A look at the math: Will growing EV adoption increase consumer rates as Cook Inlet’s natural gas production dwindles?

A view of a Chugach Electric Association power station in Anchorage


It’s no secret that there is a natural gas supply conundrum brewing for the Alaska Railbelt. The dwindling supplies in Cook Inlet announced by Hilcorp in April of 2022 beg many questions about the state’s future power production (as well as questions about how to best verify industry-supplied information) and posits an equally long list of possible solutions. According to ENSTAR’s website:

In April of 2022 Hilcorp, Cook Inlet's largest natural gas producer, announced it did not have "line of sight" to produce gas beyond its existing contracts. This effectively put ENSTAR and local electric utilities on notice that it was time to look for an alternate supply of natural gas. While ENSTAR has much of its gas needs contracted with Hilcorp until 2033, electric utility contracts with Hilcorp begin to expire in 2028. All utilities will start seeing gaps in their gas supply portfolios beginning that same year.

Solutions can be grouped into three main categories — finding ways to get more gas from Cook Inlet, bringing in natural gas from elsewhere, or fuel switching off of natural gas as supplies dwindle. Each of these options has trade offs. A question to consider in any scenario is the impact of increasing electric vehicle (EV) use in the 49th State as utilities consider implementing EV incentivization programs. As EV technology continues to improve and market demand grows in the U.S., the adoption of EVs will undoubtedly increase in Alaska. We see evidence of this trend already. What will this increase in electric demand mean for overall consumer rates as we work to address the gas conundrum over the next 5 to 10 years? 

On the one hand, increased electrification will require more electric generation, which could mean burning more gas. On the other hand, the new uses could enable greater economic use of renewable generation, because they have some inherent storage. EV batteries or heat stored in buildings and water heaters can help store renewable energy when the sun is shining and the wind is blowing and use that energy when they are not. Wind and solar power have become much more cost effective over the last couple of decades and improvements to batteries have further firmed up the use of variable power on short timescales. The lead times needed to bring new wind and solar farms onto the Railbelt power grid in Alaska are similar to the timescale over which the current gas supply is expected to stop meeting current needs. At the same time, the world has seriously started to focus on decarbonizing more than the power grid by electrifying heating and transportation. Putting these factors together, there are serious — and interesting! — questions about whether now is the right time for Railbelt drivers to switch to EVs given the impending gas shortages. 

Let’s dive into a back-of-the-envelope analysis of this question:

First of all, how much energy are we talking about needing for EVs anyway? According to a 2023 ACEP paper on load growth on the Railbelt, if 90% of vehicles electrified by 2050 (an aggressive, but plausible scenario), this would be about 450,000 vehicles. We would need almost as much electricity generation on the Railbelt to power them (4 TWh) as we use today (4.72 TWh). No one has a crystal ball, but if we do get to these numbers of EVs, it will not be all at once. Today there are about 2000 EVs on the Railbelt roads. By 2028, the extension of the state’s continued and aggressive scenarios predicts about 17,000 to 27,000 EVs in Alaska. Assuming 75% of them (roughly equivalent to the percentage of total vehicles registered on the Railbelt vs. other regions) are charging on the Railbelt, that is 13,000 to 20,000 EVs.

Display of electric vehicles at the 2023 Alaska Sustainable Energy Conference in Anchorage.
Display of electric vehicles at the 2023 Alaska Sustainable Energy Conference in Anchorage.

Looking at the high-end prediction of 20,000 vehicles in 2028, if (in a worst case scenario)  they all plugged in to a 10kW home charger at the same time, that would be 200 MW of additional power needed. Now, that isn’t going to happen, and can be assured not to with utility incentives or other mechanisms, but it gives us a back-of-the-envelope high-end number. Using some reasonable averages (a truck in a Wasilla climate driving 30 miles a day), one can use the Alaska EV Calculator and a little math to find that these 20,000 EVs would use a total of about 155 GWh per year of electricity. If we assume it all comes from gas power, given the average efficiency of Railbelt gas plants, this adds only about 1.7% extra to current Cook Inlet demand.

