Is 'Gold Standard’ for Energy Data in Trouble? - Center on Global Energy Policy at Columbia University SIPA

Guest

Adam Sieminski

Former Administrator, US Energy Information Administration

Transcript

Adam Sieminski: I believe that statistics are not Republican statistics or Democratic statistics. They’re numbers and they should be available to everybody, and we ought to have confidence that we’re doing a pretty good job in collecting, publishing, and analyzing those figures.

Bill Loveless: Everyone from energy executives to traders on Wall Street to policymakers across the country depends on accurate, timely information about energy production, consumption, and trends. At the heart of this critical infrastructure sits the US Energy Information Administration, the EIA. Daniel Yergin, vice chairman of S&P Global, has called EIA’s data the gold standard.

But while the amount and complexity of energy data is growing, federal support for ensuring robust energy data collection is waning. The agency underwent substantial staffing cuts this spring as part of the Department of Government Efficiency’s reductions. Its most recent annual energy outlook forecasts the growth of renewables, which drew a rebuke from the Department of Energy for its contrast with President Trump’s preference for fossil fuels.

So how vulnerable is the agency to losing more support from the administration? What’s at stake if EIA cannot retain or recruit people with expertise in not only traditional energy but emerging fields like critical minerals? And who else stands to lose if the agency that provides national energy data collection and objective analysis falters?

This is Columbia Energy Exchange, a weekly podcast from the Center on Global Energy Policy at Columbia University. I’m Bill Loveless.

Today on the show, Adam Sieminski. Adam is a senior advisor to the board at KAPSARC, a nonprofit energy economics and sustainability think tank in Saudi Arabia, where earlier he served as president from 2012 to 2017. Adam was the administrator of the US Energy Information Administration. Prior to joining the government, Adam spent years as Deutsche Bank’s chief energy economist and integrated oil company analyst. We talked about how the cuts have hit the agency and what they could mean for the credibility and accuracy of its reports. We discussed the logic behind the agency’s methodology in conducting its marquee annual energy outlook, and we talked about new topics the agency might consider in light of the changing energy landscape. Here’s our conversation.

Adam Sieminski, welcome to Columbia Energy Exchange.

Adam Sieminski: Thank you, Bill. It’s great to be here and I’m looking forward to our conversation.

Bill Loveless: Yeah, I am as well. It feels as though we go back some ways, me, as a reporter, calling you when you were at Deutsche Bank and at EIA, and I always appreciated your insight on these many issues.

Adam Sieminski: It’s a complicated area, energy, Bill, and there’s always something new, so it’s great to be back.

Bill Loveless: Well, the topic today is the Energy Information Administration, and there’s been a lot of attention paid to that agency lately. For those who may not be entirely familiar with the EIA, Adam, help us understand what is the Energy Information Administration? And why is it considered to be vital to US energy policy and markets?

Adam Sieminski: EIA goes back to 1977, Bill. It was created in the aftermath of the 1973-74 energy crisis. There were a lot of statistics on energy that were available at that time. Many of them were coming from energy organizations like the American Petroleum Institute, the Natural Gas Association, the Edison Electric Institute, and others. There was some skepticism in Congress because prices really went up after the embargo and everybody was upset about gasoline lines and other things, and they wanted to be sure that the data that was available to policymakers was independent and not coming from industry sources. Nobody really felt that the industry data was being manipulated, but it just had that feeling of, well, why do we have to rely on companies or associations to get data that ought to be available and available free to the public and policymakers, not just in the United States, but around the world?

Bill Loveless: And how did you use EIA data back when you were in the banking industry?

Adam Sieminski: Well, when I was on Wall Street, I was a power user of EIA’s data. It’s really useful for supply, demand, prices, storage, looking at their forecasts or their projections to try to see how they match up with your own. And as a consequence, I was able to analyze what was going on in the industry, what the trends were, and provide that information to the clients of the brokerage firms and banks that I worked for as well as the companies. So there are a lot of people that rely on EIA’s data to make business decisions, and understanding the data and being confident that it is as accurate as can be is an important part of those investment decisions.

