+44 (0)24 7671 8970
More publications     •     Advertise with us     •     Contact us
 
Technical Insight

JDS Uniphase takes a big bath (Portfolio)

Can JDSU shrug off a $50 billion loss? Meanwhile, semiconductor equipment and materials suppliers are living through the Chinese curse of "interesting times". Marie Meyer reports.
"May you live in interesting times" says an old Chinese curse. "Interesting" seems like the most polite thing that one can say about today s financial markets (see ) . Case in point: JDS Uniphase can now lay claim to the dubious distinction of generating the largest corporate loss in history $50 billion. Yet Wall Street has, for the most part, taken the news in its stride, and refrained from punishing the stock even further. JDSU appears to be engaging in a classic accounting practice that has recently acquired the nickname "big bath". It consists of gathering up lots of bad news and delivering it all in one batch as a massive one-time hit to the balance sheet. Nortel and Corning have already done the same. A maneuver best used when times are bad and expectations are low, it helps to make earnings reports for future quarters look better. Moreover, there isn t a lot of downside to the company taking the bath. While $50 billion is a huge number, most of the loss that JDSU reported was a "paper loss" that has no impact on the company s cash position or other important aspects of operations (see ). $44.8 billion of the loss stemmed from write-downs of goodwill relating to corporate acquisitions. "Goodwill" represents the difference between the asset value of an acquired company and the total purchase price. For example, JDSU paid $41.5 billion to acquire SDL, which had tangible assets valued at $2 billion. The difference between the two numbers is "goodwill". It is a depreciating asset, appearing on the earnings statement as goodwill amortization, an item that counts against earnings. By taking advantage of this opportunity to slash its goodwill figure, JDSU has put itself in a position to deliver better reported earnings in the future, without a material alteration in its "real-world" financial position. However, the reduction in goodwill should also be seen as a tacit admission by the company that it overpaid for the companies it acquired. But again, the damage done to the company is difficult to quantify, because it paid for these companies with its own stock, which was also overvalued. The three biggest acquisitions JDSU made in the last two years were OCLI, E-TEK and SDL. At the time these deals were announced they were valued at $2.6 billion, $15 billion and $41.5 billion respectively. However, each was an all-stock deal, and these "value" figures reflect the price of JDSU s stock on the day the deal was announced. If they were valued at today s share price, they would drop to $460 million, $1.2 billion and $3 billion, respectively. Obviously, write-downs of this sort would be excruciating if the company had paid in cash. And they may still be painful for any former OCLI, E-TEK or SDL shareholders who received an exchange of JDSU shares and did not cash them in. But, for JDSU itself, these write-downs can be seen as simply the cost of eliminating 90% of its competition in the fiber-optic components market leading to who-knows-what level of opportunity when recovery kicks in. Aixtron and Emcore In a topsy-turvy world where a $50 billion loss can be shrugged off, it probably shouldn t be surprising that reporting good news can spark a sell-off. This appears to be what has happened to Aixtron and Emcore, as both companies tested new 52-week lows after announcing good results for the June quarter. Aixtron reported second-quarter sales of Euro 58.9 million (Euro 1 = $0.90), roughly even with the previous quarter and 36% higher than the year-ago period. Operating income was Euro 14.4 million, up 39% over the comparable quarter in 2000 and roughly flat sequentially. Aixtron s revenues for the first six months of the year were 79% higher than last year. For the year, Aixtron is forecasting total revenues of Euro 239 million and net income of Euro 30 million, up 51% and 62% respectively over the results from 2000. For the June quarter Emcore reported record revenues of $53 million, up 10% over the previous quarter and up 76% compared with the year-ago quarter. Income, excluding goodwill amortization, was $915 000, compared with $31 000 for the March quarter. Investors are not responding on either side of the Atlantic. Shares in Aixtron have slid 46% since June; Emcore shares are down 35%. The timing of these declines is difficult to understand. Admittedly, Aixtron s sales forecast for the year indicates that the next two quarters will be flat compared to the previous two, but in the current climate that would seem to be no small achievement. There have been reports in the European press that some traders are selling because they doubt whether the MOCVD market can continue to grow at 75% per year. However, this would seem to be such an