Is natural gas a "bridge fuel"? A dead end? Or something else?

The ten biggest questions in energy & climate tech, Question 5

Hypothesis: Something else.

How should we think about natural gas’s role in the energy transition?

This is a very contentious question, which is highlighted by the fact that the two leading metaphors for gas today - which I would characterize as “bridge fuel” and “dead end” - lead to very different places.

I’m going to propose a third metaphor for your consideration: “the backstop”. But first, because this topic can be so polarizing, I want to be up-front about my personal views.

1. I take the goal of achieving a net-zero economy, by roughly mid-century, super seriously. In other words, I’m a climate hawk.

2. However, I’m also a climate pragmatist. I have very little patience for favoritism towards particular fuels or classes of technology. I believe it’s critical for society to strive towards the most affordable, reliable, safe, and secure pathway to decarbonization. This pathway could very well still include fossil fuel.

   In fact, I’m fairly confident that it will. Just look at the Princeton “Net Zero America” project - a sophisticated, well-regarded study outlining five scenarios for achieving net-zero by 2050. Only one of those scenarios ends up with less than 20% of present-day natural gas consumption by 2050, and that’s the scenario that’s explicity built to model the impact of a fully renewable primary energy system. (In other words, zeroing out fossil fuel was an endogenous choice in that scenario.)

3. It’s important to separate how we think about natural gas, the fossil fuel, from how we think about gaseous fuel infrastructure. Of course, they’re closely related today, and their futures are also related. But, one will always come with a carbon atom that needs dealing with, and the other doesn’t.

Lastly, before I propose a new metaphor for natural gas and its associated infrastructure, I want to make what I believe are the best possible cases for the current reigning metaphors.

The case for a bridge

It wasn’t so long ago when “bridge fuel” was fairly widely agreed upon, at least among climate policy wonks, as the right metaphor for natural gas’s role in the energy transition. I think a good faith argument in favor of this metaphor goes something like this:

1. Methane (CH4) carries more energy per carbon atom than either coal or oil.  Hence, substituting natural gas for its fossil fuel relatives can be a potent form of carbon emissions abatement.  In fact, the transition from coal- to gas-fired power generation in North America has arguably made a bigger carbon impact than any other emissions abatement strategy around the world to date. 

2. Realistically, it’s going to take at least a few decades to build out renewables and develop technology to address the long tail of “tough to decarbonize” sectors - many of which can fairly easily be fueled by gas.  There are plenty of plausible scenarios in which it takes much longer than a few decades to address this long tail, even if we assume good intent among policymakers in major emitting nations. We’re simply not yet that confident in how fast we’ll be able to decarbonize with clean electricity alone.  Hence, we should seek to reduce emissions as much and as quickly as possible by simultaneously building out zero-carbon primary energy resources AND transitioning as much as we can from coal to gas (and to some extent even oil to gas).¹

3. In North America, gas is especially compelling as a lower carbon fuel because the fracking revolution has bequeathed us with many decades of very low cost, domestic supply. We’ve also already built a continent-spanning network of pipelines to deliver it cheaply, with an extraordinarily high level of reliability, to consumers from coast to coast. 

4. Moreover, gas is not just a low cost abatement measure; it’s also an important instrument of national security and geopolitical power. The US and Canada in particular need to weigh these priorities against carbon goals. After all, a world in which certain adversarial nations grow too dominant is a world with serious risks to global climate policy adherence.

5. However, at some point, bridges do need to come to an end.  While we might use more gas in the short-term as we transition away from coal & oil, sticking to mid-century net-zero goals will soon require a sharp decline in gas consumption too, probably beginning in the 2030s and heading towards zero in the decades to follow.

The case for a dead end

Today, there’s a competing argument that’s growing in popularity among a subset of climate activists, who have concluded that nearly any bridge for natural gas is…a bridge too far.  According to these activists, we need to phase out gas as soon as possible; in fact, we need to begin doing so right now. This argument can pretty quickly veer into self-defeating extremist territory (e.g. “Immediately ban all gas appliances! De-growth!”). But, there’s a version of it with a pretty compelling fundamental case:

1. It’s true that burning methane emits less CO2 per joule of energy than burning coal or oil at the point of combustion.  But methane itself is a powerful greenhouse gas; in fact, a molecule of methane warms the planet about 80 times more than a molecule of carbon dioxide over the next twenty years, and 20 times more over the next one hundred years. Unfortunately, methane leaks.  Sometimes it’s allowed to leak on purpose when it’s too difficult to capture at a wellhead; sometimes it leaks by accident. Whatever the cause, the most widely accepted study of total methane emissions from wellheads to consumers estimates that the average leak rate is about 2.4%. In truth, we’re probably still underestimating that rate, especially from natural gas pipelines and underground distribution systems. But even taking a leak rate of 2.4% as given, many gas end uses might as well be burning coal from the standpoint of near-term (20-year) global warming impact.

