This chart from the U.S. Energy Information Agency was published last week in the article "EIA expands data on capacity and usage of power plants, electricity storage systems":

Monthly usage factor for utility-scale storage generators (Jan 2017-Dec 2019)

The text of the article describes what is meant by "usage factor":

Usage factors for storage generators differ from capacity factors because usage factors are based on gross generation rather than net generation. Energy storage technologies consume more energy than they store and, therefore, always have negative net generation.

The decline over time for both storage types is noticeable. While capacity of pumped storage is essentially fixed, battery capacity is growing rapidly:

Operating utility-scale battery storage power capacity has more than quadrupled from the end of 2014 (214 MW) through March 2019 (899 MW).

The article does not comment on why the usage factors (for both storage types) are noticeably declining over time. Is it a simple case of supply and demand (more storage on the grid means less demand for each MW of capacity)? This seems unlikely given the rapid increase in capacity that is both seen and projected.

What's causing the usage factor of utility-scale storage to noticeably decline over time?

3 Answers 3


Usage factor is declining because the average duration of batteries on the grid is increasing

Inspired by the report that M Juckes shared, I dug into the data source for utility-scale batteries: Table 3.4 from EIA-860.

Using the 2021 data (the most recent available), I created the plot below showing cumulative installed power, cumulative installed energy, and the average duration. Average duration (hours) is simply energy (MW*hours) divided by power (MW).

enter image description here

The results show that since the cumulative energy is increasing faster than the capacity, the average duration of batteries on the grid is increasing for the time period shown.

As a result, the usage factor is decreasing simply because it's inversely proportional to the duration.

  • I don't think that is right. Usage factor is calculated for storage as = gross generation / (capacity x hours in month). This means duration is ignored. If anything, increasing duration means usage factors would likely go up as the gross generation / capacity ratio would increase. This is also the trend you see with increasing usage rates and storage duration length over the last few years
    – ColinTea
    Nov 14, 2023 at 15:49

The operating costs of using a battery once it has been installed are generally low. In fact, they appear to be too low to report on: the recent "Battery Storage in the United States: An Update on Market Trends" report notes the need for more information on operating costs, but bases its analysis on capital costs.

Given the low operating costs, it makes sense to use batteries as much as possible. The batteries installed initially will be used almost every day. As capacity expands, some of the newer installations are not needed as often. Taking the expanded capacity into account, it is clear that the total power supplied by the batteries is increasing steadily.


Because we need capacity, not just energy.

The same amounts of megawatt-hours but more megawatts means lower usage / capacity factor.

If increasingly large amounts of variable wind power and solar power are being installed (especially wind power since solar gives just a little bit of daytime help as opposed to wind that can generate significant amounts of power nearly continuously), we have the problem that sometimes it's just not windy. If the sun isn't shining at that time then, where could we generate the electricity?

Some years ago, the electric grid was dominated by large plants, nuclear, hydro, coal, oil, natural gas, biomass, that could generate power nearly continuously. Storage was useful mainly to allow the expensive plants to run near maximum power yet still allow for daily and seasonal variation in energy usage to be handled. We needed consistently the services of adjustable power plants, so their capacity factor was large.

Today, the electric grid is increasingly dominated by wind power, plus solar as a small helper. If it's windy, we have electricity. If it isn't windy but sun in shining, we have electricity. These two are able to provide enough power maybe 70% of the time, and even 30% of the time you usually get at least some power from these if not enough. Thus, as a result, I'd expect capacity factors of all adjustable power production combined (hydropower, pumped hydro, batteries, natural gas, oil) to have 15% capacity factor / usage factor / whatever you call it (I call it capacity factor, maybe somewhat incorrectly for pumped hydro and batteries).

Out of the 15% capacity factor production combined, non-pumped hydropower probably has the largest capacity factor, far larger than 15%, because usually reservoir total energy storage capacity isn't that large and many hydropower plants don't even have proper reservoirs but are just nearly continuously producing power from river flow. At least in Finland, the total combined capacity factor of non-pumped hydro is 50% (we don't have currently significant pumped hydro and probably never will).

So what is left, pumped hydro and batteries, is expected to have very low capacity factor indeed. Their main job is not to produce significant energy. Their main job is to produce significant power. It's all about megawatts, not about megawatt-hours. Very few megawatt-hours is enough, even 500 hour long storage is plenty for pumped hydro for example. In 500 hours the weather changes and we get wind production again, allowing to recharge that 500 hour long storage. Batteries are usually between 1-24 hours, though, due to them being very expensive, so they are better today in frequency stabilization services and maybe tomorrow in storing solar power during daytime in areas near the equator where the whole year is all summer, and then using that solar power again during the night.

  • I understand why it's low, but why is it getting lower? You'd think that as solar and wind capacity increase, there'd be a higher demand for energy from storage, not less.
    – LShaver
    Nov 5, 2022 at 2:34
  • When the share of wind and solar increases, a larger share of our energy usage can come from wind and solar. However, you also need to supply whatever power is required when there's zero wind and solar. So in my opinion it's very logical: as wind and solar increases in share (assuming electricity use doesn't increase), the amount of kilowatts demanded from storage stays the same, but the amount of kilowatt-hours demanded reduces (because more is supplied by wind and solar). Therefore, capacity factor decreases.
    – juhist
    Nov 5, 2022 at 15:36
  • Yes but in theory the new wind/solar is replacing "baseload" generation, so there's an increase in variability of the energy supply.
    – LShaver
    Nov 5, 2022 at 16:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.