> Porcupine, Anyone else notice that these "extraordinary" charges
> seem to happen on a regular basis.
Nowadays a lot of companies, particularly large caps, are
restructuring as part of their ongoing business plan. On
balance, this is a good thing. Back in the days when a job with a
large cap company was impliedly "employment-for-life", companies
waited until there was a crisis before restructuring -- not the most
efficient way to go about it.
But, this underscores your point, which, as I understand it, is:
Just how "extraordinary" are charges that are part of a permanently
ongoing strategy?
> Further, it has been discussed
> that companies "bloat" the charges to help the future look better.
Well, yes, to the degree that they can get away with it. Two
countervailing forces are the IRS, and the credit rating agencies.
The former will challenge "losses" they don't really think got "lost"
(in terms of any cash expense). The latter will downgrade a
company's credit rating if too much gets "disappeared" from the
balance sheet.
> I don't buy that we should ignore these charges. Will we remember
> 3 years from now that the 1997 EPS had some "extraordinary"
> charges? I think not.
Well, once thing is for sure: Dark Side will remember!
But, kidding aside, I feel it is hard to generalize about whether it
is the future that will be artificially "brightened", or whether it
is the past that is having its artificial brightener removed.
For example, let's take XYZ Corp., again. Suppose they purchased a
piece of equipment at the beginning of Year 1 for 120mil. As you
know, under GAAP and IRS rules, XYZ will likely be taking this
expense off the income statement and balance sheets in 6 annual
increments, rather than all at once.
Say XYZ uses straight-line depreciation over a 6 year period.
Thus, XYZ plans on "depreciating" 20mil annually off the assets
column of their balance sheet, as well as from the revenues on their
income statement in the calculation of earnings (i.e., revenues minus
costs). By the end of Year 6, the entire cost of the equipment will
have been deducted in equal annual increments from each year's
revenues to arrive at a "true" picture of the profit earned by this
piece of equipment.
But, suppose at the end of Year 2, a new technology comes along that
is much more efficient than the equipment XYZ bought in Year 1.
XYZ decides to sell the 2 year old machine and get the new
model. But, although XYZ's machine is carried on its books at 80mil
(120mil - [2yr x 20mil/yr] = 80mil), now that this machine is
relatively obsolete, the best XYZ can realize for it in the used
equipment market is 20mil. Since 20mil is 60mil less than 80mil, the
company reports this as an "extraordinary" loss of 60mil.
My question is: Does writing off the 60mil in a single quarter make
future profits greater than they would otherwise appear? Or is it
instead a belated recognition that past profits are not as
great as first reported? In other words, rather than seeing this as
a future cost of 60mil, perhaps the profits in Years 1 and 2 should
be restated to deduct an additional 30mil from each of those years
(admittedly, a practice rarely followed, though I believe AOL is an
exception).
Graham might say that the reason for averaging profits (making
deductions for these "extraordinary" charges) over a number of years
is that this timing issue becomes irrelevant, since the cash got paid
out somewhere along the line.
But, it is future earnings power that is most important. One could
say that future earnings power estimates must take account of the
possibility that technological innovations will result in replacing
equipment ahead of schedule, i.e., that these "nonrecurring"
expenses are, in fact, predictably recurrent.
But, for example, I doubt that the "big bath" write-offs that
Chainsaw Al Dunlap took at Sunbeam reflect soon-to-be-forgotten
charges against future profitability, but rather a recognition that
the past was worse than the old management was admitting.
Furthermore, I think a number of these "charges" will never get paid
out in cash. For example, XYZ lays off a number of workers, and
takes a "non-recurring" charge against one quarter's earnings for 2
years of anticipated retraining and job placement services, as
stipulated in the workers' contract.
But, because of a good job market, most of these workers get jobs
elsewhere long before the 2 years is up. That portion of the
"nonrecurring" charge gets "undone" in a future item for a
"nonrecurring" gain. So, the cash doesn't actually get spent,
but this won't be known until subsequent quarters. Meanwhile, XYZ
will be taking every available tax advantage of these supposed
"losses". So, in calculating future earnings power, it is again
difficult to generalize about the actual cash consequences of these
"special" items.
My question for you, Berney, is whether the IRS charges interest on
the taxes that must be paid when prior "losses" are subsequently
"undone"?
porc --''''> |
