Technologies are tools, and humans determine their net social and environmental impacts by how they are designed and deployed. This is true for the Internet of Things, a poorly understood set of information and communications technology (ICT) solutions that are rapidly proliferating throughout society. Often joined with artificial intelligence, IoT is the enabling technology undergirding everything labeled as “smart”—homes, buildings, transport, cities. Think of a network of sensors, edge computing devices, and data gateways connected via the cloud and data centers. “Digitalization” is a term often applied to this phenomenon.
The environmental and sustainability benefits that IoT delivers have been well publicized, particularly by ICT vendors jockeying for a bigger share of a growing market. Smart homes and buildings, intelligent transportation, precision agriculture, industrial controls, electricity grid resilience, “digital water” . . . the list goes on. A common thread in all these applications is the ability of IoT to turn data into actionable analysis. The International Energy Agency has shown how IoT and other forms of digitalization can be applied to improve the efficiency and lower the climate impact of our energy system.
The potential negative effects of IoT have also received scrutiny, including rising end-of-life e-waste and direct energy consumption. Analysts have also highlighted potential rebound effects, whereby increased energy efficiency and resulting cost savings can lead to increased energy consumption in the long run. Analysis of energy rebounds by the American Council on an Energy Efficient Economy typically minimize the size of such effects, but the potential remains nonetheless.
There is a long history of ICT scary stories, especially concerning predictions of future energy consumption trends. Dating back to the California energy crisis of 2000, which some analysts errantly blamed on the growth of data centers, various “experts” have made claims that ICT devices collectively will consume most or all of the available electricity by some date in the mid-term future. Data centers have garnered the most criticism, although a recent report from Lawrence Berkeley lab shows U.S. data center electricity consumption has leveled off in recent years despite an explosion in the amount of data being processed. More recently, alarms have been raised about the energy threat posed by billions of IoT sensors projected by some date in the future.
A good analytical frame for evaluating the balance of IoT’s positives and negatives are the complementary metaphors of footprint and handprint. The footprint is the direct negative impact (energy, water, climate change) of any person, company, or society. Handprint refers to the enabling impact that technologies can have in helping a person, company, or society to reduce their footprints. ICT technologies, including IoT, definitely have a footprint, but they also present handprint benefits, perhaps more than any other sector of the economy.
Society’s goal should be to minimize IoT’s footprint and maximize its handprint. That comes down to technology design and public policy. The IEA several years ago convened the Connected Devices Alliance, a consortium of governments and ICT companies, to focus on both. One work product of the CDA is a set of “Design Principles for Energy Efficient Connected Devices” that features 10 recommendations for how IoT and other ICT device makers can minimize the energy footprint of networked devices. In parallel, the CDA issued a set of “Policy Principles for Energy Efficient Connected Devices” that highlight how policymakers can promote handprint innovations and help grow the market for IoT and other network markets.
Several groups are focused on the net benefits of digitalization. ELI itself has convened a conference and series of webinars under their Green Tech banner, with discussions focused on how smart public policies can maximize the net environmental benefits of technology. The Digital Climate Alliance, a coalition of leading ICT companies, has been promoting enabling digitalization policies in legislation on Capitol Hill. By leveraging existing resources, companies and governments alike can push for IoT to be used for good.
The Internet of Things is not a single technology but a movement, an organized campaign for the massive adoption of new standards. Sociologists and anthropologists of information technology have found that movements have a vanguard of enthusiasts and early adopters urging their organizations on the bandwagon—and sometimes confronting a counterforce of skeptics. Careers can depend on which side prevails. But movement thinking is no substitute for imagining all that can go wrong.
Neither evangelists nor agnostics can always foresee the ultimate, often unintended positive and negative consequences of new systems. Over the decades, good guys and bad guys can be reversed. Historians of transportation have reminded us of how utopian the private internal combustion engine once appeared, a solution to the health menace of horse manure and even of dead horses on city streets. In small numbers, automobiles seemed positively benign. Bicyclists had cried out for better roads, helping create cyclist-unfriendly thoroughfares. If the new vehicles began to erode streetcar use, many progressive writers applauded this blow to monopolists. Remember the song for Charles Foster Kane in the film? He “has the traction magnates on the run.” Railroads then became environmentally friendly again, until (as the New Yorker recently reported) protesters in England have been digging and living in tunnels to prevent damage to a historic forest by a new high-speed line. And vaping, promoted as high-tech harm reduction, has become a new youth addiction.
Sometimes the skeptics turn out to be wrong. The rebound effect is a special case of what safety engineers have called risk compensation, the tendency of people living in greater safety to seek out new risks unconsciously. Early in the introduction of mandatory automotive seatbelts some libertarians claimed that a sense of security made buckled-up drivers a greater danger to pedestrians. Later studies showed that seatbelts clearly reduced vehicular deaths. Risk compensation sometimes happens; people in safe financial jobs may seek out “adventure travel.” There is a whole book devoted to volcano tourism. But compensation is no iron law.
The real issue for the IoT movement is the unprecedented complexity and fragility of interdependent systems. While many people consider malware the problem, it is not the underlying weakness of the Internet of Things. That instead is a structural problem that first drew attention in the nuclear meltdowns in Three Mile Island and Chernobyl: dangerously fragile links among processes. The Yale sociologist Charles Perrow formulated this analysis in a classic book, Normal Accidents, in 1988. Conventional nuclear reactors are a classic example of tight coupling. A failure in one part of the system can create a disastrous cascade of reactions. Supply chain disruptions during the Covid-19 pandemic show what happens when individual nodes of a tightly coupled process are closed down, more than cancelling the intended efficiency of lean global organization. Shipboard safety systems have induced so-called “radar-assisted collisions,” like the error that doomed the Italian luxury liner Andrea Doria in 1956.
