PODCAST TRANSCRIPT
Associated report: "Inadequate Execution of the 7th DWINSA Lead Service Line Questionnaire Led to Flawed Data Being Used to Allot Lead Service Line Replacement Funds"

--- BEGIN TRANSCRIPT ---

KellyJune: Hello and thank you for joining us today for another podcast by the U.S. Environmental Protection Agency Office of Inspector General. I’m KellyJune, spokesperson for the EPA OIG, and with me today to is Sean Mahoney, Social Scientist for a report titled Inadequate Execution of the 7th DWINSA Lead Service Line Questionnaire Led to Flawed Data Being Used to Allot Lead Service Line Replacement Funds. Welcome, Sean.  

Sean Mahoney: Thank you. It’s great to be here. 

KellyJune: This report highlights significant data reliability issues that could impede the EPA’s ability to effectively distribute billions of Infrastructure Investment and Jobs Act, or IIJA, dollars ... dollars that were dedicated to aid in the replacement of lead service lines and related activities across the country. Can you give us a sense for what pushed the EPA OIG toward this project?

Sean Mahoney: Absolutely. We initiated this project as part of our ongoing oversight of the EPA’s implementation and use of IIJA funds.  For this evaluation, we focused on the $15 billion the IIJA gave the EPA to distribute to states from fiscal year 2022 to fiscal year 2026 to replace their lead service lines. Specifically, we looked at the $2.8 billion the EPA allotted to states in fiscal year 2023. 

But before I can talk about what we found, I need to explain how the EPA decides how much of this money each state should receive.  The EPA administers a Drinking Water Infrastructure Needs Survey and Assessment every four years. We refer to this survey as the DWINSA.  The survey was last administered in 2021, which was the seventh DWINSA. The 7th DWINSA was designed and administered before the IIJA was passed in November 2021. But, in 2018, Congress enacted the America’s Water Infrastructure Act, requiring the EPA to assess the overall cost of replacing all of the country’s lead service lines. To do that, the EPA collaborated with the states to develop a lead-service-line questionnaire, which we refer to as the LSL questionnaire. After the EPA received responses to the LSL questionnaire, it decided to use the responses to also allot the fiscal year 2023 IIJA funds for lead service line replacements. In other words, the EPA used the data submitted by the states to determine how many lead service lines each state had and alloted the fiscal year 2023 IIJA funds based on these numbers.

And so, a critical part of our evaluation was looking at whether the design and execution of the 7th DWINSA’s LSL questionnaire was appropriate to create accurate allotments to the states based on their lead-service-line replacement needs.
 
KellyJune: And what did you find?

Sean Mahoney: Our findings show inadequate internal controls and significant gaps in data reliability. We discovered that the EPA did not thoroughly verify the data submitted by the states and public water systems on the LSL questionnaire. We also learned that the EPA did not require documentation from states or water systems to support the LSL questionnaire data and, therefore, the Agency was unaware of how states collected or reviewed the data before using it to allot the lead service line replacement funds. Ultimately, we found that the EPA’s allotments of $2.8 billion in fiscal year 2023 IIJA funds for lead service line replacements were based on data that did not accurately reflect the lead-service-line-replacement needs in at least two states. This, in turn, reduced the funding available to other states that may have greater needs. Also, the issues we identify in our report do not only affect the fiscal year 2023 allotments. Because the IIJA provides lead service line replacement funds until fiscal year 2026, the findings we identify in the report have implications for fiscal years 2023 to 2026. 

KellyJune: Those are some significant findings. Can you expand on the data reliability issues you mentioned? 

Sean Mahoney: Sure. The EPA said it received responses to the LSL questionnaire from about 75 percent of the 3,600-plus water systems surveyed. Before submitting the LSL questionnaire to the EPA, states were expected to use their “best professional judgement” to review the water systems’ responses. But after the IIJA was enacted and the EPA looked to use the LSL questionnaire to allot billions in IIJA dollars – and not just to estimate the overall cost to replace lead service lines in the United States -- the EPA did not update this “best professional judgement” approach. Also, some states told us they were not aware of the expanded objective when they submitted their data to the EPA. The EPA had not yet decided to use the LSL questionnaire to allot the IIJA funds. An EPA supervisor told us that some states said they would have –quote- “done a better job ” -end quote- completing the LSL questionnaire had they known the EPA planned to use it to allot IIJA funds. One state even expressed that it was a –quote-  shock” -end quote- to learn that the EPA was using the LSL questionnaire data in this manner. And this is important because the EPA relied on states to review the data being submitted. 

