Since Donald Trump’s days on the campaign trail, the Republican president has called for a radical rollback of government regulations that he sees as overly broad and impeding business growth. Indeed, the number of regulations has grown exponentially in the last few decades, often creating a quagmire of confusion for those charged with drafting and implementing these rules. But it is possible to achieve what University of Pennsylvania law and political science professor Cary Coglianese calls regulatory excellence. Coglianese, who heads the Penn Program on Regulation, spoke recently to staffers on Capitol Hill as part of a Penn Wharton Public Policy Initiative series about how to build a framework for sensible, flexible regulation no matter which party is in the White House. He appeared on the Knowledge at Wharton show that airs on SiriusXM channel 111 to talk more about the topic.
An edited transcript of the conversation follows.
Knowledge at Wharton: Tell us about the presentation you made on Capitol Hill and the goal behind it.
Cary Coglianese: One thing I wanted to do when I went as part of the Penn Wharton Public Policy Initiative series of briefings on Capitol Hill was simply give staffers a better idea of what’s happening with regulation and how it works. The world of regulation is one of these realms of policymaking that’s absolutely critical but largely misunderstood and hidden from public view. High school students learn about how a bill becomes a law, but they don’t learn about what happens at all the hundreds of different federal regulatory agencies and how they operate. Part of my purpose was to give some insight into how this relatively obscure process works, even for people who are staffers on the Hill.
Knowledge at Wharton: It’s surprising to think that some of the people involved in crafting these regulations may not fully understand what’s going on behind it.
Coglianese: That’s one of the reasons why the number of pages in the code of federal regulations has increased over the years. It’s because Congress passes laws that call for agencies to develop regulations. Trump in December had a press conference in which he contrasted a huge pile of papers that was taller than himself that was supposed to represent the number of pages in the code of federal regulations today. About six feet away on the ground was another, smaller pile of papers that represented regulations in 1960. There was this red ribbon between them, from the low level of 1960 to this massive pile from today. He cuts this red ribbon and says, “We’re going to get back to the level we had in 1960.” Well, you can’t. One of the reasons that we have so many more pages in the Code of Federal Regulations is because Congress has passed an enormous number of laws since 1960.
Knowledge at Wharton: Isn’t that to be expected as new businesses and technologies develop?
Coglianese: That’s right. One of the things I emphasized on Capitol Hill is that if you look at the rate of growth in regulations and compare that with the rate of growth in the economy, the economy has grown much more than the rate of regulations. Look at other indicators of the more complex economy that we have. For example, the number of patent applications has skyrocketed over the years, the number of patents that have been issued, the number of airline passengers. All of these indicators of the complexity of our economy have increased dramatically, so it’s not surprising that regulations have also increased somewhat, although not at anything close to the rate of these other indicators.
Knowledge at Wharton: You had an interesting paragraph in the presentation that looked at the growth of regulation in terms of growth in other sectors. Can you talk about that?
Coglianese: The lowest level of growth has been in the pages of regulation, so we should not overstate the degree to which regulation is smothering the economy or holding back innovation. It’s true that we have more regulations today than we did in 1960. Indisputable. But we also have a lot of new technologies and, quite frankly, new problems that we weren’t even aware of. In 1960, people weren’t really focused on the dangers of tobacco smoke or environmental contamination. Civil Rights was just emerging. Consumer protection and the like, those were all nascent areas. We have nascent areas today that we need to be thinking about and watching — not to clamp down unduly, but to be able to make smart regulatory decisions when needed.
Knowledge at Wharton: Playing off of the title of your book, Achieving Regulatory Excellence, what do you believe is regulatory excellence in this day and age?
“The world of regulation is one of these realms of policymaking that’s absolutely critical but largely misunderstood and hidden from public view.”
Coglianese: One of the motivations behind the book was to begin to transfer into the world of regulation a lot of thinking that has passed through the private sector over the last several decades. We understand and have thought about quality in manufacturing in the private sector, for example. Remarkably little attention has been paid to quality or excellence in the regulatory realm. Now, some people will say, “Regulatory excellence? That must be an oxymoron.” But we wanted to take that notion seriously and try to define it a bit. It is hard to define, in part because what counts as excellent regulation will vary from to realm to realm.
Regulation is a way of solving problems. To define excellence, you first have to be thinking about concrete problems that [need] to be solved. During the Reagan administration, when the Environmental Protection Agency took a look at what was happening from the pollution from cars that were burning lead fuel, the administration moved forward and said, “We’ve got to eliminate lead from gasoline.” They did so in a time when an administration was committed to what most people thought of as deregulation.
