FTC: Cost/Benefit Analysis of Proposed Rules – A Deeper Dive

Event Video

Although primarily an enforcement agency, the Federal Trade Commission (FTC) has issued a historic number of proposed rules over the past two years. From prohibiting non-compete provisions potentially impacting 30 million employment contracts or privacy and data security rules implicating personal information online, these proposed rules will affect most sectors of the U.S. economy. This panel of experts will explore how a federal agency undertakes the cost-benefit analysis for proposed rules, comparing independent agencies to those subject to OIRA review, and provide practical tips for lawyers and economists working on agency rulemaking comments.



  • Dr. Andrew Stivers, Associate Director, NERA Economic Consulting
  • Paul Ray, Director, Thomas A. Roe Institute for Economic Policy Studies, The Heritage Foundation
  • Jonathan Wolfson, Chief Legal Officer and Policy Director, Cicero Institute
  • Paul Metrey, Senior Vice President, Regulatory Affairs, National Automotive Dealers Association 
  • [Moderator] Svetlana Gans, Partner, Gibson, Dunn & Crutcher, LLP


As always, the Federalist Society takes no position on particular legal or public policy issues; all expressions of opinion are those of the speaker.

Event Transcript



Introduction:  Welcome to the Regulatory Transparency Project’s Fourth Branch Podcast series. All expressions of opinion are those of the speaker. 


On February 7, 2023, The Federalist Society’s Regulatory Transparency Project hosted a virtual event titled “FTC: Cost-benefit Analysis of Proposed Rules-A Deeper Dive.” The following is the audio from the event. 


Chayila Kleist:  Hello and welcome to this Regulatory Transparency Project webinar call. My name is Chayila Kleist, and I’m Assistant Director of the Regulatory Transparency Project here at The Federalist Society. Today, February 8, 2023, we’re excited to host a panel discussion entitled “FTC: Cost-benefit Analysis of Proposed Rules- A Deeper Dive.” Joining us today is a stellar panel of subject matter experts who bring a range of views to this discussion. As always, please note that all expressions of opinion are those of the experts on today’s program as The Federalist Society takes no position on particular legal or public policy issues. 


Today we’re pleased to have with us as our moderator Svetlana Gans, who’s a partner with the Washington, D.C., office of Gibson, Dunn & Crutcher, LLP, where she helps clients investigate complex consumer protection, privacy, and competition related regulatory proceedings before the FTC, U.S. Department of Justice Antitrust Division, state attorneys general, and other enforcement bodies. Ms. Gans also assists on litigation matters and provides strategic counseling advice related to public policy issues. 


Before joining Gibson Dunn, she served as the vice president and associate general counsel at the NCTA, the Internet and Television Association, where she helped lead the association’s consumer protection and competition policy work. Prior to joining the NCTA, Ms. Gans served with distinction as the chief of staff to acting chairman, Maureen K. Ohlhausen at the FTC. Previously, Ms. Gans had the unique experience of serving in both litigating bureaus of the FTC, Bureau of Competition and Bureau of Consumer Protection. 


In the interest of time, I’ll keep my introduction brief, but if you would like to know more about Ms. Gans or about any of our panelists as she introduces them, please feel free to visit regproject.org and check out their impressive full bios. As a last note before I hand it over, throughout the panel if you have any questions, please submit them by the question and answer feature you’ll find at the bottom of our screen so that our speakers will have access to them when we get to that portion of today’s webinar. With that, thank you for being with us today. Ms. Gans, the floor, at least virtually, is yours. 


Svetlana Gans:  Great. Thank you so much and thank you to The Federalist Society for hosting this informative and very important program on agency cost-benefit analysis. Although the FTC has historically been an enforcement agency, they have proposed at least six new rules covering both consumer protection and competition issues in the last few years. There are several questions presented with respect to those rulemakings, including statutory authority among other questions. But this panel will focus specifically on agency cost-benefit analysis. 


We will cover three specific questions. One, what standards apply to independent agencies like the Federal Trade Commission? Two, how does the FTC go about its cost-benefit analysis internally? And three, how does it work in practice with respect to current rules that the Agency has proposed? To answer those three questions I am joined by an all star panel. Full bios are available on the Federalist Society website, so I will just give these folks a brief introduction. 


First we have Paul Ray. He leads the Heritage Foundation’s work on regulatory and economic policy as the director of the Thomas A. Roe Institute for Economic Policy. Paul previously served as administrator of the Office of Information and Regulatory Affairs. He was the federal regulation czar within the White House Office of Management and Budget. As administrator, Paul supervised the review of hundreds of regulations and led federal efforts on regulatory reform. 


Next we have Jonathan Wolfson. He’s the chief legal officer and policy director at the Cicero Institute. His research focuses on healthcare, regulatory reform, and employment policies. Before, he led the Office of Policy at the U.S. Department of Labor where he managed DOL’s deregulatory efforts and oversaw DOL’s internal policy development think tank. 


Next, we have Dr. Andrew Stivers. Dr. Stivers is an associate director of NERA. He specializes in the economics of consumer protection and privacy. Prior to joining NERA, Dr. Stivers served as a senior official in the Federal Trade Commission’s Bureau of Economics where he oversaw economic analysis of all consumer protection and privacy matters. 


Next, we have Paul Metrey. He is the Senior Vice President of Regulatory Affairs for the National Automotive Dealers Association. In this capacity he directs a team of attorneys who represent dealer interests before the federal Executive Branch that oversee dealer operations. So with those introductions aside, I wanted to ask Paul to kick us off to discuss the standards by which federal agencies like the FTC, which is an independent agency, must conduct a cost-benefit analysis. 


