—·
A mechanics-first test of 2020–2025 battery cost claims: BNEF’s pack benchmark and segment splits show how chemistry mix and EV vs non-EV pricing steer the $/kWh decline.
The “benchmark” is the closest thing the EV battery market has to a public price index. BloombergNEF (BNEF) publishes an annual lithium-ion battery price survey that aggregates reported transaction-like pricing across multiple end uses, then reports volume-weighted average $/kWh levels at pack (and, in some releases, cell) granularity. For 2025, BNEF reports a global record-low average pack price of $108/kWh, down 8% from 2024, and ties the drop to manufacturing overcapacity, intense competition, and a chemistry shift toward LFP. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/)
But the survey is also a black box in a specific way: it is not a simple spot-market exchange rate. The price you see in $/kWh is an outcome of contracting structures (long-term vs spot, forward supply, tender timing), chemistry-specific cost curves, and the mix across applications (EV transport packs versus stationary storage packs). Even BNEF’s headline number can look like a clean cost decline while the underlying drivers are shifting between “chemistry economics” and “application economics.”
That distinction matters for investigators verifying the “2020–2025 decline” narrative. If you treat BNEF’s benchmark as if it were driven only by process learning and raw-material deflation, you risk missing how much of the curve is simply mix: LFP versus NMC and EV versus non-EV. BNEF’s own 2025 release already signals the mix channel by reporting LFP packs at $81/kWh versus NMC packs at $128/kWh in 2025. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/)
A cost-curve verification for 2020–2025 should isolate two mechanically different levers:
BNEF supports both levers in public summaries, but not with full year-by-year LFP/NMC splits visible in every excerpt. That is a key limitation you should treat as a measurement problem, not as an editorial choice. For this article, we therefore use BNEF’s published annual benchmark levels as the backbone and then apply the chemistry split using the most directly reported LFP/NMC pack levels for 2025, plus the end-use segment anchor points BNEF has publicly disclosed for multiple years.
So what should you do differently as a researcher? Don’t ask “did costs fall.” Ask “did the fall come from within-chemistry cost compression, or from across-chemistry and across-application mix shifts.” Benchmarks can fall for either reason, but the policy and contracting implications diverge sharply.
BNEF describes its annual “Lithium-Ion Battery Price Survey” as covering multiple end uses (including passenger BEVs, commercial EVs, and stationary storage) and then producing volume-weighted averages. That means the benchmark is sensitive to (a) how many kWh were transacted at each point in the supply chain and (b) how the market’s kWh volume shifts across application types and chemistries.
A concrete anchor: in 2025, BNEF reports the all-segments pack benchmark at $108/kWh, while also stating stationary storage packs fell to $70/kWh (which BNEF calls out as the lowest price segment in that release). (https://about.bnef.com/insights/clean-transport/new-record-lows-for-battery-prices/) This is direct evidence that the $/kWh index is not “one market.” It is an aggregate of at least two pricing regimes.
The verification trap is a false decomposition: researchers often treat a benchmark like a single curve driven by unit costs of materials and manufacturing steps. But in a volume-weighted mix index, you can see large declines even if one segment is stable or rising, provided another segment’s volume-weighted kWh price moves down faster.
BNEF’s 2024 release shows the same structural point at the headline level. BNEF reports that lithium-ion battery pack prices dropped 20% in 2024 to $115/kWh, describing this as the biggest annual drop since 2017, while citing manufacturing overcapacity, economies of scale, low metal and components prices, and LFP adoption. (https://about.bnef.com/insights/commodities/lithium-ion-battery-pack-prices-see-largest-drop-since-2017-falling-to-115-per-kilowatt-hour-bloombergnef/)
For investigators, the actionable implication is simple: you cannot verify a “cost curve decline” without also tracking the mix of segments and chemistries that feed the index. Otherwise, the benchmark can confirm the index went down while disguising which parts of the market actually improved.
You asked for an EV battery price benchmark verification focused on pack and cell $/kWh, with chemistry and EV versus non-EV split mechanics. Public BNEF releases provide several annual anchor points; what they do not provide in every snippet is a full table for each year that includes LFP and NMC pack values simultaneously in the same paragraph. That measurement gap is real, and any “full” decomposition will require you to pull the full BNEF report tables or use a dataset that reproduces BNEF’s year-by-year breakdown.
Still, for 2020–2025, the headline anchor points are strong enough to support a mix-mechanics argument.