Knowledge at Wharton: What about the conflict between public good and the influence companies have on Capitol Hill?
Coglianese: There’s always a trade-off that you have to think about with regulation. Think about what the public benefits are, but also think about what the private costs are. You want to make sure that regulation is moving in a direction in which we’re getting benefits that justify, even outweigh, those costs. Everybody recognizes that there’s some need for regulation. There are failures in the marketplace. There are what economists call externalities and spillovers…. There’s also information asymmetries where consumers just don’t have as much information about the product, so we justify food and drug regulation on those grounds as well. Then there’s real concern about monopoly power. The concentration of market power is another justification for regulation.
Knowledge at Wharton: In your report, you focused on regulation in both the Bush and Obama administrations, correct?
“There’s always a trade-off that you have to think about with regulation.”
Coglianese: That’s right. Sometimes people wonder, what are we getting for all the regulation that we’ve seen? If you look at the estimates that have come out of the White House, both in Republican and Democratic administrations, we can see that there’s certainly a lot of costs associated to regulation. But there’s also a demonstrated, high level of benefits as well. We have this process in place since the Reagan administration to have new regulations reviewed by economists at the agencies and at the White House. The evidence is quite compelling that these significant regulations are delivering positive net benefits to society.
Knowledge at Wharton: But President Trump has talked about getting rid of two regulations for every one that he wants to bring forward. Where are we in terms of the scope of the growth of regulations?
Coglianese: I think a lot of the new innovations are probably the least regulated. If you think about self-driving cars, that’s not been very carefully or stringently regulated or overly regulated by the federal government. You think about social media, Facebook — we’re just at the early stages of thinking about what regulation might mean in that context. Machine learning, artificial intelligence, precision medicine. There are a lot of technological advances that are incredibly promising, and we certainly don’t want regulation to impede innovation that’s positive for society, but we do need regulators to kind of up their game.
One of the messages I also gave to folks on Capitol Hill is that if we want to have the positive benefits of any increasingly optimizing, smart economy that’s based upon a lot of new innovations, but we also want to avoid the real problems that these innovations are likely to have associated with them, then what we need to do is up the game in Washington and increase the human capacity and the resources that our regulators have, not necessarily to do more regulation but do to smarter regulation.
We also need good information in order to make sound, smart regulatory decisions. If we turn our backs on this and don’t study what’s happening with new innovations, then we may overlook and have some real public harms develop.
Knowledge at Wharton: You look at regulatory excellence through three main areas: procedure, management and technology. Can you explain?
“At the end of the day, I think regulatory excellence is really about people excellence.”
Coglianese: Historically, the way we’ve thought about achieving regulatory excellence in the United States has largely been procedural in that we’ve tried to create an administrative procedure act and procedures for White House review of new regulations. We’ve proceduralized what we hope will generate good regulation, and to some extent those [procedures] are really valuable. We’re probably ahead of the world in terms of our transparency in the regulatory process. But that’s not enough. We also need good management. In this regulatory excellence project, we’ve done a tremendous amount of focus in the last couple of years on what counts as good management practices in the regulatory process.
The third area is on technological innovations that regulators themselves should be thinking about using. There are advances in new sensing technology. There are new developments in machine learning that can help regulators forecast problems more precisely. It’s this last realm that I think we’ve given the least amount of attention to. I and David Lehr, a former Penn student, wrote an article called “Regulating by Robot” in which we explore the possibility that regulators might do better addressing some problems if they can use the tools of artificial intelligence to support their decision-making.
Knowledge at Wharton: Would that take out some of the personal or emotional contact in the decision-making?
Coglianese: Yes, and other forms of bias. We know a lot about the cognitive errors that people are making, and if we can design computerized tools to make better forecasts, we could really do a much better job. In fact, some research that we’ve been sponsoring here at the Penn Program on Regulation has looked at how the Occupational Safety and Health Administration (OSHA) could improve its targeting of unsafe work places through a machine-learning analysis. We’ve shown that they could target about 30% better than they’re currently doing. They can’t get to every workplace to inspect, so you want to pick and choose appropriately, and this is just one of many examples of where regulators might benefit from some of the new technologies.
Knowledge at Wharton: It has to be a constant updating process. You can’t just put a regulation into play and leave it for 30 or 40 years because of how technology develops.