Paul Ray:  Well, thank you, Svetlana, and thank you to The Federalist Society for having me. And thank you to my fellow panelists and all of you for joining. So I want to begin with the preface that I’m not an FTC lawyer. I’ve never practiced before the Commission. Indeed, I’ve never reviewed an FTC rulemaking. As you may know, FTC does not submit its proposed or final regulations to an OIRA review. 


So I’m here to talk instead about lessons from general cost-benefit analysis and how they’re applicable to the FTC because obviously the FTC has not undertaken a very, very large number of rulemakings in the past. So they’re going to be really pioneering an approach with respect to cost-benefit analysis here. And so I think there’s some lessons to be learned from the experience of other agencies. 


So first I want to offer a quick working definition of cost-benefit analysis. I’m going to use that definition “the evaluation of the effects of proposed policy for the purpose of decision.” So I want to bring out a couple of points implicit in this working definition to dispel a couple of myths about cost-benefit analysis. The first is that some people think cost-benefit analysis dictates policy outcomes, but it does not. 


As should be evident from the working definition, it is merely evaluative. It tells us what will happen if we do X or Y, but it’s not designed to, indeed it can’t dictate whether we do X or Y. The best it can do is guide an agency or the public as I’ll discuss shortly as the agency or the public deliberates about options. Cost-benefit analysis can pick out the benefits and the burdens that each given option would create and can invite the decision maker to make a reasoned judgment about whether the former or worth the latter. But it doesn’t dictate policy outcomes. 


The second myth I’d like to touch on is that -- and this actually indicates that the name “cost-benefit analysis” is a bit of a misnomer. Cost-benefit analysis if it's any good seeks to discover all the effects of a potential action, not just its costs and benefits. So this includes transfers, movements of wealth or other benefits between people. In other words, cost-benefit analysis is not committed to some sort of utilitarianism that ignores how benefits and burdens are distributed. It’s not always seeking greater optimality. It can indeed -- if it’s good cost-benefit analysis, should take into account transfers. 


Cost-benefit analysis as it’s used in the federal system serves as I alluded to briefly earlier two audiences, and it plays a different role for each. First, it is directed to the public. It apprises the public of what the government is up to so that people can respond. And second, it is addressed to the agencies themselves to guide their decision making in the first instance. 


The two goals are complementary, but they’re not the same. In particular, analysis that might be adequate to inform the public because it accurately describes a state of uncertainty may be inadequate to assist the agency to make reasoned decisions. And that’s because it may not tell the agency what it needs to know in order to judge whether it has good reason to proceed with the rulemaking. 


As many of you are doubtless aware, the Administrator Procedure Act, or the APA, requires that agencies may not issue regulations that are arbitrary or capricious. The Supreme Court has interpreted this as a rationality requirement. The agency must have good reasons for what it does. It must among other things take into account all the relevant factors when making a decision. 


Now, typically cost is a relevant factor. We know this for instance from Michigan v. EPA, which is a Clean Air Act case from 2015. But the holding, although in the Clean Air Act context, applies to the APA, the Michigan case teaches that while some statutes forbid accounting for cost, courts and thus agencies must presume absent contrary indications from the statute that cost is relevant. 


Now, Michigan dealt only with cost, but I think we can safely say that agencies also have an obligation to understand the transfer effects of their actions when those transfers are relevant to the statutory purposes. So where that leaves us is this. An agency should typically offer cost-benefit analysis adequate to understand what will happen in consequence of its rule and then determine whether it wishes to proceed with the rule in light of these effects. 


Now, uncertainty is not uncommon in agency rulemakings. Reasoned decision making in that case requires that the agency have a good reason to act in the way it proposes to act notwithstanding the uncertainty. A good reason might be that the potential downside is greater on one side than on the other. 


So to say that an agency should offer enough cost-benefit analysis to form some idea of its effects is a helpful metric, but it doesn’t specify exactly how much or exactly what kind of cost-benefit analysis is required. For that, if the FTC were presidential or a cabinet agency it would turn to EO 12866, which sets out certain standards for rules that come to OIRA. I want to emphasize the standards of EO 12866 are not required in law. They are a matter of executive practice. 


The key point is that whatever cost-benefit analysis the agency offers, it owns that analysis. It’s responsible for it upon judicial review. If the agency bases its decision on unnecessarily robust or extensive cost-benefit analysis, then its rule must live or die based on that analysis, even if it didn’t have to do that analysis in the first place. 


Now, all that I’ve said so far applies to all agencies. There are some special requirements that apply to the FTC. Section 5(n) of the FTC Act forbids the Commission to declare something an unlawful practice under that section, unless it inflicts harm on consumers that is not outweighed by benefits to consumers or competition. So some sort of cost-benefit analysis seems to be required to make a reasoned judgment about the balance of harms to consumers under Section 5. 


Section 22 specifically requires cost-benefit analysis for both proposed and final rules under Sections 6 and 18. Section 22 goes into considerable detail. It requires a statement of the need for the rule and a statement of reasonable alternatives, a description of the full set of effects, that is to say of costs, benefits, and transfers of the rule. 


Now, all of that tracks Executive Order 12866. The analysis required of the FTC is just the same as the analysis required under the Executive Order for presidential agencies. But then actually Section 22 goes further. It requires a full set of effects for not just the proposal but for the alternatives as well, and that is quite a bit more than is required of cabinet level agencies. The Commission also must give reasons for choosing the chosen alternative and summaries of comments and responses. And there that tracks presidential agencies as well. 


Now, while the requirements of Section 22 are robust, it’s important to see that the end of the section specifically exempts from judicial review—this is going to be close to a quote—"the contents of any regulatory analysis issued under this section.” So while the section’s requirements are robust, judicial recourse is not there to backstop them. 


