Will Hard Drive Prices Come Down in 2027?

What changed and why

The disagreement between these forecasts is narrower than it looks. Nobody argues that supply is about to overtake demand. The question is only how fast new capacity arrives and how much of it is spoken for before it reaches a shelf, and on that the published record is unusually explicit.

Seagate has said that the vast majority of nearline exabytes are allocated into calendar 2028, that product configurations and pricing are contracted through calendar 2027, and that customer planning conversations now extend into 2029. That is a company describing its output as sold before it is built. Western Digital has described calendar 2026 as sold out. Neither statement leaves much room for a retail price to fall while those commitments are being filled.

Morgan Stanley's June 2026 work puts numbers to it: demand growing 40 to 50% a year against supply growing 30 to 35%. A gap of that shape does not close because a new plant opens; it closes when the growth rates cross, and the model has that happening around 2028. The more optimistic reading, reported through DigiTimes, expects improvement during 2027 as capacity comes online, with the explicit caveat that AI demand may absorb it.

Both can be right about their own assumption. What neither can settle is the share of new nearline output that goes to hyperscale contracts before anything reaches retail, because the makers do not publish it. That single unknown is what separates a 2027 answer from a 2028 one, and it is why we quote the forecasts rather than averaging them into a number of our own.

Who it hits, and how hard

If you need capacity now, the forecasts are not an argument for waiting. The earliest relief any cited source allows is 2027, and the nearer-term statements point the other way. An array that is short of space costs you something every month it stays that way.

If your purchase can wait a year or more, the picture is genuinely split, and waiting is at least a defensible bet rather than a hopeful one. Watch the makers' capacity statements rather than retail prices: retail is the last thing to move in either direction.

If you are replacing a failed drive in an array, none of this applies. Buy the replacement, because a degraded array is a risk that compounds while you shop. Our failure rate table is the better input for that decision than any price forecast.

What to do now

1. Buy for the capacity you need in the next year, not the next five. Speculative capacity is the most expensive thing in this market.
2. Compare on cost per terabyte with our calculator, since the per-terabyte advantage of larger drives widened through 2026.
3. Check the live spread against flash on our SSD versus hard drive page before assuming spinning storage is still the obvious answer for your use.
4. Look at price per TB history for the line you want rather than the market as a whole, because the increases were very uneven.
5. If you are considering waiting for a bigger HAMR drive, read the HAMR roadmap first: the highest capacities are not reaching retail.
6. Set a price alert rather than checking daily, and revisit when a maker's capacity guidance changes rather than when a shelf price wobbles.

Outlook

For prices to fall, either new platter capacity has to arrive faster than demand grows, or demand has to slow. Nothing in the published record supports the first before 2027 at the earliest. The second is a question about AI infrastructure spending rather than about storage, and none of the sources above forecasts it.

The earliest honest signal will come from the makers' own capacity and allocation language, not from retail. When Seagate stops describing nearline exabytes as allocated years out, that is the change worth acting on.

We publish no forecast of our own. Everything above is someone else's published position, with the date it was published and a link to it, because a price forecast from a site that sells against those prices is worth nothing to you.