Skip to main content

Advertisement

SpringerLink
Log in

Factors Influencing Customer Participation in a Program to Replace Lead Pipes for Drinking Water

  • Published:
Environmental and Resource Economics Aims and scope Submit manuscript

Abstract

Many public water systems are struggling to locate and replace lead pipes that distribute drinking water across the United States. This study investigates factors associated with customer participation in a voluntary lead service line (LSL) inspection and replacement program. It also uses quasi-experimental and experimental methods to evaluate the causal impacts of two grant programs that subsidized homeowner replacement costs on LSL program participation. LSLs were more prevalent in areas with a higher concentration of older housing stock, Black and Hispanic residents, renters, and lower property values. Owner-occupied and higher valued properties were more likely to participate in the LSL program. Results from the two grant program evaluations suggest that subsidies for low-income homeowners to cover LSL replacement costs can significantly boost participation, but only when the programs are well publicized and easy to access. Even then, there was still significant non-participation among properties with confirmed LSLs.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Source: EPA 2022

Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Data Availability

The account-level LSL program data used in this study were acquired under a non-disclosure data use agreement with Trenton Water Works. Deidentified data aggregated at the Census Block Group level and replication code for this paper are freely available at: https://github.com/bryanparthum/lslr-paper.git.

Notes

  1. A block group is a geographical unit defined by the US Census. Our study area includes 165 block groups, 78 of which are in the urban municipality.

  2. A housing department representative indicated that there were fewer than 50 applicants for urgent rehabilitation grants in a typical year (personal communication, Farrah Gee, City of Trenton Department of Housing and Economic Development).

  3. Participants in roundtable discussions held in ten Northeastern and Midwestern communities by the US Environmental Protection Agency (EPA 2021b, 2021c, 2021d, 2021e, 2021f, 2021g, 2021h, 2021i, 2021j, 2021k) mentioned financial constraints, language barriers, landlord-renter split incentives, and lack of community trust in water utilities as barriers to LSLR. Participants in these sessions suggested subsidies, multiple types of outreach approaches (e.g., mailers, in person, online, door hangers), translation of materials into different languages, and partnerships with organizations viewed as “trusted messengers” in target communities as potential solutions to these barriers.

  4. It is also possible for individual homeowners to hire their own contractor to conduct an LSLR. The water system lacks data on whether or when any such replacements may have occurred but believes that they are rare and that any such replacements were typically undertaken for another reason besides reducing lead exposure, such as fixing a water leak or other construction/plumbing work occuring at the same time. The water system estimates that the total cost of replacement to a homeowner would be roughly $8000 rather than the $1000 cost-share offered by the LSLR program.

  5. In the July 2022 data, program registration dates were missing for 55% of accounts that had registered, and lead service line replacement dates were missing for 2% of properties that had a replacement. In these cases, we used the July 2020 and May 2021 data to determine when registration and replacements occurred.

  6. We categorized properties as owner-occupied if the first seven characters of the property address matched the first seven characters of the owner address. This indicator is extremely similar to a variable denoting a match of the entire character string of the property and owner addresses (ρ = 0.99) but allows for flexibility due to spelling mistakes or differences in the way apartment numbers are recorded across the datasets.

  7. Municipality dummy variables jointly explained a statistically significant portion of the variation in the property characteristics included in our study (p < 0.0001 in all cases).

  8. The quasi-experimental sample includes all urban properties in our final data set, except for 1338 properties that received postcards about the housing department grant program as part of the field experiment, 201 properties receiving other water system outreach mailings after the launch of the community-based grant program, 80 properties whose program registration and LSLR dates could not be determined from the account data, and 22 properties in neighborhoods that were not visited by water system contractors during both the pre- and post-grant periods (none of which were in the target neighborhood).