So as I wrap up, I’d like to propose three takeaways. First, not withstanding the lack of judicial enforceability, the FTC does have an obligation to conduct cost-benefit analysis under law and extensive cost-benefit analysis at that. That analysis is similar to but actually in some important respects as I’ve said exceeds that cost-benefit analysis requirements of Executive Order 12866. 


Second, for this reason commenters can hold the Agency to account by sending in good cost-benefit analysis in their comments. Speaking briefly about the non-competes proposal, the Commission estimated a quarter trillion dollars of increased earnings per year. And it made that estimate based on only two studies. That is extraordinary to me. If any commenter is able to send in really serious cost-benefit analysis, even one weighty study could dispute really the fundamental premises of this cost-benefit analysis. And there’s value to getting the Commission to engage with studies even though there’s no judicial backstop available to enforce quality cost-benefit analysis. 


And the reason that’s worthwhile is my third takeaway. Even though the Commission can’t be subjected to review for the contents of its analysis, the content of its reasoning that responds to its analysis should be susceptible of judicial review. So for instance as I mentioned the noncompete proposal estimates a quarter trillion dollars a year in salary increases, but it also admits that some of the increase could come from consumers. And it never makes any findings about how much of the increase would come from consumers. 


So a commenter who asserted that the Agency shouldn’t proceed if it’s not sure whether it’s going to make consumers a few hundred billion dollars poorer should be able to challenge the final rule if the Commission fails to give an adequate response to that comment. So in other words or just to sum up, while a cost-benefit analysis under Section 22 can’t itself be challenged if the Agency includes reasoning not on the cost-benefit analysis but nevertheless predicated on the findings of the cost-benefit analysis, I see no reason that it couldn’t be challenged in court. So with that, I think, Jonathan, you are up next, so over to you. 


Jonathan Wolfson:  Thanks, Paul. So I’m going to talk for a couple of minutes about some of the strategies that agencies may try to use in an attempt to avoid doing cost-benefit analysis. And then I’m going to talk about a couple of the pathways in addition to some of the FTC’s required pathways that the FTC also needs to do something that looks a lot like a cost-benefit analysis that may also have additional Administrative Procedure Act pathways for both commenting but also for the courts to step in if the analysis is not done correctly. 


So when I was at the U.S. Department of Labor, we put a number of regulations out. And one of the tasks that we were in charge of doing in every one of those because we were subject to Executive Order 12866 was that we had to put together a cost-benefit analysis that to Paul’s point included not just the costs and the benefits but  tried to evaluate all the effects. 


What are the transfers? Who’s going to be benefited? Who’s going to be harmed? What are the other consequences of the potential rulemaking? Address comments that came in about those analyses and then include a final cost-benefit analysis in the final rulemaking if we actually decided to finalize a rule. 


But sometimes agencies will attempt to get cute, and OIRA generally is able to step in in certain cases. But sometimes OIRA is also interested in moving forward with the particular regulation. And so even for cabinet agencies that are subject to Executive Order 12866, there are some little tricks that agencies find helpful to avoid doing a full fledged cost-benefit analysis. 


One of the things that agencies will attempt to say is that data are not available. And this is usually a true statement at the front end of a rulemaking where the agencies actually lack the data. Sometimes there are agencies who will claim they don’t have data when the data simply don’t align with their priors. But generally the data -- especially when you get into kind of projecting the future, the data are often not sufficient to make a good projection on the front end before you write the proposed regulation. 


However, it’s important to note that once those agencies receive projected data from outside groups as part of the comment process, those agencies no longer are able to claim that there are no data available. They are obligated at that point to evaluate the data that have been presented to them. And so this is one area where, as Paul mentioned, if you are suggesting comments to any agency’s rulemaking, you should be strongly considering whether or not you should include something about the cost-benefit analysis in your comment because the agencies, especially if they claim a lack of data, are going to be obligated to evaluate the data and the data analysis that are submitted as part of comments. 


Another strategy that agencies often try to take is they claim the projection would be difficult. It is difficult to project what the long term outcome would be from a proposed rule. But that is not an excuse for not doing the analysis. What that is is simply an opportunity for, again, commenters to provide additional information. However, agencies do need to do their best efforts, and courts will often hold them accountable for failing to take their best efforts at actually projecting what the effects of the rulemaking will be. If they are not willing to actually dig in and they simply claim there will be all benefits and no costs, everything’s rainbows and butterflies but there’s no problems that are going to occur in the world, then those agencies’ rulemakings are going to be challenged and be more heavily scrutinized. 


A third thing that agencies will attempt to do in order to try to avoid the requirements of Executive Order 12866, OIRA review rulemaking in the cost-benefit area, is to assume only transfers between groups which have no transaction costs. They claim that you’re just shuffling around the resources, and there’s actually no transaction cost for that having happened. And so they might claim, for example, that if you increase certain groups of employees’ wages in a wage and hour division regulation, that there are no transaction costs that accrue to the employers or to consumers as a result. This is simply a movement of dollars from the employer to the employee, and there’s nothing else going on. 


But this is another one of those tricks where any good economist will tell you that if you’re moving money from one person to another, there are going to be some transaction costs. And those are areas that agencies, if they are not careful, can get themselves in trouble. But those are the tricks that many agencies will attempt to follow in order to avoid doing the required cost-benefit analysis under Executive Order 12866. 


Now, it’s true that the FTC is not regulated by 12866. Executive Order 12866 does not apply to independent agencies. And the FTC has stated on a number of occasions recently that they have no obligation to submit their rulemakings to OIRA. I don’t think any of us on this call question that they are legally entitled to issue rules without going through OIRA. However, there are a lot of ways that even independent agencies have to go through a cost-benefit analysis, regardless of whether their statute requires it. 