Fortunately the history of technology suggests at least three ways to mitigate the risks of IoT. One is redundancy. Many advanced aircraft are controlled by multiple independently manufactured and programmed computers that compare results. The inevitable glitches in individual systems are outvoted. Another is firebreaks. When Tokyo was the world’s largest city in the 18th century it was notoriously fire-prone. The shoguns decreed wide streets and ordered waterways to interrupt the spread of fire. IoT systems should be able to continue functioning if they need to be temporarily disconnected from each other. That points to a third strategy. People must maintain the skills they will need when systems are periodically disrupted. Like commercial airline pilots today, driverless car owners may need to practice on simulators. Even in tomorrow’s networked everything, human attention must still be paid.
In 1989, the president of interop, a networking conference, gave engineer John Romkey a challenge: connect a toaster to the nascent internet, just becoming a part of popular culture. If successful, the engineer would get star billing at the nextconference. Romkey and his friend Simon Hackett showed up the following year with a Sunbeam toaster, a simple information database, and a flair for showmanship. They took center stage to grill a slice of bread using a single command. With this innovation, a piece of toast came to represent a slice of our future.
Ten years later, during the dot-com era, sociologist Neil Gross predicted what would come next if we were to connect almost everything to the internet. “In the next century, Planet Earth will don an electronic skin. It will use the internet as a scaffold to support and transmit its sensations. This skin is already being stitched together. It consists of millions of embedded electronic measuring devices: thermostats, pressure gauges, pollution detectors, cameras, microphones, glucose sensors, EKGs, electroencephalographs. These will probe and monitor cities and endangered species, the atmosphere, our ships, highways and fleets of trucks, our conversations, our bodies—even our dreams.”
Termed the Internet of Things, or IoT, this technology is now real. It is modernizing our businesses, cities, transportation systems, energy grids, and agriculture. It is also being proposed as the next big thing to confront our most pressing environmental challenges. There are an estimated 25 billion devices connected to the internet. The economic impact of this network, measured as value added, could be anywhere from $3.9 trillion to $11.1 trillion per year in 2025, according to the McKinsey Global Institute.
While all the hype around IoT’s economic potential is warranted, we seem to have brushed over the environmental costs––specifically, the unwanted counter-effects resulting from increased efficiencies and access to information, or what is now referred to as digital rebound. For example, how much energy is consumed by IoT devices, and how does this compare across applications? What is the indirect energy impact of IoT networks at data centers? How is IoT impacting our everyday decisions and long-term behaviors? How can we ensure that this economic boom doesn’t inadvertently become an environmental bust, consuming more energy than it saves and creating other perniscious effects? These questions are complex and involve concepts very difficult to measure or predict.
The Network for the Digital Economy and Environment is answering these tough questions. What we call nDEE is an initiative of ELI’s Innovation Lab, Yale’s School of the Environment, and the Center for Law, Energy, and the Environment at Berkeley law school. With limited empirical research on the environmental costs of IoT, there can be no action taken by businesses, technology developers, or policymakers to ensure the responsible development and deployment of this technology. The nDEE seeks to build a multidisciplinary coalition to produce research that will expand our understanding and encourage actions and policies that harness the benefits of IoT while mitigating its harms.
While IoT devices and their systems are incredibly diverse in their settings and applications, the technological structure is inherently the same, involving layers of perception, networking, and computing. Perception occurs through built-in sensors, networking happens through wireless connections, and computing translates data into specific services required by users. As IoT develops in unexpected ways, this structure will remain largely unchanged. The ubiquity of IoT, combined with its ability to connect with systems and devices anywhere, makes it uniquely powerful.
Smart transportation, for instance, is not only the fastest growing application of IoT, but it will benefit greatly when there is detailed sensory information on every vehicle on the road. The backbone consists of thousands of sensors, cameras, and Radio Frequency Identification (RFID) readers that collect data, which is transmitted through cellular routers. The system then deploys artificial intelligence to use the data to perform an action, such as changing a traffic signal due to an accident. All these components work in perfect harmony and make real-time decisionmaking possible.
With this data, IoT is already providing a multitude of functions, including real-time analysis of road conditions and congestion, finding parking spaces, and automatically paying tolls. In the future, autonomous vehicles will need to seamlessly integrate this data to plan efficient routes and ensure the safety of passengers by communicating with other IoT-enabled cars. With enough vehicles with IoT capability, some scholars predict, there will be a utopian transportation future. Traffic accidents and congestion will be almost eliminated.
Fully utilizing IoT, transportation’s greenhouse gas emissions may decrease anywhere from a low of 5 percent up to perhaps 60 percent, while fuel consumption may decrease anywhere between 30 and 90 percent. The New York City Department of Transportation is testing its Connected Vehicle Pilot Program. The department is procuring hardware and software to implement vehicle-to-vehicle, vehicle-to-infrastructure, and vehicle-to-pedestrian communication. The pilot program will demonstrate how safety-related warnings and other connected-vehicle applications can be deployed in the real world to address safety, mobility, and environmental challenges.
In a similar vein, the ability of electric grids and smart buildings and homes to communicate with each other could illuminate the balance between electricity supply and demand, leading to improved load balancing. Utilities can produce energy based on actual demand, which can refine their strategies on consumer prices and ultimately cut ratepayer costs. Conversely, consumers will be aware of the provider’s energy load and can shift use to times when electricity is cheaper. This type of temporal load balancing can reduce stress on grids during peak hours. Another type of load balancing can allow smart grids to schedule power-hungry tasks when solar and wind energy are in high supply.
