To: Cary C who wrote (13436 ) 1/31/1999 8:21:00 PM From: Sergio H Read Replies (1) | Respond to of 29382
Thanks Cary. Nice work. Here's an interesting article in today's NY Times. Gives credit to Michel Burke's thread here on SI for some excellent investigative work. nytimes.com January 31, 1999 MARKET WATCH Oh, Those Pesky Little Financial Details -------------------------------------------------------------------------------- Related Articles The New York Times: Your Money Forum Join a Discussion on Career and Workplace Issues -------------------------------------------------------------------------------- By GRETCHEN MORGENSON EW YORK -- This is a tale of two earnings reports from Micron Technology Inc., one of the world's largest makers of semiconductor chips. Micron also happens to be a very hot stock. The reports detail the results for the company's first quarter, which ended Dec. 3. One is a news release issued Dec. 23; the other is the quarterly report, or 10-Q, filed with the Securities and Exchange Commission on Jan. 5. The two reports differ in intriguing ways. The most glaring discrepancy is in Micron's sales of semiconductor memory products. The news release says these totaled $428.1 million in the quarter. The 10-Q says they totaled $409.5 million. Which number is correct? A Micron spokeswoman, Julie Nash, said: "There was a $19 million reclassification of numbers that went from semiconductor memory products sales into 'other.' These were sales that were relative to providing semiconductor revenues, but they were actually service fees." She said Micron, of Boise, Idaho, found the error and fixed it in the 10-Q. But it did not publish a corrected news release. "It was not material," she said. So it's immaterial that 4.3 percent of the company's semiconductor sales weren't semiconductor sales at all. None of this would matter if investors relied on 10-Qs rather than news releases for investment information. But that is not the case. Dan Niles, an analyst at BancBoston Robertson Stephens, has a strong "buy" rating on the stock. The revenue number he has used in recommending the stock is the $428.1 million from the news release. Niles did not know the number was incorrect until a reporter told him on Friday. He said the new, lower number would not change his positive view of the company. There is another interesting difference in Micron's two reports. The 10-Q states that net sales of semiconductor memory products increased 10 percent over the previous quarter, a result of an 18 percent increase in the chips' average selling price being partly offset by a 10 percent decline in megabits shipped. The news release, however, is mum about the decline in shipments. It is a big deal to show a decline in megabits shipped. With chip makers constantly improving the memory capacity of their products -- also known as the yield -- a drop in megabits shipped is not good news. Why did the news release not mention this? Ms. Nash again: "The press release is a summary document used to discuss what we believe is of interest to shareholders and reflect company trends." Whether this passes the smell test is something Arthur Levitt, chairman of the SEC, may want to ponder. Levitt is justifiably concerned that companies make accurate disclosures to investors. Who uncovered the Micron discrepancies? Sleuths on the Internet -- specifically, three people who post comments on the "Ask Michael Burke" and "Micron Only" threads on Silicon Investor; and Larry Woods, editor of the Tech Review newsletter in Stoney Creek, Ontario. Two footnotes: Since Micron's news release came out, its shares have risen 49 percent. Not bad for a company that lost $46.2 million in the quarter. And on Wednesday, the company filed a registration of 2.75 million shares to be sold to the public.