In a seminar held in London Penn expressed exasperation at other market forecasters who are saying the global chip market will return to single-digit growth despite the underlying circumstances of strong growth and rising average selling prices due to managed undersupply of manufacturing capacity.

Penn said that 2018 would see global chip sales would likely hit $500 billion and year-on-year growth of 21 percent. He expressed his optimism as a range from $477 million and growth of 15.6 percent on the bearish side to $510 million and growth of 23.5 percent on the bearish side of the spectrum.

In contrast, market forecasters such as IC Insights and Gartner are providing growth forecasts for 2018 of 8 percent and 7.5 percent respectively, and the World Semiconductor Trade Statistics organization’s estimate for 2018 given in November 2017 was 7.0 percent.

Penn said that barring an economic collapse: “This recovery has nowhere-near run its course.” He said he despaired of market analysts who somehow felt inhibited from saying what they believed and instead repeated a consensus what others said.

When challenged that 2017’s boom was due largely to strong memory growth based on high average selling prices and that this could turn quickly, Penn said that was true but that memory was a “trigger event” and that demand was strong across the board and that ASPs are rising across all IC types. He also said that the industry is marked by long lead times on many types of component if not by allocations.

Next: Phantom fab capacity

On the other hand 2019 is tough to call, Penn said. This is because the final figures in 2019 will depend on when in the year any correction kicks in. “But bust always follows boom.” Penn said overcapacity would be the likely trigger but this might not be due to overheated spending by chip manufacturers. He observed that as the numbers of manufacturers has reduced they have avoided speculative fab building.

However, an economic shock could collapse demand thereby triggering oversupply, or there is the potential of the phenomenon of “phantom fab capacity.” Penn pointed out that 3D-NAND yields are poor at present even while the manufacturers are racing to higher layer counts. It would only take a manufacturer to improve yield significantly and production of 3D-NAND could suddenly jump by 50 percent or a third. And as costs dropped so would prices to gain market share.

In 2017 Penn was also bullish and at the beginning of the year predicted between 8 and 16 percent growth for the global chip industry while other companies were predicting low single-digit growth. In January 2017 WSTS was predicting 3.3 percent growth and was forced to continually raise its estimate through the year.

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News


Linked Articles