DisplaySearch Blog

Blog Versus Grog! Quiver With Excitement as the Beasts Battle!

by David Barnes, Vice President, Strategic Analysis

One of the equity analysts I asked to speak at the upcoming TV Supply Chain Conference was unable to attend, but he was able to pass along a report his team issued on the “Flat Panel TV Wars.” In this report, Yuji Fujimori of Goldman Sachs Japan gave reasons for modest expectations of LCD TV stock prices. He noted several developments that I will cover in detail at the conference: increasing market power of global brands like Sony, continuing panel price declines in the near term but potential for margin improvements in the long term if brands can integrate LCD module production. DisplaySearch just released it Q2’08 Large-Area TFT LCD Shipment Report, so I’d like to comment on how this data supports his thesis.

I am famous within Philips for asserting that “pixels are free.” I made this bold statement repeatedly at a pre-merger (of LG.Philips Display, now LP Display) meeting in Eindhoven. It isn’t true, strictly speaking. The statement was valid from the standpoint of CRT and PDP positions relative to LCD but not relative to the LCD TV market itself. Within the LCD industry, more pixels for a given panel size imply less yield and more material cost for pixel drivers and backlights to support more pixels. My point at the time was that this additional cost was less than either CRT or PDP makers would experience and in that I was correct. For LCD panel makers, however, increasing pixels from HD to full HD (FHP or 1080p) increases panel cost, albeit less than it would for CRT or PDP producers. I introduced a metric to account for this cost at the US FPD Conference in 2000. It was called the PixInch price (PIP). In brief, a PixInch comprehends the area and pixel cost of a panel (for those of you who remember your high school trig, a PixInch is the diagonal inches squared times the cosine of the aspect ratio times the vertical pixel or line count). After scaling, the PixInch price for TV panels results in the following table.

Table 1   Q1’08 TV Panel PixInch Price (PIP) and YoY Changes

Source: DisplaySearch Q2’08 Large-Area Shipment Report

Looking at the time series from Q1’04 through Q1’08, there are two important takeaways. The most important insight is that 40”–49” panels are selling at a PIP discount to 26”–39” panels and that 1080p (FHD) panels are selling at a PIP discount to 720p (HD) panels. These observations imply that the average PIP will fall more rapidly than it has in the past year as suppliers bring new Gen 7 and Gen 8 capacity on-line over the next four quarters. Total Gen 7 and larger substrate capacity will rise about 40% by Q1’09. Initial Gen 10 capacity will come on line the following quarter. Total Gen 7+ capacity will more than double by Q1’10. This surge of capacity, ideal for panels in the 42” regime, will put even more pressure on prices. I find it difficult to imagine 32” panel prices staying level while the price of larger panels decline at double-digit rates. Sharp and Panasonic have already announced their intentions to allocate some new capacity for 32” products. Add to this Sony’s aggressive 2008 TV targets as described in Fujimori’s writings or in DisplaySearch reports and one can only conclude that PIP must fall more rapidly than it has in recent quarters.

A second insight is that 50” and larger LCD TV panels have been sold in small quantities to date but the PixInch price for these is less than the PIP for 26” to 39” panels. The implication is that panel makers are not earning a premium for their display area and pixel risk, and that customers are expecting suppliers to manage costs even though suppliers shipped 88% fewer PixInches of 50”+ panels than they shipped PixInches of 26”–39” panels in Q1’08. Suppliers have had to move down the learning curve very quickly. This means retailers will have to move down the ASP curve also. In prior years, retailers have been able to sell consumers up to larger LCD TV sets or to 1080p displays and thereby increase their average sale price. If 32” set prices fall as I think they must, retailers may be able to sell up, but to lower average prices than in the past. High-end consumers will certainly buy more 52” and larger sets than they did in the past, but if the monitor market is any guide, 52” sets will soon sell for the same price points 42” sets sold for in the 2007 holiday shopping season.

Such trends suggest that the LCD TV sector will come under profit pressure as very large TV sets become commodities. That is not a novel insight, but Fujimori’s writing about module integration was. When I was with LG Display (LPL), the strategic value of forward and rearward integration was frequently studied. The general conclusion was that selling cells, rather than assembled modules, reduced a supplier’s return on invested capital. Even if the panel maker obtained the same operating profit margin by selling cells, the reduction in material value-added and sales revenues threatened the supplier’s ability to reinvest and sustain its capacity share. Comments by AU Optronics executives during recent earnings calls affirm this. It is therefore logical for panel makers to avoid the foundry business model. It is also logical for their customers to prefer suppliers becoming foundries. TV brands or their OEM would like to capture module value-added or eliminate some of the module cost through integration. If you start with a clean sheet of paper, I doubt anyone would design a TV production line in which video and set control functions where on separate circuit boards when these functions could be on a single board with the panel control. I doubt anyone would design a line in which a display module was packaged as a separate entity then double handled and put into a second package that could serve the same purpose as the first. There is a lot of redundancy in the present industrial structure. Redundancy and commodity don’t mix.

I assume that Sony and other leading brands plan to design more efficient manufacturing flows and thereby preserve or improve operating margins after the next wave of capacity comes on-line. Vertically-integrated TV brands are increasing their direct and indirect control of panel capacity, which may give them the ability to redefine the industry. If they are successful, their actions may force merchant suppliers and their regional-brand customers to alter their business models, also. Over the next eight quarters, we will see the crystal cycle at work in the LCD TV sector as large panel prices decline. After that, we will see how global brands affect the following round of capital investment.

Bookmark to:

Hide Sites