Now let’s look at how this compares to renewable projects that could come online by then. The proposed Little Susitna Wind Project, if completely built out, could have 250 MW of installed capacity of power.  At a reasonable 40% capacity factor (meaning the wind farm in real conditions produces 40% of the energy that it could if the wind always blew optimally and it was never out of commission) this is 876 GWh per year of energy produced. An interior companion wind farm, Shovel Creek, is also on the drawing board, with room for 200 MW capacity, or around 700 GWh of possible yearly energy production, although the initial phase might only be half that size. As of a November 2, 2023 informational flier, the expected operational date is 2027.

When we are looking at the need for replacing gas supplies, given that gas is storable, the energy totals here are way more important than the power peaks. There are intricacies around extraction rates, storage amounts, day-ahead forecasts in contracts, etc, but looking at the yearly energy totals is a reasonable first approximation. Focusing on those numbers, 20,000 EVs on the Railbelt would need approximately one-fifth or less of just one wind farm like those proposed. If both Little Susitna and Shovel Creek Wind were built out fully, there would be 1,400 GWh of extra energy left over to help out with decreasing natural gas supplies on the Railbelt. That’s not even taking into account that many of the early adopters of EVs also have the money, roofspace and desire to install Solar photovoltaics, often sized to displace a significant fraction of the EV’s energy needs.

Now might be the time to note that switching to electric heat pumps for space and water heating is much less likely to stress the natural gas supply than lots of EVs. That’s because most heating on the Railbelt, especially in the Southcentral regions where cold climate heat pumps can function throughout the whole winter,  is already done by natural gas. First making electricity efficiently with it, and then using it in super-efficient heat pumps may actually save natural gas.

One problem with all of this fine-tuned analysis is that electrification, renewables development, and other major factors here aren’t part of some Bureaucratic Central Office 5-year Plan that can be followed to the letter. We don’t know the exact amount of gas that industry can or wants to pull out of Cook Inlet over the next decade or so, how electricity needs might change due to efficiency, population or other changes, and what or when other electricity generation projects might come online. No one can say exactly when the first molecule of LNG might need to be shipped in if that was the way we chose to move forward, and whether we need one or a million ccf to make up the shortfall — we will need the infrastructure built for that import, if that is the way we go.  (It’s even possible that increasing the demand for gas via EV’s and heat pumps could help keep both new and existing infrastructure fully utilized and thus help keep delivered costs down.) 

In the same vein, whether or not Railbelt utilities or governments incentivize EV adoption, factors such as high interest rates, high costs for EVs, lack of acceptance or a million other things could keep adoption down. On the other hand, we could also be underpredicting the natural adoption of EVs. Likewise, many factors can delay wind and solar farm growth in the state, not least of which would be the utilities signing long-term contracts for LNG that lock us in to that path, or state subsidies for pipelines or other projects that can then, with the subsidies, outcompete the cost of renewables. However, if the rumors are true that contracts are being signed right now for about double the recent costs for natural gas, then these wind and solar farms might start looking a lot more attractive.

Given all the caveats above, what the back-of-the-envelope calculation tells us is that based on current trends, even an aggressively accelerated EV adoption will only leave us needing a fifth of a new wind farm on the Railbelt to meet their energy needs five years out. So incentivizing EVs and renewables at the same time may be able to lessen Alaska’s reliance on costly natural gas AND refined transportation fuel imports as we transition to cleaner energy. There are still more questions than answers — it’s likely it will be a challenge to build out renewables fast enough to meet gas shortfalls with or without EV adoption incentives, so what do we do about that? One thing seems clear to this (admittedly a bit biased) group of electrification researchers: Unless we want to tell individuals that they can’t own EVs in Alaska, we need to start learning how to make them a resource instead of a liability. Figuring out the incentives and technologies behind smart charging to minimize costs and impacts to the grid and also provide carbon reduction benefits might be a good thing to begin early. Whether or not we want to grow the number of EVs, we really should be planning for this growth in EVs before we find ourselves reacting to it as an emergency.