However, automated load balancing at this scale is mostly theoretical. It is not known how responsive people will be to changing their habits. Some studies suggest that consumers’ energy consumption behavior is somewhat sticky and may resist rescheduling, even when certain times offer lower prices. But even relatively small cuts can add up. The Department of Energy estimates that if the electric grid were just 5 percent more efficient, the energy savings would equate to permanently eliminating the fuel and greenhouse gas emissions from 53 million cars. To take an example, the energy loss associated with many power plants can be attributed to aging infrastructure, with some assets more than 40 years old. If existing power plants were to be retrofitted with IoT systems, the expected lifetime efficiency savings would total an estimated $50 million per plant. That sounds great, but policymakers will have to consider that new IoT-based power plants of similar capacity would have an expected lifetime efficiency savings almost five times greater. IoT merely confirms the savings of new generation technologies.
With a desperate need for upgrades like this, the bipartisan infrastructure package could not have been passed at a better time. The act signals a strong push toward digitizing the nation’s utilities, transportation, and communications infrastructure. With $550 billion allocated for these upgrades, it is a given that IoT will play a key role in many, if not all, of the planned projects. The act even calls for a “Digital Climate Solutions Report” that “assesses using digital tools and platforms as climate solutions, including the Internet of Things.” No doubt there will be a plethora of opportunities for IoT. However, any assessment should give due consideration to the system-wide effects and present concerns related to the use of IoT devices. For instance, an analysis of 300 IoT applications by McKinsey found that most data from IoT devices is not used effectively. As an example, only 1 percent of data from an oil rig is regularly examined. The limited data that is actually used is mostly to control anomalies, whereas the real value lies in optimization and prediction, which would allow for significant resource savings.
Agribusinesses are also employing IoT, to reduce water consumption and fertilizer use, cut waste, and improve product quality and yield. By sensing environmental conditions like soil and air temperature, as well as humidity, cost-effective IoT devices can perform analysis such as determining the optimal time to irrigate crops or apply fertilizers or pesticides. This is particularly advantageous in controlled environments. In greenhouses, for instance, IoT devices have access to environment controls like drip irrigation systems, sprinklers to control humidity, or fans and ventilation to control temperature. According to The Nature Conservancy, such precision agriculture can enable farmers to cut water and fertilizer use by up to 40 percent without reducing yields. IoT may also find applications during harvesting, packaging, and distribution to attune farmers to the market, in hopes of reducing food waste. It is estimated that 28 percent of available farmland globally is “reserved” for food waste, as farmers commonly produce more than the market demands to avoid losing profits. Food waste on the farmer’s side is a market failure that contributes to hunger, and decomposing food in landfills is a major source of methane emissions. By tracking produce sold using RFID tags, farmers and distributors can model and predict future quantities needed in a given location, leading to an accurate understanding of demand and efficient pricing. This in turn can lead to changes in growing patterns that reduce overproduction and waste at both farms and food retailers.
It would seem that the environmental potential of IoT is unparalleled. However, policymakers need real-world piloting and testing, focused on achieving the energy and environmental resource savings that IoT promises—and avoiding its pitfalls. Even with all these benefits, IoT is not free from environmental costs. Like all electronics, the manufacturing of IoT devices is complex and resource-intensive. In fact, IoT devices are far more problematic than other electronics due to their short lifespan in situ. Battery-powered IoT devices have a limited energy supply, some just lasting a few months. Many devices are designed to fail once the battery dies. Common solutions include low-power networks and smart sleep and wake schedules. Low-power networks, however, severely limit the volume of data transmitted per day to just a few thousand bytes. Increasing the data transfer rate or using a higher-power network like 5G would drastically reduce the lifetime of IoT devices. Added computation complexity, such as security and privacy protections through data encryption, also contributes high energy overhead, resulting in a significant tradeoff between performance of IoT and its environmental impact.
While IoT devices have different uses and thus different energy requirements, there are a few common functions. Powering the microprocessor and sensors and communicating with a wireless network are universal elements, and are also the main consumers of energy. Direct energy usage by IoT devices comes from batteries inside the unit, or more rarely, from the electric grid if the device is plugged in. Extremely low-energy ones may source some of their power from energy harvesters, which provide electricity from ambient sources like solar or thermal energy. Despite the fact that IoT devices generally perform more simple and specialized functions than personal computers and servers—and thus generally require less energy to function—their sheer ubiquity more than makes up for their small size.
While there are no good estimates for the total direct energy use by IoT devices, researchers have observed that while the processing power of electronics has increased steadily, energy efficiency has also doubled roughly every 18 months, a phenomenon known as Koomey’s Law. Koomey’s Law is a derivative of the more widely known Moore’s Law, which states that the transistor count on new processors—and thus, their performance—has doubled roughly every 18 months since the 1970s. Koomey’s Law could mean that even as the number of IoT devices and their processing capabilities increase, total energy use by the devices themselves could stay roughly constant for a number of years. With the number of devices expected to grow substantially, we will certainly see how this law plays out for IoT. The direct energy demands of this technology will also be determined by efficiency innovations, computational performance improvements, high-speed network technology, and intelligent sleep scheduling of devices.
The indirect energy use of IoT networks is consumed by routers, switches, and cell towers, as well as end-application devices like cloud servers and data centers. As a whole, networks and data centers consume nearly 400 terawatt-hours per year worldwide, contributing to more than 1 percent of all global electricity use. Some models predict a doubling or tripling of this energy use by the end of the decade, in part due to proliferation of IoT devices. The energy consumption of data centers did, however, plateau between 2010 and 2018, and some researchers attribute this to Koomey’s Law. But this may change based on how IoT and its supporting infrastructure develops and influences socioeconomic behaviors in the coming years.
Behavioral changes resulting from the use of IoT are the most difficult to predict and the most understudied aspect of IoT’s impact on the environment. Within the broader scope of environmental policy, scholars have theorized and observed an unexpected behavioral consequence of efficiency gains. Technological changes that increase energy, resource, or time efficiency often have the unwanted side effect of increasing overall consumption levels. This phenomenon has become known as the digital rebound effect. There are multiple ways IoT may cause a rebound effect, many of which are rooted in behavioral economics and social factors.