So Paul’s already very carefully walked through some of the rules that the FTC has to follow because of the statutory requirements for certain analyses as part of their rulemaking. But separate and apart from those requirements, there are other federal statutes that do require some kind of analysis which has a lot of corollaries to a typical cost-benefit analysis. For example, the Regulatory Flexibility Act requires an evaluation of alternatives. 


You have to evaluate what are the possible alternative policies we could’ve come up with. You have to evaluate and project the number of small businesses, the number of small entities, that will be affected by the rulemaking. You have to quantify the impacts as much as possible. And if you have a final regulatory flexibility analysis as part of issuing a final rule, you are required to reply to comments that you receive about the Regulatory Flexibility analysis. And the agency has to figure out ways to minimize the economic impact on small entities as part of a rulemaking. 


Now, the Regulatory Flexibility Act does not limit itself to only cabinet agencies. It is a federal statute that applies to all federal rulemakings. Most agencies which are under 12866 simply point to their Executive Order 12866 cost-benefit analysis as the basis for fulfilling their Regulatory Flexibility Act analysis requirements. But FTC and other independent agencies actually have to do this analysis in the first instance, regardless of whether they’re doing it as part of a requirement under the statute for themselves. And so the FTC can point to the sections as Paul already outlined for us. They can point to their analysis done there. But under the Regulatory Flexibility Act, the FTC would be required to complete this sort of analysis. 


In addition, the Paperwork Reduction Act also requires an analysis of reporting obligations, of paperwork, of compliance activities, anything that is being done as part of a rulemaking in order to comply with it, whether that is particular forms that have to be filled out to submit to the government to demonstrate compliance, whether that is particular reporting requirements where you’re going to have to keep certain records or other compliance activities that may require a change in behavior. These are all analyses which have to be done under the Paperwork Reduction Act. And OIRA does in fact oversee that. OMB has an office that is overseeing that. And even agencies, again, that are not subject to Executive Order 12866 are required to go through the Paperwork Reduction Act requirements. 


Why does this matter? Well, ultimately this matters because if you are commenting on the Regulatory Flexibility Act analysis, the Paperwork Reduction Act analysis, then the agencies are going to be required to address those comments as part of final rulemaking. And so ultimately the Administrative Procedures Act in order to avoid the arbitrary and capriciousness finding by the courts would require even an independent agency like the FTC to go through and do some sort of cost-benefit analysis as long as there are comments that are coming in under Regulatory Flexibility or Paperwork Reduction Act, even if those agencies don’t have some independent requirement. 


And so to Paul’s point where even though the FTC statute explicitly says that the costs and benefits do not have to be addressed as part of the final rulemaking—they can’t be evaluated by the courts—the reality is that is cabin to the FTC Act. It is not ignoring the fact and it doesn’t preempt a Reg Flex analysis or a Paperwork Reduction Act analysis. And so despite being allowed not to have to do all the cost-benefit analysis under the FTC Act, under that statute, these alternative paths still permit comments as part of the rulemaking to affect and require cost-benefit analysis to occur. 


So reasoned decision making can’t ignore cost-benefit analysis in the comments, meaning that independent agencies have to do the analysis to address the comments. And this really does tell us, just to Paul’s point a moment ago, how important it is to submit good comments to the agencies because if you submit good comments to the agencies not just about their legal ability to enact a particular regulation but about the actual effects of a particular regulation, those comments can actually force even independent agencies to undergo a detailed cost-benefit analysis that ought to be scrutinized by the courts if an organization is attempting to challenge the rulemaking in the end. 


And so I think this is an important reminder that while there are some unique features to the FTC rulemaking process that allow them to avoid some of the requirements that we might typically see agencies that are cabinet agencies have to go through, there are ways that attorneys and other interested parties can force these agencies to still fulfill very similar obligations to the cabinet agencies to perform a detailed cost-benefit analysis which will be scrutinized by the courts. 


Svetlana Gans:  Great. Thank you so much, Jonathan. I wanted to turn over to Dr. Stivers to give us a view on how cost-benefit analysis is done at the FTC and any lessons learned from that process. 


Dr. Andrew Stivers:  Thanks, Svetlana. So I’m bringing a perspective of an economist and a former regulatory economist at that. Before I joined the FTC, far before I joined the FTC, I worked as a regulatory economist for the Food and Drug Administration. And my job was to write cost-benefit analyses, including Reg Flex and PRA and all that stuff. So hopefully I can give some insight into how that process works as well. 


But focusing on the FTC, as those of you who are familiar with the Agency know, it is not known as a regulatory agency. So unlike the Department of Labor or the Food and Drug Administration or some of the other cabinet level departments, the FTC’s focus really has been on case by case enforcement. And there’s good reasons for that, which we don’t necessarily need to go into here. 


But we are seeing a shift in the FTC’s priorities in terms of switching from case-by-case enforcement to trying to establish more regulations and rules that would do, I think, two things. One, it would give them penalty authority in many cases, which they’ve been trying very hard to increase their reach on monetary remedies. But it also could provide some relatively bright line kinds of per se practices that would be violative. 


So in doing that one of the things that certainly Jonathan experienced over at Department of Labor and I did at the Food and Drug Administration is there is an infrastructure than an agency needs to have to move these rules forward. And it’s often quite extensive. The FTC typically did not have a fully fleshed out, at least during my tenure, regulatory infrastructure of that nature. The regulations were dealt with within the standard kind of law enforcement structure within the Bureau of Consumer Protection at least where the enforcement divisions, particularly the Division of Enforcement often, would be responsible for moving the regulations, typically regulatory reviews of existing rules, forward. 


The current administration, and even to a certain extent before the current administration with the shift toward regulatory efforts has started to beef up its infrastructure. They’ve appointed some people to the Office of General Counsel that are going to be specifically trying to shepherd that through, which, again, in sort of a mini version mimics the larger department’s infrastructure for rulemaking. But that’s all relatively new. 