overstatement of the obvious that it is difficult to believe that this news was not factored into the share price long ago. Aixtron s management recently reiterated their estimate of "growth of an average 30% p.a. for the next few years", which would ordinarily seem like more than enough to satisfy investors. The case for Emcore s decline is even more difficult to understand, as the company would seem to be doing everything right. Aixtron has always been consistently profitable, but Emcore has not. However, this year they have, for the first time in recent memory, reported profits in two back-to-back quarters (see ). ("Profits" here being defined as operational profit, before allowances for goodwill amortization.) Moreover, the company appears to be making genuine progress in its efforts to enter the fiber-optic component market. While it is true that this sector is seriously out of favor with investors right now, at least Emcore is targeting the short-haul segment, which is showing more signs of life than the long-haul market. Other encouraging signs: the company raised $175 million via a debt offering, giving it plenty of cash to plow into R&D for that nascent opto product line. Revenues have grown sequentially over the last six quarters (averaging 10% per quarter). And Emcore also sold off its interest in the loss-making LED joint venture with Uniroyal, which had put over $4 million in red ink on Emcore s balance sheet in just the past two quarters. So why isn t Wall Street responding? Traders have indicated concern that the VCSEL product line will take off slowly, due to order push-outs from some customers. But more to the point, Emcore may be pulled to earth by its MOCVD equipment business. The centerpiece of Emcore s corporate strategy for the past four years has been to diversify beyond equipment manufacturing into potentially more lucrative areas, such as epiwafers and devices. But the pronounced weakness in the HBT and HEMT epiwafer market in the first half of the year resulted in Emcore s shipments of materials declining 10% in the June quarter, while equipment shipments rose 20%. As a result, equipment sales now constitute more than 75% of Emcore s revenues for the first time since December 1999. Thus the same forces that are depressing Aixtron s shares may also be working upon Emcore To this commentator, in both cases the timing seems very odd indeed why sell off now, when things are looking reasonably good, rather than four or five months ago when things were looking truly awful? Perhaps the fact that these companies are hitting new lows is another sign that the bottom has been reached. Epiwafer vendors report As Emcore s results demonstrate, the market for electronic epiwafers has been pretty horrendous in the first half of this year. HBT-specialist Kopin has seen its market cap reduced by two-thirds. But recently its share price has begun to rebound from the depths that it reached in April (see ). For the June quarter, Kopin reported revenues of $9.2 million down sharply from the $24.4 million reported in the year-ago quarter. Most of the decline came from the epiwafer business down from $19.5 million to $3.8 million. In contrast, the company s display business was up 8%. Kopin has been hit hard by the turmoil at Conexant, historically its largest customer. Purchases have dropped from $10 million per quarter in 2000 to $3.5 million in the June quarter, according to analysts at C E Unterberg Towbin. But Kopin is having some success attracting new customers, who accounted for 50% of revenue in the June quarter. The company landed contracts to supply InGaP HBT wafers to Alpha and Anadigics, and is sampling at Motorola and TriQuint. Kopin should also be able to begin delivering improved gross margins in the epiwafer business, as 6" wafers now accountfor 50% of shipments (up from just 10% in the first quarter). However, the company says that its capacity utilization will, at best, reach just 4050% by the end of the year. Revenues at fellow epiwafer vendor IQE have held up better, largely due to its more diversified product range. Revenues for the June quarter were GBP 13.16 million ($18.97 million), a record for the company. This was double the revenues reported in the year-ago quarter, but flat compared to the previous quarter. Operating profit before goodwill and one-time charges was GBP 1.29 million ($1.86 million). Again, this was a major improvement over the year-ago period (by 3 X), but flat compared to the March quarter. Commenting on the results, IQE CEO Drew Nelson said that the III-V epiwafer operations performed well in the first half despite a rapidly deteriorating optoelectronics market and continued softness in the wireless market. "The UK facility generated an exceptional performance, partly due to focusing on advanced product areas such as VCSELs where demand has remained relatively robust, and partly as a result of winning significant development work, which has helped to offset declining production volumes." However, Nelson also noted