulations that have long term benefits under this new discount rate.

Sabeel Rahman

Yeah, and I would say that's just that it's more accurate, right? Because a lot of those regulations have those long term benefits. If we're talking climate, if you're talking lead poisoning, say, is another example inter-generational, not quite in the climate space, those benefits are there, we just weren't counting them. Right, and that's important to your point that regulations, people care as a policy matter, as a political matter, as a legal matter, how do the benefits stack up against the costs that we can put numbers to?

David Roberts

And I think it's intuitive too. I think if you just ask average people on the street like, do you care a lot about the welfare of your children? I think this is reflective, I think, of ordinary intuitions too. Okay, so discount rate is the first big thing that's updated in A4. But there are a couple of other really big and interesting changes too. The second one I want to talk about is distribution, basically. So I think tell me if this is accurate. I think traditionally OIRA cost-benefit analysis just looked at aggregate costs and aggregate benefits without distinguishing among who, what is the nature of the recipients of those costs and those benefits.

And that has some pretty straightforwardly counterintuitive results. So, for instance, one regulation would prevent a disaster in a poor neighborhood. One regulation would prevent a disaster in a rich neighborhood. The latter clearly has higher benefits, right. Just because the property on the line is worth more. And I think it's intuitive to people that there's something wrong with that. Right? There's something wrong with that. So for the first time, this new A4 tries to introduce sort of distributional impact analysis. So maybe just tell us, what would that look like? What does that mean?

Sabeel Rahman

Yeah, and this is another one of those things that I think under the old version there are a few sentences about, oh, you might consider a distributional impact it's not really dwelled upon. And there are definitely folks in OMB who had been pushing for this for a long time, just as pure analytic to your point. You can't really say you're doing a real analytic treatment of a policy if you're not thinking about those types of very real impacts. We would all say that, or I hope you would all say that, like $100 extra dollars for Jeff Bezos is not the same thing as $100 extra dollars for really anyone else, but in particular working class folks, right?

So the new version has a much more expanded in depth discussion about distributional analysis. First, in terms of pressing agencies to really take it seriously. Second, in terms of giving just much more detail about how and when one might do that. So you should pay particular attention to distribution analysis to the new A4 talks about when you're choosing between different alternative policy designs. Suppose you have one version that might score a little better in terms of net benefits, but like, it's concentrating all the costs on people who are least able to bear those costs, and another version which is still scoring, you know, net beneficial, but is a much more even distribution of those benefits and of those costs.

That's a relevant fact for decision makers before they decide on what the final policy should be. So that's something that the new A4 says agencies should look into.

David Roberts

Does it provide a formula sort of telling you how much weight does income, or is it just sort of like pay attention?

Sabeel Rahman

It does a little bit of both. I think the general charge, as I read it in the new document is saying that you should look into this. And here are some methods by which you might do that disaggregating the impacts by the relevant constituencies, whether that's by income or taking a racial breakdown or whatever the right bucketing is. But it also gives a discussion of what's called income weighting. And this is provided not as a requirement, but as a like, here's a tool that you could use as a way to sort of shorthand estimate how much should we weigh a dollar to a poor person versus a dollar to an ultra-rich person.

The new document has put an estimate of 1.4 as an estimate for what's called the income elasticity of marginal utility. Meaning how much more is that marginal dollar worth to you, depending on where you are in the income level. And so that's a pretty new important thing to actually have that number crunched and they're available sort of on the table for agencies to use as a shorthand.

David Roberts

This is a little bit of an aside, but it's a point that I think is worth making, which is, I think when people think about at least people in our world, when they think about federal agencies, I think, tend to think about like EPA, which has a sort of staff of dozens and all these sort of PhD economists on staff, armies of analysts. But most agencies are not that big and don't have that much administrative capacity and can't sort of sit down and develop their own models for these things. This is not like a heavy hand of central wonks here. The agencies need this.

They need this guidance, they want help doing these things.

