How far will BIL dollars go in replacing lead lines across the country? An analysis of EPA’s Federal Fiscal Year 2024 allotments for lead line replacement
By Erica Galante-Johnson & Janet Pritchard
In May 2024, the Biden-Harris White House and the US EPA (EPA) announced the third year (out of five years) of allotments of lead service line replacement (LSLR) funds from the Bipartisan Infrastructure Law (BIL). Since BIL was first passed in 2021, the Environmental Policy Innovation Center (EPIC) has been paying close attention to how these funds are being rolled out to states, how far the $15 billion from BIL for lead service line replacement (LSLR) will go in ridding the nation of its lead pipe burden, and how much additional support and funding states may need.
Since 2023, EPA has linked BIL LSLR funding to service line material data [1] collected in 2021 from a sample of water systems per state as part of the 7th Drinking Water Infrastructure Needs Survey and Assessment or DWINSA (7th DWINSA). This survey is used to assess infrastructure needs every four years and to allocate Drinking Water State Revolving Funds (DWSRF), including BIL LSLR, accordingly. In 2023, EPIC urged EPA to consider updated responses to the survey since more and more water systems are identifying lead service lines (LSLs) and these responses are critical to determine how much states will receive for LSLR. EPA initiated an unprecedented one-time update to survey responses in late 2023 to better align funding with states’ lead burden (as measured by number of lead service lines) [2]. We don’t know yet what EPA will do with future BIL LSLR allotments, and whether future surveys will collect additional information to capture the rapid pace at which LSLR is occurring. According to the EPA, a total of 67% of the surveyed water systems provided a response to the recent update, increasing the overall response rate to 78% when combined with the 7th DWINSA data.
In EPIC’s initial reaction to the FFY 2024 allotment announcement, we noted a handful of states receiving more funding (e.g. Minnesota, New Jersey) while others saw decreased LSLR funding (e.g. Texas), though most saw no change in funding. These changes in funding mirror the EPA’s updated number of estimated lead service lines posted in May 2024. We can now determine whether the one-time update helped align LSLR funding with states’ lead pipe burden and how far these funds will go to replace toxic lead lines over the five years of BIL.
Figure 1: Changes in estimated lead service lines by state following the one-time update to the 7th Drinking Water Needs Survey and Assessment (DWINSA).
LSLR funding capacity is not commensurate with state lead burdens
Each state receives a minimum 1% allotment (and 1.5% across all territories) from the BIL LSLR funds through the Drinking Water State Revolving Fund (DWSRF), as directed by the Safe Drinking Water Act (SDWA) [3]. With the remaining funds, EPA - since 2023 - has calculated BIL LSLR allotments based on states’ estimated percentage of LSLs relative to the national number of projected LSLs [4]. This LSL-specific allotment helps align BIL funding (beyond the minimum 1%) with estimated lead burdens more appropriately. However, based on our analysis, BIL LSLR funding is still not commensurate with state needs, despite adjustments following the 7th DWINSA update.
States with the highest lead burdens can only replace a fraction of their LSLs
Despite receiving the largest BIL LSLR allotments, states with the highest lead burdens remain underfunded and will only be able to use BIL LSLR to defray the cost of a small fraction of these replacements. For example, Illinois is the state with the highest estimated lead burden and it received the largest portion (8.41%) of the BIL LSLR funding, accordingly. In fact, Chicago, Illinois has nearly 400,000 LSLs, which is more than the total number of lead lines in 44 individual states. However, despite gaining 4.7% more funding, this adjusted allotment is still only enough to replace approximately 86,000 (8%) of more than one million estimated LSLs in Illinois over the five years of BIL [5].
With funding falling short in the most overburdened areas of the country, vulnerable populations – in particular, children, low-income, and communities of color – will likely continue to be at risk of lead-contaminated drinking water for much longer. This is especially true for communities relying primarily on BIL LSLR funding and if funds are not prioritized to replace LSLs in the most vulnerable communities first, as EPIC has advocated.
Figure 2: Percentage of lead service lines that can be replaced solely with BIL LSLR funds by states receiving more than the minimum 1% allotment in FFY2024. Only states receiving more than the minimum 1% allotment are represented with additional percentages (over 1%) indicated in the x-axis and the number of lead service lines that can be replaced with this additional funding represented in the y-axis. To calculate the number of LSLs that could be replaced we assumed an average cost of $12,500 per lead service line replacement as reported by Kutzing et al. (2023).