Bill Loveless: Yeah. Can you give an example of a business decision that would rely to some extent on EIA data?

Adam Sieminski: Well, companies have to make their own forecast of what the outlook is for the business that they’re in, but companies are making decisions all the time. Should they invest in another oil or natural gas well? Should they upgrade the refinery to produce a different kind of product than they’re currently producing? What will the demand for gasoline be over the next 10 years or 20 years, the life of the equipment that they’re putting into place? What’s the effect of the increasing demand for electric vehicles and hybrid vehicles, and how does that play into the outlook for fuels? That’s the kind of thing that companies struggle with all the time. For consumers, it’s often things like, what’s the outlook for gasoline? Should they be looking at a car that gets better mileage than the one that they currently have if they expect that prices are going to be rising over time?

Bill Loveless: Well, this summer you wrote a piece in the online publication The National Interest entitled “The Death of Energy Statistics We Can Trust.” Among other things, you refer to “unwarranted concerns about how EIA’s energy projections are prepared.” What did you mean by that?

Adam Sieminski: EIA has traditionally used a forecasting methodology that uses what is called a reference case. So they try to look at what existing law and regulation are, what are the trends that are going on in fuel consumption, and use that kind of thing. And they base the main projection on the idea that the existing law and regulations that are in place will remain in place. That’s an important idea that is not well understood. For example, Bill, when I was the EIA administrator between 2012 and 2017, there was a lot of concern about EIA’s forecast for renewable energy. They were low. The reference case projection often underpredicted the extent to which solar and wind would gain market share. The reason for that is the financial support that was available in the tax code for investment in solar and wind was set to expire.

It often expired at the end of each congressional term. And EIA did not want to get into the business in its reference case of saying that, well, of course Congress is going to extend these renewable tax credits and we should build that into the reference case. The danger of doing that is that in addition to all of the other things that you have to think about in making a projection, you’d also be making guesses about what an administration or Congress might do, and that just compounds the difficulty associated with that. So EIA decided that it would be much better in the reference case to use existing law and regulation, and then as that law and regulation was changed, it would then be built into the next reference case. And EIA has been consistent, so it consistently underpredicted solar because those credits kept being extended. EIA tried to take that into account by also publishing side cases or a secondary case that said, well, what if there was no sunset on the wind and solar tax credits? What would you get then? And those forecasts turned out to be more accurate. And people said, well, why don’t you use these more accurate ones as your reference case? And the answer is because then you’d be getting into the business of predicting legislative affairs, which is really not what I think a statistical agency ought to be doing.

Bill Loveless: So what you’re saying is that it’s much more reliable for EIA to simply take into consideration the policies and laws that are in effect at the moment when it’s collecting data—and this is for its annual energy outlook—than attempting to speculate what might happen based on perhaps some politics or legislation that’s being considered in Congress?

Adam Sieminski: That’s exactly right. And one of the other things that we always felt was that by showing what would happen—there would be less solar if the tax credits were removed—that we were actually providing a helpful service for policymakers to say, if you do extend these credits, you’ll get more of this form of energy, if that is what you think is in the best interest of the taxpayers and voters that you represent. And I always looked at it as a way of actually helping inform those decisions rather than the other way around.

Bill Loveless: Right. Well, when this most recent annual energy outlook came out, the 2025 report came out this spring, it drew a rebuke from the Department of Energy, the Trump Department of Energy, for its outlook on fossil fuels and renewable energy. That outlook was largely based on data gathered during the Biden administration. It projected rapid growth in alternative energy and declines in US reliance on coal, oil and natural gas. The Department of Energy expressed some dissatisfaction with those findings because it runs counter to President Trump’s energy policy and its preference for fossil fuels. I’m not sure the administration was doing anything more than just drawing the distinction between what the report said based on the data it had collected during the prior administration and perhaps its views on how things should be done going forward. But did this seem to raise some concerns with you?

