U.S. Natural Gas Capacity Must Peak Soon To Achieve Sustainable Pathway

by Adam James

As you may have noticed, the natural gas industry has undergone a bit of a boom in the last few years.

In April, generation from natural gas tied that from coal for the first time ever. The boom has some energy analysts hailing a natural gas “miracle.” But others are very concerned, largely because of the local environmental impacts and the long-term climate impacts. Letting natural gas displace coal is a net-benefit on carbon emissions, but there are still unanswered questions about methane release and the impact on renewables.

What we know for sure is that the climate threat is real and must be taken seriously in charting a path forward for energy policy. The amount of natural gas we burn for electricity is critical in determining whether or not we’re on the right path.  This post compares a few climate pathways related to natural gas, and then presents a recommendation for developing natural gas capacity — including balancing additions and retirements.

The bottom line is this: in order to meet climate targets, the United States needs to build only planned additions and, starting in 2023, retire all natural gas plants over 45 years old. This will allow the United States to develop renewable alternatives and not waste excess natural gas capacity through uneconomic retirements later in an effort to meet climate goals. Of course, there are little adjustments around the margins, but this approach represents the most likely solution given the data available.

Before diving into an explanation of this point, I’ll summarize some of the more interesting conclusions from the different projections of natural gas capacity from Energy Information Administration (EIA), EPA, and National Renewable Energy Laboratory (NREL):

  • There is a gap between the business-as-usual (BAU) pathway for natural gas capacity and where we need to be to hit climate targets, and that gap widens significantly over time. In 2025, capacity of natural gas exceeds the recommended climate target by 13.9 GW. In 2030, natural gas capacity exceeds the climate target by 83.6 GW. In 2035, it exceeds the climate target by 143.8 GW. Assuming no further retirements, by 2050, natural gas capacity exceeds climate targets by 218.8 GW.
  • Coal retirements won’t make up the difference in this increased natural gas capacity. The coal retirements needed to continue business-as-usual natural gas development and meet the NREL scenario’s 80 percent CO2 reduction would be unlikely. Specifically, it would require taking 240 GW of coal out of the system before 2035, a feat that the NREL scenario does over 40 years to ensure reliability.
  • Aggressive retirements of natural gas capacity later on to meet the climate target are pretty unlikely. Over half our current natural gas capacity is under 12 years old. 38 percent of our current natural gas capacity was built between 2000 and 2004. An additional 15 percent was built between 2005 and 2011. That means 53 percent of capacity is basically brand new. These plants aren’t going anywhere anytime soon — retirements would be incredibly uneconomical.

This post will walk through where we need to be to address climate change, where we are headed, and quantify the gap between the two. Next, it will dismiss a few of the common excuses for our business-as-usual path of natural gas development, including that coal retirements will make up the difference and that we can retire old natural gas later on. Finally, it will present a reasonable approach to natural gas development that will ensure we can meet a climate target of 80 percent emissions reductions by 2050.

Assessing the Status Quo: Where do We Need to be and Where Are We Headed?

What are the implications of the increased utility investment in natural gas for the fight against climate change? To answer this question, we must compare current trends of natural gas capacity investment to projections for the levels of natural gas capacity needed to meet climate targets.

The “climate target” here is an 80 percent emissions reduction below 2005 levels in the power sector by 2050 (or reducing power sector emissions by about 2000 MMTCO2e a year). The three data points used to model how to reasonably get to 80 percent emissions reductions are the EIA and EPA projections, respectively, of implementing the Waxman-Markey bill (which projected an 83 percent GHG reduction below 2005 levels by 2050) and the excellent “Renewable Electricity Futures” study from the National Renewable Energy Laboratory (the core scenario would achieve 80 percent emissions reductions from the power sector).

In these studies, analysts from NREL, EIA, and EPA charted a course from our present day generation mix to a low-carbon fuel mix that ensures climate safety. In each case, the analysts show the year-by-year capacity additions and retirements for each fuel sources that would be necessary to ensure reliable electricity delivery during the transition. Because these pathways each look a little different, I have also shown an average of their recommendations.

For business-as-usual projections of natural gas capacity, this post uses the EIA Energy Outlook. Using these projections, we can ascertain what the planned capacity additions are for natural gas out to 2035.

