The price of lithium, one of the key mineral commodities of lithium-ion batteries, has risen dramatically over the past couple of years. Under ordinary circumstance those costs would be passed along to manufacturers and, ultimately, consumers. But with respect to li-ion batteries, the current business cycle does not seem to be ordinary.

Many analysts see continued declines in the overall cost of energy storage, including li-ion battery storage costs, despite rising lithium prices. And despite rising demand, some analysts think lithium prices are in a bubble that could soon pop.

A report done last year for the World Energy Council by DNV GL, PwC and others estimated that energy storage costs could fall by as much as 70% over the next 15 years. And Lazard’s most recent levelized cost of storage (LCOS) report noted a decrease in the median cost of using lithium-ion storage technologies — 12% for peaker replacement, 24% for transmission investment deferral, and 11% for residential use cases, compared to the previous year.

Falling cost is one of the key reasons analysts cite for their expectation of the continued rapid growth of energy storage, particularly li-ion battery storage. In its third quarter the U.S. Energy Storage Monitor, GTM Research and the Energy Storage Association said they expect the U.S. energy storage market to grow to 260 MW in 2016, up from 226 MW in 2015.

A report from Technavio last year put the compound annual growth rate (CAGR) for li-ion batteries at 72%, hitting 3,130 MW by 2020, compared with a 67% CAGR for the lead-acid battery segment, which Technavio expects to reach 990 MW by 2020. A first-quarter 2016 report from Navigant Research noted that li-ion batteries have become the leader in utility-scale energy storage applications because they offer “the best mix of performance specifications for most energy storage applications.”

In August, BMI Research said it expects li-ion battery technology to “solidify” its dominant position over the coming decade based on the ability of the technology to perform storage tasks across multiple sectors. Meanwhile, BMI expects vanadium flow batteries – a technology that has been cited as a contender to disrupt li-ion’s dominance — will become an attractive storage option in applications that involve isolated or unreliable power systems where the technology can help displace the need for expensive diesel-fired power and support intermittent wind and solar power generation.

Behind the robust outlook for li-ion battery storage is growing demand for storage of all sorts, from utility scale to electric vehicles, and that has caused a sharp rise in the price for the underlying commodity, lithium.

Lithium has a lot of commercial and industrial uses, such as the manufacture of ceramics and glassware, but the fastest growing use segment is batteries, which represented about 44% of lithium demand in 2015 and is expected to reach about 55% by 2020.

In April London-based consultancy CRU Group noted that global consumption of lithium rose at an annual rate of about 11% between 2010 and 2015 and that the prospects of future demand has led to an increase in Chinese spot prices from about $7,000 per tonne in September 2015 to over $20,000 per tonne. Looking ahead, CRU sees robust demand extending through to 2020. In a subsequent blog post, CRU said that in a market where the lithium price curve has turned “almost vertical, it is hard to decide on the most likely future price path for lithium.”

In a December blog post, Benchmark Mineral Intelligence – a London based consulting firm and not to be confused with BMI Research, which is owned by Fitch Group – said it does not expect lithium prices to crash in 2017, but expects further price increases.

“The difference between lithium and other minerals and metals is that it is underpinned by a real demand story,” BMI noted.

In December 2015, Goldman Sachs called lithium “the new gasoline.” And in November Credit Suisse analysts said they expect shortages in lithium, as well as two other minerals used in batteries, graphite and cobalt.

But not all analysts share that outlook. In October, in a report called “Lithium: the Big Short,” AllianceBernstein analysts noted that while demand for the metal could double, “the lithium frenzy should all end in tears.”

AB analysts argued that the mainstream reasoning for the continued rise in lithium prices focuses solely on demand and “totally ignores the supply side of the equation.” In addition, barriers to entry in lithium production are low and technological innovations, namely, a proliferation of li-ion battery megafactories, will change the dynamics of the market.

The AB analysts say there are few supply side restraints on lithium that would justify a continued bull market in the metal. Far from being rare, they note that lithium is plentiful and one of the most under utilized commodities in the world with reserves that could last for 431 years. That compares with 39 years of copper reserves and 15 years of zinc reserves.

AB also notes that while most of the world’s lithium is produced in China, Chile, Argentina and Australia, high prices are drawing new production from places such as Mexico and Serbia that could result in “overwhelming surpluses.”

The fact that rising prices are increasing supplies was also noted in an October report from BMI Research. BMI commodities analyst Molly Shutt noted that world lithium production in 2015 was 32.5 thousand tonnes with Argentina, Australia, Chile and China accounting for over 90% of output.

Shutt expects Australia to remain the top producer but to increase supply to 85.1 thousand tonnes by 2020, from 18.8 thousand tonnes in 2016. She also expects Chile and Argentina to ramp up production. And, she says, there is plenty of room for new players, such as Bolivia to enter the field. She noted, however, that Bolivia’s production will probably not take off as quickly as new production in Argentina because of regulatory restrictions.

And while Tesla’s “giga” battery factory in the Nevada desert has attracted a lot of attention, it is only one of 14 mega factories for batteries under construction, including nine in China. Those factories could soak up a lot of demand. In the case of Tesla, they could also increase supply.

In September 2015, Pure Energy of Vancouver, British Columbia, signed an agreement to supply lithium to Tesla’s Nevada giga battery factory from brine deposits in Clayton Valley, Nev. The agreement calls for Pure Energy to supply Tesla with the mineral for five years at prices below current market rates.

Lithium, as well as minerals such as cobalt, nickel and graphite, play an essential role in manufacturing batteries. Tesla CEO Elon Musk has downplayed that role, comparing lithium to “the salt on the salad.” He added that lithium represents only about 2% of the material in Tesla battery cells.

Navigant Research estimates lithium materials make up around 10% of overall battery production costs. All things being equal, a doubling in the price of lithium would mean a 10% increase in battery production costs. But in an April 2016 blog, Navigant analyst Scott Shepard wrote that “price increases from materials may be easily absorbed by battery makers as costs are cut elsewhere through economies of scale or energy density improvements”

It should also be noted that it takes several years for a new mine or increased production to go from the ground to the market. The metal has to be transported and refined, creating a lag in cost pass-throughs.

“In the short term, it is very likely that rise in lithium prices will adversely impact energy storage costs,” Ravi Manghani, director, energy storage at GTM Research, says. He noted that lithium costs typically make up single digit percentage of battery pack costs and an even lower percentage of energy storage system costs. So even if lithium costs double, it would not result in more than a 5% to 10% increase in costs.

In a market that is increasingly commoditized, a 5-10% price increase could make or break stationary storage and EV economics, “but we believe that in medium-to-long term, this push-pull will even itself out,” Manghani says.

In addition, as Manghani noted, “most large companies hedge themselves with longer term contracts and are well protected from short term price fluctuations.”