Skip to main content

Keystone XL Spotlights History of Environmental Analysis

In April, the Obama administration once again punted on whether or not to approve the Keystone XL oil pipeline. The State Department, which is responsible for permitting the pipeline because it crosses an international border, postponed the decision, saying that more time was needed because of ongoing litigation that could alter the pipeline's route.

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


In April, the Obama administration once again punted on whether or not to approve the Keystone XL oil pipeline. The State Department, which is responsible for permitting the pipeline because it crosses an international border, postponed the decision, saying that more time was needed because of ongoing litigation that could alter the pipeline’s route.

The project was hamstrung for years as State worked on the pipeline’s environmental impact statement, which was released in January. The EIS process was established in 1969 when Congress passed the National Environmental Policy Act (NEPA), which forces agencies to consider the environmental consequences of their decisions.

The EIS for Keystone XL polarized supporters and opponents, with business interests praising it and environmental groups criticizing it. To get a better understanding of the history and controversies surrounding one of our nation’s oldest environmental laws, I asked Rutgers University’s Michael Greenberg, author of The Environmental Impact Statement After Two Generations: Managing Environmental Power, several questions about NEPA and its history. His responses have been edited for brevity.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


There has been some confusion over what an EIS does and doesn’t do. Could you clarify?

The EIS is not a mechanism to force a decision, as some might wish. As mandated by law, each EIS is designed to discuss several environmental concerns: the environmental impact of the proposed action; adverse environmental effects that cannot be avoided if the action occurs; alternatives to the proposed action, including no action; and the relationship between local short-term uses and long-term productivity.

We’ve had several decades experience in this country doing EISs. Are we any better now that when we first started?

Much better. We have improved statistical, sampling and mapping tools. Now we have a scoping stage that can be helpful by getting meaningful stakeholder input, and finding out the questions that should be asked so that we get good answers. We also now have much more cooperation among federal agencies, and EPA guidelines on how to develop data to answer certain questions.

The first EISs were short and in many cases not too informative. Some later ones have become massive, and perhaps contain too much data; they can overwhelm the reader, even obfuscate the important impacts.

Putting the data aside, the major limitation of the EIS mechanism has not changed. The EIS forces agencies to generate data and consider ideas; it does not require the head of the federal agency to accept the scientific-grounded outcomes of an EIS. In other words, it provides information but is not the final decision.

For example, assume we are building a new road that people want, but the road is going to go either through or around a mountain. A scientific analysis may conclude that going through the mountain will cause the least environmental impact. But the agency might chose to go around the mountain, and thereby upset farmers close by, because it costs too much to go through the mountain.

So the process for doing an EIS has become more uniform over the years or is there wide divergence across entities?

In some ways yes, and other ways no. Like all of us, agencies learn from mistakes. Whatever was done to make a decision and avoid controversy becomes the guidelines for the next one. I think the scoping process has helped all of them avoid being surprised, but not always.

On the other hand, reducing the size of each EIS and moving forward rapidly puts pressure on the agencies and the analysts to perhaps be somewhat less thorough than they otherwise might be. And some proposed actions are reviewed with an Environmental Assessment (EA), which is a shorter, less thorough EIS. I wonder how often those decisions were the right call.

The issue here is time and cost. The EA is used about 10 times as often as the EIS because EISs have become very time consuming and expensive. Sometimes full-blown EISs are not warranted.

We have to rely on the integrity of the lead agency and their partners to decide what level of analysis is appropriate. As you know, however, these days, many people do not trust the federal government.

The State Department has been criticized on the Keystone Pipeline EIS for hiring a contractor with conflicts of interest. Is it common for agencies to hire third parties to do this work, and does it inevitably lead to concerns about conflicts of interest? Can such concerns be abated?

Given the massive size and specialized expertise required to prepare an EIS, you need to work with the best experts, and most of them have a history of working with government and private clients. For example, I have worked for a private utility that was trying to locate a nuclear plant; that application failed partly because our research showed that the site was not a good choice.

I worked for the Nuclear Regulatory Commission to assess siting issues, most notably to figure out why was the population was growing more than expected in some areas hosting nuclear power plants. The answers were tax payments to the local governments that reduced property taxes and increased local services. These two examples illustrate that analysts can work for several parties involved in the EIS process.