As you can see in Figure 1, the U.S. is actually projected to be close to being on track to meet the climate scenario up until 2020, when new additions of installed natural gas capacity kick in, and natural gas capacity takes a dramatic uptick right when it ought to be declining.

Well, we can always retire more coal or more natural gas later on, right? Right?


Let’s look at coal first. While some could argue that this additional natural gas capacity is okay for climate because it will eat up some of the carbon allowance these projections provided for coal, our analysis shows that the retirements of coal needed to make that true would be incredibly dramatic and steep; retiring about 240 GW of coal before 2035. To put that in context, the NREL pathway phases out that same amount of coal capacity, but does so between now and 2050 to ensure reliability. The point is that we cannot stick to the climate target and justify our business-as-usual natural gas installations on the basis very aggressive — and unlikely — coal retirements.

So what is the main conclusion of this data? We are projected to blow past our ability to hit a climate target starting in 2017 unless we are prepared to later sacrifice reliability or retire plants uneconomically. The gap between the BAU pathway for natural gas capacity and where we need to be to the hit climate target is real, and widens exponentially over time. In 2025, capacity of natural gas exceeds the recommended climate target by 13.9 GW. In 2030, natural gas capacity exceeds the climate target by 83.6 GW. In 2035, it exceeds the climate target by 143.8 GW. Assuming no further retirements, by 2050, natural gas capacity exceeds climate targets by 218.8 GW.

That ever widening gap can only be bridged by natural gas capacity retirements. But where are those going to come from? The next section takes a quick inventory of the natural gas capacity stock to gauge the likelihood of retirements.

The simple fact is, most of our natural gas fleet is very, very new — and it isn’t going anywhere.

Since retiring plants early is uneconomical (utilities want to get the most out of their existing investments), it is important to assess our existing natural gas stock. As Figure 2 below shows, over half of the existing natural gas capacity was built between 2000 and 2004 (green). An additional 19 percent was built after 2005 to the present date (red), making 70 percent of our natural gas fleet under 12 years old.

Further, as Figure 3 shows, the construction of this installed natural gas capacity was an outlier compared to other construction trends. Solar, for instance, hardly registers. Even coal, with its spike in the 1960’s and 1970’s, pales in comparison. What does this mean? Well, these plants are highly unlikely to go anywhere anytime soon — so when we are thinking about building additional natural gas capacity, we should remember that the vast majority of our existing capacity is very, very new.

Of course, building excess natural gas capacity doesn’t necessarily exclude retiring old natural gas. But it would make it uneconomical, as utilities would be left with stranded costs. Similarly, you can shut down certain fuel sources in a manner not recommended by these analysts, but there is no guarantee that electricity will be delivered reliably.

Looking Ahead: Charting a Reasonable Course for Natural Gas

There is a way forward on natural gas that would enable us to stay on the NREL pathway to 80 percent renewables, and thus, 80 percent carbon dioxide emission reductions. Building only planned additions and, starting in 2023, retiring plants over 45 years old would keep us on well track, as Figure 4 shows below.

The key years to look at here are 2020 and 2025.

In 2020, we will either build out the EIA projected capacity for natural gas, and give up on meeting the climate target, or we will build only planned additions. That will mean natural gas capacity remains stable out to 2025, at which point retirements of plants built over 45 years old will need to start taking place.

This may not be a popular view, but for those of us who want to address the climate challenge, it is important to recognize that the investments we make today will have a serious impact on the decisions we will be able to make in 10, 20, and 30 years. By working backwards from where we know we need to be, and considering real constraints like reliability and utility return on investment from generation, we can get an idea of our year-by-year goals for natural gas and how that fits into the climate picture.

Adam James is a Special Assistant for Energy Policy at the Center for American Progress.

19 Responses to U.S. Natural Gas Capacity Must Peak Soon To Achieve Sustainable Pathway

  1. Overall, a useful post, but I would like to take issue with this statement: “These plants aren’t going anywhere anytime soon — retirements would be incredibly uneconomical.” There’s nothing uneconomical about retiring existing capital. The money has been spent, and the problem is a political one–the person who’s spent the money on the plant wants to get his money back, and will fight like hell to do so. But when capital is obsolete (because of climate constraints or competition) then it should be retired, and the investors will take a bath. I agree with your overall points in the post, but it’s important to be precise about what’s going on here.

