Notice & Comment

The Real World is Messy, by Cynthia Giles

*This is the seventh and final post on a symposium on Cynthia Giles’ “Next Generation Compliance: Environmental Regulation for the Modern Era.” For other posts in the series, click here.

Anyone involved in the debates about environmental policy knows how uncommon it is for environmental lawyers, economists, and practitioners to agree. So, I am thrilled that the foundational idea of Next Generation Compliance: Environmental Regulation for the Modern Era (Oxford University Press) resonates with these reviewers, who are both implementers and academics, and all thoughtful, experienced, and informed participants in environmental regulatory design.

The thesis of the book is that there is a huge gap between the public health and environmental protection objectives of most environmental regulations and their performance in actual life, and that more reality-grounded rule design is the only way to bridge that divide. We have long been laboring under the unfounded beliefs that compliance with environmental rules is good and it is up to enforcement to make sure it is. Next Gen lays out the overwhelming evidence that those beliefs are wrong. It includes the most complete accounting anywhere of the rampant noncompliance in most environmental programs. We will not protect public health, reduce climate threats, or address environmental injustice, without rules that are designed to perform much better in the complicated and messy on-the-ground world were implementation happens, or doesn’t.

A theme in several reviews – particularly from economists and the experimentally minded – is the importance of rigorous experimentation for determining which kinds of Next Gen strategies are most effective. They point out that statistically valid sampling and robust collection of performance data for retrospective analysis are important for a reliable evaluation of regulatory strategies. While the concept behind those ideas is compelling, it is not surprising that these approaches haven’t been widely adopted. Explaining why, and what we might do about it, is an opportunity to show how Next Gen tackles problems a different way.  

If anything is obvious from the extensive evidence and case studies in the book, it is that companies don’t reliably do things just because it’s a good idea or a regulation directs them to do so. Neither do governments. Repeated calls for more statistical sampling to determine effectiveness of government programs, and pressure for greater embrace of retrospective analysis, have not led to either of those ideas being generally accepted.

The reason: there are large disincentives for governments to engage in statistical sampling and retrospective review. Statistical sampling to determine compliance requires investigating a randomly selected subset of facilities. Of course it does; that’s why it tells us something about the larger population. But in so doing, it requires that scarce resources be used looking into facilities that we don’t suspect of violations, instead of investigations where we have reason to be worried. Valid sampling is for a noble purpose, but the bigger the size of sample required, the higher the opportunity cost in lost productive inspections will be – a major deterrent for already grossly under-resourced regulators. The same unaffordable opportunity costs plague retrospective reviews.

Both statistical sampling and retrospective reviews have an additional barrier: robust program evaluation requires a willingness to discover that the government-designed program isn’t effective. Economists praise that outcome as providing essential feedback, and they are right. But in the real world lots of people and institutions don’t race to jettison established ideologies and find out that they have been wrong.

In short, there is a lot of pressure against statistical sampling and retrospective reviews, and potential ambivalence about critically examining government’s own performance. In these circumstances, it is predictable, if not inevitable, that statistically valid studies and robust look backs will be rare. That’s the same reason that compliance is poor in so many programs. Sure, companies are supposed to comply. But there are also plenty of reasons that compliance runs off the rails, including incompetence, higher priorities, lack of resources, unexpected events, and resistance to change. Governments are subject to the same pressures and mess that occur in every aspect of life.

The insight of Next Gen is that we need to address these problems in a different way. Rather than committing to particular methods, especially methods with powerful negatives, we should look at alternatives that go around the barriers. Here’s one Next Gen idea, discussed in the book, that could drive change. Rather than binding ourselves to statistical sampling as our primary knowledge tool, how about rules that require data about nearly all regulated sources in near real time? We won’t need to sample because we have near-census data. That broadscale information has tremendous compliance power because it means facilities will themselves know what their performance is, the first step toward taking action to improve, government will have comprehensive data to inform its understanding about what works, and the public will have an inside look at the outcomes, and facilities, that affect them. All these forces together – real time measurement, regular reporting, greater transparency – will do two things: push toward better compliance results and provide information that can be used for more sophisticated analysis about effectiveness. More complete and real time data make it easier to try new ideas because there is feedback on implementation without unacceptable measurement expense.

This compliance-driven strategy also helps to solve many of the challenges that deter government from program evaluation and retrospective reviews. Instead of the time and money required to collect data after the fact for a retrospective analysis, we can have the data already in hand, as Joe Aldy points out. Through this method program evaluation becomes relatively easy to do, both by government and outside sources, so it will happen more. And it turns the tables on the conflicting motivations. Once compliance failures become apparent to everyone because there is robust public data, guess what? Experiments to figure out what works to produce better outcomes will suddenly be attractive, because bad performance revealed in publicly available data will produce demands for action. Where there used to be resistance and reluctance, we can have government programs actively seeking out experimental data and program effectiveness information.

This example is illustrative of the Next Gen approach. Don’t insist on pushing the rock uphill and waste time complaining about lack of progress. Design the system so gravity works for you. Make the desired action a natural outgrowth of the program structure. Work with, not against, the natural inclinations of companies and governments. Don’t think that bludgeoning or pleading with people will lead to the desired result across the board – the evidence is overwhelming that doesn’t work. Instead, set up the design so their own motivations push them toward the desired actions. Build compliance in.

It isn’t just environmental rules either. The same faulty assumptions, and the same compliance disasters, plague many other regulatory programs. Health care, tax law, airline safety – you see implementation fails across the board, often with devastating results, as many reviewers point out.

Enforcement is essential. You can’t have an effective regulatory program without it. But enforcement alone will never fill the giant compliance gap created by a poorly designed rule.

The book is about how we can do a lot better. Everyone who cares about environmental results, or effectiveness in any kind of regulatory program, can find ideas to consider in Next Generation Compliance. Many thanks to the reviewers for engaging so thoughtfully with the ideas in the book. And even more so for their impressive body of professional work that insists on including the messiness of the real world in the policy debate.

The book is available for free download at

Cynthia Giles is the author of Next Generation Compliance: Environmental Regulation for the Modern Era and former senate-confirmed Assistant Administrator at the US Environmental Protection Agency’s Office of Enforcement and Compliance Assurance.

Print Friendly, PDF & Email