By Leif Fredrickson and Sarah Lamdan
Following up on our recent report on enforcement at the Environmental Protection Agency (EPA), we wanted to dig into the numbers to see what was happening in different EPA regions and programs across the nation. The results are startling, showing huge declines in environmental enforcement actions across most regions and programs. These enforcement actions — which include orders to institute anti-pollution measures, fines, requirements for clean up and so on — are critical to getting companies and other regulated entities to comply with environmental laws, which in turn help assure a clean and healthy environment.
In our report, A Sheep in the Closet: The Erosion of Enforcement at the EPA, we showed that, based on the EPA’s enforcement and compliance database, enforcement has significantly declined in Fiscal Year (FY) 2018. (Fiscal years run from October 1 to September 30). The reasons for the declining enforcement numbers and the negative health and safety impacts related to decreased environmental law enforcement are elucidated in news coverage of our report, and in the report itself, which is based on extensive interviews with agency staff and internal documents we obtained.
Among other things, EPA’s internal documents include a detailed analysis of the agency’s declining enforcement broken down by EPA region and by the major statutory programs (e.g., the Clean Air Act, the Clean Water Act, and so on). We wanted to include the EPA’s internal analysis in our report to show that the agency, not just our group, had found steep declines in enforcement across regions and programs.
But that analysis of regional and program declines in enforcement was only based on mid-fiscal year numbers. And since we have released our report, people have asked about the regional implications of the national decline rates — are these declines happening in every EPA region? Are they affecting some areas worse than others? To answer some of these questions, we decided to look at full fiscal year data for different regions and programs.
Large Regional Declines in Enforcement
Our original map breaking down declining enforcement by region was intriguing to us and others. We wanted to understand what enforcement declines looked like for the full fiscal year by region (rather than just national trends), so we created a map with full fiscal year numbers (below).
Percentage decline in civil case conclusion from FY 2017 to FY 2018. As the map shows, there were declines in every region, and most regions showed huge, double-digit declines. Source: ICIS FE&C Data Set, downloaded from Enforcement and Compliance History Online, Data Downloads, Environmental Protection Agency, accessed December 3, 2018, https://echo.epa.gov/tools/data-downloads.
As the map shows, the number of civil cases concluded declined in every region. Civil case conclusions are the number of civil (as opposed to criminal) cases the EPA has completed — whether through settlements, administrative orders or court orders.
Another way of measuring enforcement is to count the number of cases started, or what the EPA calls “case initiations.” The two are linked, of course, because a case must be initiated before it can be concluded. Today’s case initiations are tomorrow’s conclusions. However, data on case initiations are not publicly available, nor is detailed information on criminal cases. Civil case conclusion are, nevertheless, a fundamental measure of EPA enforcement. Civil cases conclusions provide a glimpse into how many cases are being initiated and provide a way to measure how diligently the EPA is pressing to conclude cases. In the opaque realm of EPA enforcement, civil case conclusion data provide evidence of how the EPA is pursuing environmental law violations.
In several regions, the drop in civil cases settled is enormous. Region 4 saw a decline of 77%, and three other regions (1, 3, and 8) saw declines of about 55%. Three more regions (2, 5, 9) saw declines of 40% or more. Regions 6 and 10, had relatively smaller declines, although in absolute numbers the declines were still quite large (32% and 37% respectively). Region 7 saw the smallest declines, the only one in the single digits, but the EPA region nevertheless closed out fewer civil enforcement cases in 2018 than in 2017.
Percentage decline in civil case conclusions from FY 2017 to FY 2018. As the table shows, there were declines in every major region except Region 7. Source: ICIS FE&C Data Set, downloaded from Enforcement and Compliance History Online, Data Downloads, Environmental Protection Agency, accessed December 3, 2018, https://echo.epa.gov/tools/data-downloads.