Coglianese: A lot of people will look at how many regulations are coming out of Washington and not realize that a good number of those regulations are in the category that I call housekeeping or maintenance. Here’s a surprising statistic in light of all that the Trump administration has said about its regulatory rollback: For the first year of the Trump administration, 3,222 new final rules were published in the Federal Register, which is the daily publication where all new regulations appear. That’s about the same level as we would find in the Obama administration. Now, are all these 3,000 rules highly consequential? Are they path-breaking? No. A lot of them are just what regulatory agencies need to do day in and day out to maintain that stock of regulations and make sure that it’s working well.
Knowledge at Wharton: I wanted to take you back to the management piece. How important is that part of the entire process?
Coglianese: I think it’s central. At the end of the day, I think regulatory excellence is really about people excellence. We think about regulation as this technocratic, maybe obscure thing. When people hear the word regulation, probably their eyes glaze over and it just sounds pretty boring and dull. Fundamentally, regulation is about people, the regulators trying to shape the behavior of other people. Why? To serve still yet other people: consumers, the broader public, for example. You can’t really have excellent regulation if you overlook the people and the management of regulatory organizations.
One of the things we’ve developed is a framework for regulatory organizations and their leaders to use to assess how well they’re doing on three critical attributes of regulatory excellence that we identified from an extensive amount of research. Those three attributes fall into the categories of integrity, so a regulator not only needs to lack corruption but be even-handed and willing to play fair. A regulator needs to be competent and needs to be able to put in place rules that are sensible. Everybody gets that.
The third is, they need to be empathic. They need to engage with the public and listen attentively. It doesn’t mean that they have to do everything that they’re told by the public or by industry, but they do have to listen attentively with an aim of learning how to create smarter regulation.
Knowledge at Wharton: Do you think these three attributes exist in Washington right now?
Coglianese: I think it’s something that you do find happening in some agencies at some times. But it ebbs and flows. There are very serious questions right now about, for example, the integrity of the top leadership at the EPA, with all sorts of allegations and investigations that are taking place right now by Congress and so forth. However hard somebody’s trying to work at improving at a competency level, if you don’t have integrity and if you’re perceived as illegitimate, it’s going to make it very hard to get your job done. I think it is making it hard for Scott Pruitt, head of the EPA, to get his job done, and it may well drag down the administration some and be a contentious issue in the next election. If Republicans lose in the House and maybe the Senate in the fall, that’s going to make it much harder for this administration to achieve its objectives on the regulatory front because these administrators are going to be hauled in front of Congress for hearings, and it’s going to slow things down.
Knowledge at Wharton: You have to be diligent in the process, but don’t people have an expectation that the process could be more efficient?
Coglianese: The process is not quick. For many years, folks on the left criticized the process as being too ossified and slowing things down. Of course, now I think the folks on the left are happy that the process is ossified. When Trump claims to have really rolled back regulations, he hasn’t yet. The administration has only started the process. They haven’t finalized it, so it does take some time.
Knowledge at Wharton: Is it even possible for the administration to pull back as many regulations as the President has vowed?
“One thing that we’re seeing increasingly is a recognition that regulation needs to be more flexible.”
Coglianese: It’s not a possibility whenever the underlying statutes that Congress have passed have called for and even required regulations. The Volcker Rule, for example, didn’t just get dreamed up by folks in the Treasury Department. Congress said in the Dodd-Frank Act, “You shall create this rule,” and it gave a deadline. This is a real barrier to any kind of regulatory rollback. But even when agencies do have some discretion and can change their minds from administration to administration, they have to go through a process of proposing a rule, taking public comment, justifying the final rule and making sure it will meet the satisfaction of the courts. In this regard, some of the early efforts by the Trump administration have met with remands and repeals from the courts, which have said, “You haven’t really done your homework yet. You have to do your homework to roll back rules.”
Knowledge at Wharton: What do you think is the future on regulation? With all these new technologies coming forward, will there be a different or increased level of regulation?
Coglianese: For each of those problems, what form regulation takes will need to be different. It’s got to be customized to fit the problems that are at hand and the technologies and the innovation. One thing that we’re seeing increasingly is a recognition that regulation needs to be more flexible. It needs to try to leverage the private sector’s awareness of problems and try to pursue what I’ve called management-based regulation, where the regulators are not issuing edicts on high about exactly what to do or how to run a business. Rather, they are telling private-sector managers, go and analyze the hazards created by your operations and develop your own plans for addressing those problems so there are requirements. The requirement is to develop a management plan or a management system that deals with the problem. And for new problems, for problems that are highly variable or changing rapidly because technology is changing, that’s a sensible approach. That is essentially how the U.S. Department of Transportation has so far approached autonomous vehicles. They haven’t issued this so much in firm regulations, but they have a guidance that calls for this kind of management planning by the automakers.