Making Money In Flat Panel TVs And Monitors-Part 1
LCD
Flat-Panels: High Growth, Few Equity Plays
We are poised
for a spectacular boom in sales of flat-panel LCD TVs and monitors, with a
number of fabs coming on line over the next couple of years — all Asian, mostly
the existing major consumer electronics companies, though increasingly in joint
ventures with others.Â
On December 27, this column discussed growth trends in technology for 2005 and beyond, of which digital television (DTV) was one.
Digital
television (DTV) sales were a front-runner amongst the various consumer
electronics categories in 2004. According to the Consumer Electronics
Association, DTV sales in the U.S. reached $10.7 billion on 7.3 million units —
an increase of 78 percent in dollar sales and 63 percent in unit sales over
2003. One of the biggest drivers of DTV is flat-panel, ultra-thin displays
(liquid crystal display (LCD) and plasma). Total sales of LCD TVs, one of the
fastest growing segments in consumer electronics, exceeded $2 billion in 2004
and are expected to surpass $3 billion in 2005. By 2010, as many as 50 million
homes could be digital; which represents a 5-year compound annual growth rate
(from 2005) of around 75 percent.
One big
catalyst that’s looming out there is the ending of analog broadcasts. The FCC
and congress still have to figure this one out. I doubt it’ll happen by 2007,
but there’ll be a lot of pressure to make it happen as soon as possible. That
could turn into a one-time bonanza that will look a lot like 1990s telecom
deregulation. Not only will most of the country need new TVs and/or converters,
but lots of existing analog spectrum will become available for other purposes
which a variety of people are looking at.
Companies /
joint ventures that have existing known commitments to build new capacity for
flat-screen LCD panels in the coming years include:
- LG.Philips Corp. based in
Korea, - AU Optronics Corp. of
Taiwan, - Samsung/Sony Corp. (a
Korean-based JV known as S-LCD Corp.) - Sharp Inc.,
- Hitachi Corp./Matsushita
Electric Industrial Co./Toshiba Corp. (a Japanese JV in the works), and - NEC Corp./SVA Group joint
venture based in Shanghai.
The new
Samsung plant will process up to 45,000 substrates per month, each of which is
1870mm x 2200mm — that’s approximately 74†x 86″. Depending on the size of the
screens Samsung makes, it could get somewhere between 2 and 8 fairly large
screen TVs from each substrate. If the plant is making 42″ diagonal screens — a
standard size — then it would get up to four screens. For 37″ screens, Samsung
could squeeze six screens on the same substrate; for 30″ diagonal screens, maybe
eight. All this, of course, before allowing for any manufacturing losses, which
could be high at the beginning but should fall well below 10 percent as the
plant matures.
At 45,000
substrates per month, that means the Samsung plant could produce around one
million 42†panels per year; more if smaller screens are produced. Even allowing
for some manufacturing losses, that’s a pretty big number; and that’s just one
of several plants of that size that are in development by Asian manufacturers.
LG.Phillips has another fab about the same size coming on line at about the same
time. I’m not quite sure of the capabilities of all the other fabs coming
online, but they’re probably not all that different. Even with a very
conservative manufacturing yield, each of those plants could produce over a
million LCD panels each year. That said, some of the new capacity will most
certainly replace older fabs that were designed exclusively for smaller screens.
So the net increase in total capacity will be somewhat smaller — somewhere
around 4 million units per year in total.
All the new
LCD capacity being built is Asian; to the best of my knowledge there are no U.S.
or European companies building new LCD fabs. The last European company involved
in making any kind of screens was Philips, which has entered into a joint
venture with LG — LG.Philips Corp. —for manufacturing large flat-screens. I
don’t recall any major U.S. company manufacturing TVs or screens of any kind in
the U.S. in a decade. Texas Instruments could be a beneficiary of DLP (digital
light processing) technology in bigger displays where it is a cost-effective
option; although unfortunately it’s really buried in TIs other, bigger
businesses.