KellyJune: But didn’t the EPA check the data also?

Sean Mahoney : Yes, it did, but the Agency did not have sufficient internal controls and data quality checks in place to identify data anomalies in the state-submitted data. Also, the EPA did not require any documentation to support the states’ responses to the LSL questionnaire. While the EPA did ask water systems to retain records to support their LSL questionnaire responses, the Agency did not request these records. We were told that the Agency did not want to place a high burden on states to develop such documentation and chose to focus more on achieving a higher response rate by – quote – “balancing the value of the requested information with the burden of providing it” Without any documentation to support the LSL questionnaire data received, the EPA couldn’t fully determine how the LSL questionnaire was developed or how states reviewed the data before sending it to the Agency. 

KellyJune: Hmm, That’s interesting. And in the report there is specific mentions of inaccuracies in the data EPA received from Texas and Florida. What happened there?

Sean Mahoney : In Texas, a water system serving the city of Houston intended to report all of its service lines as being of unknown material, but inadvertently reported that it had over 300,000 service lines made of lead. The state didn’t catch the error before it submitted the figure to the EPA. The EPA didn’t catch it either, even though that one water system, which served just one city, reported nearly 99 percent of the total reported lead service lines in Texas. The Houston water system noticed the error in May 2023, after the fiscal year 2023 lead service line replacement funds were allotted.  And while they told the State of Texas about the error, the state did not immediately share this information with the EPA. The EPA did not learn of the error until December 2023. The Agency adjusted Texas’ fiscal year 2024 allotment accordingly, but the state’s fiscal year 2023 allotment was over $117 million more than it should have been. 

KellyJune: That was a big error, did Florida have similar data issues?

Sean Mahoney: Now in Florida, most water systems reported all their service lines as being of unknown material. After the EPA asked states to try to minimize the number of service lines they were reporting as unknown using information to remove service lines that are likely not lead, Florida used its contractor to develop a methodology to estimate the number of lead service lines in the state’s medium and large water systems. And then Florida submitted that data directly to the EPA. However, that contractor-produced methodology was not uniformly applied across all water systems, and neither the state nor its contractor could explain a second pattern that we identified in the state’s data. If Florida’s contractor-produced methodology was uniformly applied, the LSL projections for the state would have decreased by about 33 percent, which means Florida would have been allotted fewer IIJA funds in fiscal year 2023 for its lead service line replacements. 

KellyJune:  Well if these were estimates, then neither Florida nor the EPA would know for sure whether the numbers were off, correct?

Sean Mahoney: Well, we communicated directly with eight water systems in Florida. According to the data that Florida submitted to the EPA, these eight systems had over 220,000 lead service lines, which accounts for nearly 40 percent of the total reported lead service lines for the state. But the eight systems told us that, as of January 2024, they had NOT identified any lead service lines. Plus, several systems told us they do not believe they have many, if any, lead service lines. 

Also, to go back to my point about how the data Florida submitted to the EPA was inflated because the contractor’s methodology was not consistently applied:  that methodology was only applied to one of these eight systems. But even if it had been applied to all eight systems, it would have estimated that they had a total of over 100,000 lead service lines – which we know from talking to these systems is likely not the case. So, the methodology itself would not have produced lead service line estimates for these systems that reflect their understanding of the lead service line issue in their jurisdiction. The methodology appears to inflate the number of lead service lines in Florida. 

KellyJune:  And what was the result of these data errors and inflations? 

Sean Mahoney: These flaws made a big impact on the IIJA allotments for fiscal year 2023. We did not look at every state that submitted data to the EPA so we can’t say whether other states made errors that further skewed the allotment of funds. However, Florida received the largest allotment of all states in fiscal year 2023, over $254 million dollars. Texas received the fifth-largest allotment, over $146 million dollars. These amounts are not proportional to their lead service line replacement needs. This means that states with greater lead service line replacement needs received reduced allotments. 

