Environmental Enforcement in Market-based Economies
Eric V. Schaeffer, Director, Environmental Integrity Project

John Mutter: Thank you very much, Professor Pachauri. Thank you all, the first three speakers of the panel for the afternoon. We can now take a brief break. Let me tell you how the rest of the afternoon will go. There will be two more speakers in this panel who will the convene for a discussion moderated by Al Fishlow, after which we have the pleasure of being able to listen to Professor Peter Singer, Professor of Bioethics at Princeton University, who will give a presentation to end the day today. You will want to listen to Peter Singer, so please come back.

Take your seats for the final session of the afternoon, the final session to include the two remaining speakers from this panel, moderated by Al Fishlow, followed by a discussion among the five panel members, followed by a special keynote presentation to end the day by Professor Peter Singer. Thank you. Al?

Albert Fishlow: We resume this afternoon with Eric Schaeffer who is Director of the Environment Integrity Project which is a non-profit public interest group dedicated to improving enforcement of US environmental laws. And EIP has helped to focus public attention on the problems of violations at refineries, power plants, municipal waste water treatment plants, and other facilities. And Schaeffer had previously served as Director of the Office of Regulatory Enforcement of the US Environmental Protection Agency from 1997 to 2002. He's a recipient of the Presidential Rank Award, recognizing his outstanding achievement as a member of the senior executive service. It's a great pleasure to be able to introduce Eric Schaeffer who will be speaking about Environment Enforcement in Market-Based Economies.

Eric Schaeffer

Thank you. I think I'm one of the few lawyers on the panels that you're hearing from, for which you're probably grateful. I guess one question I've been asking myself is what am I doing here in a sea of economists and scientists? I am here basically to bring an enforcement perspective some of the issues you'll be talking about. Tomorrow I think you have a panel on market-based approaches to environmental protection. There's no doubt that those are very powerful movements in this country, in other countries, and also in international institutions like the World Bank which are very influential.

Maybe starting briefly with definition and there are lots of variations on the theme, but a typical market trading program will have a couple of elements. There will be a ceiling or a budget for pollution from a particular sector, let's say sulfur dioxide from power plants. Within that ceiling and inside that sector the right to emit so many tons of pollution of that pollutant, sulfur dioxide in this case from power plants, is divvied up. If you pollute less than the number of allowances you hold you get to sell the emission credits to somebody else who's fallen short. So you can see the appeal of a system like that, it has a logical elegance, and I think it's based on an assumption that it creates less work for government, and I'm going to challenge that, and that it's a lot more efficient for industry, and I probably won't take that issue on.

Another question I'm not going to get to is the values that go into setting up a market-based system. There is a very hot debate over whether that kind of approach is appropriate for let's say mercury, which is a neurotoxin that falls fairly close to the polluting source, should that source be able to buy the right to continue polluting from somebody else far away? I don't think so, but I'm not going to get into that debate, I'm instead going to focus on practical issues of enforcement.

Now one of the problems with staying in enforcement for as long as I have is your instinct for suspicion gets overdeveloped. I see my son's homework differently, I see everything differently after having been in enforcement so long. And one story to tell you how that happens, we had some cases against pretty much the entire diesel truck engine manufacturing industry in this country, and let me tell you what that was about. The diesel truck makers with their engines had to install catalytic converters, and they had to meet an emissions standard. How do we determine compliance? Sample engines were taken to a lab in Ann Arbor, placed on a dynamometer, and then run through a loop which was meant to simulate a mix of city and highway driving, so much at 50 miles an hour, so much at 65, braking, low speed, etcetera. And the loop was a standard loop. What did the engine manufacturers do? Well, we have computers on board engines now. They taught the computers to understand when the test was running. When they saw the cycle repeat, they understood we're in Ann Arbor. The catalytic converters always worked when they were in Ann Arbor because they knew they were being tested. Once they understood that loop wasn't being repeated, the catalytic converters were turned off, especially in highway conditions to boost power. Now we confronted the industry with those facts, and their response essentially was “Was that wrong? We thought we just had to pass the test, you know, we didn't know that we had to comply with the emissions standards.” I think they were somewhat sincere about that. It didn't keep us from taking them for about eighty million dollars in penalties in some of our biggest cases. Thank you, that one felt good, a lot of nitrogen oxide emissions at stake there as well. But I guess it does raise for me the question of what happens when that kind of thinking meets market-based environmental standards? You have to kind of worry about that.

A couple of concerns I'm gong to raise, basically two points I want to make. One is the idea that market-based approaches mean less work for government is by and large false. Often it can mean a lot more work. If you think about it you're not only tracking emission standards at individual plants, you're trying to follow transactions. Who sold what to who? Does the person own the allowances? Was the purchase complete, and so on. Second, and equally important, maybe even more important, you absolutely have to have accurate emissions monitoring and very reliable data to make, I would argue, any market system work, whether it's economic or environmental. We are a long way short of meeting that standard today. And we just have to address those problems if these approaches are going to get anywhere.