    It’s hard to stop high carbon infrastructure from being built, but even harder to stop it after it’s built (and generating profits), which is why we need to focus on limiting construction of such infrastructure as quickly as we can. We already have enough natural gas capacity to fuel the transition to renewables, and now it’s time to focus much more on the fuels of the future in our current investments.

    Of course, we can start by retiring the older gas plants, of which there are many.

  2. NewDude says:

    The simple truth is that we’re not going to choose our own peaking. We’re going to continue with BAU or worse until renewable energy and storage is economical, regardless of legacy costs or externalities, for both the developed and the developing nations.

    We ought to be investing in geo-engineering avert haphazardly doing it when the shit hits the fan in the future. It may sound like Doctor Evil’s solution for AGW but it’s the only politically possible choice we have.

  3. Steve Davis says:

    Seems to me that “investors taking a bath” and “incredibly uneconomical” aren’t all that different. If your point is that policymakers aren’t or needn’t be concerned about the sunk costs of energy infrastructure, I’d say you’re dead wrong. Even setting aside cynical views of special interest politics, it’s all one economy, and the reality is that policymakers will avoid policies that entail costly early retirements.

  4. Ken Barrows says:

    Don’t worry. NG will peak soon if the rig count trend continues. It has plummeted over the last year. Look at the rig counts and you can figure out where NG is going.

  5. Mike Roddy says:

    Why should beancounters and investors make the call about whether the United States decides to continue to bake the planet? Are we so enthralled to the process of becoming wealthier that little details like a living planet are not “serious”?

    When World War II started, Americans didn’t ask whether building more tanks was cost effective. It’s actually easier now- renewables are close to competitive in a real market, due to fossil fuel (including gas) externalities.

    We cannot blow this one. Wake up, suits and investors, and try to dig a little deeper. These kinds of arguments are embarrassing the human race.

  6. thomas says:

    Thank you for the depth of the article, well digestible for the layman….your pie chart is a little skewed in re your stat of ~53% of NG gen <12 years old.

  7. Unless they are trumped by the costly effects of global warming. Remember, this discussion and any discussion of energy capacity needs to happen in that context. Policy makers might not want to force those losses, but they might not have any choice.

  8. Mark E says:

    This is a useful post but ya’ll have missed the biggest flaw in its reasoning: A fatally bogus concept of what is “economical”, and how to tell which plants are “profitable”. Did anyone add the value of ecosystem services to that spreadsheet? Global loss of life due to heat stress? Somewhere on that spreadsheet should be the value of entire cultures on pacific islands.

    I embrace the post’s main thesis, e.g. that we really need to base today’s investment decisions on our desired stabilization targets. It’s just that this is not the only revision we have to make to our current fatal reductionist notions of economics and profit.

  9. Mark E says:

    no links; no cussing, first post of the day I think … why did the spam filter catch THAT one?

  10. jim moore says:

    Coal causes 3 dollars worth of health and environmental damage for every dollar worth of electricity it produces. Killing all of the coal power plants of the next 10-15 years should be the first goal.

  11. Adam says:

    Steve beat me to the punch on this one. I realize there is more nuance to “uneconomical” than this post allows for. All the information on plant amoritization etc. was too much to include here.

    Basic point is that new facilities create a built in constituency to defend them. Later on, that gets framed as action on climate vs jobs, or action on climate vs reliability, or action on climate vs cost. Regular readers of this blog know those dichotomies are BS, but policymakers and mainstream folks don’t always. At the end of the day, my thinking is that we just want to avoid creating more opposition to renewables through bolstering the incumbents than is absolutely neccesary.

    A policy down the road which shuts down natural gas plants is going to be hard, period. So my personal view is that we should clearly define the outer limit of capacity generation and then use policy now to keep it within those bounds. Just my personal view.

    Also, another thing to add that this post DOES NOT account for that is important. There is a difference between nameplate capacity and capacity factor. This is a relatively crude assessment based on projected capacity margins, etc. But you could do a much more comprehensive analysis on these questions.