  9. Specifically, we excluded properties that met at least one of the following criteria prior to the intervention: the property was located outside of the urban municipality; the property was located inside the target neighborhood for the community-based grant program; the homeowner-side service line was confirmed to not contain lead; a LSLR already occurred; the water system account was inactive, vacant, or a non-valid address according to water system records; the property was not owner-occupied according to assessor data or the municipality’s registry of rental properties; occupants had previously refused water system staff or contractors access to the property; the property was located in a Census block in which no properties had previously registered for the LSLR program.

  10. Prior to randomly assigning addresses to the treatment and control groups, power calculations were estimated using a sample size of 1500 treatment and 1500 control with an assumed baseline take-up rate of 10%, resulting in a minimum detectable effect of 3.2%. The 10% take-up rate was based on the rate of program registration among urban households before postcards were sent.

  11. These results are not driven by multicollinearity with other socioeconomic variables included in the analysis. While the shares of Black and Hispanic residents in the urban municipality are strongly negatively correlated (ρ =  − 0.88), there is low correlation among the other explanatory variables included in the analysis. The variance inflation factors for share Black and share Hispanic in the urban regressions are 6.3 and 7.6, respectively. Variance inflation factors for all variables in all regressions shown in Table 1 are below 3, indicating relatively stable coefficient estimates.

  12. Appendix Table 5 reports results from a regression in which we pooled the urban and suburban samples and included interaction terms between an urban indicator variable and all other explanatory variables to assess the heterogeneity across locations in associations between property and neighborhood characteristics and LSLR program participation. The results show that the magnitudes of the associations between property and neighborhood characteristics and LSLR participation are often significantly different across urban and suburban municipalities, even though Table 1 indicates that the sign and statistical significance of most characteristics’ association with LSLR program association is similar.

  13. The unadjusted p-values of the effect of the community-based grant program on program registration, inspection, and replacement are 0.023, 0.055, and 0.032, respectively. When applying the Holm-Bonferroni adjustment to account for the possibility of a higher rate of falsely rejecting the null when testing multiple hypotheses, the p-values become 0.069, 0.055, and 0.064 (Holm 1979). While there is a marginal decline in statistical significance, the results remain statistically significant at the 10% level using a two-tailed t-test. Because the three LSLR program participation outcomes are highly correlated, and the Holm-Bonferroni approach does not account for correlation across outcomes, the approach has low power to detect false null hypotheses (List et al. 2019). We provide the adjusted p-values for illustrative purposes to demonstrate the robustness of our results to a conservative approach for addressing multiple hypothesis tests.

  14. This upward trend is less apparent for the program registration outcome, which was less affected by the COVID-19 pandemic; the water system began registering customers in 2018 and continued to encourage sign-up throughout 2020, anticipating that the program would eventually resume. The water system slowed its efforts to register new properties later in 2021 and 2022 as it became clear that replacements would pause later in 2022 due to funding constraints. In addition, program registration was not strictly required for residents to have inspections and replacements.

  15. The first stages of the IV regressions are reported in Appendix Table 13. They confirm that assignment to the postcard group is very strongly predictive of receiving a postcard.

  16. In addition, appendix Table 11 confirms that property and neighborhood characteristics are well balanced across the treatment and control groups.

  17. The unadjusted p-values of the effect of the housing department grant program on program registration, inspection, and replacement are 0.36, 0.17, and 0.87, respectively. Applying the Holm-Bonferroni adjustment for multiple hypothesis testing does not change our inability to reject the null hypothesis of no effect of the housing department grant program on LSLR program participation.