For instance, IoT has shown great potential to cut production costs in industry through increasing efficiency. The result is that industries can produce more goods at a lower cost. Since some of these lower costs are passed on to consumers in the form of lower prices, demand for these goods can rise. This increase is known as the income effect. In manufacturing, this means that while IoT can improve energy efficiency in the production process, these gains may be offset or outweighed by an increase in production overall, creating an energy rebound. Using IoT to improve energy efficiency can actually have an undesired impact on total energy use, or at least a smaller positive impact than expected.
Additionally, there are other environmental concerns the manufacturing process may create that aren’t balanced by efficiency improvements. For example, an IoT system in a factory may significantly reduce electricity use from machines on standby mode, decreasing the factory’s costs and resulting in increased production levels. However, the IoT system may not decrease the amount of non-energy-related pollution generated or the volume of raw materials consumed per unit. Thus, while increased electricity use from increasing production may be countered by better energy efficiency, other environmental costs may not be.
The rebound effect is also created through substitute and complement services. A good is a complement of another if the demand for one good increases when the demand for the other increases. For example, peanut butter and jelly may be complements of each other since they are often consumed together. This theory can be applied to IoT applications as well. If an IoT system supplements rather than replaces existing behavior, and thus acts as a complement rather than a substitute, then consumption may be drastically increased through both traditional activity and novel IoT activity. For instance, online shopping could be a complement to in-store shopping, and IoT may boost both types of purchases. Or, more likely, it may be found that AI models and IoT systems complement each other. As IoT systems proliferate, more AI models are trained to capitalize on the data generated from them. Training some AI models can emit as much carbon as five cars in their lifetimes. Thus, the rebound effect for complements is much greater and more likely to result in environmental backfire than substitutes. Unfortunately, the study of complements in the context of the digital rebound effect is nearly nonexistent despite its likely implications.
Another lesser-known rebound effect is the skill rebound, which in effect reduces the need for qualifications or skills to perform certain activities, thanks to digitization. With the autonomous vehicle example, driverless cars could mean anyone, regardless of age or driving ability, could get in a car and “drive,” resulting in more cars on the road.
Rebound effects seem to be the rule rather than an exception and cannot be ignored when assessing the total environmental impact of IoT or any other technological innovation. There have been early attempts at estimating the direct rebound effects of specific programs and policies, which have been found to be 10 percent or less. However, it should be recognized that these estimates are based on varied assumptions and methods, resulting in some uncertainty. More recent research is focused on assessing the accuracy of existing methodologies and proposing solutions that would ensure scientifically robust assessments. As IoT faces a constant push and pull between its positive and negative effects, sound research will be critical to our understanding of IoT’s rebound effects. These effects should be considered an open question, one that should be continually asked, especially given the rapid pace of digitalization, which has been further accelerated by the Covid-19 pandemic.
Will IoT become yet another burden on our planet, or will it be its long-awaited savior? Tipping the scales on this duality will be the policies and standards that frame the IoT ecosystem as a whole, or IoT governance. Due to the distributed, decentralized, and global nature of IoT, there are no clear governance organizations or definitive goals or guidelines. Many researchers have advocated for a distributed model of governance, where responsibility is spread both vertically by hierarchy and horizontally by geography or sector. International organizations, national governments, and individual corporations may all have a hand in managing IoT systems.
While such a complex, global, and hierarchical IoT governance system is still in the initial stages of framing, existing governance systems and institutions may help guide its development. IoT governance can rely at least partially on established entities like international standards-setting organizations, or SSOs. The International Organization for Standardization, International Electrotechnical Commission, and International Telecommunication Union play a large role in the governance of information and communications technology by setting global, generally voluntary technical standards.
The work of these organizations and others like them has led to extremely effective governance of the internet, a sector closely related in scope and nature to IoT. Thus, internet governance can be instructive for IoT governance. Governance of the internet is multilayered and hierarchal, with international standards and protocols established by SSOs (i.e., Wi-Fi, HTTP), regulations and laws enacted by governments (i.e., General Data Protection Regulation), privacy policies and technical limits set by companies, and even individual restrictions like parental controls. The same governance structure will likely be applied to IoT.
The environmental benefits and harms of IoT are often seen as just a technological issue, rather than a governance issue. Optimists believe that IoT has the potential to be a boon for the environment, so much so that they think technological improvements will eventually arc toward sustainability without the need for regulation.
The other issue is that there can be no governance without facts. Simply put, the problem is not well defined, and current research of IoT and the environment is lackluster. Many academic papers frame IoT as an economic boon, while overlooking its environmental costs. Robust research at this intersection is crucial for technological improvements. Good governance will thus catalyze research that provides powerful empirical data on IoT’s second-order and third-order impacts; promote the exploration of methodologies to better understand and estimate the system-wide impacts; and facilitate an inclusive and interdisciplinary community of practice.
Despite the proliferation of billions of IoT devices since the 1990s, researchers and industries have only recently begun to pay attention to their large-scale benefits and harms. Predictions that IoT can single-handedly save or destroy the environment are at the very least premature. Some IoT applications will persist and propagate, while others will enjoy only momentary hype. The ones that prove durable will have effects beyond those that were intended and be subject to diverse and global economic, behavioral, and cultural influences.
The current research gap, particularly in the quantifiable environmental effects and long-term direction of IoT, leaves much of the future of IoT and the environment up to fate. But because we have yet to define its future, this destiny is malleable. The study and discussion of IoT today will be critical in developing the fundamental capabilities and priorities of IoT tomorrow. We still have time to ensure that the environmental impact is not a side effect, but a primary feature of the IoT revolution. TEF
LEAD FEATURE The Internet of Things brings billions of electronic devices into our daily activities, the places we live, and the natural environment. Do we know if we’re making the planet smarter—or outsmarting it?