Making LCD
panels is a lot like semiconductor manufacturing: the technology dictates the
minimum size necessary to achieve economy of scale. The ultimate cost depends on
how many chips can be squeezed onto each of the 300mm or 240mm platters going
through the plant and the manufacturing yield (i.e., how many of them are good).
The major
difference between LCD panels and semiconductors is that in the semiconductor
space, newer processes usually allow manufacturers to “shrink” a chip over time
using smaller line sizes. So over time they are putting more and more chips on a
platter while simultaneously improving yields. In the LCD space, panels are sold
by the square inch, so the cost-cutting dynamic isn’t so strong over the long
term. However, in the short term the improvement in manufacturing process and
yields is going to result in a huge drop-off in prices, which benefits
consumers. This should allow LCDs to completely eclipse plasma screens for 42″
and smaller sizes, and most likely will replace standard CRT (cathode ray tube)
TVs in “conventional” size sets. For the biggest screens, plasma may continue to
hold an edge for a much longer time, though projection sets using Texas
Instruments’ DLP (digital light processing) technology are looking very good for
a fraction of the price.
LCD manufacturing is, by definition, a capital-intensive, volume business that
requires huge economies of scale to make it work well.
 For this reason, LCD is more likely to be used for medium-sized
panels. Larger LCDs are possible. Sharp, Japan’s leading maker of liquid crystal
display panels, unveiled a 64†LCD screen late last year. Samsung bested Sharp,
and displayed a 102†HDTV plasma TV at the Consumer Electronics Show in Las
Vegas in January; problem is that the cost was up into the low six figures. Even
allowing for some decline over time, it’s hard to see how they’d be able to
compete at any point in the foreseeable future. There just isn’t going to be
enough volume at those sizes and price-points to push things down the cost curve
at anything more than a snail’s pace.
Plasma and
OLED (organic light emitting diode) technologies are well behind LCD. Plasma
screens have a limited lifetime, deteriorating after three to four years. OLED
is a relatively nascent technology. The
biggest OLED panel made them so far is 21″Â and there are still issues
of lifetime mostly related to blue LCDs, though there has been some progress on
this. There are also IP issues — Kodak owns key patents. Ultimately, there is
some real potential here but so far OLED is not far enough along to make the
bigger screens needed for TV.
I suspect the
really big screens (100″ or so and bigger) will migrate towards a variety of
projection systems using DLP; sizes between 50 and 100″ will see a mix of DLP
and plasma; LCDs will take over most things in the mass market of 42″ and
smaller, with OLED taking some share initially in the smallest sizes. OLED will
also start taking some sales in computer monitors. Most of this is a function of
cost issues, and may change over time.
The
expectations (from EETimes) are that growth in the number of plants will slow
after 2005; but it’s a slowing in growth, not a cessation of growth. But
considering how much new capacity is in the pipeline right now, that’s not
surprising. Consequently, equipment orders for LCD plants are already
forecast to fall off this year and next.
The
improvement in manufacturing processes and increased capacity will result in a
substantial drop-off in prices — great for consumers and, by extension, unit
sales volumes. According to EETimes, cost could be down below $1,000 for a 42″
panel by the end of the year. The panel itself is the major cost to a
large-screen TV, though the proliferation of panels may actually lead to
shortages in other critical components that haven’t had as much
investment behind them. And
this is just a taste of the amounts of money being spent on manufacturing
LCDs:Â
Over the
long-term, the prices on these LCD panels are likely to follow
Moore’s Law pretty
closely, which won’t be great for manufacturers’ long-term profit margins or
total dollar sales. The flat-screen TV business is very similar to the computer
monitor business, which was discussed in this column
on
January 3. While profit margins, at around 10 percent, are currently
more attractive than the 2 to 3 percent for PCs, long-term profits are no more
sustainable for TVs than any other piece of electronics. On the bright side,
however, this should make large-screen LCD TVs more affordable and benefit
consumers.
Melanie Hollands
melaniehollands@yahoo.com