What I'd like to do is first is to give you a couple of excerpts from a World Bank paper, this is probably enforcement speak, but I kind of consider admission against interest. This is a study of market-based programs in actually Latin America and the Caribbean finding, I think not surprisingly, that these programs actually can cost more than traditional command and control. That doesn't have to be a bad thing, it just means if you don't design for that, if you don't have the resources for it, then you ought not to get into the business, because you will be taken, you will be taken. And in fact they found in this study that the data were so poor and the monitoring was so poor that many of the permitting programs on which these market-based approaches were built essentially didn't exist, there was no there there. So I give the World Bank credit, and this is not an official paper, but this is a brave study by people who favor market-based approaches. Often you get some pretty ideological statements, I guess on both sides, but from proponents I think in particular. Here's a hard look at what's actually happened in practice and what some of the flaws are in design that really need to be fixed. Not surprisingly they found enforcement in many cases non-existent, which meant you weren't getting the benefits, and I think equally important law abiding companies were being shafted. You're basically paying for pollution controls that your competitors don't have to meet, and that can't be a good thing.

These problems are not limited to South America or the developing world. Obviously there are particular challenges there. We've got them here in the US despite our decades of experience, and in many, many academic conferences that have been devoted to market-based systems we are still struggling with some fundamental problems.

I'm going to start with one of our better known laws and also data systems, the Toxics Release Inventory, Right to Know Program, you may have heard it by that name. This was established after the Bhopal accident, requires companies to report a wide range of toxic chemicals as they are released to the environment, into the air, into the water, and also as they're generated in waste. As you [audio dropout] program has done an excellent job allowing you to search, you can search by company, you can search by county, you can compare one refinery to another, I'll give you one example there. If you want to know what the benzene emissions are and who the largest emitters are and what parts of the country those big emitters are located, you can find that out. So the ability to manipulate this data is pretty advanced. Unfortunately the data that you're manipulating is not always very good, which seems to be a kind of classic internet problem these days.

There really are no rules that determine the quality of the data that gets loaded into these reports. So, for example, a power plant near by house reports to this program that it released about 340 pounds of mercury in 2004. But if I dig into the State files I found out that they actually ran an emissions test and found that the number was 790 pounds. Now, I immediately thought well, you know, that can't be legal. And it sort of is, because their report of 340 pounds, less than half what they actually found when they tested their emissions, is based on an industry-wide average, and the rules are so unclear they apparently can select that lower number, it's obviously more advantageous to them, but apparently not illegal.

A better program is EPA's acid rain database. If you do any work with power plant emissions I urge you to get online and try it out. This is sort of the entry, the portal, to the system. It allows you to select data and manipulate it according to all kinds of different categories. Just to give you one example, you can pull annual emissions for specific units. If you've got the time and the inclination you can get hourly emissions data, you can get daily emissions, you can get it by unit, you can get it by state, there are lots of ways you can pull this information together. We actually use this information to monitor compliance. You can also go to the allowance page that you saw earlier and you can find out who's buying and selling pollution allowances, in other words who's done better than they should've and who's done worse, and that can be very useful as well.

To give you a contrast for power plants where the data systems aren't working well, or rather the monitoring systems are plainly inadequate, since 1977 there has been a limit on the amount of particulate matter that a power plant can release every hour, it's an hourly emissions limit. And you read it and you think boy, that's pretty stringent. How do they test that? Once every five years a facility will run a test. And they pass. It's really hard when you get to pick the time to take the test, if you think of passing a speeding test, you know, if the idea were basically once every five years you'd arrange you drive past the patrolman, you'd probably mostly make it. Incredibly, some people fail, incredibly. But obviously if you have an hourly limit that you're checking every five years, you don't have a program that you can build emissions trading on, you just don't have the information.

Enforcement information is available online. This is the ECO data system. Again, you can search by facility, you can go to state level, you can search by program, you can find out the compliance status of a company, and you can also find out what enforcement actions have been taken. Now, I was actually at the EPA when we were desperately trying to push this out the door, and this is one of the few things that I think the Bush administration has done in the right to know area that I would give them credit for. However, I have to concede that a lot of the data in the system is not terribly accurate, it's not very accurate. We took a calculated risk at EPA, we decided if the data doesn't get out the door, if it remains the province of the bureaucracy, it's never going to get any better, nobody will pay any attention to it, it will never be used. In fact, when you release data tagging a company as being in noncompliance you get a lot of attention, and you get a lot of feedback, both from the company and sometimes from the community around the plant. So I do think the system is evolving, and the information is getting a lot sharper than what we've been used to. But I need to warn you if you get on here and you try to use this information you're going to need to check, check with EPA or check with the State.

Here's an example, and you can open the individual files and see detailed records. I don't think it goes far enough over, but you can find out penalties paid, and how much has been invested in compliance. If you want to get a copy of the particular consent decree, if you're that interested, you can do that as well, that information is kept online. So there's an example of data that's promising, but still has a long way to go.

I guess the concern that I want to close with, or leave you with, is I guess in my view the discussion and debate around market-based approaches is very, very theoretical. And I understand that, and I think to some extent that reflects you know, maybe the way - I have to pick my words carefully here - maybe the way economists are trained to think, and that's a good thing, I'm not saying that's a bad thing. What I wish for is a lot more attention to the fundamentals of how you implement these programs and how well they can be enforced. What I'm afraid we are headed for if we don't do that is we're going to have a theory that's basically at war with the facts. So I leave you with that, and thank you for your attention.