At the one year anniversary of President Biden’s taking office, many are focused on what he didn’t accomplish—the passage of the Build Back Better Act, with its $550 billion investment in clean energy to combat climate change. The bill came to a halt when Senator Joe Manchin (D-WV), a key swing vote, announced his opposition.
The drama continues, as Manchin has suggested that he might back various climate provisions in scaled-back form. “The climate thing is one that we probably can come to an agreement much easier than anything else,” he said. President Biden is also managing expectations: “I think we can break the package up, get as much as we can now, and come back and fight for the rest later.”
Lost in the shuffle of what didn’t happen is the historic nature of what Biden did accomplish when he signed the bipartisan Infrastructure Investment and Jobs Act in November, which the White House accurately described as “a once-in-a-generation investment in our nation’s infrastructure and competitiveness” that will help “tackle the climate crisis.”
Environmental practitioners understand that infrastructure policy is climate policy. You can’t have one without the other. You can’t have electric vehicles without charging stations; you can’t have wind and solar without new transmission lines; and you can’t sequester carbon without pipelines to deliver the captured CO2. The IIJA makes historic investments and changes in all these areas, which will put our system of environmental review and permitting to the test and keep environmental practitioners busy for years to come. Here are only a few of the highlights.
Transmission lines: The IIJA seeks to bring more renewable energy to the grid by making massive investments in transmission infrastructure and technology and enhancing Department of Energy regulatory authorities. For example, the IIJA enhances DOE’s authority to designate National Interest Electric Transmission Corridors and it adds to the Federal Energy Regulatory Commission’s authority to issue construction permits for transmission lines within those corridors, overriding state law impediments where necessary.
Critical minerals: Also vital to the transition to a less carbon-intensive economy is access to “critical minerals” such as copper, lithium, nickel, cobalt, and rare earth elements used in items such as electric vehicle batteries and solar panels. Currently, the United States is heavily dependent on foreign sources. The IIJA attempts to increase domestic production by providing funding, enacting key regulatory and policy changes, and improving the slow pace of permitting for mining projects on public lands.
Offshore wind: The administration has set an ambitious goal of deploying 30 gigawatts of offshore wind energy by 2030. The IIJA supports this goal by, among other things, providing funding for upgrades to ports and the electric grid that are essential to unlocking offshore wind as a reliable energy source.
Hydrogen: According to the International Energy Agency, “hydrogen is an increasingly important piece of the net-zero emissions by 2050 puzzle” because it is a versatile energy carrier that can be used as a fuel option for a variety of applications. To help realize hydrogen’s potential, the IIJA defines “clean hydrogen,” establishes a charging and fueling infrastructure grant program, and requires the secretary of energy to develop a national clean hydrogen strategy and roadmap.
Carbon capture: The technique known as carbon capture and underground storage refers to the process of snaring CO2 from emissions sources and either reusing it or storing it permanently underground. Congress has expressed the belief shared by many that large-scale deployment of CCUS “is critical for achieving mid-century climate goals.” To that end, the IIJA increases DOE’s funding and authorities. Notably, the law also clarifies that the secretary of the interior may grant a lease, easement, or right-of-way for long-term sequestration on the outer continental shelf, thus resolving a longstanding legal question.
NEPA review: Whether the IIJA’s historic level of investment in infrastructure will facilitate an energy transition depends in part on the speed with which federal agencies can get money out the door and projects constructed. The IIJA’s amendments to Title 41 of the Fixing America’s Surface Transportation Act (FAST-41), which establishes an interagency coordination process for certain “covered” infrastructure projects, are intended to further expedite this work. The IIJA enhances the FAST-41 process by, for example, expanding applicability and imposing accelerated deadlines for agency reviews. The new law reflects a bipartisan acknowledgment that further streamlining of environmental review and permitting regimes is key to facilitating the needed energy transition.
New Infrastructure Law Will Require Help to Achieve Goals
For the United States to quickly eradicate harmful lead in drinking water, we need full federal funding and a hard deadline to get every lead service line pulled and replaced with safer tubing.
I applaud the Biden administration’s Infrastructure Investment and Jobs Act, a bill that would provide $15 billion in dedicated funding for lead service line replacement. This will give America’s older cities an opportunity to replace lead pipes to protect the health of residents, especially children.
While lead services lines are found throughout the country, they are mostly clustered in neighborhoods with older homes and multi-family units, and therefore disproportionately impact communities of color. With full federal funding and a mandate for cities to set replacement timetables to qualify for such support, we may soon see the eradication of lead in minority communities — an outcome we have already achieved in Newark.
To date, we have replaced more than 21,000 known lead lines in 30 months, an unprecedented achievement. The project continues to test homes without recorded lead lines to make sure we capture them all.
We did this by developing a strategy, finding the funding, and making it a citywide priority to get our project done. Part of our operational plan included keeping residents informed and asking for their cooperation through community meetings, mailings, and robocalls. We also created an apprenticeship program within the project, providing employment in good union jobs for Newark residents.
When the city’s lead levels spiked, we immediately made water filters available to residents as a short-term fix and changed our anti-corrosion system. But from the very beginning, we knew the permanent solution was to replace all lead service lines as quickly as possible. That was our strategy: Get it done as fast as we could, and engage the residents in rebuilding the city’s infrastructure. Residents were supportive and part of the process at every turn.
With a $75 million city bond, we began replacing lead lines in March 2019, with a 10-year plan that asked each homeowner to pay $1,000 toward construction costs. I was not satisfied with this. We needed to do it faster and for free. Our amazing federal legislators helped by pushing for more resources, including the introduction of the Water Infrastructure Funding Transfer Bill, which provided more flexibility for states to fund infrastructure projects.
The game-changer came in August 2019, when the Essex County Improvement Authority backed a $120 million bond for us to accelerate the program and eliminate the cost to residents.
