Harry: three comments
>>Figuring cap spending/income does not take into consideration the
>>positive cash flow from depreciation.
When I was discussing capital spending/income, I was indeed
talking about spending net depreciation.
>> The 1996 estimates may be low since they do not include the latest earnings
>> report, but the trend is clear.
As a percentage of earnings, cap spending net depreciation is
0.6620 0.5332 0.5531 0.6233 0.4042
It is hard for me to see any strong trend in these figures, unless we
believe that 1996 ests. are representative of the future, a dubious
assumption at this point IMO. Even so, 40% of earnings into capital
spending is a lot. Compare CSCO.
>>As income and depreciation increase faster than
>>growth in cap. spending, earnings available to the owners increases
>>dramatically.
It is true that at some point depreciation will catch up to capital
spending. This will happen either from a fundamental change in
capital spending requirements, or from maturing markets.
As Intel matures, represented by slowing eps and capital-spending growth,
it is true that free cash flow as a percentage of net earnings will
increase. But for a while, before depreciation slows down to match capital
spending, eps could take a big hit, and I bet that will not do wonders for
Intel's stock price, especially if it has aquired a bloated PE in the
meantime.
As far as capital spending requirements, I don't see any reason to
believe that developing and equipping, say, 0.18um processes will be
less expensive than 0.35um.
Maybe you can offer some insight here. [Cheerleading does not count!]
Anyway depreciation has to track capital spending over the long run. Over
the short run, it has to lag, and when large capital spending is required
to achieve large growth rates, this has the effect of overstating net
earnings.
--joel |