One of the other things that’s important to understand is that typically although the Bureau of Economics -- the Commission has a Bureau of Economics -- the Bureau of Economics while I was there did not have a big role in doing any of the required analysis for rulemaking. We made comments on rules. But we did not do the PRA analysis. We didn’t do the Reg Flex analysis or the small business analysis. 


My understanding from kind of reading the tea leaves is that is going to change. I think that the Agency recognizes that it’s taking on more of a regulatory role. And to do so it needs to up its game. And so my understanding is it is going to have the Bureau of Economics doing at least some of the regulatory analysis that’s going to be required. 


Because of that and to echo both Paul and Jonathan’s points, data that is submitted will be taken seriously. The folks that work in these agencies take the rulemaking process quite seriously. And it can be a little bit stylized. Like here are a bunch of hoops that we have to jump through. But if you submit data, it will be scrutinized. It will be taken seriously. It will be analyzed. There will be a response. 


And again, the responses can vary. Sometimes it’s really just we don’t believe this. Off we go. Other times it really does trigger a more in depth analysis. 


I think one of the points that Paul made that the cost-benefit analysis doesn’t determine the rule is absolutely right. When I was at the FDA, typically what would happen for any kind of substantive analysis is that there would be a proposal put forward typically from the Enforcement Bureau, or sometimes the political level would say, hey, we want to move this initiative forward. There would be a proposal put forward. And often as seems to be the role of economists in government, they would come to the economists. We’d be like, well, here’s some problems. Here’s why we think maybe this isn’t the most cost effective way to do it. Here’s why we think the market mechanism is not going to work. And then there would be a dialog back and forth, sometimes a contentious one. 


But ultimately what would come out of the rulemaking process would be heavily influenced by the experts and the data within the Agency. So given the shift as I understand it to allow the Bureau of Economics to have more of a role in the analysis of these rules, and even without that, we really should expect an internal process if data is submitted, if data’s available, where the rule that comes out the other end will be influenced. 


Now, sometimes that simply means it’s going to be hardened against some of the criticisms put against it. But often it also means that it is a better rule, maybe not a great rule, but a better rule, less costly, more beneficial, higher net benefits, all that stuff. So that’s I think the main thing that I would kind of echo from the previous speakers is that the data matters. 


And the submissions particularly as they’re substantive matter -- simply having a comment that says we don’t like this rule or we think it’s bad, there’s thousands of those in some of the rulemakings. And they typically got batched up. Somebody formulates a response saying, well, no substantive data was submitted. And they move on. So having real data is crucial. 


The last point I’ll make is in terms of actually doing these analysis as I think Paul had suggested. When you start off, often you don’t have a lot of data. The kinds of questions that an agency is grappling with, especially an agency like the FTC which has gotten away from some of its research and advisory priorities over decades since it was founded, there’s not necessarily a repository of data that they can use. And the research that’s done in the academic field is sometimes related but a little bit tangential. The research that’s more likely to be applicable is going to be within industry often and not publicly available. 


So the Agency doesn’t necessarily have access to the data that really would inform the rulemaking. But as that rulemaking goes on, the economists, the analysts, they come up with creative ways to find data, to use data to create analyses that really do inform the rulemaking. Again, as Paul suggested, informing doesn’t mean determining. So just because an economist says, hey, this is going to be really costly doesn’t mean there won’t be a rule. And I’ll end it there. 


Svetlana Gans:  Great. Thank you so much, Andrew. I wanted to turn it over to Paul with respect to the motor vehicle dealers rule in particular. I know that your association weighed in. Andrew talked about kind of the theoretical inner workings of the Agency cost-benefit analysis. And I thought perhaps you could comment on how it is playing out in proposed rules and what some of the issues are that you’ve seen in recent proposals and kind of your takeaways from those proposals. 


Paul Metrey:  Certainly. I suppose I should start out just by applying a little bit of context here. A lot of the other rules that we’ll talk about are of general application. If we talk about things like the non-compete clause proposed rule and some of the others, it cuts across multiple industries. Of course, this is specific to motor vehicle dealers. So many of the listeners may not be plugged into it. 


But maybe just to put in context before we get to the cost-benefit analysis just a couple features of how this one has been launched might be helpful. So this was proposed last summer in June. It came out on the FTC’s website, and they proposed really a series of new duties and restrictions on motor vehicle dealers when they interact with consumers. It’s something that really is cut out of whole cloth, so unlike a lot of the rules we comment on when they’re amending an existing rule, this is something that they really were starting from scratch with. It did not have the benefit of years of experience of already having been in the marketplace and affecting regulated entities. 


Many of the issues involve things like what they call add-on products and the way that consumers and dealerships communicate with each other, whether orally or in writing. And it talks about a number of those. And there’s many peculiar things here and really many procedural shortcomings. And we think that that carried over to the cost-benefit analysis. 


Let me just tick off a few things. First, this was never listed in any semiannual regulatory agenda as required. It was stated that this was first considered after the last one was published. It never was put into that document. 


In addition, it was never preceded by anything, not an advanced notice of proposed rulemaking which we think was required by law, but also not a request for comment, not a request for information. For those of you that are familiar with the CFPB’s 1071 rulemaking under Dodd-Frank, not an outline of proposals for which they were trying to seek comment, there was nothing on the front end of the process to try to draw in information. In addition to that, there was no stakeholder input. We met with the FTC, certainly had nice discussions with them. Neither the rule nor any of its particulars were put before us for any type of comment or reaction. 


The proposed rule also includes 49 questions on all kinds of things, including a lot of data. That probably goes to much of what Andrew was talking about. That of course can be a healthy exercise, although we know that many proposed rules do not drift far between the proposed rule and the final rule. And much of the information that was sought should have informed these proposals. In fact if we were actually to go through it—we certainly don’t have time here, although we detail in our comments—I think one would see many of the questions asked are for basic information. And the answers for those certainly one would need to have a handle on to come up with a meaningful proposal in those areas. And unfortunately that did not happen. 