KellyJune:  I want to go back to something you said earlier, you said that the findings identified in the report have implications for fiscal years 2023 to 2026. What are the potential impacts of these inaccuracies on the total funding scheme?

Sean Mahoney: Good question. The EPA estimates that it will cost between $50 billion and $80 billion dollars to replace the approximately 9.2 million lead service lines nationwide. The IIJA will fund $15 billion dollars, which is about 18 to 30 percent of the total cost, making every dollar critical to help states with replacement costs. As I mentioned previously, inaccuracies and flawed data affect the entire country, not just the states with errors in their data. Inflated projections in some states leave fewer IIJA funds available to other states, which may actually have lead-service-line replacement needs that merit a greater allotment. And remember, the $15 billion dollars in IIJA funds for lead service line replacements are to be distributed over five years, so the potential impact isn’t limited to just one year either.  In fiscal year 2022, which was the first year that IIJA funds were available, the 7th DWINSA results were not finalized, so the EPA used the 6th DWINSA to distribute the lead service line replacement funds to the states. But the 6th DWINSA focused on general infrastructure needs and did not include the LSL questionnaire, so the EPA used the 7th DWINSA’s LSL questionnaire for the fiscal year 2023 funds to allow states to receive financial assistance commensurate with their lead-service-line-replacement needs. However, as we discussed, the data from the LSL questionnaire did not accurately reflect the lead-service-line-replacement needs of at least two states. In addition, according to the EPA, service line material information is rapidly evolving as water systems develop inventories of their service lines to meet statutory requirements. As the states begin to more concretely identify how many lead service lines they have, the EPA will have access to more and better data to understand the lead-service-line-replacement needs of states.


KellyJune:  So, what has the EPA been doing to take into account this new information?

Sean Mahoney: In the fall of 2023, the Agency allowed states to update their questionnaire responses or to submit responses if they had not done so previously. The updated responses were incorporated into the fiscal year 2024 funding calculations, which resulted in changes to allotments for 25 states. These adjustments led to funding variations ranging from as high as a 128 percent increase in funds to a reduction as large as 80 percent.  And yes, the allotments for Texas and Florida both changed in fiscal year 2024.  Texas received $117 million less than it did in fiscal year 2023, and Florida’s allotment decreased by around $26 million. Our findings show that Florida’s fiscal year 2024 allotment is still too high, though. We learned that the EPA made several efforts during and after the one-time update to collect new data from Florida and Texas, but the states’ environmental departments did not submit any updates. The EPA reached out to some systems directly to get updated data, but much of the data used for the fiscal year 2024 allotment to Florida was the same as the fiscal year 2023 data that we reviewed. So it is critical that the EPA improve its internal controls and data verification process. The Agency has work to do to ensure its fiscal years 2025 and 2026 allotments reflect the lead-service-line-replacement needs of states.

KellyJune: This report includes three recommendations to help the EPA address the issues identified. What are those recommendations, and has the EPA acted? 

Sean Mahoney : This report really shows the importance of ensuring data quality and reliability when allotting IIJA funds. While the EPA did take steps to improve the data for the fiscal year 2024 allotments, we recommended the Agency develop a process to identify unreliable lead service line data and to identify actions to address this data and determine whether further data updates are needed. If the EPA determines updates are necessary and appropriate, we recommended that it adjusts the IIJA allotments for fiscal years 2023 to 2026. And we specifically recommended that the Agency assess whether Florida’s allotment is appropriate for the remaining two fiscal years. Ultimately, we want to see the IIJA lead service line allotments match the states’ needs in accordance with congressional intent.

KellyJune: Sean, this report is an important reminder that proper oversight and accurate data is essential for the successful implementation of IIJA funding. Thank you for being with us today and sharing your report.

Sean Mahoney : Thank you for having me.

KellyJune: And to our listeners, thanks for joining us. The full report from today’s podcast, as well as other OIG reports and project notifications, are on our website at www dot epaoig dot gov.  

We urge anyone with information about fraud, waste, abuse or mismanagement involving an EPA or Chemical Safety Board operation or program to contact the EPA OIG Hotline by calling 888-546-8740, emailing OIG-DOT-hotline-AT-epa-DOT-gov, or submit a complaint form via our website. 

--- END TRANSCRIPT ---