References

  • Abadie A (2021) Using synthetic controls: feasibility, data requirements, and methodological aspects. J Econ Lit 59(2):391–425

    Article  Google Scholar 

  • Abadie A, Athey S, Imbens G, Wooldridge J (2022) When should you adjust standard errors for clustering? Q J Econ qjac038

  • Aizer A, Currie J (2019) Lead and juvenile delinquency: new evidence from linked birth, school, and juvenile detention records. Rev Econ Stat 101(4):575–587

    Article  Google Scholar 

  • Aizer A, Currie J, Simon P, Vivier P (2018) Do low levels of blood lead reduce children’s future test scores? Am Econ J Appl Econ 10(1):307–341

    Article  PubMed  PubMed Central  Google Scholar 

  • Allcott H, Greenstone M (2017) Measuring the welfare effects of residential energy efficiency programs. In: NBER Working paper #23386. National Bureau of Economic Research

  • American Water Works Association (2005) Strategies to obtain customer acceptance of complete lead service line replacement. In: White paper

  • American Water Works Association (2022) Considerations when costing lead service line identification and replacement. In: Report prepared by CDM Smith, November 2022

  • Angrist JD, Pischke JS (2008) Mostly harmless econometrics: an empiricist’s companion, 1st edn. Princeton University Press, Princeton

    Book  Google Scholar 

  • Baehler KJ, McGraw M, Aquino MJ, Heslin R, McCormick L, Neltner T (2022) Full lead service line replacement: a case study of equity in environmental remediation. Sustainability 14(1):352

    Article  Google Scholar 

  • Banzhaf S, Banzhaf MR (2023) Impact of in utero airborne lead exposure on long-run adult socio-economic outcomes: A population analysis using U.S. survey and administrative data. PLoS ONE 18(11):e0293443.

  • Bettinger E, Long B, Oreopoulos P, Sanbonmatsu L (2012) The role of application assistance and information in college decisions: results from the H&R Block FAFSA experiment. Q J Econ 27(3):1205–1242

    Article  Google Scholar 

  • Bhargava S, Manoli D (2015) Psychological frictions and the incomplete take-up of social benefits: evidence from an IRS field experiment. Am Econ Rev 105(11):3489–3529

    Article  Google Scholar 

  • Brown MJ, Raymond J, Homa D, Kennedy C, Sinks T (2011) Association between children’s blood lead levels, lead service lines, and water disinfection, Washington, DC, 1998–2006. Environ Res 111(1):67–74

    Article  PubMed  Google Scholar 

  • City of Newark (2019) Mandatory replacement of lead service lines

  • Dave DM, Yang M (2020) Lead in drinking water and birth outcomes: a tale of two water treatment plants. J Health Econ 84:102644

    Article  Google Scholar 

  • Deshpande M, Li Y (2019) Who is screened out? Application costs and the targeting of disability programs. Am Econ J Econ Pol 11(4):213–248

    Article  Google Scholar 

  • Egan KB, Cornwell CR, Courtney JG, Ettinger AS (2021) Blood lead levels in U.S. children ages 1–11, 1976–2016. Environ Health Perspect 129(3):037003

    Article  PubMed  PubMed Central  Google Scholar 

  • Finkelstein A, Notowidigdo M (2019) Take-up and targeting: experimental evidence from SNAP. Q J Econ 134(3):1505–1556

    Article  Google Scholar 

  • Fowlie M, Greenstone M, Wolfram C (2015) Are the non-monetary costs of energy efficiency investments large? Understanding low take-up of a free energy efficiency program. Am Econ Rev Pap Proc 105(5):201–204

    Article  Google Scholar 

  • Fowlie M, Greenstone M, Wolfram C (2018) Do energy efficiency investments deliver? Evidence from the weatherization assistance program. Q J Econ 133(3):1597–1644

    Article  Google Scholar 

  • GAO (2020) Drinking water: EPA could use available data to better identify neighborhoods at risk of lead exposure. In: GAO Report to congressional requesters, December 2020. GAO-21–78

  • Gerarden TD, Newell RG, Stavins RN (2017) Assessing the energy-efficiency gap. J Econ Lit 55(4):1486–1525

    Article  Google Scholar 

  • Gillingham G, Tsvetanov T (2018) Nudging energy efficiency audits: evidence from a field experiment. J Environ Econ Manag 90:303–316

    Article  Google Scholar 

  • Giraudet L-G (2020) Energy efficiency as a credence good: a review of informational barriers to energy savings in the building sector. Energy Econ 87:104698