Adam Sieminski: No, Bill, the way I was looking at it, it was in a sense suggesting that EIA was probably on the right track. They got criticized for a long time for under-forecasting renewables, and then more recently they’ve been criticized for over-forecasting the penetration of renewable fuels, and maybe that suggests that they’re getting it just about right. I thought that the easiest answer to the criticism of EIA’s methodology was that, well, don’t worry about it because as the executive orders take effect, as new laws and regulations begin to influence the supply and demand statistics, it will show up. I mean, it’s a self-correcting kind of methodology. With new policies from the Trump administration, EIA’s next forecast will reflect those changes, and it seems to me that that’s a pretty good way of doing it.

Bill Loveless: Yeah. Do you think the reference case remains—that the way the EIA conducts these annual energy outlooks will remain the same going forward?

Adam Sieminski: I hope so, because I think it actually gives the policymakers, regardless of what party they’re coming from, the best information about the trends that are taking place. If you try to build in other expectations, you can very quickly go wrong. And in fact, the International Energy Agency in Paris made that mistake when they gave up using their kind of business-as-usual or reference case and began to assume that the pledges that were being made by countries concerning carbon dioxide and greenhouse gas reductions would all take place easily and on time as the political statements made, when in fact it proved much more difficult and costly to do those things. And so IEA’s projections of the lower need for investment in oil and gas drilling activities turned out to be very misleading and wrong.

Bill Loveless: Yeah. Of course, IEA has been coming under criticism from some quarters, including some members of the administration, some members of Congress, as well as a bill in Congress that would curtail US funding for IEA. I mean, by and large, when you look at numbers from each of these agencies, do you think that EIA’s data is perhaps more reliable than what you see coming from IEA?

Adam Sieminski: I think that’s certainly the case for their projections domestically for the United States, and EIA also has an international energy outlook. They hadn’t done one in a while. That’s a funding issue, resources, but I think that it’s something that, again, the methodology that you use to make these forecasts, if it’s consistently using existing law and regulation as a reference case and then showing what would happen if law and regulation changed—it could change either to foster more renewables or to foster more hydrocarbon energy—that at least then you would know where you stood in order to make the decisions.

Bill Loveless: In your experience, was there ever a time when political leaders were uncomfortable with EIA’s findings? And if so, how did you navigate the situation without compromising the agency’s credibility?

Adam Sieminski: When I was there, the biggest criticisms that we got usually came from associations and groups rather than a lot of critique from the Congress or the administration. Prior administrators ran into difficulties when EIA’s forecast for gasoline prices was too low. When some crisis of supply or demand caused imbalances in the market and prices went above what politicians never seemed to worry that much if your fuel forecast for gasoline or propane was too high or too low, rather they like low prices, they get upset if the forecasts don’t predict those exceptional cases when things move up very quickly. And that has happened to prior administrators. When I was there, EIA was criticized for under-forecasting renewable energy, and I spent a lot of time trying to explain the idea of the reference case forecast and why if you really wanted to understand renewables during that time period, you had to look at the alternative cases with no sunset in the tax code. EIA was also criticized by some groups for not believing as they did that peak oil supply was coming and that we didn’t have enough oil and gas reserves to fuel increasing demand—that the supply of oil and gas would surely be declining and create grave political and economic problems. And EIA spent a lot of time explaining the effects of technology and prices on improving the addition of oil and gas reserves could change the supply curves and make it possible to meet growing demand for fossil fuels, which turned out to be right.

Bill Loveless: Yeah. It took a lot of explaining of what that data, what was behind the data that it was reporting. EIA relies on specialists in fields ranging from statistics to energy engineering. You describe in the piece you wrote what you called “misguided workforce reductions” at the agency, largely the result of cuts prompted by DOGE or the Department of Government Efficiency. How vulnerable is the agency to losing talent now, and what’s at stake if EIA cannot retain or recruit people with expertise in not only the traditional fields but emerging ones like critical minerals?