Accelerating Declines in the Second Half of Fiscal Year 2018
We also wanted to review and update the numbers because the EPA has responded to our report by claiming that our numbers are based on preliminary, mid-fiscal year data, and that the agency expects that its final 2018 numbers will show “a significant improvement from mid-year” [i.e., the mid-fiscal year].
Curious whether the EPA’s data showed that enforcement actions did improve in the second half of the fiscal year compared to the first, we went ahead and pulled data from their site and analyzed it.
The short answer is: No, EPA enforcement numbers did not improve from mid-year 2018 to the end of the year. In fact, based on civil case conclusion data, enforcement got worse in the second half of fiscal year 2018, not better.
In every region except Region 7, the decline in civil enforcement case conclusion was even more severe when comparing all of fiscal year 2018 to all of fiscal year 2017, rather than just the first halves of those fiscal years. This especially obvious when you compare the above map to the one below, from our report.
Percentage decline in civil case conclusions from the first half of FY 2017 to the first half of FY 2018. Comparison with the map above shows that these declines got even more severe in the second half of the fiscal year (with the exception of Region 7). Source: Office of Enforcement and Compliance Assurance, “FY17 to FY18 Mid-Year Analysis for Regional Evaluation,” June 16, 2018 [document in EDGI’s possession].
Overall, there were 1,435 case conclusions in fiscal year 2018, but 935 of these came in the first half of the fiscal year. This means that nearly two-thirds of the EPA’s civil case conclusions happened in the first half of 2018, showing clearly lagging civil enforcement work, especially in the second half of 2018.
Enforcement Decline by Environmental Program
Looking at how enforcement has declined by program (i.e., specific environmental laws) shows the same overall pattern that enforcement declines by region show, namely that the declines are across the board. Every major program saw substantial declines. The bedrock programs that protect the air (the Clean Air Act, CAA), the water (the Clean Water Act, CWA, and the Safe Drinking Water Act, SDWA) and the land (the Resource Conservation and Recovery Act, RCRA) all saw declines of over 40%.
Even the Superfund program (Comprehensive Environmental Response, Compensation, and Liability Act, CERCLA) — which forces polluters to clean up the polluted land and had been one of the few anti-pollution programs former EPA administrator Scott Pruitt’s said who would champion — saw a decline of 24%.
Other programs that deal with toxic and hazardous substances, such the Toxic Substances Control Act (TSCA) and the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) saw declines of 47% and 60%.
Last, but not least, the Emergency Planning and Community Right-to-Know Act (EPCRA) — a foundational law for environmental justice that warns people about hazardous substances nearby — saw a 48% declines in case conclusions.
Percentage decline in civil case conclusion from FY 2017 to FY 2018. As the table shows, there were declines in every major declines in every major environmental program (statute). Source: ICIS FE&C Data Set, downloaded from Enforcement and Compliance History Online, Data Downloads, Environmental Protection Agency, accessed December 3, 2018, https://echo.epa.gov/tools/data-downloads.
A Closer Look at Regions 4 and 8
Region 4 and 8 were two of the regions with the largest declines in the number of civil case conclusions from FY 2017 to FY 2018. As with most other regions and the overall nation, these regions showed declines in enforcement in the second half of fiscal year 2018, not increases. Region 4 had 89 case conclusions in the first half of the fiscal year, but only 47 in the second half. Region 8 had 34 case conclusions in the first half of the fiscal year, but only 19 in the second half.
Looking back at data from prior fiscal years shows that the number of civil cases concluded in FY 2018 in Regions 4 and 8 is historically low. Region 8’s 53 cases are the lowest since at least 2006. Indeed, they are not even half as much as the next lowest year (FY 2017 with 118 cases). Region 4’s numbers are even worse. There were 136 civil case conclusions in FY 2018, by far the lowest number since at least 2006. The next lowest year was FY 2015, with 376 cases — over two and a half times as many as in 2018.