2. Making new gas infrastructure investments today will unaovidably lock in strong social & political incentives to continue consuming gas for several decades to come. Capex spent on gas assets needs to be amortized over a 20-30 year financial life; workers hired & trained to maintain those assets will anticipate a 30+ year career; local governments may become dependent on tax revenues; etc. Hence, continuing to invest in gas assets will make it more difficult to wean ourselves off of gas, in perpetuity. (Put more concisely: A natural gas ‘bridge’ will probably be self-extending.)

3. We already have pretty good, lower-carbon substitutes for gas in most big gas consuming sectors, namely: power generation, industrial heat, and building heat. Although the substitutes we have today aren’t perfect, we don’t need them to be perfect to begin deploying them aggressively. They’ll get better, as technology tends to do when it is manufactured & deployed at scale.

4. Hence, we should stop thinking of gas as a bridge to some radically different energy future. We’re already in that future, and we need to start pragmatically building the terminal end of the bridge. We should be extremely skeptical of any investments prolonging the life of gas infrastructure, let alone investments in new gas infrastructure.

The case for something else: a backstop fuel

I see compelling elements in both of these arguments, and frankly I’m not yet certain either of them are wrong. It’s still entirely possible that gas turns out to be either a “bridge” which concludes several decades hence, or a much more imminent “dead end”. However, I tend to find another way of thinking about gas - another metaphor - more persuasive: “the backstop fuel”.

In many ways, the case for a “backstop” is very similar to the case for a “bridge”. However, I think it better trains the mind on how to make the most of gas while taking decarbonization goals seriously, and while remaining agnostic regarding just how much gas we ought to continue using, for how long.

Here’s the argument:

1. The case for a bridge (see above) is correct right up until the last point, which concludes that our use of gas must come to an end by around mid-century. Of course, there will be some distant future period in which humans stop consuming natural gas, but that future might not arrive until well beyond mid-century.

2. There’s no plausible decarbonization scenario in which natural gas consumption doesn’t decline in the next few decades. My personal best guess is that gas consumption will decline by a factor of three to five, as it’s displaced with alternative sources of carbon-free primary energy.

3. However, in certain sectors, it will be extremely valuable to maintain some amount of gas, and access to gas delivery infrastructure, for a few reasons:

   1. Fossil gas is a tremendous pressure release valve for the cost of power grid decarbonization and electrifiction. Both of these endeavors are likely to face exponentially rising cost functions.

   2. Fossil gas is a uniquely strong contributor to energy security and resilience. If either of these attributes of our energy system is seriously compromised by decarbonization, the whole endeavor would be thrown into a tailspin.

   3. Gas infrastructure is also a uniquely low-cost and resilient means of delivering energy. Maintaining access to those pipes for an as-yet-unknown blend of natural gas and alternative gaseous fuels (e.g. RNG, synthetic e-methane, and perhaps a touch of hydrogen) could prove extraordinarily valuable. At the very least, for now, we should consider it an important source of option value for energy delivery.

4. Most importantly, we can deal with the methane emissions problem really cheaply! In fact, the International Energy Agency estimates that achieving near-zero emissions from oil & gas production would be practically cost-neutral over time, with an up-front investment totaling less than 3% of the net income of the global oil & gas sector in 2022 (about $100b). About 40% of emissions abatement measures would actually be financially positive for the industry, because the value of the gas they’d preserve would exceed the cost of the equipment required to contain it.

   None of these gas leak abatement measures require betting on new technology. Most methane emissions can be avoided by replacing worn pipes, fittings, and gas-driven pneumatics. There is, however, an opportunity for new solutions to better identify leaks in real-time, and to provide scientifically rigorous monitoring & quantification of leaks in order to make sure the industry is holding emissions below a negligible threshold. At my firm, Energy Impact Partners, we’ve invested in one such set of solutions developed by a company called Project Canary. Project Canary provides tools for methane leak monitoring from wellheads to gas distribution networks, along with analytical tools for independent analysis and validation. We’re also increasingly confident in the use of satellite data for spotting leaks across wide areas, such as hundreds of miles of gas pipeline infrastructure.

   Critically, in the US there’s now a hefty ‘stick’ compelling companies to reduce their methane emissions (in addition the ‘carrot’ of being able to market more responsibly sourced gas). In the US, the Inflation Reduction Act now requires the EPA to use more rigorous, empirically driven methods for quantifying the emissions from sizable sources, and imposes steep fines on large emitters. The scale of these fines relative to the price of gas and the cost of abatement measures suggests that they ought to be highly salient to oil & gas investors.