    Article  Google Scholar 

  • Glik D, Eisenman D, Zhou Q, Tseng C, Asch S (2014) Using the precaution adoption process model to describe a disaster preparedness intervention among low-income Latinos. Health Educ Res 29(2):272–283

    Article  PubMed  Google Scholar 

  • Hainmueller J (2012) Entropy balancing for causal effects: a multivariate reweighting methods to produce balanced samples in observational studies. Polit Anal 20:25–46

    Article  Google Scholar 

  • Hastings J, Weinstein J (2007) Information, school choice, and academic achievement: evidence from two experiments. Q J Econ 123(4):1373–1414

    Article  Google Scholar 

  • Hexemer A, Pintar K, Bird T, Zentner S, Garcia H, Pollari F (2008) An investigation of bacteriological and chemical water quality and the barriers to private well water sampling in a Southwestern Ontario Community. J Water Health 6(4):521–525

    Article  PubMed  Google Scholar 

  • Holladay J, LaRiviere J, Novgorodsky D, Price M (2016) Asymmetric effects of non-pecuniary signals on search and purchase behavior for energy-efficient durable goods. In: NBER working paper #22939, National Bureau of Economic Research

  • Holm S (1979) A simple sequentially rejective multiple test procedure. Scand J Stat 6(2):65–70

    MathSciNet  Google Scholar 

  • Hull C, Anderson N, Saxena S (2022) What will it take to remove all lead service lines? Common barriers to getting the lead out of drinking water. In: Federal Reserve Board of Chicago. May 2022

  • Iacus S, King G, Porro G (2012) Causal inference without balance checking: coarsened exact matching. Polit Anal 20(1):1–24

    Article  Google Scholar 

  • Karlsson N, Loewenstein G, Seppi D (2009) The ostrich effect: selective attention to information. J Risk Uncertain 38:95–115

    Article  Google Scholar 

  • Kreutzwiser R, de Loë R, Imgrund K, Conboy M, Simpson H, Plummer R (2011) Understanding stewardship behaviour: factors facilitating and constraining private water well stewardship. J Environ Manage 92(4):1104–1114

    Article  PubMed  Google Scholar 

  • Kuffner A (2022) Providence water gets $3.3M more in federal grant money to replace lead service lines. Provid J

  • Lanphear BP, Hornung R, Khoury J, Yolton K, Baghurst P et al (2005) (erratum 2019) Low-level environmental lead exposure and children’s intellectual function: an international pooled analysis. Environ Health Perspect 113(7):894–899

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  • Linos E, Prohofsky A, Ramesh A, Rothstein J, Unrath M (2022) Can nudges increase take-up of the EITC? Evidence from multiple field experiments. Am Econ J Econ Pol 14(4):432–452

    Article  Google Scholar 

  • List JA, Shaikh AM, Xu Y (2019) Multiple hypothesis testing in experimental economics. Exp Econ 22:773–793

    Article  Google Scholar 

  • MacDonald K, Tippett M (2020) Reducing public exposure to common, harmful well water contaminants through targeted outreach. J Water Health 18(4):522–532

    Article  PubMed  Google Scholar 

  • Manson S, Schroeder J, Van Riper D, Kugler T, Ruggles S (2022) IPUMS National historical geographic information system: version 17.0. Minneapolis, MN: IPUMS

  • Miranda ML, Kim D, Reiter J, Overstreet MA, Maxon P (2009) Environmental contributors to the achievement gap. Neurotoxicology 30(6):1019–1024

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  • Morgan KL, Rubin DB (2012) Rerandomization to improve covariate balance in experiments. Ann Statist 40(2):1263–1282

    Article  MathSciNet  Google Scholar 

  • Mullainathan S, Shafir E (2013) Scarcity: Why having too little means so much. Times Books

  • Pieper KJ, Tang M, Edwards MA (2017) Flint water crisis caused by interrupted corrosion control: investigating “ground zero” home. Environ Sci Tech 51(4):2007–2014