Adam Sieminski: Well, at the start of this year, 2025, EIA had about 350 full-time employees working on collection and analysis of vital energy statistics. There have been several things that have changed those employee count numbers. There were firings associated with the Department of Government Efficiency. There were also buyouts of people who were encouraged to take early retirement and that kind of thing. And in the fiscal year that starts this October, EIA is limited to 246 people. That’s a pretty significant reduction in workforce from 350 to 246. My concern is that with the hiring freeze that’s still on, with normal attrition beyond the firings and buyouts, that EIA might find it difficult to even get up to its allowed number. EIA actually was treated well in the budget figures for fiscal 2026. They kept the budget flat and the workforce number at 246, although it’s a lot lower than where it was. I think the agency can still accomplish most of its work.

But the effect of all of the workforce changes, I think, has been a bit demoralizing. Some areas within EIA have been hit harder. For example, natural gas statistics—it just so happened that more people left from that division within EIA than others. Now that can be changed and people can be reassigned and so on, but it makes it difficult. And by the way, the weekly natural gas storage report of EIA is a principal federal economic indicator. It’s one of the key statistics that is used to help forecast the direction of the US economy, and with increasing natural gas production, with exports of natural gas, having that natural gas storage report is really important. And what I would hope, Bill, is that all of the workforce reductions and concerns about methodology and so on don’t end up creating a problem for the collection of these vital statistics.

Bill Loveless: Yeah. And of course, natural gas is an area of primary importance to the administration given its interest in seeing gas production increase, seeing LNG exports increase in coming years. Adam, do you hear from people at EIA who express some concerns they have?

Adam Sieminski: I’ve tried to be very careful about not getting deeply involved in EIA’s internal business. I don’t think that would really be appropriate. I have had conversations with people who’ve already left who were knowledgeable about how things were going in the agency, and that was very helpful to me when I was writing the article about my concerns about keeping the data and analysis of EIA independent from political influence.

Bill Loveless: Well, President Trump’s nominee for EIA administrator, Tristan Abbey, told the senators in his confirmation hearing that he would fulfill his statutory role of providing objective energy data and analysis to the American people and to policymakers. He said he would steward carefully the reputation that the agency has built. He’s got a big task ahead of him given these constraints with the loss of personnel and all. At least to the extent that we hear him in his testimony, he’s acknowledging the traditional role that the agency has played.

Adam Sieminski: I think Tristan actually is a good choice for the job. He’s worked on the Republican staff in the Senate Energy Committee. He acted as an independent energy consultant. He’s very knowledgeable. I think he values EIA statistics in the same way that a lot of other people do that try to use the data in a way that can help illuminate what’s going on in the markets and so on. Along with a couple of other former EIA administrators, I wrote a letter of support for Tristan’s nomination and I attended his hearings. I thought he did a pretty good job in answering the questions that were put to him by both sides of the aisle. And I hope that the slowdown in nomination confirmations that is affecting everybody, not just the EIA administrator, is dealt with so that some of these critical positions can be filled and get the agency back on track.

Bill Loveless: Yeah, of course, as you note, he has not yet been confirmed by the Senate. And of course that presents another obstacle to an agency when its political head has not been confirmed and is in place and doing his work. Abbey acknowledged during that hearing that energy systems are changing rapidly, and he called for expansion of EIA’s global data collection and analysis and establishment of a critical minerals outlook. It seems like that global data collection interest of his is already maybe of some additional concern. As I understand it, EIA, because of staff shortages, has said it’s not going to be producing an international report this year. Is that right?