4. If you’re convinced that we can deal with methane emissions, then we can continue to burn some amount of natural gas and still achieve net-zero emissions (hopefully net-negative emissions), economy-wide. Large point sources of gas combustion emissions have the potential to be mitigated through carbon capture & sequestration (or CCS). Smaller, distributed emissions sources like household furnaces & office building boilers are much tougher to manage, as there’s no plausible strategy for capturing and removing carbon from all of these sources at scale. However, if fossil gas consumption in these applications can be substantially reduced, then it’s quite plausible we’ll be able to dislace their emissions by assembling a large enough portfolio of “offsets”, in the form of big off-site carbon removal projects. (Note, I’ll write a lot more about CCS, including carbon removal, in a subsequent post, Question 10: Will CCS make a lot of these other questions moot?)

5. Hence, it makes sense to strategically maintain natural gas infrastructure, and even in some cases invest in new infrastructure, to preserve the backstop value that gas can provide through the energy transition, and beyond.

Some examples of gas as a backstop

The first kind of “backstop” we might need from gas is a macro-level backstop for our primary energy supply. In the initial post in this series, I posited that we’re unlikely to build all of the electric transmission capacity we’ll need to meet our primary energy needs with wind & solar resources alone. In the second post in this series, I expressed my belief (or, at least my hope) that we’ll be able to fill in some of our remaining primary energy needs with nuclear power; but, I also shared some reasons to be skeptical of the pace with which we’ll be able to deploy nuclear.

Natural gas with CCS, particularly in power generation, is one of the few remaining options for zero-carbon primary energy.

Now, to be clear, CCS has yet to be demonstrated on natural gas based power generation at scale. In fact, CCS has yet to be successfully demonstrated on any power generation at scale. The few big, serious attempts so far have been on coal-fired power generation, which ought to be an easier target than gas-fired generation because the concentration of CO2 in the emissions flue gas is about twice as high. But the first round of attempts from the late aughts & early 2010s have been pretty big failures. See, for example, the infamous Petra Nova project.

I’m pretty confident that CCS is on track to become technically viable in gas-fired power generation. (Sorry to punt again, but I’ll discuss this in more detail in a future post.) For now, I’ll note that there are a number of jurisdictions which have the right subsurface geology for sequestration, and strong political support for continued fossil gas extraction - e.g. the US gulf coast, or our neighbors to the north up in Alberta, Canada.

The second kind of “backstop” role that gas is already playing, and I believe will continue to play, is an energy reserve for periods of exceptionally high energy demand or low supply from other resources. I’ve written previously about the opportunity for energy storage technology to play this role, and I do believe that storage will be able to deal with most of the natural intermittency of both energy demand, and renewable energy supply. However, there are going to be some periods, in some regions, that are just exorbitantly expensive to cover with renewables + storage, and will probably end up being cheaper to fuel with a small amount of natural gas. (I’ll note that one challenge to this strategy is that CCS is fairly capex-intensive; so, adding CCS to a natural gas flue to operate for just a few hours per year would also be exorbitantly expensive.)

One especially promising variant on this renewables balancing role for gas is a concept pioneered by EIP portfolio company Enchanted Rock. The basic idea is to deploy natural gas power generation in a highly distributed manner. Distributed gas generators can serve the function I just described, filling in for renewables on the grid during periods when doing so with storage would be especially costly. What makes them extra valuable is that they can also play a role providing resilience for specific energy consumers, like schools or grocery stores, or for whole segments of the grid thath are especially prone to losing power due to extreme weather, like remote towns. Moreover, in some cases they can also reduce the need for costly new electric transmission and disttribution infrastructure.

Lastly, a third “backstop” role for gas - the most important one, in my opinion - is to heat buildings in cold climates during extended periods of freezing temperatures. I want to emphasize, for the record, that I am a big believer in the potential for heat pumps of all varieties, and believe that they will ultimately be a 70-90% solution for the vast majority of building heat (including hot water). However, the last 10-30% of heating demand in some regions could end up being punishingly expensive to satisfy with heat pumps, and I suspect that a blend of gaseous fuels (including, probably, some fossil gas) will remain an important part of the energy mix for buildings. (More on this in the next post in this series, Question 6: How will we keep warm in the winter?)

Gas is polarizing, but it doesn’t need to be

Natural gas has become one of the more polarizing players in the drama that is decarbonization. In my role at EIP, I do a lot of presenting to energy companies and climate tech investors alike, and I’ve been warned that certain types of audiences will immediately tune out when they hear the phrase “bridge fuel”.

It’s telling that I’ve received this exact warning before presenting to radically different audiences, on both sides of the argument about gas’s future!

It doesn’t have to be this way. I’m convinced that conceptualizing natural gas as a “backstop fuel”, in good faith, can help ‘bridge’ the divide. (See what I did there?) It’s certainly not going to be easy, but I believe we can determine a role for gas that maximizes our chances of getting to net-negative carbon emissions, while simultaneously preserving the cost, resilience, and security benefits of keeping some gas - and even more importantly, some gaseous fuel infrastructure - in the mix.

Enjoy this post? Please check out the rest of the Ten Biggest Questions in Energy & Climate Tech

1. Oil to gas switching: for example, from diesel to compressed or liquified natural gas for trucks & ships.