Sabeel Rahman

Yeah, I think that's such an important point and I think this is also not a flipping of the switch, right? Like having the new guidance, you also then need to your point kind of to have some time for the agencies to get used to this, to build some muscle, to build some capacity. And so I think getting this out now with a couple of years still left in this first term, to actually then have some runway for agencies to start using these new approaches, see what's working, see what's most helpful. I think that's the kind of work that I would anticipate Ricky and Revesz and the team to be digging into going forward.

David Roberts

One other question about this. When we're talking about distributional impacts, are we just thinking of income or are there other indicators?

Sabeel Rahman

Right, so absolutely income, but also could be broken down in different ways. And so you have to look at in the guidance itself, it actually has this whole section saying that the agency, depending on the policy, should really be thoughtful about what is the most informative and relevant set of breakdowns. And it might be more than one income, race, geography, sexual orientation, there are a bunch of different ways that one might break it down. You obviously can't do every category for every policy that would be multiply really fast. But I think the point of the guidance is to say think about what are the constituencies and communities who are most likely to be impacted differently by the policy.

And then devise an approach, ideally quantitative, but if not even using qualitative assessment of what you can to think about rigorously, how are those different communities being impacted? And then having that inform the policy choice. Because at the end of the day, it's not a make work exercise. Right. The point of this is are we making good policy that serves the public and that is attuned to the very real disparities that we have in our country, right? That's the issue. So this gives a framework to do that, but it's really going to then be up to the agencies working with OMB and OIRA folks to make it real.

David Roberts

Kind of already asked this once, but I want to return to it. So if I'm an agency reading this, do I read this as I have to convince OIRA that I thought about this? Or is this a gentle suggestion from OIRA that I can take or not take? Is this now going to be a sort of a requirement for new regs?

Sabeel Rahman

It's a great question in part because I think different OIRAs have different styles depending on the administration, right? When I was there, we very much saw ourselves as working hand in glove in partnership with the agencies to make good policy, right? But that said, OMB's role is also to sort of kick the tires on whether it's the budget or the regs. And so I do think it's not a you must do this, but it is a very, very strong suggestion about the kinds of things that one should look into. And look even now, under the old A4, it's not like every regulation does every single thing that A4 talks about because there's so much in there.

So in that sense, there's lots of tailoring, lots of flexibility on what's needed. But it is very much a like, this is the bar, this is what we're going to be looking for. And when OMB comes asking for stuff, the agencies know that they got to pay attention to that.

David Roberts

Okay. So first, updated discount rate, second, the strong suggestion that agencies do some distributional impact analysis. The third thing, which is also quite interesting, I think, for insiders, because this has been a point of contention for a long time, is the suggestion that agencies take the international impacts of regulations into effect. And it was, we mentioned earlier, the Trump administration's sort of ludicrously low social cost of carbon. Part of how they came to that ludicrously low number is very explicitly ostentatiously, even saying we don't give a damn about how our regulations impact other countries, that we just don't care.

The only numbers that are feeding into our calculations are how does this benefit or cost Americans? I think just as sort of ostentatiously says, no, that's real dumb and also morally horrific. So what exactly does A4 say about international impacts? Is it similar to distributional in that this is a strong suggestion?

Sabeel Rahman

I think of it as opening up the aperture to allow first stuff that had been squeezed out before. Right? So not every rule is going to have massive global implications, but many of them will, and those ought to be taken into account. And so I read this section for those falling along at home. The geographic scope of analysis, there's a lot of great language in there about the different ways in which global effects might come into play. So it might be that there are non US citizens who are living abroad and face certain impacts that might have parallels in the US.

So they're good proxies. There might be experiencing an externality of US decisions. Climate comes to mind there as well. So there are a lot of different kind of variations laid out in the guidance. The point is that the guidance says that you should think about the global effects and incorporate them into your analysis. It even talks about how, if it makes more sense, include that as a separate thing, right, that you have your traditional analysis, but then you could also sort of provide a separate analysis of the impacts abroad, if that's the more sort of feasible way to get at it.

But you really should be thinking about it.

David Roberts

And is this similar in that you've provided a formula that agencies can use or not use? And if there is a formula, I'm wondering sort of like how much less do we value a foreign life relative to an American life? Is there a number there on how much discounting we're doing geographically?