For a program like ours to succeed, there must be cooperation at every level of government. In our case, EPA solved the mystery of our lead exceedances by determining that our corrosion control system had waned. The New Jersey legislature passed and Governor Phil Murphy signed a law which allowed us to use public money to improve private property, and the Newark Municipal Council adopted an ordinance that allowed us to replace lead service lines without a homeowner’s permission.
This was crucial because 75 percent of Newark residents rent their homes, and many live in multi-family units built before the city outlawed lead lines in 1952. Many of these homes have absentee landlords, so tracking them down for permission would have been arduous and time-consuming.
The passage of this ordinance allowed us to replace lead lines block-by-block in an organized manner. We were able to replace as many as 100 lines a day, keeping street closures and parking interruptions to a minimum.
These important shifts in law point to the overriding philosophy of our program, which was — simply put — the will to get it done and give our residents the best drinking water in America.
To date, my administration has invested more than $190 million in enhancements to our water and sewer system, including upgrades in monitoring technology, filtration, environmental systems, and delivery infrastructure. Most were done before our first lead exceedances, and these upgrades continue today.
This will to get it done must be imposed by leadership. Newark’s Water and Sewer Director Kareem Adeem has been a force of nature, pushing his staff and our contractors to complete this project quickly and efficiently, with the least amount of inconvenience to the residents.
Essex County Executive Joe DiVincenzo’s willingness to help Newark and use the county’s AAA bond rating to secure the $120 million bond is a great example of governmental mutual aid. So was the quick passage of the infrastructure bill that let us tackle this public health problem head-on.
I hope our story inspires other governments. Full lead line replacement does not have to be an eternal infrastructure nightmare. With federal funding and imposed deadlines, as well as cooperation at all levels of governance, we have the power to eliminate lead exposure for the health and safety of current and future generations.
In its American Jobs Plan, the Biden administration set an aggressive goal of replacing all of the nation’s lead service lines in 10 years. By achieving this important target, we can protect both children’s brains and adults’ hearts from the harm caused by lead in drinking water. Using estimates from the Environmental Protection Agency, we calculate that the socioeconomic benefits exceed $22,000 per replaced LSL.
The Environmental Defense Fund has extensive experience advocating for LSL replacement, including serving as a founding member of the Lead Service Line Replacement Collaborative, a group of 28 national public health, water utility, environmental, labor, consumer, housing, and state and local governmental organizations. Based on our experience, I believe success will involve five actions.
First, the administration says that based on EPA estimates, it will take $45 billion to achieve 100 percent LSL replacement. The American Water Works Association indicates it will cost 33 percent more. The bipartisan Infrastructure Investment and Jobs Act that passed the Senate and awaits approval in the House provides $15 billion in dedicated funding through State Revolving Funds, with about half as grants and the balance as loans. LSL replacement projects may also apply for traditional SRF opportunities, but they will be less competitive without mandates under the Lead and Copper Rule, a regulation published by EPA that protects consumers from these substances.
Clearly, more funds are needed. Congress must not only finalize pending legislation but also increase the amount, including providing EPA with funds so the agency can directly support important lead pipe replacement projects where a state SRF is unwilling or unable. If communities do not receive the needed funding, they will have to raise rates to cover the shortfall, making it especially difficult for the many cities that struggle with affordability.
Second, EPA must move quickly to distribute the new dedicated funds to states, using a revised allocation formula based on the number of LSLs in a state. The agency must also help states achieve the twin goals of driving down the average cost to fully replace the lines and prioritizing environmental justice communities that are disproportionately burdened by lead risks. We need to emulate leaders like the investor-owned utility American Water and utilities in Cincinnati, Newark, Lansing, Denver, Green Bay, and Pittsburgh that have focused on both goals with significant success. And we need to question inflated cost estimates by utilities that lack a successful track record of fully replacing LSLs.
Third, EPA must follow the example set by Illinois, Michigan, and New Jersey, states with 25 percent of the nation’s LSLs, and strengthen the Lead and Copper Rule to set a timeline to eliminate the lines and ban partial LSL replacements. Partial replacements unfairly force families to choose between paying to replace the portion of the pipe on their property, or risk increased exposure to lead. With federal funding available, the deadline should be 10 years, with a mechanism to handle utilities that need more time on a case-by-case basis.
Fourth, utilities need to immediately set a goal of fully eliminating LSLs and manage their distribution systems with that goal in mind. There is no need to wait for a revised Lead and Copper Rule to force the issue. They know enough about which neighborhoods have LSLs to get started right away to train crews from those communities to replace the lead pipes and drive costs down safely and efficiently. These efforts can take place while simultaneously developing a comprehensive service line materials inventory. To support the effort, they must recognize that full LSL replacement serves a public good and, where necessary, use rates paid by customers to replace lead pipes without cost to individual property owners or residents.
Finally, community leaders, especially elected officials and public health agencies, need to support programs to eliminate both lead pipes and lead-based paint hazards so we can truly have lead-safe housing for all — without income, racial, or ethnic disparities. Key opportunities include engaging community-based organizations — Denver Water is a model here — and integrating online maps that show the likelihood of a home having either lead pipes or lead-based paint. Other effective strategies include tailoring communications so folks understand the risk from all sources of lead that threatens children’s brains and adults’ hearts.
For thirty years, the country has tried to manage LSLs through corrosion control, replacing lead pipes only as a last resort. Proactive LSL replacement needs to be an integral part of any program, with optimized corrosion control managing other sources of lead in drinking water, such as solder and brass.
No one buys or rents a home because they wanted a lead pipe. No water system wants lead pipes either. LSLs are a legacy from decades ago that the nation needs to eliminate. Rather than assign blame and waste limited resources, it’s time to get LSLs out as efficiently, safely, equitably, and quickly as possible.