Funding for states with the lowest lead burdens exceeds their needs
On the other hand, many states with low lead burdens (e.g. western states) continue to be capable of replacing far beyond one hundred percent of their lead lines after the allotment adjustments using BIL LSLR funding. This is true even when low-burdened states receive the minimum 1% mandated by SDWA. For example, Alaska is the state with the lowest estimated LSL number and therefore receives the minimum allotment (1%). However, given the state’s comparatively low LSL estimates (just over one thousand), this minimal funding is enough to cover over ten times the total number of estimated LSLs.
Figure 3: Percentage of lead service lines that can be replaced using only Bipartisan Infrastructure Law - Lead Service Line Replacement (BIL-LSLR) funds based on FFY24 allotments and EPA’s new methodology for projecting lead service line numbers. An average cost of $12,500 per lead service line replacement as reported by Kutzing et al. (2023) was used to calculate the percentage of service lines each state could replace.
To explore how many lead service lines states can replace using varying replacement costs, please see EPIC’s interactive, online map.
Changes in overall state funding is weakly linked to LSLR funding capacity
Increases or decreases in the third round of BIL LSLR allotments following the most recent DWINSA update resulted in varying, often minor, changes to a state’s capacity to fully fund LSL replacement with BIL dollars.
For example, North Dakota, Nebraska, New Hampshire, and Maine all saw modest increases in their ability to fund LSL replacement. This is because, while their allotment remained the same as the previous funding round, their LSL estimate decreased, per the DWINSA update. Therefore, BIL LSLR funding in these states could be stretched further as they have to replace fewer lines, though how far the funding goes varies broadly. While Nebraska can only cover roughly a quarter of replacements with BIL LSLR funds, the other four states could replace over more than half of their service lines.
In other states like Georgia and Kentucky, the percentage of lead lines that could be replaced using BIL LSLR funds was markedly reduced despite receiving increased allotments in this funding round. Though these increased allotments are no doubt linked to higher LSLs estimates in these states, these funds do not meet the need, indicating BIL LSLR allocations are not commensurate with lead burdens.
In cases like Utah, LSLR funding capacity was reduced because they received the same minimum 1% allotment while their estimated LSLs increased. However, this increase was not enough to place Utah over the EPA’s cut-off to receive funding above the 1% mandated minimum allotment, which is 9,000 LSLs or more.
Surprisingly, Texas’ LSLR funding capacity increased despite the dramatic reduction (80%) in its BIL LSLR allotment in FFY 2024. In fact, after the adjustment based on the DWINSA update, Texas joined the states receiving the minimum 1% allotment, and yet this state’s LSLR funding capacity surpasses 100 percent. This turn-around is due to corrections in the inventory numbers reported in the initial 7th DWISNA – which some believe resulted from a possible data entry error – revealing the state’s estimated lead burden to be much lower than previously expected in the corrected inventories.
Figure 4: Changes in state LSLR funding before (FFY23) and after (FFY24) updated 7th DWINSA lead service line estimates.
This misalignment between LSL-specific allotments and replacement capacity could indicate that the current formula is not suited to distribute BIL LSLR funds in accordance with state needs. Inequitable allotment outcomes appear to be driven, in large part, by the SDWA directive that each state be allotted a minimum 1% of appropriated DWSRF funds. However, states are only awarded the amount of allotted funds they can demonstrate a need for by listing priority projects within their annual intended use plan (IUP). Any unrequested could be available for reallotment to other states, mitigating some of the observed inequities.
Despite the historic BIL LSLR investment made by the Biden-Harris Administration and Congress, future funding should take special consideration of the disparate distribution of lead burdens across the country to ensure funding reaches the communities that need it the most.
This may require Congress to reconsider whether the minimum 1% SDWA requirement, which was initially conceived in relation to general DWSRF funds, should be applied equally to LSLR funding distributed through the Drinking Water State Revolving Fund (DWSRF). Future funding could also be directed through alternative funding streams such as Water Infrastructure Improvements for the Nation (WIIN) grants which are designed to help disadvantaged communities reduce lead in their drinking water.
Reallotments may help fill some of the funding gap in heavily lead burdened states
It may be the case that states do not request their share of BIL LSLR funds or opt for partial funding, leaving dollars to be reallotted to more heavily burdened states [6]. For example, in November 2023, EPA realloted $219,469,603 that had initially been alloted to seven states (Alaska, Hawaii, Maine, Nevada, Oregon, South Dakota, and Washington) that declined all or part of their FFY22 BIL LSLR allotment. These funds were reallotted to the remaining 43 states, using the FFY23 allotment formula, in accordance with the SDWA and related federal regulations [7].
