Hard Tech Sanity Check: Asian P/E’s Too High
Times are tough for Asian technology stocks. But as long as there is volatility, there is money to be made. TechBuddha has recently been very active on the short side of the market for a number of reasons: 1) NASDAQ valuations are in free-fall, 2) Asian techs are still expensive relative to US techs, and 3) Asian techs have decidedly bleaker prospects than their American and European cousins. Let’s quantify these musings with a look at the Asian ADR (American Depository Receipt) scene.
The title of this article implies that we are looking at companies that have “e” in the first place. Some profitable Asian techs with ADR’s include: Chartered Semiconductor (CHRT US), Taiwan Semiconductor (TSM US), and United Microelectronics Corp. (UMC US).
Each of these stocks has a slightly different story, but all are susceptible to certain expectations overhangs that we have built up for them over the past 2 years. A quick review of Asian valuations leads one to conclude that we still have a long, bitter road ahead.
Let’s start with the semiconductor stocks. The first potential danger is earnings expectations: the street is being too optimistic for 2001 and pegs 2002 earnings growth at 65-80% for CHRT, TSM, and UMC. The second danger is that Asian semiconductor stocks are 2-4x more expensive (on a P/E basis) than US stalwarts like AMD. One (CHRT) is actually 25% more expensive than Intel (INTC US), an astounding fact indeed.
Why the negativity? Why can’t we just accept the oft-cited spiel: Asian foundries will be the beneficiaries of outsourcing by Japan and explosive growth in demand for telecommunications IC’s. Three reasons: 1) they produce commodities; 2) they have excess capacity; and 3) they are value destroyers.
Before progressing further, let’s look at the projected P/E ratios for 2001: CHRT 52x; TSM 27x; UMC 20x. And don’t forget, these companies are going into a down year, facing earnings declines of anywhere from 40%-60%. They have no revenue visibility and are revising capacity utilization downward at a rate of 15% per month. To make matters worse, Nokia (NOK US), one of the world’s biggest consumers of semiconductors, has revised down its sales estimates for 2001. Other producers of consumer electronics are similarly pessimistic.
Semiconductors are commodities, just like pork bellies and wheat. Similarly, the foundries that make them have neither intellectual property, nor any particular competitive edge over one another, save for access to the capital needed to build the next fab.
Being a commodity producer is not a problem as long as supply/demand works in your favor. Unfortunately, this is seldom the case and will certainly not be the case in 2001 and 2002. Who is to blame? China of course. The investment dollars currently pouring into Chinese foundries will ensure that semiconductors are in surplus for at least the next three years. China will deflate the rest of the world once more, as it has done in the past with textiles, computer monitors, and plastic toys. China is the reason why TSM, CHRT, and UMC will someday trade at one-third of their current multiples.
But some say that semiconductors will be the first techs to recover from the bear market. This is only partly true. Being cyclical stocks, semiconductors tend to move 18 months before their fundamentals. What frightens us, however, is that the industry is still building capacity as we head into a demand downturn. That sort of behavior lends one to believe that the semiconductor trough is a bit further off than mid-2002.
So why have Intel (INTC US), AMD (AMD US), Applied Materials (AMAT US) and others been rallying as of late? We believe it due to misplaced hope that techs have finished their decline and that the electronics industry will be saved by a V-shaped recovery in the US. Sorry Charlie, we don’t buy it.
However, when the recovery does peep above the horizon, you can bet that these stocks will experience sharper, more sustained recoveries than their Taiwanese and Singaporean cousins. That, in turn, owes much to their ability to add value through the creation of intellectual property. INTC and AMD design and build CPU’s while AMAT designs and builds the machinery that makes semiconductors. To put it bluntly, CHRT is to INTC as Yue Yuen is to Nike, as Coke Amatil is to Coca Cola, as…well you get the idea. So why the 50x P/E multiple?
Why, indeed. An especially worrisome sign is that CHRT’s ADR trades at a 5-10% discount to the local stock (CSMF SP). Meaning that demand for the stock among foreign investors is flagging and the local market hasn’t yet gotten the hint yet. It doesn’t look good for CHRT.
As for TSM and UMC, yes they are cheaper than CHRT. But let’s not stop there. If you are not a QFII (translation: Big Kahuna), you are not investing in the local Taiwanese stock market. Instead, you are buying TSM’s ADR, which trades at a 42% premium to the local share. The premium has always been there, a function of the eagerness of foreign investors to participate in the Taiwanese stock market.
What we find hard to understand is this: if you can buy AMD at 12x earnings, why buy the rather illiquid TSM ADR at 21x earnings? If you don’t have a good answer to that question, then you shouldn’t be paying a premium for TSM. Or for UMC, for that matter (UMC trades at a 14% premium).
As a final note of caution, let us point out that the P/E ratios cited are derived from mainstream analyst estimates. There is no reason why the “e” in the denominator will not move down from their current position. In fact, they will almost inevitably do so as the markets continue to slump and fundamentals deteriorate further. This presents us with an ugly scenario: falling share price and rising P/E. That’s when the fear will really set in. Trust us, you don’t want to see the curtain call on this show.
techbuddha.com |