The “verification” lens now becomes clearer: 2022 shows a reversal (up to $151/kWh), then a sharp drop in 2023 and 2024, and then a smaller but still downward 2025. That pattern fits a contracting-cycle and capacity-overhang story more than a smooth global learning curve story.
In 2025, the chemistry split is directly observable in BNEF’s public summary: LFP packs at $81/kWh versus NMC packs at $128/kWh. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/) That $47/kWh spread is huge. If market kWh shifts toward LFP, the benchmark can fall even if the absolute engineering cost of both chemistries is moving modestly.
A proper verification would ask whether LFP became cheaper because of within-chemistry learning or because of manufacturing overcapacity and competition concentrating on LFP supply. BNEF’s 2025 explanation points to overcapacity, competition, and adoption of LFP as the drivers despite rising metal costs. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/)
To isolate the mechanics, you need a chart that does not claim to prove the full 2020–2025 decomposition, but does show what chemistry spread alone can imply. Public BNEF numbers provide LFP and NMC pack levels for 2025 in the summary, plus we can anchor the all-segments pack benchmark levels by year.
This visualization is therefore intentionally asymmetric: it shows (i) the all-segments benchmark path for 2020, 2022, 2023, 2024, 2025, and (ii) the only directly reported LFP and NMC pack levels in the accessible text excerpt—those for 2025. For earlier years, the chart does not insert guessed LFP/NMC values. It is a “mechanics illustration,” not a year-by-year chemistry reconstruction.
Below is a verification-oriented visualization with what we can support in public BNEF text: the 2025 chemistry spread and the all-segments pack benchmark path. (For earlier years, you would need BNEF’s full tables or a licensed dataset to fill year-by-year LFP/NMC pack values without guessing.)
The bar for 2025 shows why chemistry mix alone can swing an index number: LFP pack pricing at $81/kWh and NMC at $128/kWh are far apart. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/)
Meanwhile the year bars show that the all-segments benchmark traveled from $137/kWh in 2020 through $151/kWh in 2022, then down to $108/kWh in 2025. (https://about.bnef.com/insights/clean-energy/battery-pack-prices-cited-below-100-kwh-for-the-first-time-in-2020-while-market-average-sits-at-137-kwh/, https://about.bnef.com/insights/commodities/lithium-ion-battery-pack-prices-rise-for-first-time-to-an-average-of-151-kwh/, https://about.bnef.com/insights/clean-energy/lithium-ion-battery-pack-prices-hit-record-low-of-139-kwh/, https://about.bnef.com/insights/commodities/lithium-ion-battery-pack-prices-see-largest-drop-since-2017-falling-to-115-per-kilowatt-hour-bloombergnef/, https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/)
The second lever is application segmentation. BNEF’s survey covers multiple end uses and, crucially, public releases often highlight that different segments move differently.
In the 2025 BNEF summary, BNEF reports that stationary storage pack prices fell to $70/kWh, which BNEF frames as a major decline relative to 2024 and the lowest-priced segment. (https://about.bnef.com/insights/clean-transport/new-record-lows-for-battery-prices/) That matters because stationary storage is not contractologically identical to EV battery supply. Storage projects can behave more like infrastructure procurement with different tender cadence and different risk allocation than OEM vehicle programs.
Energy-Storage.News, reporting on BNEF’s release, also cites transport segment BEV pack prices at $99/kWh in 2025, and again emphasizes that stationary storage is the steepest decliner with $70/kWh. (https://www.energy-storage.news/li-ion-battery-pack-prices-fell-8-since-last-year-despite-metals-prices-rising-bloombergnef-says/?jet_blog_ajax=1&nocache=1765526002)
Even if you treat these as media summaries of BNEF releases, they support the same investigative conclusion: when the benchmark drops, it can be because one application segment is under a different supply-demand balance.
BNEF explicitly connects the 2025 decline to “continued cell manufacturing overcapacity” and “intense competition” alongside LFP adoption. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/) When overcapacity exists, suppliers often price aggressively in the most procurement-responsive segment, which can be stationary storage if tender demand holds up or if OEMs are contractually buffered through multi-quarter pricing.
This is how month-to-month swings in real pricing can exist even when annual benchmarks appear smooth. The annual survey is a summary. The underlying contract netting is continuous.
So what should you do with EV versus non-EV? Treat the benchmark like a weighted portfolio. If you want “EV battery price truth” for OEM purchasing, you need to model the gap between the all-segments price and the EV-specific segment price path (when publicly available), because the rest-of-market pricing can drag the average down.
