Thanks for the message Tad. Here are my thoughts:
1) "Over time, supply growth will accelerate or decelerate depending upon perceived rates of return." You also need to include cost of capital. Bear with me here, this may be a bit long, but it is a necessary discussion as many people on SI think that the world begins in New York and finishes in San Francisco.
Many Asian nations have developed a similar economic growth model to that employed by Japan in the post WWII period. They were characterized by a macroeconomic environment that encourages savings by citizens by suppressing domestic demand. The easy credit supplied by the banking sector, as a result of a growing deposit base, was guided by bureaucratic edict to strategically targeted sectors of the economy in the form of ultra-low interest rate loans. The Japanese succeeded in developing a healthy export sector and the Koreans followed suit. Favorite sectors were autos, steels, chemicals and, yes, semiconductors.
The problem is that banks, having to now meet stricter BIS ratios have had to cut back lending drastically, including to their much beloved chip companies. The Japanese hit this roadblock sooner than the Koreans. In fact, the Korean semiconductor makers had access to such loans that were 7-8% below equivalent market rates. Cost of capital had been kept artificially low in these countries to support the development of their targeted industries but it had another unexpected impact - the development of overcapacity. We see it in semiconductors, steel, chemicals, autos etc. If your cost of capital is cheap, so to is your expected rate of return. Samsung, LG & Hyundai had access to almost unlimited sources of cheap capital - no wonder they could build capacity at breakneck speed and go all out for market share. They were not required to have any real return on their investment.
What has changed for Asian countries is that now due to financial liberalization and global markets (not to mention answering to your share holders rather than your debt holders) is that regional differences in cost of capital are becoming less distorted. In effect, we are finally seeing a level playing field. We are only just emerging from the archaic system and the effects will only be felt over time. What it means is that Korean semiconductor makers will be subject to the same cost of capital and return on capital as their peers in Japan, Taiwan, and here. That is exactly what MU/IMF/USTR wants. So do I.
As a result we have seen a large drop in capex by the Koreans over the last year and only a very slight recovery this year - only 20-30% of levels that we saw only two-three years ago. Anyone else who expects a sharp rebound are deluding themselves and unaware of the underlying change that we are going though in Asia. With capex levels still subdued, bit supply growth will be suppressed as we are already seeing. What growth you saw in the past was the result of variables that are no longer applicable. This leads to my next point.
2) Your previous post had DRAM bit supply (I assume that is what you meant when you said "market") growing at 95%. I find this figure to be not only the highest I have seen, but unrealistic given that we have seen massive capex cutbacks, no new fab announcements, and have seen Motorola, TXN, Siemens, LG, Fujitsu, Hitachi, Nippon Steel Semiconductor, Oki Electric either announce, about to announce, imply or deny rumors that they will be withdrawing from production of commodity DRAM. Taking this into account, we are seeing some 20-30% of DRAM supply being taken out of the market. I would like to hear your underlying assumptions for your case.
3) As for MU. They will gain market share on the fact alone that their competitors are leaving the industry. The acquisition of TXN facilities will also boost their share, though I doubt it will be as quickly as they think - the TXN facilities have to be completely gutted and installed with new gear in order to become productive, that does not take months, it takes years. But then again, what do I care if there are powerful industry fundamentals that helping MU to do this. You are correct in pointing out that there is premium that is paid by investors just because MU is a U.S. stock. If I had the choice I'd prefer to buy Samsung or Toshiba. That being said, I believe that MU is a close proxy for the industry, regardless of its growing pains.
There also needs to be a "survivor premium" on this stock as well. Not only because of the fact that it is the only U.S. domiciled company but because it has survived a massive industry shakeout (comparable only to 1986). In addition, the industry structure is such that we are no longer just discounting next year's profits but are discounting an industry structure which will make profits much more sustainable over the long term. As we move to the 128meg/0.18 generation I can only see Samsung, Toshiba, NEC, MU and perhaps LG/Hyundai competitively supplying commodity DRAM.
Hope this provides food for thought. I encourage you to continue posting as I believe that this thread is sorely in need of reasoned debate. |