Sabeel Rahman

Right. So not in A4. The discussion A4 is more qualitative in terms of just guidance to the agencies. And I think, like with distributional analysis and discount rates, there'll be rules where there will be trade offs that have to be weighed, I think, compared to right now, where we don't have consistent analysis of what those trade offs might even be, it's hard to even make good judgments about them. Right? And so I think the idea here is let's take that on, do the analysis, do the work. And then there are all sorts of reasons why a policy might come down one way or another.

It's hard for A4 to be prescriptive in that way, but A4 can say you need to take these issues seriously, right?

David Roberts

At least think about it.

Sabeel Rahman

Yeah.

David Roberts

So those are the three big things. And like I said earlier, roughly those same issues are reflected in A94. When they're talking about public investment, I think maybe one thing worth picking out on A94 and talking about specifically is and this was a little bit of a mind blower to me A94 as it exists has a 7% discount rate for investments in public infrastructure. Which just seems to me crazy, because public investments in particular are designed for long term benefit. That's the whole point. Like infrastructure investments and stuff like that are like we have a whole history in our country of huge investments we made that paid off handsomely over the long term but wouldn't have penciled out under a 7% discount rate.

So where the heck did that 7% discount rate come from in our public investment considerations. And how does A94 change that?

Sabeel Rahman

Big difference between A4 and A94 is that A94 is focused exactly on, as you said, David, the kind of government's expenditure on public investments. It's always been sort of related to A4, but a little bit different. The 7% number in A94 was aligned with A4 originally, and that A4 had a 3% number and then also had as an upper bound for those types of capital investment related rules, had that 7% in there. But that also is really in sore need of updating. And so that's what A94 does. I think what you'll see in the new version, without getting too deep into the weeds, is that the new version is more aligned with the new version of A4.

The numbers are a little bit different to sort of take into consideration the particularities of what, say, the budget side of OMB has to take into account when they're dealing with federal investments in buildings and stuff like that. It's a little bit different from regulation. So that's where the divergence comes from. But that old 7% number, I think, is reflective of where we were with the thinking on this stuff in the early two thousands.

David Roberts

Is there a new number for discount rates for public investments or is it not that simple on this side?

Sabeel Rahman

Yeah, it's a little bit more complicated for the investment side, but that's also going for the same kind of peer review and public comment. So I think we'll know more in the coming months about where A94 lands. Exactly. But what's exciting and important is that A94, the last revision was in, I think, 1992, not quite as old as A4, but still pretty old.

David Roberts

Yeah, that's wild. And A94 also gets into distributional stuff and international stuff.

Sabeel Rahman

Really important, yeah, really important distributional and global effect language in there.

David Roberts

And part of this, I think part of the point of all this, one of the points of all this was to align A4 and A94 better, so we have some sort of coherent they had kind of drifted apart in a way that made no sense.

Sabeel Rahman

Totally. And full props to the OMB Chief Economist team who they sort of are the keepers of A94 in the sort of OMB ecology. I know they've been working really hard on all this.

David Roberts

So those are the three big changes. Are there other changes in A4 that you think are worth calling out? Like there's something having to do with risk and risk tolerance and risk assessment that I don't even know how to create a coherent question about, but feel free to talk about it.

Sabeel Rahman

Totally. Yeah. No, I appreciate that. I think there are three other buckets of things that I'll just highlight briefly because they're important and also, I think really brings this up to 2023. So one is what you were just alluding to. There are a bunch of things that will mean a lot more to our economist and economic modeler friends who might be listening, but they really amount to kind of bringing A4 into line with sort of cutting edge of economic theory. So how do we take into account uncertainty and uncertain effects that we don't know with perfect certainty these estimates, right?

Let's factor that in. Taking into account risk and risk aversion, right? People are willing to pay more to be protected from the risk that something might go haywire. It's kind of typical understandable human behavior, but that's not always baked into the models. So there's some technical stuff. That's bucket one. The second bucket is there's a lot of really interesting what I think of as like macroeconomic structure, things that are baked into this new version. So normally reg reviews, reg analysis would be you're looking at the new regulation almost at a micro level. Like, you're just looking at that regulation and you're kind of holding the rest of the world kind of constant.

But the new A4, it talks about, for example, business cycles. There might be regulations that have different benefits and costs when we're in a recession versus when we're not. If you think about, for example, social insurance policies, if they're designed to be countercyclical, those benefits really only kick in under certain conditions that may not be around when the regulation is being written, right? So it incorporates that. It incorporates a lot of great new thinking about market concentration and competition that's been a big focus of this administration, being attuned to the ways in which concentrated ownership of industries can lead to higher prices, less stable production, kind of all the antitrust, anti-monopoly stuff that is happening.

So that's baked into A4 much more. So these kind of like big macroeconomic conditions.

David Roberts

Macroeconomic context, which you would think is like totally like duh, of course.

Sabeel Rahman

Right. And then the last bucket is also another like you'd think, of course it's not rocket science, but it's a really important shift is that the new A4 has a lot more language and guidance about what to do with those impacts that you can't put numbers to because they're obviously real. And there and it even talks specifically about things like civil rights and civil liberties, democracy, equity. These are goals for good public policy. They may not have number values and in some cases ought not to have number values.

David Roberts

Or the welfare of other species. If I can just ...

Sabeel Rahman

Yeah, there's a section there about ecosystems and ecological impacts as part of the hard to quantify discussion. So there's tons of really important implications of that that you could imagine for everything from ecology to equity that I think this new A4 is much, much better at.

David Roberts

And those, I think, are really sticky to deal with because there's no formula. I mean, even a candidate formula, right? There's just no way to come up with a formula for how, like, you know, how much should EPA weigh beauty or whatever. Or whatever. But as you say, those things exist and matter. So is this just basically OIRA saying to agencies, take note of think about these things, like, take these things into consideration. Is that all there is to it?

Sabeel Rahman

I think it's two things. One is that take into consideration and bring those considerations into your analysis. Because a lot of times agencies are thinking about that stuff, but they've struggled in the past at times to bring those very real considerations into an impact analysis, given how narrow the old A4 used to be. And so then you had this kind of weird, right? Like, we know that part of what we're trying to do is protect ecosystems or protect the dignity of disabled persons who a curb cutout is expensive to have on every sidewalk, but absolutely critical to just look at basic human dignity if you're a disabled person of a particular kind.

Let's take that obvious real factors that are in any human decision about this kind of stuff and let's actually give it a proper pride of place in the analysis, and then it's up to the agency to sort of make the all things considered best decision. And that's something that would be worked out through the review process with DOT might come to a particular view about what the rules should be for lavatory access for disabled persons on an airline. That's going to have costs. It's going to have qualitative considerations about basic civil liberties, human life and safety and dignity.

And then if we can get everyone on the same page through the review process that this makes sense, this is good, then that should be the way we go.

David Roberts

Right, so this is or saying we're going to give you wide-latitude to think about these things and incorporate these things.

Sabeel Rahman

Yeah, absolutely. And a lot of times I think people who are more attuned to the hard sciences might feel like this stuff can be squishy and amorphous. I always found it very straightforward. Give us your reasons. Like any good like any good piece of writing. Give your reasons, give your evidence. And just because it's qualitative doesn't mean you don't have reasons and you don't have evidence. So talk about it.

David Roberts

And even if you think it's arbitrary to pick a particular number, it's quite clear that the number is not zero, right. The default has been zero, which is clearly against our common values.

Sabeel Rahman

Absolutely.

David Roberts

Okay, so that's a lot of changes. A lot of changes to pack in this technical circular. One other thing I saw at one point the term cost-effectiveness analysis. Is there some effort to replace the whole sort of notion of cost-benefit analysis with something else? What's the deal there?

Sabeel Rahman

Yeah, I would put in the theme of much more flexibilities right, to get the right kind of analysis for the policy at hand. So I think people are familiar with traditional benefit cost-benefit analysis where you kind of monetize everything on both sides of the ledger and then you sum it up and then you're done. But especially when you're talking about things that might not be quantified quite as well. Right. There are other variations that the new A4 talks through in more detail. So cost-effectiveness is one suppose you have a kind of easy to understand qualitative goal that may not have an exact number that you can measure, but you can measure the costs.