Also, not only was there not anything on the front end to kind of lay this out but we put in, as did the SVA Office of Advocacy, a request for extension of time in order to respond to many of those questions and other issues that had been put out there. And I think this hits on Paul’s second point that he made when he was talking about takeaways. We want to use that opportunity to provide very good cost information. We stated that there was a reputable firm that told us it would take four months to prepare a detailed cost study that could aid the Commission. 


So we asked for a 120 day extension of time. SVA Office of Advocacy did as did others. The extension that was granted, nothing, not even so much as 30 days. So we did not have the ability to try to inform that process by answering those questions. The 60 day comment period, which for this type of rule is exceedingly short, was not extended at all to try to accommodate that process. 


In addition to that, there were all kinds of Regulatory Flexibility Act and Paperwork Reduction Act flaws. I think some of the prior comments were very good. I know Jonathan touched on it in terms of things that are required from any agency, even when you step outside of the Executive Order. And there were just a number of things that were not done here. 


The FTC stated that there was no significant economic impact, but then they did not take steps to forgo an initial Regulatory Flexibility Act analysis. There was no 605(b) certification stating this, no factual basis statement, no certification and statement to SVA’s Office of Advocacy. And they did not conduct the RFA with the proposed rule, and they did not send a copy to SVA. 


And importantly, and this was flagged on earlier, nowhere did they consider alternatives. There’s not any part of this 128 page document that appeared on the website where alternatives were looked at in any meaningful sense. And by not doing that certainly it short circuited the process that we’re all talking about. And again, SVA brought this to light. 


Now, there was an attempt, at least a purported attempt, at a cost-benefit analysis under Section 22 of the FTC Act, although it is filled with untested assumptions and figures that quite frankly are pulled out of thin air. And it’s really not a serious attempt to do the type of things that all the speakers have been talking about are important here to really inform this process. 


Now, we detail this at length in our comments. And if anyone wants to go into them, please pay attention to appendices 18 to 20. This is laid out in great detail. 


However, one thing I think is indicative of the problems with this, with the cost-benefit analysis that hopefully will put this in perspective, so the Commission states that over a ten year period discounted by 7 percent this is going to save consumers $31 billion. And so you would say where did they get that? And you might say, well, in a program this short there’s no time to flesh all that out. Here we can flesh it all out because it’s unbelievably thin, and it really quite frankly is very hollow. 


So to come up with the 31 billion figure, what do they look at? They look at three variables. They say that the Commission assumes that this will save the average consumer three hours of time when they shop for and research vehicles and subsequently make a purchase. 


Now, I would tell you if we had time to get into the meat of this, I think one would say if anything this would cut the other direction. There’s all kinds of new disclosure obligations and forms that have to be signed. It’s very difficult to imagine how this saves any time. But nevertheless, the Commission says it assumes three hours will be saved. 


What is that based on? Nothing. What is it supported by? Nothing. Is there a footnote? Is there an explanation? Is there an analysis? Nothing whatsoever. They just simply assume it. They then say you can measure the value of someone’s time by looking at Bureau of Labor statistics. For nonwork time, that comes out to $22 an hour. 


And then they look at the number of transactions that occur between franchise and independent dealers and consumers. And that figure is quite frankly wildly overinflated to the tune of 46 and half percent. And really it’s even more than that if we had time to break it down. 


So there’s three simple figures: three hours of assumed time savings, the value of time based on BLS statistics, and an inflated number of transactions. They put it together. It comes to $4.1 billion per year. They extrapolate it out over 10 years. They apply a 7 percent discount rate. They say it's going to save $31 billion to consumers. 


And then what they also say in that context after they measure the costs, which are really incomplete and grossly understated -- if we had time, we could go through that as well. Again, it’s in our comments. But they say that the cost to dealers would simply be over 10 years 1.4 billion. So they take the 31 figure. They multiple the 1.4 figure. It comes to 29.7 billion in savings over 10 years, discounted. This made its way into its press release, and it made its way into news stories like consumer reports and many others. 


And the FTC says at the end of the notice of proposed rulemaking “Consequently, this preliminary regulatory analysis indicates that adoption of the proposed rule would result in benefits to the public that outweigh the costs.” And I would on that just simply share—I know we have to move on to questions—but an observation, we asked the American Action Forum to look at this. And they stated the following, and if can just read a quick two sentence quote to you. 


“Upon greater scrutiny of some key assumptions, the monetized cost-benefit balance is potentially closer to even, if not on the net cost side, than the roughly 30 to 1 benefits to costs ratio the Agency’s current calculations suggest. Having billions of dollars of potential impact hinge on a few flawed assumptions should give FTC pause and perhaps spur a reevaluation of its rulemaking’s overall direction.” So again, we think that that neatly summarizes a number of flaws in the cost-benefit analysis, not to mention other procedural flaws and we believe violations of a number of statutes. So many of these concepts apply to this rulemaking. We think that’s how it fits. Again, we flesh this out in much more detail in our comments. 


Svetlana Gans:  All right. Thank you so much, Paul. Paul Ray or Jonathan, Andrew, any feedback on kind of Paul’s points that he made? 


Dr. Andrew Stivers:  I guess the one point I would make is that the auto dealer rulemaking is likely to be one of the easier cost-benefit analyses to approach by the Agency given especially for example in relation to the proposed privacy rulemaking or the noncompete rulemaking. In the case of auto makers, you’ve got a relatively defined industry, a very well researched industry generally speaking and relatively speaking. So that should be a relatively easier bar to meet by the Agency. And I think Paul’s comments about the potential failings of the analysis that was put out is potentially telling about what we might expect in the future. 


