Valuations have risen for many semiconductor-equipment producers - but some are still relatively cheap

Companies that make semiconductor manufacturing equipment are poised to benefit from the expansion of production capacity to support generative artificial-intelligence technology.

Artificial-intelligence mania has seemingly disrupted the normal order of the typically cyclical semiconductor industry. And with chip makers adding capacity to meet surging demand, those who make semiconductor capital equipment - known as semicap makers - appear positioned to benefit even more than they already have.

Mizuho analyst Jordan Klein told MarketWatch last month that this was the first time in years that investors didn't seem so worried about demand stability for semicap makers, which make machines, software and other complex tools required for chip manufacturing.

The enthusiasm is already starting to be reflected in the market. In early January, we published a list of 25 companies across developed economies in the "semiconductor equipment and testing" industry group, as determined by LSEG. This included all stocks in that industry group among components of the PHLX Semiconductor Index SOX, the S&P Composite 1500 Index XX:SP1500 and the MSCI World Index.

The thesis was that these were "safer" plays on the AI trend. Regardless of which hyperscalers profit the most from their spending infrastructure to support the development of AI, or whether or not Nvidia (NVDA) continues to dominate the related critical market for graphics processing units, the semicap makers will enjoy increasing demand for their products and services.

Since the end of 2025, all of the stocks on the list except for one have risen at least by double-digit percentages.

While AI demand is the main reason for the surge, chips are also getting more complex, and that has spurred further momentum for companies that provide equipment and services to semiconductor manufacturers, Klein said.

And while hot memory companies like Micron Technology $(MU)$, Seagate Technology $(STX)$ and Western Digital $(WDC)$ have been reluctant to add more capacity out of fear that demand will eventually slow down, they are making some moves to expand manufacturing, and that should require more equipment.

Micron announced late last month that it was building a new advanced wafer-fabrication facility at its complex in Singapore dedicated to manufacturing NAND, a type memory that has become crucial to supporting more sophisticated AI models.

See more: Micron's stock is surging as the company looks to cash in on a 'desperate' market

While the outlook seems rosy for chip-equipment makers as long as AI demand stays hot, investors may be wondering what to do with these stocks after their furious year-to-date rallies.

In the span of a month and a half, analysts' ratings and price targets for our basket of 25 semiconductor-equipment stocks haven't changed much, but forward price-to-earnings ratios have increased dramatically for some of them. So let's take an updated look at the entire group, sorted by their price changes in 2026, excluding dividends.

The table includes forward P/E ratios. These are Friday's closing prices divided by consensus earnings-per-share estimates among analysts polled by LSEG. For comparison, consider the weighted forward P/E of 21.7 for the S&P 500 SPX, 23.7 for the S&P 500's information-technology sector XX:SP500.45 and 25.6 for the iShares Semiconductor ETF SOXX, which tracks the PHLX Semiconductor Index.

Ultra Clean Holdings $(UCTT)$ tops the list, with its stock rising 119% this year and its forward P/E increasing to 34.6, from 16.8 at the end of December. The stock closed at $25.33 on Dec. 31, and at that time the consensus price target for the shares was $35. That implied 38% upside over the following 12 months.

The consensus price target for Ultra Clean has risen to $38.75, but that is now 30% lower than Friday's closing price of $55.39.

When we published the list in early January, all but five of the 25 stocks had consensus price targets above the Dec. 31 prices. But now there are 13 stocks trading above their updated price targets.

Then again, there are majority "buy" or equivalent ratings for nine of those 13 stocks. This suggests that although the Wall Street tradition is to set a 12-month target, the analysts have longer-term horizons in mind when assigning ratings.

Getting back to valuation ratios, it is interesting to see that for Kulicke and Soffa Industries $(KLIC)$, the forward P/E has declined even though the stock price has risen by 57%. That shows consensus EPS estimates have been rising much more quickly than the share price.

Another with a declining forward P/E is Lasertec (JP:6920), but that stock has risen only 4% so far this year.

Stocks on the list with forward P/E ratios below that of the iShares Semiconductor ETF include Screen Holdings (JP:7735), Photronics $(PLAB)$, Amkor Technology $(AMKR)$, Axcelis Technologies $(ACLS)$ and Veeco Instruments $(VECO)$.

Earnings commentary within the industry has been encouraging. Lead times for semiconductor capital equipment are usually six to nine months, meaning customers are ordering in advance, Mizuho’s Klein noted. He added that companies such as Applied Materials, Lam Research and KLA have sent positive signals about demand going forward.

Earlier this week, Applied Materials guided for more than 20% growth in its semiconductor-systems business for 2026.

Read: Applied Materials’ stock jumps after earnings. The CEO just made a bold prediction for the chip sector.

Applied Materials is “the largest semicap with the broadest portfolio,” Bank of America analyst Vivek Arya noted, writing that the company is therefore positioned for “strong leverage” in the multiyear growth cycle for wafer-fabrication equipment. Demand for such equipment is being driven by the need for leading-edge foundry and logic chips, and for dynamic random-access memory, he wrote in a Thursday report.

RBC Capital Markets analyst Srini Pajjuri also highlighted Applied Materials’ leadership when it comes to equipment for DRAM and high-bandwidth memory, as well as for advanced logic chips and packaging technology — all of which are in high demand to support AI. He wrote in a note to clients on Thursday that he expects spending on DRAM to keep rising as semiconductor manufacturing capacity expands and as companies launch greenfield projects starting next year.