In early February, an executive at Sharp was quoted in the Japanese press that the company was thinking about investing in a display fab in the USA. Separately, (or is it separately?), Apple’s CEO, Tim Cook has announced that Apple plans to create a $1 billion fund to support ‘advanced manufacturing. Further, a couple of weeks ago, Terry Gou of Foxconn was filmed leaving the White House.
He visited again before saying in a statement that Foxconn is planning a ‘number of expanded investments in America’.
So, what are the prospects for Apple investing in a display fab in the US? In the end, as I talked about in a recent Display Daily, one of the challenges of the display business is that the marginal cost of production is close to the marginal cost. (Why LCDs are Not Just Big Chips) That is to say, too much of the cost of an LCD is related to the direct cost and materials of the display. In the past, David Barnes has said that the business model of LCD makers is really that of a material reseller, but with the disadvantage that you have to spend a few billions on a fab in order to get into the market.
Of course, much of the material costs of an LCD display come from the backlight and other components. That is one of the drivers for LCD makers moving some module assembly to Europe in the past. By buying the backlights and other materials much closer in time to the point of sale, the panel makers were able to release a substantial amount of working capital. If they make the modules in Asia, there is a very long supply chain and the materials have to be financed for several months. Of course, Europe has high tariff barriers against TVs, which also has a big influence on supply chain decisions in the display business..
Will OLED Supply also Be a Material Supply Business?
That would change, of course, if you moved to making OLEDs or, perhaps, microLEDs. Although OLED has some expensive materials, the key to profitability seems to be technology and manufacturing capability and the consequent yield. Now, it could be argued that the US ought to able to be competitive in these areas, as it is in semiconductors and other high technology areas. If you make glass displays, you also need a relatively local glass supply, but if you make the OLEDs on plastic, that’s not such an issue. You might need a stock of substrates, but for making plastic OLEDs, these are re-cycled and act just as carriers.
Having a panel industry in the US would help with logistics costs. Bob O’Brien of DSCC has posted a blog that suggested that there might be a logistics saving by putting a TV factory somewhere in the middle of the US (although in a previous post, he came to the conclusion that a US fab was unlikely). A saving in logistics cost might well be the difference between profit and loss in the fiercely competitive TV set market, he concluded.
Forecast Average Daily Long Haul Truck Traffic on the US NHS in 2045 – Click for higher resolution
Of course, another factor in the equation is that unless its negotiations with Hisense change significantly, Foxconn will not have the rights to sell Sharp TVs in the US until 2020, so there is little point in having a fab until then. Having said that, it would take some time to set up a full supply chain for a panel fab from a green grass site.
Some sources have suggested that rather than building a G10 or G11 fab for TV panel production, Foxconn is more likely to build a smaller fab, perhaps a G6, using, perhaps, polysilicon (LTPS) technology, which would allow it to address other markets, such as mobile devices for supply to Apple and others, or to make displays for the automotive business. That certainly might be less of a profit risk than building a fab for TV products.
However, the reality of all of these big fab decisions is that the cost of capital is one of the main factors in the decisions. Although interest rates are low in the US, there is always a risk they could rise. It seems, from what I’ve seen, that Chinese factories are able to get capital at, effectively, no cost. That’s hard to compete with.
Bob
