In the wake of the Flint, Michigan water crisis, a significant shift has occurred in how public water systems across the United States report lead contamination levels, as revealed by pioneering research led by the University of Massachusetts Amherst. This study sheds light on the previously unnoticed susceptibility within the nation’s water monitoring framework—specifically, the manipulation of lead level data around critical federal thresholds. Through the development of sophisticated statistical techniques, researchers are now able to discriminate between genuine data rounding practices and intentional “threshold manipulation,” an advancement that challenges prior assumptions in environmental data reporting.
At the core of this groundbreaking research is the realization that traditional analytics have often conflated natural rounding conventions with deliberate data distortion. Assistant Professor Tihitina Andarge from UMass Amherst clarifies that their innovative approach meticulously separates these phenomena, unveiling a more precise understanding of reporting behaviors among water systems. Unlike existing methods, their statistical models rigorously assess the likelihood that reported lead levels cluster suspiciously at just-beneath-threshold values, thereby identifying potential incentives to underreport contamination levels.
Focusing on the decade from 2011 to 2020, the interdisciplinary team comprising Andarge, Professor David A. Keiser, Dalia Ghanem of UC Davis, and Gabriel E. Lade from Ohio State University, thoroughly analyzed lead concentration data submitted under the Environmental Protection Agency’s Lead and Copper Rule (LCR). This rule mandates public water systems to measure whether the 90th percentile of lead in water samples exceeds established federal limits, prompting remediation and public alerting if thresholds are breached. The self-reporting nature of this regulatory framework, however, creates fertile ground for statistical anomalies amounting to threshold avoidance, a vulnerability this study sought to explore.
Approximately 50,000 water utilities nationwide rely on self-monitoring and reporting lead levels in their distribution systems. Two critical benchmark concentrations—0.005 milligrams per liter and 0.015 milligrams per liter—dictate the frequency of testing and necessity of corrective action. Those exceeding the lower threshold are subject to intensified testing protocols, while surpassing the higher threshold triggers costly infrastructure interventions and mandatory public notifications. This dual-tiered regime inadvertently incentivizes utilities to strategically report lead levels just below regulatory cutoffs, thereby evading stricter oversight.
Prior to the Flint crisis, statistical analyses identified that nearly 3% of medium-sized water systems and half a percent of smaller systems suspiciously reported lead concentrations precisely matching federal thresholds. Such precise rounding at pivotal points strongly suggests deliberate threshold manipulation rather than arbitrary measurement imprecision or natural data variations. Interestingly, while small systems exhibiting this pattern were predominantly located in Alabama, the medium-sized systems exhibiting manipulative tendencies were geographically dispersed, suggesting a systemic issue within water reporting protocols.
The national outrage and emergency declaration surrounding Flint’s toxic water crisis in 2016 catalyzed sweeping changes in regulation and compliance enforcement. Post-Flint data illustrates a remarkable disappearance of these problematic clustering patterns, indicating heightened transparency and improved adherence to accurate lead level reporting. The EPA responded decisively with new guidance discouraging dubious compliance practices, including the avoidance of testing higher-risk homes and manipulation of sampling procedures, aiming to restore public trust and mitigate underreporting risks.
David Keiser emphasizes the imperative of ensuring water systems measure lead concentrations accurately, as reliable data is crucial for enabling timely corrective action and protecting public health. By exposing the prior vulnerabilities in the self-reporting system, this study places focus on the fragile interplay between regulatory thresholds and data integrity, highlighting how loopholes can undermine environmental safety efforts without vigilant oversight.
While the research stops short of accusing any entities of deliberate fraud, it underscores essential weaknesses inherent in the current framework of drinking water quality surveillance. The authors caution that without sustained federal attention and methodical scrutiny, some water utilities may revert to exploiting incentives to understate lead levels, thus placing populations at renewed health risks.
In response to evolving risks and knowledge, the EPA revised the Lead and Copper Rule in 2021 and again in 2024, broadening its coverage to encompass over 90% of the U.S. population served by public water systems. These updates aim to tighten monitoring requirements and close loopholes to prevent recurrence of lapses seen before Flint. Given that even low-dose lead exposure is correlated with neurodevelopmental deficits in children and cardiovascular disease in adults, the stakes of accurately reporting lead levels remain extraordinarily high.
Beyond the immediate implications for water quality monitoring, Keiser points out that the advanced statistical methodologies developed could have far-reaching applications in addressing threshold manipulation across disparate domains. These include monitoring ambient air pollution levels or even scrutinizing academic testing results where similar incentives might distort reported figures. This cross-cutting utility highlights how sophisticated data scrutiny can enhance transparency and accountability in regulatory environments broadly.
Funded by the National Institutes of Health, this influential work published in the American Economic Review: Insights not only illuminates previously hidden patterns in environmental reporting but also pioneers an analytical toolkit that future researchers and policymakers can leverage to safeguard public welfare. As environmental data grows increasingly central to public health and policy discussions, innovative approaches like these will be critical to detect and deter subtle forms of manipulation that can compromise system integrity.
Subject of Research: Not applicable
Article Title: Threshold Manipulation in Lead Level Reporting by U.S. Public Water Systems: Insights Post-Flint Crisis
News Publication Date: 1-Sep-2025
Web References:
- American Economic Review: Insights article DOI link: http://dx.doi.org/10.1257/aeri.20240258
- EPA Lead and Copper Rule: https://www.epa.gov/dwreginfo/lead-and-copper-rule
- EPA Safe Drinking Water Act: https://www.epa.gov/sdwa
- EPA 2016 Sampling Guidance: https://www.epa.gov/sites/default/files/2016-02/documents/epa_lcr_sampling_memorandum_dated_february_29_2016_508.pdf
References:
Andarge, T., Keiser, D. A., Ghanem, D., & Lade, G. E. (2025). Threshold Manipulation in Environmental Data Reporting: Evidence from Lead Levels in U.S. Public Water Systems. American Economic Review: Insights. DOI: 10.1257/aeri.20240258
Keywords: Economics research, Environmental issues, Public policy, Lead contamination, Data manipulation, Environmental monitoring, Safe Drinking Water Act, Statistical methods