Right. So we want to increase safety in the workplace. We can proxy that in different ways, but the proxies are all kind of imperfect and we know how much it would cost to increase the safety requirements or slow down production so that people aren't hurting themselves in a horrible way. Right. So then you can sort of do a cost effectiveness comparison of how much boost to your goal are you getting for some higher cost? And then the policymaker can say, yeah, that's justified because that goal is really important. We've done some work to know what the costs are, and we think this is the overall more cost effective way to get to that goal compared to some other variation.

David Roberts

Right. This is something that debates over cost-benefit analysis have been batting back and forth for a long time, which is the sort of premise of cost-benefit analysis is you let the cost-benefit analysis tell you what your goal is, whereas cost-effectiveness is, here's the goal. Now let's work backwards what is the most cost-effective way to reach that goal, which is a very different way of approaching.

Sabeel Rahman

Yeah, absolutely. And I think if I could pin down sort of the ethos of this overall A4, I think you just put it really nicely, David. The analysis should be informing the policy, not the other way around. Right. And so there's another variation on this that's called breakeven analysis that's also talked about in the new A4. It's a similar kind of idea that let's say we can't actually put a hard number to the benefits, we can't monetize it for whatever reason, but can we figure out sort of like, what's the threshold that if we think the benefits are above that number, then we know we more than break even.

So even if we can't put a hard number to the benefits, we know with good certainty that the benefits are above this level of what the cost might be. And so these are all imperfect, right. I think for folks who are wary of quantified and monetized cost-benefit analysis, I think these will not get you all the way there from that critique. But what it does do is give you a lot more options to say, let's stop shoehorning good policy judgment into an old straitjacket process.

David Roberts

Right. Another of the Biden administration's big priorities, big pushes, is to bring the public more fully into these deliberations.

Sabeel Rahman

Yep.

David Roberts

And I just wonder, I have a couple of questions about that. One is just what does A4 say about that? In what way do you recommend that agencies do this?

Sabeel Rahman

I'm really glad you asked this because this is also something I feel super strongly about and I think they've done some good work on this. So alongside A4, the President also issued, when the new A4 came out, an Executive Order, which included some new requirements around public participation. So one, it's a much stronger emphasis for the agencies to do more proactive early engagements with impacted constituencies as they're designing their rules. So by the time a rule comes into OIRA review, a lot of it is I don't want to say cooked already, but a lot of work has already been done.

And a lot of times it's much more impactful and meaningful to have robust public participation earlier in the process, especially if you're talking about underserved constituencies or impacted groups, right? Those are the voices that you need to hear early on. So there's a general charge under this administration in a number of different Executive orders actually, to press agencies to do more of that proactive early engagement. Then there is a switch to the OIRA review process itself to open up that process to more participation as well. So right now, an interested party could request a meeting with OIRA when they're reviewing a rule to give their views.

It's kind of a wonky thing. Not that many people know about it.

David Roberts

How is that different from public comment?

Sabeel Rahman

It's very similar to public comment, but it basically takes place during the OIRA review process before the rule goes out for public comment. And the rationale is there might be nuances or details that if you're from an impacted community, you might have some additional information that might be worth making sure is emphasized or shared as an input. Those are not backs and forths. Right. They're basically listening sessions. But the new Executive Order continues that practice and charges OIRA with making that much more accommodating and inclusive, especially to historically marginalized communities, communities whose voices may not have sort of K-Street lobbyists, right. Easy for them to go.

And then the last thing that it does, it also puts a lot more meat on this process that's called petition. So communities, civil society groups, individuals can petition agencies to push them to take action on an issue that maybe the agency hasn't thought about before or hasn't been as proactive on those petitions, then kind of sit there. The new process creates some more coordination so that when someone brings a petition, OIRA is also aware of that.

OIRA can then also sort of check with the agency, like, hey, this petition, what's the response to this petition? How are we responding to the needs that folks are bringing to you? So these are a lot of little pieces here and there but the sum total of it is to try to improve that participatory and inclusive aspect of the rulemaking process.