Svetlana Gans:  Jonathan or Paul, any comments? 


Paul Ray:  I’ll say just that I’ll certainly glom onto what Andrew just said. The analysis in the non competes rule is quite extraordinary thin and not without reason. I’m sympathetic to the plight of the poor economists at the Commission who have to try to figure out the effect on wages across the entire economy extrapolating from just a handful of studies. But even though we might sympathize with their plight, that doesn’t mean it’s good cost-benefit analysis. So I hope that the Agency takes very seriously the comments that it receives in response to that rule and tries to come up with a more rigorous assessment. 


Jonathan Wolfson:  The only thing I’ll add is Paul mentioned how long the analysis their outside economist told them it would take. And I think this is just an important reminder that the sooner you can start your economic analysis if you want to do that as part of your comment the better. So if a FTC commissioner gives a speech and says, we’re considering rulemaking to do X, Y, Z and you know that that will have an effect on your industry, that that’s when you start the process of building your economic analysis instead of waiting until you see the proposal and build an analysis that’s catered just to the rulemaking, having something where you can evaluate or even have the pieces in place to have that analysis ready to go because it does take a long time. 


We’re in a group of mostly lawyers here I’m assuming on the call, and the majority of us even on this panel are lawyers. But the economic analysis piece is vital to winning a lot of these fights. And the sooner you get that piece started, the better. So I’m just going to point that out. 


Paul Metrey:  Yeah, and maybe --


Dr. Andrew Stivers:  Oh, I’m sorry, Paul. 


Paul Metrey:  I’m sorry. Yeah. Just to follow up, so we certainly -- our economist keeps very robust data. A real challenge here, though, was just the nature of these proposals. As I indicated there was nothing on the front end. So a lot of times you’re getting some sense of -- even in that 1071 rulemaking that the CFPB is doing they had preceded it with a number of questions. They had an RFI that went out. There were a number of other things where one would say, okay, I think they’re going this direction or that direction. 


Also keep in mind 1071 of course was implementing a statutory provision, so they had the framework of a statute to at least base that on. In this context, nothing was directed by Congress, and they had not proceeded by indicating they were going to do anything. So trying to pre-position data on this one was a particular challenge, even though certainly take the point in general that that’s always a helpful thing to do. 


Dr. Andrew Stivers:  I was just going to point out we did an analysis for a proposed rulemaking last year, and it took us about six months start to finish. And luckily the group that wanted us to perform the work did approach us early. So we weren’t under a time crunch, but that work does take some time, especially if you want to kind of choose the good, cheap whatever the three pick two kind of rubric there is. 


Svetlana Gans:  So we do have a question from the audience, and I’ll just read it. It was mentioned by Paul Ray that the cost-benefit analysis does not necessarily decide which proposal prevails. What are the other major normative criteria used in addition to the CBA to evaluate different proposals? 


Paul Ray: Yeah. Great question. So the Agency is supposed to achieve the objectives that Congress had when it enacted the statute. The Agency only exists to execute the will of Congress with a couple of exceptions for what you might call core presidential agencies, the Department of Defense, the Department of State, necessary to execute the president’s constitutional duties. But otherwise agencies exist to carry out the direction of Congress. So the touchstone for what the agencies should choose and the touchstone to which the courts will look is the objectives of the statute. 


Now, very often the statutes have multiple objectives. Indeed, one of the great justifications advanced by the courts for the Chevron doctrine with which many of the viewers will be familiar is precisely that agencies are better than courts at balancing among the multiple priorities multiple ends that any major statute is going to have in view. So the agencies have a lot of discretion in deciding which of these ends, which of these touchstones to maximize. Cost-benefit analysis can indicate how effective the agency would be with a particular proposal at achieving, say, objective X or objective Y. It can’t tell the agency which of the objectives to pursue. 


Now, cost-benefit analysis can cabin the acceptable outcomes for the agency in the following way. If cost-benefit analysis indicates that the agency’s proposal would be worse than obvious alternative proposals at achieving any of the objectives that Congress had in mind, well, then in that way cost-benefit analysis could place limits on agency discretion. But just considered by itself without relation to the ends that Congress has in mind, cost-benefit analysis can’t cabin agency discretion. In other words, cost-benefit analysis is not a kind of algorithm in which you input certain data and it pops out a rule in which you must proceed with. It can inform agency discretion, but almost always there’s a vast deal of discretion left to the agency in any significant rulemaking. 


Jonathan Wolfson:  I just would add to that -- that’s exactly my view as well. Economics often in the context of regulatory efforts either Congress has said do X, and the question is, well, how do we do X? And part of the cost-benefit analysis is what’s the most cost effective way to do X? But it can’t really inform should we do X in the first place. In sort of the broad question like should we have child labor laws, well, that’s not a question economics is going to answer. There might be some value in trying to understand, well, where’s the cutoff? 18, 21? You know, where do we allow labor to occur? 


Within the context of deciding these questions, there’s also a huge amount of detail. When Congress directs an agency to write a rule, typically my experience is Congress say very vaguely do this thing that we think is very important. So one of my experiences was with calorie labeling in restaurants. 


So Congress said write a rule saying that restaurants have to put calories on their menus. What the heck is a restaurant? And we spent a huge amount of time informed by the economics of trying to figure out what is the set of firms that we really should include here to maximize the potential benefit to the consumer but kind of minimize cost that just aren’t going to have any effect. So those kinds of decisions really can be informed by the economics. I think that’s especially important here. 


As Paul suggested, there’s not that directive from Congress. Both Pauls suggested -- there’s not necessarily that directive from Congress. It’s sort of a free form, hey, we’re going to do X. What X is and how it’s made up is really entirely up to the agency. 