    Great comments all around though, thanks for reading!

    -Adam (@adam_s_james)

  12. Stranded investment hurts us all. That’s a chief reason we have public utility commissions–to prevent stranded investment. Lost capital bankrupts people. Bankruptcies cascade through the economy. That’s what happened in the Great Depression, and almost happened in 2008.

  13. Why further complexity is preferable to simply stopping what we’re doing is beyond me. It provides the illusion of solution, I suppose. But, similar to gas, geo-engineering would simply give permission to continue the status quo. Not to mention that it would introduce a global-scale unintended-consequences risk. It is a very, very, very bad idea.

  14. That is a common misperception. The rig count doesn’t matter much for unconventional gas. They are re-used. One rig can sit on a well pad for a long time and drill multiple new fracks. Gas production increased 25% in 4 years while the rig count fell by more than half.

  15. WW2 sacrifice is the model. How do we inspire it? People react to militarized dictators with bad haircuts and funny mustaches. They don’t react to graphs.

    This is the key to action. How to convince people that we need to make the sacrifice?

    It’s not about the bankers–it’s about all of us. Stranding investment makes all of society poorer. Your 401K will take a big hit.

    Of course, it will take a order-of-magnitude BIGGER hit if we don’t take the hit now.

  16. A carbon price is the answer.

  17. I had myself convinced for a while that NG could displace coal and give us a start on carbon reductions. I embraced the shale phenomenon. After about two years, I saw that the additional NG was just that–additional. Much of the coal being displaced from power plants in this country is being shipped overseas. So no net gain on carbon.

    I basically concluded you can’t reduce fossil fuel usage by expanding it. Brilliant, huh? It was one reason I left my consulting job. They were just too gung-ho on NG. I couldn’t do it anymore.

  18. Brad Mushovic (@BradMushovic) says:

    Thanks for your work on this James. Agree that if an overbuild of natural gas occurred it could materially impact when and how we get to an even cleaner and more secure electricity future. Fortunately, advancements can undercut the business-as-usual assumptions / trend lines, fundamentally changing the NG trajectory (and potential capacity overshoot).

  19. Kim Feil says:

    I this urban drilling is to continue…we need to build int he necessary costs to do this by people or risk air quality and health effects which are costly too. I live at ground zero for urban drilling. We have about 60 padsites in our 99 sq mile town here in Arlington TX. The following requests won’t cover public protections on the huge buildout and the associated human errors or accidents. We had a drill spill in Lake Arlington, our drinking source, a couple of years ago and have had maybe a dozen emission events over the last couple of years that I am aware of. The following are the items lacking in our oil and gas drilling ordinance….

    1. Use electric (not diesel) rigs in urban areas.
    2. Video tape ALL cement casing pours when it comes back up to the top through the annulus so that there is proof of an even pour and ensure all wells have electric bond log tests.
    3. All drilling mud farming (private & commercial) and *brine “road spraying” is subject to open records of water & soil test results.
    4. We need the industry to invent technology to keep the toxic, silica dust on the padsite-those pathetic pillow case looking socks aren’t getting the job done.
    5. Mandate ventless, emission free flowback tanks.
    6. We shouldn’t have to wait 2.5 years for the EPA mandated Green Completions and allow venting or flaring in urban or rural areas….methane losses should be prevented-period.
    7. The pipeline should be in place FIRST before fracturing so that flowback doesn’t sit in the ground for months festering some unknown, man-made hydrogen sulfide-like stale water flowback.
    8. The setback away from people should be substantial. Rural method drilling is not acceptable in urban areas. A environmental tester/PHD said that the health effects are being seen downwind from about 1,800 – 2,500 feet.
    9. Zero tolerance for underinspected, or faked Waste Disposal Injection Well casing pressure tests. Cureenlty we risk eventual migration of toxic fluids into our drinking supplies.
    10. State entities overseeing oil and gas should regulate how close the old wells are to new wells.

    Lets elect LOCAL officials that will ensure a protective Oil & Gas Drilling Ordinance cause OBAMA AIN”T GETTING IT DONE!. This is in our “collective” power so we must all become active to be proactive because reactive measures are too costly.