David Roberts

Well, one other question about that, which is I'm sure you're aware of the sort of larger conversation going on around liberalism these days which is that it's become slow and that NIMBYs are stopping everything and it's hard to build anything and it's hard to move quickly. And basically we've become sclerotic. And sometimes that critique takes the form of saying basically like the public has too many ways of inserting itself and exercising a veto here and we need to streamline things and do things faster. And this seems intuitively on the surface at least, to tack against that.

So how do you think about the kind of need for moving expeditiously relative to the need for public participation?

Sabeel Rahman

Right, it's such an important question. I'm glad you asked it. I think there are two things to bear in mind. One is just on these proposals here. These are all very much sort of inputs. These are not decisional or veto types of discussions. Right. So I think things still move along and the overall time frame of the regulation OIRA has a clock under the old Executive Order which has stayed in place for how long it ought to be taking on rules.

David Roberts

Although can we just say it has frequently exceeded that alleged clock. I remember all the complaining about OIRA that used to go on when Cass Sunstein was in charge under Obama. Like OIRA was frequently charged with slow walking these regulations.

Sabeel Rahman

I wasn't there during that time. But one of the things that I'll say is when there's a delay, usually it's because there's actually a dispute as opposed to someone's just kind of holding it up. But one thing I'm really proud of with this administration and this OIRA is that if you take ARP, for example, we set an OMB, a two week cap, a two week ceiling on any this is American Recovery Plan at the height of the economic freefall right in spring 2021. All that money had to get out the door because the economy was in such grave danger.

People were hurting right during COVID, so we set and we kept to a two week turnaround. Every single piece of ARP policy came through OIRA, went through OIRA review and was done in two weeks or less because we had to, right? And I say that to say where there's will and focus and dedication, the process moves. And one thing I really like about this administration's approach is that it has tried to balance we got to get stuff done with. We got to also kind of have evidence and do things robustly. So to come back to participation then I think that the critiques are important and well founded.

But to my mind it's not a choice between participation or effectiveness, it's a question about what kind of participation to make the policies effective. So you're not going to have good policy if you don't hear from the people you need to hear from. But b, there's a way to hear from them in a way that is efficient. Right. This is why that upstream early engagement is so valuable because you get everyone together, you get all the inputs you need and then you design the policy right the first time around and then you don't have to have 50,000 kind of nipping at your heels types of conversations downstream.

People are going to disagree, that's fine, you can litigate disagreements at the next election or with the next regulation or whatever, but that doesn't mean we shouldn't be hearing from people.

David Roberts

Yeah, and there is a value just to being heard. I think a lot of people, they might not say this explicitly, but I think a lot of these groups will feel like there's progress being made just if they are convinced that they are genuinely having a voice in this.

Sabeel Rahman

And that's something I saw a lot of in the time I was in government too, that it is really important the government serves the people and it's hard to convince people you're serving them if you're not actually hearing from them or seeing them. Right. And that doesn't mean that you have to do everything that everyone says because we have a big country. But I do think that's an important piece of this.

David Roberts

What's the process now? So this is A4 and A94 have been put out for both public comment and peer review, which are separate processes. So what happens next? What's the road?

Sabeel Rahman

So over the next maybe couple of weeks, there's still some time left in the public comment period. So if you have listeners who might have views about either of these documents after listening to this, they should absolutely weigh in, particularly if they have expertise on some of these issues. So public comment is ongoing that will close in a couple of weeks ...

David Roberts

Kind of pause there on public comment. I'm just curious if you have any sense or guesses about how public comment is going to unfold. My gut instinct is because most of these changes will have the effect of making it easier to pass big regulations in the public interest, that industry is just going to knee jerk, be against them and rail against them in public comment. Is that sort of how you expect public comment to shape up this sort of like public interest groups versus industry yet again? Or is there more nuance to it?

Sabeel Rahman

I hope it ends up being more nuance than that because there's a lot here. Obviously. I think that's anytime you're talking about regulation, we live in the world, we have to be attuned to that dynamic. But I think my hope would be that there's enough here that is evidence based, empirically, rigorous, and just like obvious updates, right? Like what you're talking about on discount rate and on distribution analysis and so on, that I hope we'll get a range of comments. It'll be particularly important, too, to sort of get comments from the field, as you were saying, from economists, but also anthropologists and sociologists and people who are working it in community on the grassroots level about what kind of distribution analysis will actually help, you know, make sure their voices are counted right.