Though this reallotment of funds could indeed help states most in need, it can be a lengthy process. States have two years to request allotted funds [8], and reallotment of unrequested funds can only happen once this two-year availability period has ended [9]. This delays LSLR efforts in states that might have replaced LSLs at a faster pace if they had received a larger portion of the BIL LSLR funds in the initial allotments, rather than waiting for the reallotment process.
Moreover, while the reallotted funds can benefit communities in more heavily burdened states, it does not resolve the underlying inequity:
As a result of the 1% minimum allotment requirement, states with relatively low lead burdens receive an allotment that is more than sufficient to enable them to replace 100% of their estimated LSLs. At the same time, the BIL LSLR funding available to the most heavily burdened states remains woefully inadequate to their need, even after taking reallotted funds into consideration.
Unreported and unknown service line materials affect BIL LSLR allotments
State BIL LSLR allotments are determined based on the proportion of a state’s estimated LSLs relative to the national total. To obtain this number, EPA uses the DWINSA responses to estimate the proportion (by sampling weight for medium and large systems) of LSLs in the state which is then multiplied by the total number of connections in the state [10].
EPA assumes the same proportion of the state’s estimated “unreported” and “unknown” lines to be lead, and this assumption is factored into the total estimated LSLs. At face value, this is a reasonable assumption. However, assigning the same weight to lead line estimates derived from “unreported” or “unknown” as estimates derived from “reported lead” lines when calculating total number of LSLs may end up awarding disproportionately larger allotments to states with incomplete inventories. This is particularly true in cases where the number of either unreported or unknown lines is high and a large proportion of lines with “known” materials are reported as lead, since the same proportion of unknowns/unreported will be assumed to be lead [11].
Consider, for example, that a state with several cities that were developed prior to the 1950s which contain a substantial number of LSLs, and the water systems serving these cities report these LSLs; they know that the BIL LSLR funds provide an opportunity to remove these lead pipes, and hence they are updating and reporting their inventories. At the same time, a large portion of the state’s population are served by water systems in suburbs and other newer cities that were developed after the 1970s, have little to no lead in their distribution systems, and have placed less priority on updating their inventories. EPA’s formula assumes the same percentage of lead in systems with large numbers of “unreported” pipes as in the older cities reporting a significant percentage of known LSLs.
To ensure LSLR funds are reaching the communities where they are most needed moving forward, it is imperative to avoid perverse incentives from not completing service line inventories. To this end, we will explore the implications of accounting unknown and unreported lines in LSL estimates for BIL LSLR allotment formulas more in-depth in future analyses.
BIL LSLR funds alone are not enough to replace all lead service lines
Replacing all of the nation’s estimated 9 million lead service lines can cost between $72 billion [12] to over $90 billion, with individual replacements typically ranging between $8,000 [13] and $12,500 [14] on average. These costs are far higher than the available funding through BIL LSLR funding alone. To deal with this challenge, municipalities across the country are adopting a variety of strategies to pay for lead service line replacement, including stacking different funding and financing sources to cover costs. Some strategies to supplement BIL LSLR funding include making use of other federal and state funding (e.g. Detroit), using customer rates, and leveraging municipal bonding (e.g. Denver). EPIC is also analyzing additional cost-saving measures states and utilities can take to stretch their funding at low or no cost as well as recommendations to avoid LSLR spending pitfalls. States can further optimize their use of BIL LSLR funds by maximizing the use of allowable set-asides to cover activities such as lead inventories. EPA has strongly endorsed the expansive use of LSLR set-asides in its May 2024 memo on Implementing Lead Service Line Replacement Projects Funded by the Drinking Water State Revolving Fund.
Sluggish funding could hamper achieving complete lead service line replacement over the next decade
To meet the Biden-Harris Administration’s goal of removing all lead pipes in 10 years, it is crucial that BIL LSLR funding be deployed quickly, efficiently, and equitably, to track shifting needs and the number of known lead service lines is updated as inventories are completed. This may require a more flexible allotment process and formula that can react to these rapid changes rather than relying on data that is only updated every four years, as the current DWINSA process entails. The 8th DWINSA, which currently presents the next opportunity when the allotment formula could potentially be updated, will not happen until 2025. This would only allow adjustments for the final round BIL LSLR allotments, if timing allows. Removing policy barriers to the flow of funds from the federal to the local level as well as assisting communities in accessing the funding are also important in the effort to replace the nation’s lead pipes. EPIC will continue to track where BIL LSLR funding bottlenecks are, what may be causing them, and what states and municipalities can do to overcome these barriers in addition to working directly with communities to connect them to the funding.
