‘The urgent need for power drives demand for energy storage’: Dax Kepshire, Energy Operating Partners

Dax Kepshire is a founder of Energy Operating Partners, an energy storage advisory firm, and has been working in the energy storage sector for the past 18 years. After earning a PhD in engineering sciences, he founded SustainX in 2007, one of the first grid-scale energy storage startups in the US. In 2015, he joined AES Energy Storage, a unit of US utility AES Corporation, where he built technical sales globally.

After the merger with Siemens Energy Storage that formed Fluence Energy, he held various senior leadership positions and ultimately became its COO for the Americas region. In this interview with Gas Matters, Kepshire discusses his expectations for battery cost declines and explains why he believes US energy storage will see a 100% compound annual growth rate until 2026.

We have seen a big drop in lithium-ion battery cell prices over the past couple of years. Were prices this low expected a few years ago?

I would say that these prices were considered to be kind of the holy grail of energy storage. Some had often said that below USD 100/kWh would be difficult to attain at a module level.

And as you point out, with prices at USD 60/kWh or lower at a cell level, we’re now breaking through that threshold of USD 100/kWh at a module level by a considerable margin.

And are you expecting further cost declines in 2025, or do you see this as a temporary dislocation?

I think about things mostly at a system level – it’s just where I sit in the industry. So year over year, the grid market might have seen a 35% price decline.

In my view, that is mostly driven by three factors: we’ve got rapidly declining battery costs, so those are contributing to the decline in cell costs and at the module level too.

Then we have larger battery enclosures. Over the last year, the industry has transitioned from 1 MWh to 5 MWh as standard enclosures. That drives that component down significantly by approximately 40%.

I think we can expect a further decline in cell and module costs.

And then you’ve got lower energy performance certificate (EPC) costs coming from larger components that are being installed, and I estimate that’s been on the order of a 35% decline year on year.

Moving into the future, I do not have a crystal ball of course, but what we’re going to see is larger ampere hour cells, so the industry is going to transition from around 300 ampere hour cells to 500-plus. I think at that fundamental level we can expect a further decline in cell and module costs.

Looking forward at enclosures, I think production volumes will increase and that’s going to help drive some declines there from manufacturing efficiencies. And then I think the EPC side of things will be relatively flat.

So next year, my two cents would be that perhaps we’re going to see something in the order of a 10-15% year over year price decline but we’re not going to see something dramatic.

And I think that’s consistent with the signals we’re seeing. [Chinese manufacturer] BYD has given their suppliers a target to stretch and reduce prices by 10% next year.

For storage, to what extent is the rollout of battery capacity in the US sensitive to fluctuations in battery cell prices? Prices were obviously higher last year but the pace of deployment was very high in places like California.

I would say it is sensitive because at the end of the day investors are getting a return, so if the prices go up the returns get smaller and when the prices go down the returns get bigger.

The industry is pretty efficient. In terms of procurement, particularly within contracted projects where there are offtake agreements, people typically find a price that gives a reasonable return threshold and that also gives the offtaker some reasonable benefit.

I think where you really see the declines benefitting investors the most is merchant markets, where maybe because of the price decline, it actually means that a merchant project is now something that makes sense or their returns become significantly larger.

Given the lithium-ion cell price declines we’ve seen, and especially for lithium iron phosphate (LFP), is this likely to boost the adoption of stationary storage?

Yes and no. The yes to that is that I think it’s opening new market applications, it’s driving better returns for investors, so I think that’s helping.

Analysts have a long track record of misestimating the growth of industries that are truly disruptive.

I think the bigger thing that’s going to drive demand is going to be the urgent need for power. In the US there’s a lot of demand for data centres, electrification in general, so you’re seeing load growth like you haven’t seen in a considerable amount of time, perhaps ever, in some jurisdictions over 10% per year and that’s compounding.

That’s really where storage can shine because you can build and permit these systems much faster than you can traditional gas-fired generation, so I think that’s where we’re going to see the driving force for installations in the next several years.

You have predicted that the US will cumulatively add 321 GWh of grid-scale energy storage by the end of 2026, which is a 100% compound annual growth rate and is higher than most forecasts. What are you seeing that you think most other forecasters are missing?

First, if you look at industries that are truly disruptive and growing exponentially, analysts have a long track record of misestimating that growth. And if you look to solar, and you look at previous predictions from the International Energy Agency (IEA) and other industry analysts, you’re going to see they’re dramatically missing it – where solar is growing about 100% per year and they’re forecasting it’s going to be 30%.

At the end of the day, analysts tend to be more conservative. When I look at the historic growth rates, and what’s in the interconnection queues, for example over the next few years, I do believe that those forecasts of about 100% annual growth are realistic.

I also think that because of the upcoming tariffs, we’re going to see an acceleration of projects and imports of goods to try to mitigate the tariff risk. I think through mid-2026, we’re going to see a really strong number of projects being built.

Chinese suppliers are obviously very competitive. Do you think that the US’s upcoming tariff rate hike in 2026 on Chinese lithium-ion battery imports for energy storage – from 7.5% to 25% and on top of the existing 3.4% general tariff on lithium-ion battery imports – is not a sufficiently high tariff to cause a big dent in those imports?

At the 25% level, I believe it won’t. If you go significantly higher than that, to 60% on all Chinese imports as has been perhaps proposed by President-elect Trump, that might be a different situation.

For storage to compete with combined-cycle turbines, you’re going to need much longer storage durations.

But again, we’re seeing really rapidly declining battery costs and other system level costs, so my honest opinion would be you could almost absorb the 25% tariff with just cost declines. So a 25% tariff turns into about a 15% total system cost increase. 

I’m interested to get your views on the ability of batteries to replace some gas peakers. I understand that one of the issues for batteries is that the typical runtime for peakers in a place like California is often longer than the typical battery duration of 4 hours, so your storage in MW may need to be multiples greater than the peaker’s MW. What’s your view on batteries versus peakers and could we start to see more competition on that front?

I actually think that peaker replacements was one of the first markets that really helped batteries get traction. Those projects were typically on the order of four hours, and if you look at peakers, most of them have very low capacity factors and don’t run for much longer than that anyway.

And so that’s why we saw a lot of the growth in California – a need for peak capacity. Batteries were able to come in and provide utilities with that capacity. So I do continue to see storage being a suitable replacement for peakers, particularly when combined with solar.

I think combined-cycle gas turbines are a different story, which in my view is what’s next: how does storage begin to compete with combined-cycle turbines? And for that you’re going to need much longer storage durations to really replace baseload power.

China has a key role across the battery supply chain. Tariffs perhaps might change that, but so many of the components needed for batteries come from China. Do you expect to see a shift whereby China becomes less dominant, or is it already well-established and difficult to change?

I think it’s pretty well-established and difficult to change. If you look at the bankruptcy of Northvolt, for example, that’s an indication of how difficult it is to achieve.

It’s very hard to compete in such a highly-competitive market. It seems the most practical way to go to market is to partner with a either a Chinese or perhaps a Japanese or South Korean company that has particular expertise at manufacturing batteries at scale. But China is dominant, particularly for LFP batteries. - EO