Adam Sieminski: I think they did say that, and I hope that Tristan, when he’s confirmed, can change that. The EIA administrator has a lot of flexibility in terms of how EIA funds are spent. And as I said, the budget proposal for EIA is fairly generous relative to how many other agencies were treated. So I think that he could use the funds that he’s got available to make some changes and get that work done. I think it’s really important that EIA has a good solid international forecast to go along with the annual energy outlook, which covers the domestic scene, because energy, certainly oil and natural gas, is a global business, and you really have to know what’s going on all over the world in order to understand the trends that are occurring in the industry. The other thing that I hope Tristan would be able to do is to beef up EIA’s work on electricity supply and demand. We have this surprising to most people trend in growth in electricity demand for data centers, artificial intelligence and crypto even, and being able to get a good handle on what’s going on in that area I think is going to be critical to consumers in terms of their cost of electricity and the reliability of household electricity.

Bill Loveless: I guess it comes back to, do you think that EIA is going to be able to recruit the people, the experts, the analysts, it will need to do these important reports just given the uncertainty at the agency at the moment?

Adam Sieminski: Well, I think that could be an issue across the government. EIA, I think, still has a very strong reputation for giving its employees a good working environment and the ability to do things. So I think that if EIA were allowed to hire, it wouldn’t be a huge problem, but it’s going to be not easy. I suspect that the administration in general still has a policy in place of tightening up on the federal government and trying to push more things back on the states and individuals. I think in the case of statistical agencies, I think we want to be really careful about that because I think that the collection of these statistics that everybody uses really does depend on broad-based consensus and federal implementation. EIA specifically in its founding documents from Congress has the legal authority to enforce collection of its statistics, so they can take companies to court if they aren’t providing the data or if they’re doing so in a way that seems to be non-business-like. And that’s a really important aspect of EIA’s credibility as an agency and the reliability of the data.

Bill Loveless: This all comes at a time when government statistical agencies seem to be under increasing scrutiny. The president just recently fired the head of the Bureau of Labor Statistics over his displeasure with the numbers that they released. Does that send a sort of frightening signal to folks who are doing this sort of work in government or maybe considering applying for a job to do this in the government?

Adam Sieminski: I think that the independence from a data standpoint of agencies is very important. Again, you come back to the credibility of the data. If you’re trying to make business decisions on investment and location of facilities, labor initiatives and so on, you have to have an understanding that the data that you’re using to build the cases that you’re looking at are solid. And I think that at the Bureau of Labor Statistics, it was fire the messenger. The data suggested that employment figures were not as robust as had been expected, and there were revisions made to some of the earlier data. That happens all the time, by the way. And it’s not just labor statistics. The state of Texas collects data on oil and gas production in Texas, and their data has lags associated with it and often has to be corrected as all of the data comes in.

You get better details on this stuff and then it gets published. So the idea that you can get it perfectly right the first time you get the reports coming in is really an expectation that goes against, I would say, common sense. You could do more to get stuff right away, but it would be really expensive. So there’s this trade-off between how much do you want to pay to collect the data, how important is it for the specific figures that you’re looking at, and when revisions are made, is it transparent what the reasons were and so on. I think BLS was doing an okay job, and they just got caught in what I think are unfortunate political crosswinds.

Bill Loveless: Yeah, I was just wondering if those crosswinds could affect more statistical agencies such as EIA going forward.

Adam Sieminski: Well, it’s certainly possible, and that’s Bill, one of the reasons I wrote the article, and I was delighted that The National Interest was willing to publish it. They’re considered a right-of-center organization, and that’s really the audience that I wanted to reach. I believe that statistics are not Republican statistics or Democratic statistics. They’re numbers, and they should be available to everybody, and we ought to have confidence that we’re doing a pretty good job in collecting, publishing and analyzing those figures.

Bill Loveless: So Adam, if you were speaking directly to a member of Congress like you used to do when you were a government official, or you’re speaking to the public about why investing in the EIA matters, what’s the one message you’d want them to remember?