that capacity utilization was weak at the company s US operation (the former QED), which has much larger exposure to the wireless market. Picogiga has yet to announce its full results for the June quarter. However, the company has announced that its sales were Euro 4.57 million during the quarter, which is a sequential increase of 10% over the first quarter of 2001. This welcome news would seem to be yet one more indicator that the worst of the slump in the wireless sector is now over. The company also reported that its backlog at the end of June was valued at Euro 10 million, Euro 2 million higher than it was at the end of last year. Linh Nuyen, chairman and CEO of Picogiga, noted that the company has Euro 18 million in cash available, some of which will be spent on "intensified" R&D programs that will allow the company to diversify out of its traditional reliance on the HEMT market. Modulight, the company s Finnish optoelectronic start-up, will begin operating in September. The company is also branching out into phosphide and nitride compound semiconductors. AXT s roller-coaster ride continues Shares in AXT continue their up-and-down odyssey. Having dipped as low as $13.25 in April, the company rocketed up to $40 in May, close to its 52-week high. Since then, however, there has been a steady trend downward into the $1517 range (see ). As suggested earlier, this author does not have any great insights into investor behavior, but here is one plausible interpretation of what has happened. The share-price rise began in mid-April in anticipation of the company s first-quarter earnings announcement. And indeed, when it came out on 25 April, it showed that AXT had recorded record revenue for the quarter, up 8% sequentially over the previous quarter. Given that most semiconductor companies were reporting declining revenues at that time, investors were thrilled. The company got a further boost by announcing the availability of new telecom laser and 850 nm oxide VCSEL products. And on 1 May TRW announced its plans to spin off its InP operation into a new company (Velocium). This created the prospect of more demand for AXT InP products. (AXT s InP revenues have been growing 25% per quarter and now represent a quarter of AXT s total substrate sales.) On 21 May AXT shares closed at $40.67, but have declined steadily since then. Presumably, investors slowly grasped the possibility that the strength of AXT s first-quarter results may have been skewed somewhat by the company s "supply guarantee program". Last fall, when GaAs substrate prices were rising and supplies were running thin, AXT entered into long-term fixed-price supplier agreements with several of its largest customers. Participating customers were assured that they would receive an agreed supply of wafers at predetermined delivery dates and prices. In return they made partial, non-refundable advance payments to AXT and agreed to give 90 days notice before changing any orders. AXT reported that nine customers agreed to purchase $91.6 million-worth of GaAs and InP substrates under the program in 2001 more than 50% of AXT s production, if current trends hold up. While this program was intended to protect customers from potential shortages, the 90 days notice provision became a liability for them in the first quarter in the face of the unexpected and severe recession in the GaAs industry. Dale Pfau of CIBC World Markets believes that the supply guarantee program "artificially" insulated AXT from the overall weakness in the GaAs semiconductor market during the first half of 2001, as some customers were required to accept shipments of GaAs wafers that they really didn t want or need. Now, he warns, "this situation may accelerate the downturn in revenues for AXT during the second half of 2001, as orders are pushed out and canceled". Despite the fact that AXT s results for the June quarter were in line with estimates (revenue of $41.3 million, income of $5.2 million), the company s share price has failed to rebound. Pfau has warned that AXT may now lag behind the recovery of the GaAs semiconductor market, and indeed AXT executives themselves have also advised of sequentially lower GaAs substrate revenues in the second half of the year partially offset by higher InP revenues in the fourth quarter. This is certainly an "interesting" and unexpected side effect of a sales program that seemed like a good idea at the time.
×
Search the news archive

To close this popup you can press escape or click the close icon.
×
  • 1st January 1970
  • 1st January 1970
  • 1st January 1970
  • 1st January 1970
  • View all news 22645 more articles
Logo
×
Register - Step 1

You may choose to subscribe to the Compound Semiconductor Magazine, the Compound Semiconductor Newsletter, or both. You may also request additional information if required, before submitting your application.


Please subscribe me to:

 

You chose the industry type of "Other"

Please enter the industry that you work in:
Please enter the industry that you work in: