Fig. E1 shows the 12/12 worldwide monthly growth rates for IC sales in dollars, units and ASP for January 1997 to June 2009 inclusive. They need to be looked at in conjunction with the other 12/12 and rolling 12-month charts provided in the Market Summary section of this report.
June’s total semiconductor sales came in at US$19.3 billion, heralding a US$51.7 billion second quarter, up 16.9 percent on Q1-09 (down 20 percent on Q2-08). This compares with Q1-09 that was down 15.3 percent on Q4-08 (down 30 percent on Q1-08) and confirms our 2009 forecast upwards revision, reported in last month’s Report and at our July Mid-Term Industry Forecast Seminar, that the worst of the chip market recession is now over.
We can now expect a seasonally strong Q3 (albeit not too strong) of around 12 percent growth on Q2-09 (down 16 percent on Q3-08) followed by a normal year-end slowdown in Q4 at plus 3 percent (up 14 percent on Q4-08) confirming our minus 14 percent forecast for the year as a whole. At last it is now back to industry normal abnormality.
There are wild fluctuations when looked at on an individual monthly basis meaning no single month’s data is a good indicator of the underlying trends. Each month is thus just another peg in the ground, especially during a period of rapidly changing conditions.
June’s minus 25.8 percent year-on-year growth thus looks closer to our original minus 28 percent forecast for the year, rather than the minus 14 percent we reforecast last month, but this does not take into account (a) the prospective second-half-year rebound and (b) the fact we will be measuring future 12:12 growth rates against a dynamic whereby the 2009 numbers are trending up whereas the 2008 numbers were trending down, amplifying the impact of the 2009 positive monthly trends. We should start to see this upward trend kick in again with the release of July’s WSTS data.
There are still several wild cards however in play. Units are now much better aligned with real demand but ASPs are all still over the map, hardening in memories but weak in logic. So too is near-term fab capacity, with tight-geometry 300mm capacity now getting tight but ‘loose-geometry’ 200mm capacity still plentiful. This will send mixed signal on pricing over the second-half of the year, which in turn is likely to lull the industry into a false state of complacency.
The July move into positive territory of the Front-End Book-to Bill ratio may have finally broken the 34-month spell of a book-to-bill less than parity (i.e., since Sept 2006 aside from the 2 two-month blips), the actual spend numbers are still derisory in absolute terms. Spending is still currently more to do with linebalancing adjustments than capacity build out and will do nothing to alleviate the 2010 capacity shortage.
The Cap Ex billings run rate is circa $800m/month, supporting a chip sales rate of $16b/month; that is barely 5 percent of sales. So, either we have suddenly got 3x mega-efficient at building ICs (we have not) or we are building ourselves a massive capacity problem down the road (we are). The foundries (i.e., TSMC) will be the beneficiaries.
Fresh data points are now arriving each week indicating that the global electronics industry is rebounding from its 2008-09 financial meltdown. DRAM and PC sales are up with the impetus for renewed growth and recovery coming from Asia.
The IMF is currently forecasting a return to world GDP growth in 2010 at +2.5 percent, up from its +1.9 percent estimate made earlier this year, but the world could just as easily tip into a second global recession triggered either by the current sharp rise in oil prices or downstream inflation caused by the current excess liquidity and the longer-term need to increase interest rates everywhere.
Interest rate rises will hit everyone very hard indeed, especially those firms and individuals over-extended in debt, currently saved only by interest rates at near zero levels. We are thus nowhere near out of a moribund economy woods, indeed it is more likely to get worse before it gets better making a W-shaped economic recovery the most likely scenario, unless the economic balance of power has shifted to Asia as the new engine of economic growth for the 21st century.
Table C1 shows the quarterly semiconductor equipment sales trends for the period Q1-2008 through Q2-2009 inclusive. Total Q2-2009 equipment sales were US$2,666 million, down 13.3 percent from Q1-2009, which in turn was down 34.8 percent from Q4-2008. This represents the fifth successive double-digit quarterly fall in Cap Ex spend, unprecedented in the history of the chip industry.
Q2-2009 wafer fab equipment sales were down 67 percent on Q2-2008, the fifth consecutive quarterly high double-digit drop, with further declines in prospect, albeit at a likely slower rate, bringing 2009’s Cap Ex in at between 50-60 percent down on 2008. Cap Ex levels are now running at levels not seen since the early 1990s when the overall chip market was one-third its current size.
The quarterly trends are not much better with Q2-2009 front end Cap Ex down 17.2 percent versus Q1-2009. This was on top of the four previous quarterly declines of 35.3 (Q1 vs Q4), 22.6 (Q4 vs Q3), 20.3 (Q3 vs Q2) and 27.2 (Q2 vs Q1) percent respectively.
It should not be forgotten that these cutbacks were not triggered by the current chip market recession; the first two quarterly drops, namely Q2 and Q3-2008, took place against a backdrop of strong IC unit growth, i.e. well before the Q4-2008 chip market collapsed. In other words, these cutbacks were premeditated not diagnostic which makes the current capacity dynamics different from before.
The cutbacks were thus a clear intent to engineer tight capacity, a strategy that would by now have bitten home had it not been for the cruel interruption of the Q4-2008 market collapse. This time is different … this has NEVER happened before.