Adam Sieminski: It’s actually a pretty clear one, but I would say: look, Congress in the 1970s created the gold standard of energy statistical collection and analysis, and that has served the country very well for a long time, and it would be a real shame to lose that level of confidence and excellence for reasons that I think are not necessarily going to help the people that are the users of the data. And those users are often those congressmen themselves. They have to make decisions quickly sometimes on what to do in an energy crisis. And they too need to be confident that the data that they’re looking at is good, and that if they ask for a study on something that they’re worried about, that they are going to get the best that can be done. And EIA, I think, has fulfilled that role pretty well for 50 years, and it would be a sorry day if we gave up what is the envy around the world of others who are trying to make the same kinds of decisions.

Bill Loveless: Well, you were a user of the data. You managed the data for a period of time. You know that agency pretty darn well. Adam Sieminski, thanks for joining us today on Columbia Energy Exchange to discuss this important topic.

Adam Sieminski: Thank you, Bill.

Bill Loveless: That’s it for this week’s episode of Columbia Energy Exchange. Thank you again, Adam Sieminski, and thank you for listening. The show is brought to you by the Center on Global Energy Policy at Columbia University School of International and Public Affairs. The show is hosted by Jason Bordoff and me, Bill Loveless. The show is produced by Mary Catherine O’Connor, Caroline Pitman and Kyu Lee. Greg Vilfranc is the sound engineer. For more information about the show or the Center on Global Energy Policy, visit us online at energypolicy.columbia.edu or follow us on social media @ColumbiaUEnergy. If you like this episode, give us a rating on Spotify or Apple Podcasts. You can also share it with a friend or colleague to help us reach more listeners. Either way, we appreciate your support. Thanks again for listening. See you next week.

Everyone from energy executives to traders on Wall Street to policymakers across the US depend on accurate, timely information about energy production, consumption, and trends. At the heart of this critical infrastructure sits the US Energy Information Administration (EIA). Daniel Yergin, vice chairman of S&P Global, has called EIA’s data the “gold standard.”

But while the amount and complexity of energy data is growing, federal support for ensuring robust energy data collection is waning. The agency underwent substantial staffing cuts this spring — part of the Department of Government Efficiency’s reductions. After the EIA’s most recent Annual Energy Outlook forecast the growth of renewables, the Department of Energy criticized the findings.

So how vulnerable is the agency to losing more support from the administration? What’s at stake if EIA cannot retain or recruit people with expertise in not only traditional energy but emerging fields, like critical minerals? And who else stands to lose if the agency that provides national energy data collection and objective analysis falters?

This week, Bill Loveless speaks to former EIA Administrator Adam Sieminski about the state of play at the EIA and what is at risk if support for the agency continues to erode.

Adam is a senior advisor to the board at KAPSARC, a non-profit energy, economics, and sustainability think tank in Saudi Arabia, where he earlier served as president. He was the administrator of the EIA from 2012 to 2017. Prior to joining the government, Adam spent years as Deutsche Bank’s chief energy economist and integrated oil company analyst.

Our Work

Relevant
Publications

Energy Policy

AdapTech: Resilience-Driven Innovation in Energy Systems

Op-eds & Essays with Tatiana Mitrova • The National Interest • September 05, 2025

Hydrogen

Trump’s One Big Beautiful Bill Act Has Given the US Hydrogen Industry Certainty — but Where Will the Green Energy Come From?

H2 projects will have to compete for a shrinking pipeline of zero-carbon electricity with energy-intensive data centres.

Op-eds & Essays with Anne-Sophie Corbeau • Hydrogen Insight • July 29, 2025

Trump’s One Big Beautiful Bill Act Has Given the US Hydrogen Industry Certainty — but Where Will the Green Energy Come From?

Energy Policy

The Social Cost of Carbon Is Gone — and That May Be Good News for Future US Climate Policy

President Trump has ended the federal government’s use of the "social cost of carbon" (SCC), an official estimate of the harms caused by carbon dioxide emissions.

Blog by Noah Kaufman • July 23, 2025