Svetlana Gans:  So maybe, Andrew, I’ll ask you. How should an agency go about a cost-benefit analysis when they’re dealing with a nationwide or economy-wide rule? Are there certain things that must go into it in your experience?


Dr. Andrew Stivers:  There’s not a hard and fast set of things that go into it. There’s a little bit of, well, what data do we have? And you start there, and then you kind of build out from that point. The economists in particular are really trying hard to inform the process, and they’re going to be trying to bring forward whatever evidence they can find that might help inform that process. 


In terms of a nationwide analysis of non-competes, there’s all sorts of landscape questions that I think the FTC has tried to do a little bit of so far. But how extensive are these things really? And we have a bunch of media reports. We have some advocacy work that’s been done in the area. I’m not sure there’s really a nationally representative survey of how extensive these things are. And this is one of the points Paul might bring up. 


If you’re bringing a rule to OIRA and you’re trying to have some sort of survey estimate of how prevalent is this practice, OIRA’s going to hold your feet to the fire in terms of the statistical validity of that survey instrument. And it can be quite frustrating as a practitioner trying to actually get the data out to deal with a desk officer at OIRA who’s like, no, here’s what are standards are. And you’re like but nobody answers the phone anymore, so we can’t do a phone survey. 


So having that kind of baseline information is really important. But there isn’t a hard and fast rule. And I think the FTC’s prior practice has sort of illustrated that. They haven’t been under a lot of scrutiny, even with respect to PRA and Reg Flex analysis. My understanding is the OIRA desk officer was like yeah, it’s fine. Like they never really gave them much hassle. 


That and some of the other agencies that do rulemaking that aren’t getting a lot of national press -- again, the quality of the analysis, the CBA that’s produced, varies radically across agencies. Department of Labor, FDA, Department of Defense do some of the best ones frankly in my view. And then it kind of goes downhill from there. I won’t name any names, but yeah, there’s a wide range. And all of them are -- they’re all cost-benefit analysis. There’s no real requirements. 


Svetlana Gans:  Jonathan or Paul, anything from your end on kind of the requirements or the elements that should go into a good cost-benefit analysis? 


Jonathan Wolfson:  So I think one of the things that you want to do if you are in the agency and you are evaluating is you want to look at what are sort of the key people who are going to be the players when this rule goes into place and then try to think about what are the real world effects that are going to take place, not just kind of what do we want to happen but what are kind of the unintended consequences. And part of that is having enough information about the groups of people who you’re talking to. 


So when Andrew was talking about at FDA doing the calorie regulations, knowing what is the work that’s going to be required at a certain type of restaurant -- right, if a food truck counts as a restaurant, how much work is it going to be for a food truck to put calorie counts on when they’re making kind of everything from scratch in the morning versus a chain restaurant that kind of has a very strict formula for what those calories are going to be? So trying to figure out who are the people who we’re actually going to regulate and what are the real world lives and livelihoods of those people going to look like I think is a really important factor and something that sometimes gets lost in kind of the government centric world where it’s kind of like we just regulate and that’s what our agency does. And we don’t actually have to think about who are the actual people who are going to live under the rulemaking that we put into place. 


Svetlana Gans:  Great. So we’re just at almost time, so I just want to give each of the panelists 30 seconds with any concluding thoughts about FTC rulemaking or tips for practitioners who are working on comments at these various agencies. Maybe just do like 30 seconds per person, and Paul Ray, I’ll start with you.


Paul Ray:  I won’t even use all my time. Subject the FTC to OIRA review. It would help a lot. It would result in much better cost-benefit analysis. 


Svetlana Gans:  Jonathan? 


Jonathan Wolfson:  So there’s two things I would recommend to practitioners. It’s to be specific in your comments. Don’t give the comment of we think this is a bad rule but be specific. Explain why you think it’s bad. It’s going to cause costs for us to rise. It’s going to cause the product quality to go down. 


And then challenge the agency’s assumptions in the rulemaking proposals. Be willing to say you’re making this assumption based on faulty data. It’s based on a misunderstanding or a misapprehension of how the world actually is working. And so those are two really important things practitioners should do in FTC rulemakings but in all rulemakings. 


Svetlana Gans:  Great. Andrew?


Dr. Andrew Stivers:  Just submit as good a data as you can come up with. The Bureau of Economics will tear into your data if they think it’s bad. So submit the best you’ve got. 


Svetlana Gans:  All right. And Paul?


Paul Metrey:  In addition to all the legal requirements that any agency has to meet in this arena, hopefully the agencies appreciate the policy reasons behind them too. A cost-benefit analysis should help an agency put something in the marketplace that’s going to carry out its intended effect. And to the extent it’s short circuited or some of these due diligence steps that have been talked about today are missed, aside from the legal issues, it’s just more than likely to produce something that is not going to be effective in the marketplace. And that should certainly be revisited. 


Svetlana Gans:  Great. Well, please join me in thanking all the panelists for a wonderful discussion this afternoon, and I’ll turn it over to FedSoc to take us out. 


Chayila Kleist:  Absolutely. On behalf of both myself and the Regulatory Transparency Project I want to thank all our experts for sharing their time and expertise today. And I want to thank our audience for tuning in and participating. We always welcome listener feedback at [email protected]. And if you’re interested in more from us at RTP, you can continue to follow us at regproject.org as well as find us on most major social media platforms. Again, thank you all for being with us today. Until next time, we’re adjourned.


Conclusion: On behalf of The Federalist Society’s Regulatory Transparency Project, thanks for tuning in to the Fourth Branch Podcast. To catch every new episode when it’s released, you can subscribe on Apple podcasts, Google Play, and Spreaker. For the latest from RTP, please visit our website at regproject.org. That’s R-E-G-project.org.