Intel CEO Brian Krzanich sold more than $39 million worth of company stock after Intel learned of a fundamental design flaw in its products, but before the general public was made aware.
Why it matters: The SEC may take a hard look at Krzanich's windfall, particularly the part where he changed the rules governing his stock sale schedule.
June 4, 2015: Krzanich adopts a new Rule 10b5-1 trading plan. Such plans are regularly used by company executives to establish automated stock sale calendars, so as to eliminate accusations of insider trading.
April 22, 2016: Krzanich adopts a new Rule 10b5-1 trading plan.
February 10, 2017: Krzanich adopts a new Rule 10b5-1 trading plan.
June 2017: Google security researchers inform Intel and other large chipmakers of security vulnerabilities related to longstanding processor designs. Per Intel:
October 30, 2017: Krzanich again changes the terms of his Rule 10b5-1 trading plan. The prior two changes had come after 10-month periods. This one came after an 8-month period.
November 29, 2017: Krzanich sells more than $39 million worth of Intel stock, in accordance with the revised trading plan adopted just weeks earlier. It represents the sale of all but 250,000 shares, which is the minimum amount that Intel requires Krzanich to hold.
January 3, 2018: Intel and other chipmakers publicly disclose the security flaw.
An Intel spokeswoman says that Krzanich's October 2017 trading plan change and subsequent stock sale were "unrelated" to the chip design flaw, but declined to provide any alternate explanation.
Intel also says that it does not expect material financial impacts from the design flaw, although it remains too early to know for sure.
While Intel shares took a hit from yesterday's revelation, they continue to trade higher than where Krzanich sold last November. Such fluctuations, however, would be irrelevant to an insider trading investigation.
Bottom line: Executives are typically allowed to sell shares via automated trading programs even if they know of non-public material information, but that's much different from changing the terms of the program once in receipt of such knowledge. For Krzanich, the key could be whether regulators agree with Intel's contention that the design flaw is non-material.