When President Biden vowed to take the bold step of including lead service line replacement in the bipartisan infrastructure deal, I was elated. If enacted, this will serve to eliminate lead from the nation’s drinking water system. We all know that clean water is a human right, but we are equally aware that this is not happening in many low-income communities. As we look toward strategies to achieve this foundational goal, we must first focus on the integrity of government, equity among communities, and cost factors that might avoid or mitigate incomplete measures.
Let’s agree to not act like an ostrich and bury our collective head in the sand by studying this issue forever. We cannot continue to live in a country that repeatedly ignores violations of the Safe Drinking Water Act. Communities cannot afford the time wasted when local officials analyze what would be the best way to attack this problem without taking action. There are a number of ways to identify lead service lines, but the use of predictive modeling through tools such as artificial intelligence may be the most expedient. By beginning with environmental justice communities that are already overburdened by a myriad of inequities, we will be able to expedite solutions to a host of challenges they face daily.
As an educator in an urban community in Newark, New Jersey, I am acutely aware of the trials many of my students and their families face. A recent op-ed in The Hill by Dr. Mona Hanna-Attisha and Erik D. Olson reminds us of the negative impacts of lead exposure and the economic sense it makes to invest $45 billion in this effort. I have witnessed the deleterious effects of lead exposure in many forms in my classroom. The developmental delays are significant. Lead in water is a silent interloper in communities that are already overwhelmed with societal ills. We can and must do better.
While the plan to remove LSLs within 10 years may seem like a lofty goal to some, I submit that we can push to be proactive and complete it sooner. Local politicians and communities can work together to identify LSLs, and plan and execute their removal. By using Newark as a national model, we can see that holistic lead remediation is possible. Community pressure, coupled with the political will of all local partners, allowed the removal of over 22,000 LSLs within two years. That is unprecedented.
Education and information dissemination are imperative in this effort to help all those affected by lead contamination in water. Since toxic water is often tied to unsafe housing and cognitive delays, we must support those who are most vulnerable to lead exposure. This includes aggressive campaigns to notify at-risk community residents of LSL removal plans in several languages, media campaigns, and door-to-door efforts, among other actions to notify the public of the government’s intention to remediate the problem.
But this is not enough. There must be an inclusive plan to regularly inform the constituency of the schedule and plan to fully replace the lines. We must also implement a plan to educate the public about lead exposure and removal; in short, we must train the trainers. In communities that are most affected by lead — the disenfranchised communities — trust must be earned. That requires EPA and local governments to partner with communities in efforts that meet them where they live. Consider how expeditious and telling it would be if communities were trained to partner with governments and educate each other, as well as advocate for their right to clean water.
As I stated in an article that I co-wrote with NRDC Chief Science Officer Dr. Kristi Pullen-Fedinick in May 2020 (nrdc.org/experts/kristi-pullen-fedinick/covid-context-lead-water), lead can rob a community of its economic potential. It increases the need for social and educational services, and severely impacts the ability of a community to grow and thrive. Our president has proposed a sweeping change to the status quo. When this infrastructure bill is realized, the impact on the disenfranchised, as well as our nation as a whole, will be great.
Furthermore, if we analyze the data from EPA and note the intersection of SDWA violations with communities of color, we will acknowledge that these communities must be identified and acted upon first for LSL replacement. Strict oversight of violations and implementation of safe LSL removal and replacement will bring this plan to fruition.
$45 billion is a start; some experts believe that it will cost $60 billion. We must begin by recognizing that this is not a red or blue state issue, but a necessity for everyone to have a basic human right: clean water to drink. As Frederick Douglass stated, “Power concedes nothing without a demand. It never did and it never will.” People are beginning to understand what that means. The government needs to meet their demand for clean water with the full removal and replacement of LSLs in a swift and timely manner.
Replacing all lead pipes in the United States is compelling, but complicated. Lead pipes are the single greatest contributor to elevated tap water lead levels, or WLLs. Though not the only source of lead in public and private plumbing, where present such pipes usually dwarf the contributions from other lead sources.
The health impact alone compels us to replace lead pipes. Lead is toxic to everyone, although not all Americans have the same lead exposures. For most Americans currently, lead in drinking water probably poses the greatest exposure risk, weighing particularly heavily on low-income communities and communities of color. Even the typically low levels of lead in U.S. tap water can affect the blood lead levels of local children — that is, slightly higher WLLs translate to slightly higher pediatric blood lead levels. These levels are not benign. Research shows that even in communities where drinking water meets federal and state requirements, higher WLLs are associated with poorer outcomes in, as diverse examples, dialysis patients and children’s performance in school.
Lead’s adverse impacts are evident at even the lowest levels. Effects are detectable at every level measured, and no threshold has yet been determined. Children aren’t the only ones at risk — for example, there’s also no threshold for lead’s impact on the blood pressure of pregnant women or older men. For many of these effects, the dose-response relationship is non-linear, showing a steeper slope at lower exposures. This means that the lower lead concentrations in drinking water that serves the general population cause greater damage, not less.
Maintaining public infrastructure also provides a compelling reason to replace pipes. Except in Chicago, which mandated the installation of lead service pipes until the federal ban in 1986, most lead pipes in the United States date to the Victorian era, when cities were expanding rapidly to accommodate the Industrial Revolution. Public water systems serving burgeoning urban centers chose to install higher-grade lead pipes because they would last longer. Those lead pipes are now 100 years old and far beyond their useful lives. The American Society of Civil Engineers estimates that in the United States, a water main break occurs every two minutes. ASCE also estimates that over 20 percent of the water leaving a water treatment plant never reaches a customer due to leaking pipes. Beyond addressing public health, lead pipe replacement would also improve critical public assets.