    Article  CAS  Google Scholar 

  • Renaud J, Gagnon F, Michaud C, Boivin S (2011) Evaluation of the effectiveness of arsenic screening promotion in private wells: a quasi-experimental study. Health Promot Int 26(4):465–475

    Article  PubMed  Google Scholar 

  • Reuben A, Caspi A, Belsky DW, Broadbent J, Harrington H, Sugden K, Houts RM, Ramrakha S, Poulton R, Moffitt TE (2017) Association of childhood blood lead levels with cognitive function and socioeconomic status at age 38 years and with IQ change and socioeconomic mobility between childhood and adulthood. J Am Med Assoc 317(12):1244–1251

    Article  CAS  Google Scholar 

  • Reyes JW (2015) Lead exposure and behavior: effects on antisocial and risky behavior among children and adolescents. Econ Inq 53(3):1580–1605

    Article  Google Scholar 

  • Robbins M, Saunders J, Kilmer B (2017) A framework for synthetic control methods with high-dimensional, micro-level data: evaluating a neighborhood-specific crime intervention. J Am Stat Assoc 112(517):109–126

    Article  MathSciNet  CAS  Google Scholar 

  • Salkever DS (1995) Updated estimates of earning benefits from reduced exposure of children to environmental lead. Environ Res 70:1–6

    Article  CAS  PubMed  Google Scholar 

  • Severtson D, Baumann L, Brown R (2006) Applying a health behavior theory to explore the influence of information and experience on arsenic risk representations, policy beliefs, and protective behavior. Risk Anal 26(2):353–368

    Article  PubMed  Google Scholar 

  • Shadbegian R, Guignet D, Klemick H, Bui L (2019) Early childhood lead exposure and the persistence of educational consequences into adolescence. Environ Res 178:108643

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  • Sussman J, Hayward R (2010) An IV for the RCT: using instrumental variables to adjust for treatment contamination in randomised controlled trials. BMJ 340:c2073

    Article  PubMed  PubMed Central  Google Scholar 

  • Transande L, Liu Y (2011) Reducing the staggering costs of environmental disease in children, estimated at $76.6 billion in 2008. Health Affairs 30(5):863–870

    Article  Google Scholar 

  • Triantafyllidou S, Edwards M (2012) Lead (Pb) in tap water and in blood: implications for lead exposure in the United States. Crit Rev Environ Sci Technol 42(13):1297–1352

    Article  CAS  Google Scholar 

  • U.S. EPA (2013) Integrated science assessment for lead. EPA Office of Research and Development, National Center for Environmental Assessment, Research Triangle Park, NC. EPA/600/R-10/075F.

  • U.S. EPA (2019) Strategies to achieve full lead service line replacement. October. EPA 810-R-19-003

  • U.S. EPA (2021a) Economic analysis for the final lead and copper rule revisions. Office of Water, EPA 816-R-20-008

  • U.S. EPA (2021b) LCRR community roundtable summary—Newark NJ. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0273. Accessed 8 December 2022

  • U.S. EPA (2021c) LCRR community roundtable summary—Newburgh NY. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0275. Accessed 8 December 2022

  • U.S. EPA (2021d) LCRR community roundtable summary—Milwaukee WI. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0273. Accessed 8 December 2022

  • U.S. EPA (2021e) LCRR community roundtable summary—Malden MA. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0274. Accessed 8 December 2022

  • U.S. EPA (2021f) LCRR community roundtable summary—Washington DC. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0278. Accessed 8 December 2022

  • U.S. EPA (2021g) LCRR community roundtable summary—Pittsburgh PA. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0087. Accessed 8 December 2022

  • U.S. EPA (2021h) LCRR community roundtable summary—Chicago. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0281. Accessed December 8 2022

  • U.S. EPA (2021i) LCRR community roundtable summary—Flint and Detroit MI. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0284. Accessed 8 December 2022