The survey’s messaging also provides a clue about what not to assume. In 2025, BNEF says pack prices fell to $108/kWh “despite rising metal prices,” and attributes the decline to overcapacity, competition, and LFP adoption. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/) That is a direct counterexample to the simplistic “metals drive the curve” story.
Similarly, BNEF’s 2024 release credits overcapacity and intense competition as key drivers alongside low metal/component prices and LFP adoption. (https://about.bnef.com/insights/commodities/lithium-ion-battery-pack-prices-see-largest-drop-since-2017-falling-to-115-per-kilowatt-hour-bloombergnef/)
From a contracting perspective, overcapacity affects negotiation leverage. If cell suppliers have idle lines, they can push lower $/kWh into contracts where acceptance is faster or where qualification cycles are already in place. That can be the non-EV segments first, then spill into EV if OEMs seek cost relief to protect margins.
This also helps explain why 2022 bucked the decline trend. BNEF reports that pack prices rose to $151/kWh in 2022, and says LFP packs were 20% cheaper than NMC cells that year on average. (https://about.bnef.com/insights/commodities/lithium-ion-battery-pack-prices-rise-for-first-time-to-an-average-of-151-kwh/) The point is not that LFP vanished in 2022, but that the index can still rise if other cost and supply-demand pressures dominate.
Benchmarks embed negotiating power. Overcapacity changes the slope of that power curve, and chemistry adoption changes the sensitivity. When those two coincide, the benchmark can drop faster than metals would justify.
Your brief asked for minimum two real-world cases; the deeper investigative requirement asks for minimum four. The challenge is evidentiary discipline: public sources rarely disclose contract $/kWh figures in a way that can be directly compared to BNEF. So these cases focus on documented outcomes and timelines that plausibly connect to the benchmark mechanics (chemistry substitution, tariff-driven supply re-routing, or procurement program shifts). Each case is still evidence-limited; where direct pricing data is missing, that’s stated plainly.
Even without contract-level $/kWh, a case can still be analytically useful if it changes one of the variables that determines the benchmark: chemistry share (LFP vs NMC), segment mix (EV vs stationary storage), or the intensity/timing of contracting (when suppliers push discounting into tenders). The cases below are therefore written as “variable movers,” not as substitute datapoints for the benchmark.
In 2025, a report describes LG Energy Solution securing a large LFP battery contract amid U.S. tariff pressures and notes that the LFP batteries are inexpensive and long-lasting, with Tesla described as using LFP made by CATL. The report frames the contract as a response to tariff reshaping supply chains. (https://biz.chosun.com/en/en-industry/2025/08/04/OWWLDJ6OMRFIVMIY4LFR4LOERI/)
Timeline: contract reported in August 2025.
Outcome: supplier reallocation toward LFP under tariff constraints.
Direct implementation data (actual $/kWh paid) is not provided in the reporting excerpt, so you should treat this as a “mechanics” case, not a benchmark-point case. The evidentiary value is that tariffs can act as a chemistry-share variable mover (pushing procurement to the tariff-friendly supply base), which in turn can push the benchmark down in a mixed index even if metals do not fall.
A separate report states Tesla agreed to a $4.3 billion contract with LG Energy Solution for LFP battery supply from LGES’s U.S. Michigan plant for energy storage systems (not a direct EV pack quote), with an August 2027 to July 2030 supply window described. (https://battery-tech.net/battery-markets-news/tesla-strikes-4-3b-deal-with-lg-energy-for-lfp-batteries/)
Timeline: deal described as public reporting in 2025; supply window 2027–2030.
Outcome: chemistry and geography shift toward domestically produced LFP inputs.
Again, the source excerpt does not provide $/kWh. The investigative value is the contracting mechanism: long-duration agreements can “smooth” what parties pay relative to annual benchmarks, and energy storage procurements can anchor LFP pricing pressure earlier in time than some EV OEM contracts—affecting the segment-mix component of the index when those storage volumes scale.
BNEF’s own 2024 release credits manufacturing overcapacity, economies of scale, and LFP adoption as drivers of the 20% pack price decline to $115/kWh. (https://about.bnef.com/insights/commodities/lithium-ion-battery-pack-prices-see-largest-drop-since-2017-falling-to-115-per-kilowatt-hour-bloombergnef/)
Timeline: BNEF released the analysis in December 2024.
Outcome: structural supply pressures, not just metal costs, were sufficient to move the benchmark.
This case differs from the first two because it is “benchmark-to-structure,” not “company-to-contract.” The analytical use here is to justify treating “negotiation power” and “contract timing” as endogenous to the benchmark—i.e., the index is not merely reflecting exogenous commodity cycles.