Like, I think we want to cast it open so it's not just the same, you know, conventional wonks as as much as we do want to hear from them too.

David Roberts

Industry lawyers.

Sabeel Rahman

Right. I think a bigger set would be lovely because this is some of the source code of an executive branch that can honest to God serve the public interest and serve all of us. I really believe that. So I hope we hear from more people. So first there's comments. Once that comment period wraps up, the peer review should be happening. I'm not sure exactly when, but I assume they'll be doing it in parallel to that because peer review can take a long time.

David Roberts

And is this peer review just like the same that academics are familiar with?

Sabeel Rahman

Yeah, exactly. It's meant to be a similar process. So the government actually has a pretty standard peer review process for technical documents that this review, I assume, will be following. And that's modeled on sort of scientific, academic peer review procedures. So that will go for a couple of months as well. And then these documents should be finalized. The timeline they gave was no longer than a year. Basically by play it forward a few more months, maybe it's a little sooner, maybe it's a little later, but I think it's going to move pretty quickly, especially once the comment period closes, just because this is way overdue and is important.

David Roberts

And you think these will be finalized in time to actually inform the writing of substantial regulations from the Biden administration?

Sabeel Rahman

I hope so. I think that certainly is the goal with this charge coming out on day one and then now the full proposal out here at the start of year three. So I really hope that's the case.

David Roberts

And once they're in place, I guess I'm wondering how resilient they are. Like, if there's a DeSantis administration in 2024, is there anything stopping them from just ripping these up and going back to something older? Do these have any resilience against political chicanery or an administration can do what it wants?

Sabeel Rahman

That's a tricky question in a world where when you look at what's happening in places like Florida and elsewhere. I think what ... I'd say two things about that. One is that it is meant to have staying power. And part of the point of making sure this document goes through peer review and goes through public comment and goes through all the things that a long lasting, non political technical document ought to go through. This is that kind of enterprise. And so the old A4 lasted for 20 years, and this new version is very much an update to that.

It's not junking the old enterprise at all. So I think the hope would be that that would continue. Now, that said, when you have people on the right continuing to organize around things like Schedule F, the Trump administration's plan to junk most civil service protections, for example, there's a lot of stuff that is brewing on the right. Just to say that is really aimed at destroying a well functioning, evidence based and transparent bureaucracy. But that's a broader question. That's not an A4 question. That's a broader question for all of us to say that, okay, yes, we're having a debate about policy and about all sorts of kind of horrifying other things that are happening too, on the far-right.

But at some point, we got to say, if we got to have a government that serves the public understanding that the public doesn't always agree on a lot of policies, we can do that, right? We can do that with evidence and with transparency and with good procedures that allow for participation and evolution of ideas. There's a way we can do that. It's possible to have a government in this country that is effective and that deals with our complexity and our diversity, but not if you have bad actors who come in with a desire to bring a wrecking ball, right?

David Roberts

If they want to come nuke the administrative state, A4 is not going to stop them.

Sabeel Rahman

Right. And I think it's sort of for all of us who care about these issues, I think it's important for us to care about wonky stuff like A4. But I also think it's important for us to care about those kinds of existential threats to the project of shared collective government in the first place.

David Roberts

Yes. Alright, well, that seems like a great place to wrap up Sabeel Rahman, this has been so helpful, so clarifying, and I love getting into the wonky guts of stuff because, as you say, it's a source code. It's going to affect everything that comes out of the government after this. So it's really great to get a clear view of it. Thanks so much for coming on.

Sabeel Rahman

Yeah, thanks, David, for having me. This was great.

David Roberts

Thank you for listening to the Volts podcast. It is ad-free, powered entirely by listeners like you. If you value conversations like this, please consider becoming paid volts subscriber at volts.wtf. Yes, that's volts.wtf, so that I can continue doing this work. Thank you so much, and I'll see you next time.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.volts.wtf/subscribe