The Dell Debacle: How Serious is the Damage?

Is there an Intel connection?

The admissions Dell CFO Don Carty made on behalf of his company would appear to corroborate claims of a class-action shareholder lawsuit against Dell filed last February. The suit itself offers limited evidence, though it claims that during roughly the same period of time as the now-confessed accounting improprieties, Dell senior executives including former CEO Kevin Rollins and chairman Michael Dell personally conspired to manipulate stock value for their personal holdings by re-engineering company expenses.

That re-engineering also enabled executives, the suit claims, to sell their holdings prior to 1) actually incurring the costs that were allegedly deferred, and 2) breaking bad news to shareholders. Though the numbers have yet to be validated, the suit claims that Rollins sold as much as 98.6% of his own Dell shares prior to precipitous news.

If the lawsuit's allegations are correct, Dell executives conspired to manipulate both good and bad news to their respective personal advantages. This raises the question of just who may have been manipulated besides securities analysts. In other words, who else was Dell trying to fool?

The lawsuit cited revelations made by Japanese investigators that Dell, among others, may have been the recipient of hundreds of millions of dollars in rebates from Intel, presumably for selling Intel-based PCs in high quantity and possibly for selling Intel exclusively. These rebates, the suit claimed, "had secretly been used to boost Dell's operating profits and margins."

As we now know and as Dell has now admitted, its operating profits were indeed improperly boosted. Receiving rebates may not however constitute an illegal boost of profits, if they were recorded using standard accounting principles (GAAP).

But as history has shown, companies can use certainly non-standard accounting principles to adjust receipts so that they qualify for such things as quarterly rebates. A 2004 SEC probe of Lucent Technologies (now part of Alcatel) discovered that in 2000, that company was inflating its accrued income by counting the future replacement prices of equipment it sold to its own customers, as current revenue. For example, it recorded a sale of $135 million of software to Winstar Communications for that year. But Lucent gave Winstar $35 million in credits towards the purchase of the next version of the software, and another $45 million towards a new Lucent network on which that software would run. Otherwise, it would have recorded only $55 million for the original sale.

When a witness is discredited in a courtroom trial, the cross-examining attorney often asks, if what you just said is false, how can we trust anything you've said before? So the following questions arise: Could Dell have been adjusting its sales figures to qualify for rebates from Intel? Knowing that Dell did adjust its expenses to inflate its apparent income, could that inflation have also been to fool Intel? Or rather, what else is Dell hiding, and who else has it been hiding it from?

The reliability of Dell's accounting for its own income between fiscal years 2002 and 2006 could call into question whether its own sales reports are accurate for those years. If a company can shift the time period in which it received revenue (or might receive revenue), it would probably also have to shift the apparent delivery dates for the items it sold to produce that revenue. So it therefore becomes possible that Dell might have booked computers as having been sold when they were not yet sold, borrowing a few pages from Lucent's playbook.

If that's the case, then Hewlett-Packard may have been the global leader in computer sales and shipments for longer than was originally realized.

Not everything offsets itself

Now that the improprieties have been discovered and out in the open, CFO Carty suggested that it may be an academic matter to simply shift back the improperly adjusted expenses to the quarters and accounts to which they originally belonged. Doing so would be like taking bricks from one wall and replacing the holes in the other...just as easy as having taken them from the other wall in the first place.

"At a high level, the restatement is expected to reduce net revenue for each fiscal year within that restatement period, by less than 1%," Carty said yesterday. "Net income and earnings per share are expected to be reduced by 5% or less in three of the four years, and are expected to be increased by about 1% in the other year. Changes for the quarters will vary, with some changes expected to be somewhat greater than those ranges."

Specifically, the quarters for which we should expect the biggest corrections to come will be the company's fiscal Q1 2003 and Q2 2004, with reductions in net income between 10% and 13%. For Q4 2005, the reduction should be about 7%. Then for Q2 2005, Q3 2006 and Q4 2006, net income should be expected to increase by 5% - 7%. For all the other quarters affected, expect variances of 5% or less, Carty said. "As I said earlier, these restatements are not expected to have a material impact on the current balance sheet or cash flows," he added.

The first question Carty received from analysts yesterday asked whether he was willing to admit these improprieties constitute fraud, or whether this is just a case of "profit fraudulently recognized." "Obviously if you recognize revenue or an expense inappropriately, it's inappropriate," he responded, "whether it's going to reverse itself or not. But I think the answer to your question probably lies in the magnitude of the cumulative change here. The fact that some of these things here are reversing themselves doesn't obviate the fact that we had serious control deficiencies. That should not have happened. But I think you can see by the fact that, in the end, we don't have a significant financial impact on our cumulative [profit and loss], and we don't have one on the balance sheet, does suggest that most of this reverses itself."

It was the old "little fib" defense, the kind that a child makes when caught red-handed. So the analyst pressed Carty, was everything that the investigation turned up something that can be chalked up to reversible time shifts?

No. "During the time period of this investigation...there was one event that is impacting these numbers that came out, that really actually wasn't a part of the investigation - it was part of a transaction that occurred overseas where we did find evidence of fraud, and we did find some evidence of revenue that had been booked that had to be completely undone. But that is the only one that didn't reverse itself."

That transaction at least fits the profile of the Intel rebates discovered by Japanese investigators, and cited in the shareholder lawsuit. It has yet to be explicitly identified as such, but Don Carty definitely painted a picture of something that walks like a duck.

It was, as many a father might tell his red-faced child, a nice try.