Real Differentiation
May 16th, 2007By David Barnes, Vice President of Strategic Analysis
Panel makers have been struggling to differentiate their products for years. Leading TFT LCD producers have been able to make larger displays than new entrants have, but that is about the extent of product differentiation. Variations such as MVA or IPS have not garnered price premiums; they have only complicated second sourcing. As described in the Q1′07 issue of the Display Economics Report, TFT LCD is a commodity industry where producers have not earned their weighted average cost of capital.
I believe we are beginning to see real differentiation in capacity growth and market allocation, however. From 1997 to 2004, most TFT capacity was versatile enough to fabricate a variety of mainstream displays for PC and industrial applications. Newer fabrication plants were more cost effective than older ones, but older plants were still supplying mainstream products. Beginning in 2004, leading producers committed funds to TV panel capacity and converted older plants to serve small/medium display applications. By 2006, leading producers had brought very large substrate plants on-line that are inefficient for mainstream PC displays. There are limits to the percentage of substrates that have more than 12 panels on them, because cell sealing and cutting tools have throughputs that vary with the number of panels. If a producer tries to fill a generation 7 or 8 line with monitor product, the cell shop will become log jammed and plant productivity will fall below acceptable levels. From Q4′06 to Q4′08, we estimate that capacity in generation 7 and larger plants will increase 143%, while capacity in older plants will increase only 16%. Owners of such capacity will therefore make a different range of products than others will. The PC display market is already mature; there is little need for additional capacity for such applications. The LCD TV market is still developing, and the newer, larger plants will shape the products offered.
This change makes my former supply-price model obsolete. I was able to predict the rate of price decline and crystal cycle timing for many years by simply measuring the acceleration of capacity. According to this model, TFT LCD makers should be experiencing an unprecedented era of profitability. The model fails to consider that capacity grow has past its inflection point and is now decelerating. Those who attended our US FPD Conference in March saw my new model of price development, which indicated a slower rate of price decline.
Profit indicators by generation using the DisplaySearch Q1′07 supply-demand system
The preceding chart applies my former supply-price model to three sizes of substrates: generations 4 and 5; generations 5½ and 6, and; generations 7 to 10. It indicates positive price and profit movements for makers of notebook and monitor panels. Capacity growth in this segment has almost ceased, and it becomes negligible after 2007. This is what drives the old model to predict good times for panel makers. Looking at generations 5½ and 6, which have substrate sizes optimal for large monitor and 32″ TV panels, we see indications of substantial price decline in Q3′07. At this time, however, panel makers are negotiating for higher prices or less of a price decline, at least. Their interest in price hikes is clear; they have been losing money at recent levels. They may succeed somewhat and avoid the indicated decline. In late 2008, the chart shows another rapid price decline. The price of 42″ and larger TV panels may fall sharply at that time. On the other hand, as mentioned during my talk at the conference, TV brands may find panels in short supply at the start of 2009 because even less capacity growth will occur. Long term, this is positive for panel makers.
We may therefore see real differentiation from now on. Panel makers rich in large TV panel capacity will experience one set of market dynamics. Panel makers with a greater portion of PC panel capacity will experience another. Those with a greater portion of small/medium panel capacity will serve distinct markets and operate differently than large panel makers do. Prices and profits may no longer rise and fall in phase for all producers. Some producers may find the new profit zone and succeed better than most. There will be three distinct markets and there could be a leader in each one, as there is today in the semiconductor industry.
We all need new models!