  • U.S. EPA (2021j) LCRR community roundtable summary—Benton Harbor and Highland Park MI. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0277. Accessed 8 December 2022

  • U.S. EPA (2021k) LCRR community roundtable summary—Memphis TN. Aug. 23. https://www.regulations.gov/document/EPA-HQ-OW-2021-0255-0279. Accessed 8 December 2022

  • U.S. EPA (2022) Guidance for developing and maintaining a service line inventory. Office of Water (4606M), EPA 816-B-22-001, August

  • Wallander S, Ferraro P, Higgins N (2017) Addressing participant inattention in federal programs: a field experiment with the conservation reserve program. Am J Agr Econ 99(4):914–931

    Article  Google Scholar 

  • White House (2021) Fact sheet: the biden-harris lead pipe and paint action plan

  • Zartarian V, Xue J, Tornero-Belez R, Brown J (2017) Children’s lead exposure: a multimedia modeling analysis to guide public health decision-making. Environ Health Perspect 097009-1–097009-10

Download references

Acknowledgements

The views expressed in this paper are those of the authors and do not necessarily represent those of the U.S. Environmental Protection Agency (EPA) or of Trenton Water Works. In addition, although the research described in this paper may have been funded entirely or in part by the U.S. EPA, it has not been subjected to the Agency's required peer and policy review. No official Agency endorsement should be inferred. The protocol for the randomized controlled trial was approved by the University of Chicago Institutional Review Board. The authors thank Trenton Water Works for providing the data and participating in the project; Caitlin Fair, East Trenton Collaborative, and Farrah Gee, City of Trenton Department of Housing and Economic Development, for providing information about Trenton lead service line replacement grant programs; Emma Hopkins, EPA ORISE research fellow, for research assistance; Ludovica Gazze, University of Warwick, and Jackson Reimer, Wharton School at the University of Pennsylvania, for valuable comments and contributions; Erik Helm, EPA Water Economics Center, and Michael Goldberg, Amina Grant, and Kory Wait, EPA Office of Ground Water and Drinking Water, for helpful comments on previous drafts.

Author information

Authors and Affiliations

  1. U.S. EPA, Office of Policy, National Center for Environmental Economics, Washington, USA

    Heather Klemick, Ann Wolverton & Bryan Parthum

  2. CDM Smith, Edison, New Jersey, USA

    Kristin Epstein & Sandra Kutzing

  3. Accenture Federal Services, Arlington, Virginia, USA

    Sarah Armstrong

Authors
  1. Heather Klemick
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ann Wolverton
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Bryan Parthum
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Kristin Epstein
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Sandra Kutzing
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Sarah Armstrong
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

HK is the primary and coordinating author. KE and SK provided insight into the LSLR program, data, and local knowledge of the study region. All authors contributed equally to the research design and implementation of the RCT. HK, AW, and BP analyzed the data, estimated the regression models, and developed the analysis. HK, AW, BP, and SA wrote the paper.

Corresponding author

Correspondence to Heather Klemick.

Ethics declarations

Competing interests

H.K., A.W., and B.P. declare they have no financial interests. S.K. and K.E. are employed at CDM Smith, which was a paid contractor of Trenton Water Works, and K.E. was employed at Trenton Water Works when the study was initiated. This study was not part of Trenton Water Works’ contract with CDM Smith, nor did the results of the study affect the contract. S.A. worked at the University of Chicago Energy and Environment Lab, which received funding from the U.S. Environmental Protection Agency, when the study was initiated, and currently works at Accenture Federal Services, which also receives funding from the U.S. Environmental Protection Agency; that funding did not support this study, nor was it affected by this study’s results.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Klemick, H., Wolverton, A., Parthum, B. et al. Factors Influencing Customer Participation in a Program to Replace Lead Pipes for Drinking Water. Environ Resource Econ 87, 791–832 (2024). https://doi.org/10.1007/s10640-023-00836-9

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10640-023-00836-9

Keywords

Search

Navigation