However, these compelling arguments are complicated by costs and questions of who will assume responsibility. There are an estimated 6 to 13 million lead pipes buried in U.S. cities. Replacing them costs on average $3,000-5,000 each; again, Chicago is the outlier, with an estimated cost of $25,000 per pipe. Cities that have replaced all their lead pipes, such as Lansing, Michigan, and Madison, Wisconsin, reduced their unit costs through improvements in technology and productivity. Currently, over 100 cities in at least 13 states have committed to replacing all their lead pipes.
Remediation is complicated further by a lack of data on the number of lead pipes that exist. In a recent national survey, only 10 states could estimate how many lead pipes they have. More alarmingly, almost half of states — 23 — admitted that they still don’t know the location or number of lead pipes in their jurisdiction. (This, despite the fact that the 1991 Lead and Copper Rule required water utilities to assess the materials within their systems, including lead pipes.) States must first commit to finding the lead pipes that exist before they can begin to replace them.
Ensuring safe drinking water also requires corrosion control treatment, effective monitoring, and enforcement. Lead is a corrosion by-product; utilities must control corrosion while waiting for pipe replacement funds and to control lead from other plumbing sources. Monitoring tells water utilities how much lead is leaching into water. The fiasco in Flint, Michigan, taught us that enforcement oversight is necessary.
This brings us to the biggest complication: EPA’s and the water industry’s failure to accept responsibility for controlling lead exposures from public water. Flint got national headlines, but Flint is not unique. In 2015 and 2016, USA Today and the Natural Resources Defense Council both found that thousands of public water systems serving millions of Americans in all 50 states had lead violations. EPA’s own audits documented that over 90 percent of lead violations are not reported by states. Protecting U.S. public drinking water should be a responsibility shared by EPA, state oversight agencies, and local water utilities. But the behavior of those three entities mirrors the three monkeys: see nothing, hear nothing, say nothing. Without newspaper coverage and one valiant pediatrician, we never would have known what happened in Flint.
The Biden infrastructure bill, if passed, will provide a down payment for these efforts, but it won’t fund replacing all U.S. lead pipes. And it won’t ensure lead-safe public drinking water. Only a committed coalition of federal, state, and local leaders can ensure that — ones willing to take responsibility and act. The time for the buck to stop is now.
Our nation’s water and wastewater infrastructure is essential to public health and safety, environmental protection, and community growth. American Water appreciates the Biden administration’s recognition of the need to invest in our country’s water systems, and its focus on lead service line replacement, or LSLR.
American Water has been a leader in developing practices for addressing lead in drinking water systems for many years. As the country’s largest and most geographically diverse investor-owned water utility, we regularly plan for the future and invest in renewing our systems’ infrastructure. Our future includes leveraging new applied technologies to enhance services related to water quality, water pressure, energy efficiency, and water efficiency. Lead pipe must not be a part of that future.
What will it take to accelerate LSLR programs across the country? It will take commitment and tenacity of purpose at all levels to overcome barriers. American Water knows that addressing the issues around lead in drinking water, including removing lead service lines, is a shared responsibility. This includes customers, regulators, health officials, and water utilities, among others.
We have direct experience in answering this question. Since 2017, we have replaced about 29,000 lead lines across our service areas. Beyond financial resources, we also need to recognize that this undertaking requires skilled tradespeople, excellent communicators, safety-focused contractors, and effective construction project managers. Many water utilities possess such talents and are well positioned to drive successful LSLR programs. Individual homeowners across the country cannot do this on their own.
On a federal level, EPA’s revised Lead and Copper Rule includes a number of important improvements over the existing LCR and should accelerate future replacement programs. However, to achieve an impact level closer to 100 percent, all utilities with lead service lines — not just mandated utilities — would need to implement replacement programs.
Funding Drinking Water State Revolving Loan Fund programs, which were established under the Safe Drinking Water Act, is also key to accelerating the elimination of lead pipes. All water systems should have access to these programs, regardless of ownership type, and state programs should work to eliminate other perceived barriers that may exist. Even so, these programs are neither robust nor comprehensive enough to reach every utility with lead. Additional tools are needed if lead service lines are to be quickly eliminated across the country.
Combining utility construction dollars with state revolving loan funds is an effective strategy to accelerate this work and make a broader impact. American Water has worked with numerous state public utility commissions, consumer advocates, state and local representatives, and other stakeholders to drive solutions that allow this work to be funded within utility infrastructure replacement programs. Note that our utility LSLR programs are not driven by a LCR regulatory requirement, since we already comply with water quality standards. Rather, they are driven by a desire to serve the long-term interest of our customers and the communities they live in, and help advance the elimination of lead service lines in areas where they still exist.
Experience in multiple states has shown us that constructive regulatory practice can support LSLR programs. Legislation in Indiana and Pennsylvania serves as a notable example. These states identify customer lead service line replacements as improvements eligible for inclusion in infrastructure cost recovery mechanisms for ratemaking purposes. Another example can be found in Virginia, where American Water uses a combination of utility construction dollars and funding from a program established by the Virginia Department of Health. The department’s Drinking Water State Revolving Loan Fund program provides limited funding specifically designed to accelerate the removal of lead pipes for both public and private portions.
To make LSLR programs efficient, it is better to focus available dollars directly on funding replacements as opposed to creating elaborate programs that drive up administrative costs. Customer loan programs may initially sound attractive, but they can be costly due to customer credit issues, defaults, and the difficulty in enacting property liens. Efforts are better directed to finding solutions at the local and state levels that can help streamline LSLR programs. These actions could include bundling construction contracts, and streamlining processes for street opening permits and plumbing permits, just to name a few options.
American Water holds water quality and safety paramount. We take critical steps during the water treatment process to reduce the potential for lead to leach from pipes into the water. We are well aware that it will take time to reach 100 percent replacement, and note that there are other actions that utilities take to reduce the potential risk from lead service lines. Our experience shows that achieving these goals will require commitment and tenacity of purpose by all stakeholders involved, and leveraging the expertise and leadership of utilities will be key to this success.