The U.S. Department of Energy’s Vehicle Technologies Office cites an Argonne National Laboratory cost estimation tool (BatPaC) and reports an estimated $153/kWh (2022) on a usable-energy basis for scaled production, reflecting the modeling logic of a cost curve tool rather than a market benchmark. (https://www.energy.gov/eere/vehicles/articles/fotw-1272-january-9-2023-electric-vehicle-battery-pack-costs-2022-are-nearly)
Timeline: DOE post in January 2023.
Outcome: illustrates why analysts can get conflicting “battery price” results depending on whether they mean modeled manufacturing cost versus BNEF’s survey benchmark.
This case is crucial for investigator methodology: do not mix benchmark definitions without labeling them. It also sharpens the decomposition problem—if the thing you’re verifying is a market benchmark, you must account for mix and contracting; if you’re verifying a modeled engineering cost, you must account for design assumptions, utilization, and manufacturing learning.
Now we can state the verification-ready conclusion in mechanics terms.
A defensible verification cannot claim a fully quantified 2020–2025 attribution without the underlying year-by-year chemistry and segment tables. The publicly accessible snippets accessible here do not provide LFP and NMC pack price levels for every year 2020–2024 in the same way 2025 is provided. Therefore, a rigorous year-by-year chemistry time-series decomposition cannot be constructed without (a) purchasing and extracting BNEF table values or (b) using a licensed dataset that reproduces BNEF’s year-by-year LFP/NMC numbers. I will not fabricate missing year-level chemistry values.
Even with missing historical LFP/NMC values, you can still test whether mix is plausibly sufficient to generate a meaningful share of the 2020–2025 benchmark decline by running a sensitivity exercise anchored to 2025’s chemistry spread:
This produces a bounded, testable requirement on mix-shift magnitude without pretending you know the exact year-by-year LFP and NMC pack prices.
If you are building the chart.js time-series for 2020–2025 chemistry pack prices by LFP versus NMC, treat it as a data procurement task: obtain the year-by-year chemistry pack price values directly from BNEF’s annual survey tables for each year. Then:
This is the “verification” workflow that turns the benchmark from narrative into a measurable accounting identity.
BNEF expects continued declines through 2026 and beyond, but the mechanism is not only “learning.” It is still overcapacity, competition, and LFP adoption, all of which can flip faster than learning rates. In 2025, BNEF reports overcapacity and competition as the key driver of falling packs despite rising metals. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/)
Recommendation: The European Commission and national procurement agencies that buy large stationary storage and EV fleets should require bid comparability to include chemistry-tagged $/kWh and segment-tagged pricing in tender evaluation, not only benchmark references. The goal is to prevent contracts from being indexed to an all-segments average that can fall due to non-ev market pricing, leaving EV procurement exposed.
You might ask why the EU specifically. Battery policy is increasingly formalized in regulation and reporting requirements; procurement comparability is the practical complement. A related EU simplification action references due diligence reforms for batteries (context for battery policy governance), showing that the EU is actively changing battery compliance administration. (https://consilium.europa.eu/en/press/press-releases/2025/07/18/simplification-council-adopts-law-to-stop-the-clock-on-due-diligence-rules-for-batteries/)
Forecast (next 12–18 months): Through late 2026, expect the all-segments BNEF benchmark to remain downward or range-bound, but with larger divergence risk by chemistry and segment. The logic is that competition and overcapacity can keep prices falling, while chemistry mix and application tender timing can create whiplash in EV procurement $/kWh relative to the benchmark average. This forecast is consistent with BNEF’s 2025 framing that the decline is driven by competitive overcapacity and chemistry adoption rather than metals. (https://about.bnef.com/insights/clean-transport/lithium-ion-battery-pack-prices-fall-to-108-per-kilowatt-hour-despite-rising-metal-prices-bloombergnef/)
Investigators and practitioners should stop chasing a single $/kWh headline—and instead verify who paid what for LFP versus NMC, and whether those kWh came from EV tenders or from non-EV procurement.
Battery pack and cell benchmarks in 2025–2026 are diverging by chemistry, region, and contract timing, turning $/kWh from a trend into a risk variable.
Even as headline pack $/kWh benchmarks fall, LFP versus NMC mix, lithium volatility, and region-specific manufacturing premia can make OEM contract prices move differently.
A rebound in LFP-first benchmarks can ripple through global EV battery $/kWh expectations via tenders, CIF conversions, and policy pass-through.