As a reporter you should ask me to describe my history, but I would urge you not to assume that my work is biased without first reviewing what I have done. In every environmental assessment the alternatives have pros and cons. Every expert should be able to discuss both. If they cannot, then I would worry about their competence or motives. I would then ask more pointed questions.

Also, contractors should not go off into a corner and write whatever they feel like. The agencies are supposed to oversee the work. They should prevent the contractors from deviating from the objectives or biasing the data. I like to examine the extent to which scientists without a stake are involved in reviewing the process, from start to completion.

About a week before releasing their EIS, the Wall Street Journalquoted a State Department official saying that the report would be “crafted in a way that gives the president wide leeway to make a decision.” Another official said, “[T]he report is expected to be relatively vague, so Mr. Obama would be able to cite it to support a decision for or against the pipeline.” Is it normal to have an EIS take political matters into consideration?

The EIS is not supposed to be a political analysis. It is supposed to be a set of scientific analyses that inform the political process. I didn’t see this EIS. However, as noted above, the Secretary of State has the power to decide from among any of the options considered in the EIS. So the statement need not have been made.

I don’t know what “relatively vague” means. A vague EIS sounds like an oxymoron to me. I have seen EIS statements that are thousands of pages long. You could criticize them for having too much information, but vague is not one of the labels I would attach to EISs that I have read or worked on.

Does an EIS vary depending on the entity that puts it together—a state or federal agency, for instance? In other words can an EIS looking at the exact same thing come out dramatically different depending on who is in charge?

Different federal departments have developed their own processes and cultures for preparing EISs. For example, the Department of Transportation (DOT) and Department of Energy (DOE) have been writing them for many decades, and they have their own style. An agency should theoretically not be able to run away with an EIS and do whatever it wants.

I call to your attention the EIS for the Alaska oil pipeline. Multiple agencies were involved. The Alaskan oil pipeline was under the Department of the Interior but DOT, EPA, US Coast Guard, and Army Corps all played roles. As I recollect, various departments from the State of Alaska were involved also. These large EISs are supposed to be multi-party assessments.

Several members of Congress have sought to remove the pipeline EIS process from State. In one example, a law was debated to give the authority over to Commerce. Several Democrats are now asking for a review of State’s competence to see if another agency should have been the lead, such as the EPA. Do you think we would have had a different product if another agency had been in charge?

What I know about this subject is from reading about it. I cannot say who should be in charge. I can say that when State took the lead, I thought, “Why not the Department of the Interior?” I am assuming that State is getting a great deal of input in design, execution and review of products. I certainly hope that they have been getting help.

EPA plays a role in every EIS. They get a chance to review the EIS and call for more information. I would prefer to have them as part of the team rather than its leader.

Agency experience on a type of project is important. I reiterate that multiple federal agencies and their state progeny should be working with lead agency, in this case the State Department. The agency in charge is key, but for complicated projects, it’s supposed to be a partnership.

When a final decision is made on an environmental topic, what comes into play besides the EIS in the final decision?

When you have worked on environmental issues as long as I have, you come to the realization that the EIS is merely a part of the information brought to the table. The EIS is one set of data, but economic and social advantages and disadvantages are always included. And, of course, politics.

The EIS presents some economic impact data, but these typically are, in my experience, not well done. And the EIS considers environmental justice, but I have some trouble with their environmental justice presentations. I think, but am not certain, that my concerns about these two parts are due to the evolution of the document as an environment first analysis.

Any final thoughts?

Yes, I urge the readers to recognize the reality of what an EIS is versus what some would like it to be. NEPA is the most widely emulated U.S. law, so the U.S. government must have done something right. In fact, the law was an eye-opener because it forced agencies for the first time to formally consider environmental issues. In 1970, that was a very big deal.

The bottom line is that, even if an EIS follows all the requirements, the agency leadership can make a decision that is not necessarily the most environmentally friendly.

About Paul Thacker

Paul D. Thacker is a Fellow at the Edmond J. Safra Center for Ethics at Harvard University and a consultant to multiple nonprofits and foundations. He is a former congressional investigator for the United States Senate Finance Committee where he led investigations of corruption of science and medicine. In 2007, he was profiled in Expose: America's Investigative Reports, a PBS series on investigative reporters.

More by Paul Thacker