A year ago, a decent $150 smartphone was an easy sell. Enough storage for photos and apps, enough RAM to run two or three things at once, and a price that made sense for someone buying their first phone or replacing a cracked screen. That math is falling apart in 2026, and the reason has almost nothing to do with phones themselves.
Memory chips, the DRAM that lets a phone multitask and the NAND flash that stores everything on it, have become suddenly, brutally expensive. Not because factories are producing less of them, but because a much bigger customer showed up and started buying nearly everything the industry can make.
The Industry Is Calling It “RAMmageddon” for a Reason
Counterpoint Research now expects mobile DRAM prices to nearly triple in the second quarter of 2026 compared to the final quarter of last year. TrendForce has tracked NAND flash contract prices jumping as much as 70 to 75 percent quarter over quarter earlier this year, with DRAM up around 60 percent over the same stretch. Zoom out to a full year-over-year view and DRAM pricing is up roughly 172 percent, while Kingston has reported NAND costs climbing 246 percent since the start of 2025.
Those aren’t typo-level numbers. They’re the kind of price swing that normally only happens during a genuine supply catastrophe, a factory fire, an earthquake near a fab, a trade embargo. This time the cause is demand, not disaster. It is the same story we’ve covered before with the AI data-center building spree that’s already draining the world’s chip supply: hyperscalers are pouring money into servers built to train and run AI models, and those servers are hungry for memory in a way ordinary computers never were. Analysts now estimate AI data centers could consume close to 70 percent of all high-end DRAM produced this year. Every gigabyte a cloud provider locks up in a long-term supply contract is a gigabyte that isn’t going into a phone.
Industry watchers have started calling the whole mess “RAMmageddon”, and it’s not just phone buyers feeling it. Camera makers, laptop vendors, and console manufacturers are all fighting over the same shrinking pool of chips, and manufacturers like Sony have already had to pause memory card sales just to keep up with what they can actually source.
Why Cheap Phones Take the Hardest Hit
Here’s the part that makes this especially rough for budget devices specifically. Memory isn’t just one line item on a smartphone’s bill of materials, it’s one of the biggest ones, and it’s a much larger share of the total cost on a cheap phone than on a flagship.
On a sub-$400 device, memory now accounts for roughly 60 percent of the total cost of building the phone. Drop down to the sub-$99 tier, the phones sold to first-time buyers across Africa, South Asia, and Latin America, and that figure climbs past 64 percent. A flagship phone can absorb a memory price spike by trimming margins elsewhere or quietly passing along a modest price bump that a premium buyer won’t blink at. A $90 phone has no such cushion. There’s no fat left to trim when the single most expensive component in the device just got dramatically more expensive.
The market research firm Omdia now forecasts that shipments of budget smartphones will fall more than 22 percent in 2026 alone. Counterpoint and IDC both put the decline for the entire global smartphone market at 13.9 percent this year, which both firms describe as the worst annual drop on record, a mix of the memory crisis and separate geopolitical disruptions layering on top of each other. The sub-$150 segment specifically is facing what Counterpoint bluntly calls a “virtual collapse,” a direct hit to budget-focused brands like Transsion and Xiaomi that built entire business models around selling enormous volumes of inexpensive hardware.
What Happens to the Phone in Your Pocket
Manufacturers aren’t just sitting still while their margins evaporate. Rather than raise sticker prices to levels that price-sensitive buyers simply won’t pay, many vendors are freezing memory configurations at last generation’s levels. A phone that might have shipped with 8GB of RAM and 128GB of storage a year ago is now more likely to ship with 4GB and 64GB at the same price point, or the same specs at a noticeably higher price.
Either way, the buyer loses something. Less RAM means more stuttering when you switch between apps or leave a browser tab open too long. Less storage fills up faster with photos, app updates, and the operating system itself, pushing people toward constant “storage full” notifications and, increasingly, the secondhand market, where a two- or three-year-old phone with generous 2023-era specs starts to look like a genuinely smart buy compared to a cramped new one.
The ripple effects extend beyond the cheapest phones too. Counterpoint expects average smartphone selling prices across the entire market to rise 6.9 percent year over year in 2026. Gartner’s estimate is even steeper, projecting phones could cost roughly 13 percent more than they did in 2025. If you’ve been holding off on an upgrade, waiting for prices to settle probably isn’t the winning strategy this year.
Is There Any Relief Coming?
There are early signs the most extreme price spikes are cooling. TrendForce’s most recent data shows contract price growth slowing into the third quarter of 2026, down to a still-painful 13 to 18 percent for DRAM and 10 to 15 percent for NAND, compared with the 60 to 75 percent jumps seen earlier in the year. Part of that is simply consumers hitting their affordability ceiling: device makers are running out of room to pass costs along, which takes some of the pressure off the raw negotiating table.
But cooling growth isn’t the same as falling prices, and nothing about AI’s appetite for memory looks like it’s shrinking anytime soon. As long as data centers keep signing long-term supply deals to lock in DRAM for the next generation of AI training runs, phone makers, especially the ones building the cheapest devices for the customers who can least absorb a price hike, will keep negotiating for what’s left over. For now, the budget smartphone isn’t dead. But it’s noticeably smaller, slower, and pricier than it was a year ago, and the AI boom sitting in some distant data center is the reason why.