Methodology
To compare differences in lead service line replacement funding capacity before and after the 7th DWINSA update, we calculated the percentage of LSLs that could be replaced throughout the five years of BIL LSLR making use of the FFY23 and FFY24 allotments, respectively. To do so we employed EPA’s state level lead service line estimates based on the original and updated 7th DWINSA and the EPA’s FFY22-FFY24 BIL LSLSR allotment amounts. We assumed the average cost of full (i.e., private and public side) per-pipe replacement to be $12,500 [15], to be conservative.
Based on these numbers, we calculated the percentage of pipes that would be covered by adding the allotted amounts in previous funding cycles and the allotment amount of the FFY considered multiplied by the remaining years of BIL, assuming this allotment is held constant, as described in Figure 5:
Figure 5: Lead service line replacement (LSLR) funding capacity calculated as the percentage of LSLs a state can replace using BIL LSLR funding exclusively based on Federal Fiscal Years 2023 (top equation) and 2024 (bottom equation) allotments. An average cost of $12,500 per lead service line replacement as reported by Kutzing et al. (2023) was used to calculate the percentage of service lines each state could replace.
Where FFY allotment corresponds to the allotment amount received in each respective federal fiscal year (i.e., 2022, 20223, 2024) and LSLR funding capacity is the percentage of lead service lines a state could replace over the five years of BIL funding assuming a per-pipe replacement cost of $12,500.
[1] US EPA, Office of Water. 2023.Drinking Water Infrastructure Needs Survey and Assessment: 7th Report to Congress p.23. https://www.epa.gov/system/files/documents/2023-09/Seventh%20DWINSA_September2023_Final.pdf
[2] The definition of ‘lead burden’ in this blog is measured by the number of lead service lines only, though the total lead burden for public health measures must inevitably include all sources of lead, including in air, dust, paint, dirt, and households.
[3] 40 CFR § 35.3515 as revised August 7, 2000. https://www.ecfr.gov/current/title-40/chapter-I/subchapter-B/part-35/subpart-L
[4] Ibid.
[5] The percentage of lead service lines that can be replaced solely with BIL LSLR funding was calculated by adding allotted amounts in previous funding cycles and the allotment amount of the federal fiscal year considered which we then assumed to remain constant for the remaining years of BIL funding. We assumed the average cost of full (i.e., private and public side) per-pipe replacement to be $12,500 as reported by Kutzing et al. (2023). For more detail see “Methodology”.
[6] Ibid.
[7] 40 CFR § 35.3515 as revised August 7, 2000. https://www.ecfr.gov/current/title-40/chapter-I/subchapter-B/part-35/subpart-L
[8] Cite to EPIC’s Flow of Funds Explainer: https://static1.squarespace.com/static/611cc20b78b5f677dad664ab/t/66687fbf8565a43af2120425/1718124486556/FundingFlowExplainer_2024.06.11.pdf
[9] 40 CFR § 35.3515 as revised August 7, 2000. https://www.ecfr.gov/current/title-40/chapter-I/subchapter-B/part-35/subpart-L
[10] US EPA, Office of Water. 2023.Drinking Water Infrastructure Needs Survey and Assessment: 7th Report to Congress pp.42-46.
[11] Ibid p.43
[12] US EPA Economic Analysis for the Proposed Lead and Copper Rule Improvements. Office of Water (4607M) EPA 815-R23-005 (Nov. 2023), https://www.regulations.gov/document/EPA-HQ-OW-2022-0801-0712
[13] Elin Warn Betanzo, Safe Water Engineering, LLC and Vanessa Speight (2024).Lead Service Line Replacement Costs and Strategies for Reducing Them. https://www.nrdc.org/sites/default/files/2024-08/lslr-costs-strategies-reducing-them.pdf
[14] Kutzing, S., Murray, N., Sayre, A., and Gawlik, E. (2023) Planning for Service Line Material Identification and Lead Service Line Replacement Costs. Journal of the American Water Works Association, 115(6), 22-33. https://doi.org/10.1002/awwa.2123
[15] Kutzing, S., Murray, N., Sayre, A., and Gawlik, E. (2023) Planning for Service Line Material Identification and Lead Service Line Replacement Costs. Journal of the American Water Works Association, 115(6), 22-33. https://doi.org/10.1002/awwa.2123