Real-Time Data Was Used to Regulate Water Use. It May Have Worked Too Well.

One area where it has enormous potential is in keeping track of—and limiting the use of—scarce resources. That’s something economists and environmentalists have been advocating for years. Now, an analysis of a pilot project conducted by researchers affiliated with the University of Chicago’s Energy Policy Institute, which I help run, offers evidence of the promise—and potential shortcomings—of using real-time data to encourage conservation.

City officials in Fresno, Calif., were grappling with how to conserve water in the face of a historic drought. The traditional technique: water cops. The city’s inspectors would drive around to spot households watering their lawns during times when watering was prohibited. But when the researchers analyzed historical data collected by the city, it showed that water cops were missing most of the violations. In 2016, for example, 68% of households illegally watered their lawn at least once during the summer months. Yet, the part-time water cops issued fines to just 0.4% of households, according to the analysis. 

The system wasn’t working. So, city officials joined with the research team to test a new approach during the summer of 2018. Under the new system, smart water meters were used to detect who was using water during prohibited times, and that data was then used to enforce the rules. Those who violated the rules were first warned, and then fined—with fines increasing with each additional violation. Since this was set up as an experiment to see what worked and what didn’t, a segment of households continued to be monitored by the water cops.

The new system of automated detection worked, the researchers’ analysis found. The share of households fined for noncompliance increased sharply to 14%. Perhaps more important, many residents changed their habits. The threat of being detected caused more households to comply with the water-conservation rules, with total violations dropping by 17%. With more people following the rules, the city was able to decrease water consumption by about 3% over the summer. 

The results of the experiment point to the immense promise of low-cost monitoring paired with enforcement. In fact, had the program been scaled and had the water savings measured in the months after the experiment persisted at the same levels, it would have saved the city 394 million gallons of water annually, according to the research team. Their analysis suggests that if scaled citywide, the pilot project would have achieved one-fifth of the annual reductions in residential water use that Gov. Gavin Newsom encouraged from California residents.

Yet, despite these impressive results, the program wasn’t scaled.

The researchers say city officials received a swarm of complaints from violators and didn’t renew the pilot. The researchers’ conclusion: The monitoring technology was too good at detecting violations, and the pressure from complaints made it difficult for policy makers to move forward.

(Editor’s note: Water utility officials in Fresno declined to comment. The city of Fresno’s communications office didn’t respond to requests for comment.)

The lesson from this experiment is that regulators will need to strike a fine balance if they want to use technology to enhance enforcement. Many environmental challenges, including California’s recurring droughts, are only becoming more acute, and historical approaches to enforcing environmental rules that could help are already failing to keep up. The evidence shows that technology can make a significant impact. And yet, an important lesson from this pilot program is that using real-time data to enforce regulation in a consistent and effective way, increasingly a possibility due to technological advancements, may be politically infeasible—even when the net benefits are very high. Individuals, households, and industry may all have an interest in something less than regulation that is 100% effective.

Write to Mr. Ori at [email protected].