Penalties were also very low in these regions compared to previous years. Region 4 levied about $2.95 million in penalties in FY 2018, the lowest since at least 2006. The next lowest was $7.99 million in 2009. Region 8 levied $3.03 million in penalties in FY 2018, which was higher than a few of the previous years going back to 2006, but still lower than the average $3.8 million in annual penalties collected by the Regional office between fiscal years 2006 and 2017.
Region 8 saw very large enforcement declines in FY 2018 in the Clean Water Act and the Safe Drinking Water Act. (Note that regions are not required to report SDWA data for a quarter until the end of the falling quarter. Thus it is possible there is more SDWA data yet to come in). Region 4 saw very large declines in FY 2018 in the Clean Air Act, the Clean Water Act and the Resource Conservation and Recovery Act.
Civil case conclusions for Regions 4 and 8, including breakdowns by some major environmental statutes and including penalties levied in both nominal and inflation-adjusted (2009) dollars. Source: ICIS FE&C Data Set, downloaded from Enforcement and Compliance History Online, Data Downloads, Environmental Protection Agency, accessed December 3, 2018, https://echo.epa.gov/tools/data-downloads.
How Reliable Are These Numbers?
Before concluding, it is worth talking about the reliability of the data we are using. The data comes from the Integrated Compliance Information System (ICIS). This is the database the EPA uses to track its own enforcement actions. It is also the data that is used to produce the annual reports for the EPA’s Office of Enforcement and Compliance.
When the EPA conducted an internal analysis of enforcement actions in June, 2018, as we note in our report, they used numbers from ICIS. In June, they were looking back at cases from before March 31, 2018, when the first half of the fiscal year ended. That had given EPA regions extra time to input data into ICIS about cases from that period before the end of March. As the EPA wrote in its internal document analyzing the mid-fiscal year trends, “Regions are expected to enter data into ICIS within 10 business days of the event… The FY18 mid-year-data should be of good quality as the regions have had two months to complete data entry after the mid-year close (March 31, 2018).” The document also stated that “end of year data certification that prompts data quality reviews” occurs in “October 2018.”
We downloaded the data used here on December 3, more than two months after the end of the period in question (September 31) and well after the “data quality reviews” that began in October. The EPA has also indicated that it may put out its annual report on enforcement in December. If the EPA is in the report writing stage, its ICIS numbers should be very close to final. For these reasons, the numbers presented here from ICIS will be a close approximation of the final numbers reported.
That is not to say that the ICIS numbers presented here will be exactly the same as the final numbers published in the FY 2018 Annual Report on enforcement and compliance. The raw data from ICIS will be “cleaned” in various ways, such as removing cases that have been double counted, fixing data errors, and filling in missing data. The exact numbers, however, are not as important as the approximate number and the trend over time. Past correlations between the raw ICIS data and the final numbers published in annual reports show that the numbers are very close (the annual report numbers usually slightly lower) and the trends are virtually identical. You can see this in the graph below, which graphs data on Administrative Penalty Orders (APOs) — the most common type of enforcement action.
Correlation between APOs, the most common type of enforcement action, as measured using raw ICIS data and data from the Annual Reports of the EPA’s Office of Enforcement and Compliance Assurance (OECA). Source: ICIS FE&C Data Set, downloaded from Enforcement and Compliance History Online, Data Downloads, Environmental Protection Agency, accessed December 3, 2018, https://echo.epa.gov/tools/data-downloads; OECA, Annual Reports, various years.
If the numbers used here are not a close approximation of the final Fiscal Year 2018 numbers — the numbers that will be published in the annual report — then there is something very problematic in how regions are adding data to ICIS or in how EPA is disseminating that data to the public. If that is the case, we must ask the EPA to explain and fix this problem. It is incumbent upon the EPA to provide the public with accurate enforcement data and statistics.
If, as is much more likely, these numbers are a close approximation of the final FY 2018 numbers — that they accurately capture the trend — then we must press the EPA to explain why enforcement has deteriorated so badly and how the agency plans to reestablish itself as a serious enforcer of environmental protections for the land, air, water and people in all regions of the United States.