What's Behind the Texas Instruments Disappointment
Bob Faulkner Dec 09, 2009 8:45 am
The problem wasn't demand, but a miss-match between demand and inventory.
Few things are more disappointing then when something doesn't live up to
expectations. If there was a lot of anything going into last night's mid-quarter
update from Texas Instruments (TXN), it was expectations. And why not? Over the
last few weeks we've had a fairly steady stream of semiconductor companies raise
guidance for their December quarters as demand proves to be better than
anticipated.
To be clear, Texas Instruments raised the midpoint of its prior guidance from
$2.90 billion to $2.96 billion. Nothing wrong with that, but it wasn't enough.
The Street was clearly looking to see the top end of the revenue range ($3.02
billion) bumped up as a signal that demand was stronger than anticipated. Alas,
the Street was disappointed and the stock fell in after-hours trading.
But there's more here than meets the eye and, as an investor, you should be
aware of what's going on beneath the surface. The demand is there but the
company wasn't able to deliver.
If I'm a customer and placed an order with Texas Instruments today, if it
weren't in inventory, I'd have to wait until early March to get it. That's about
how long it takes to turn a bare silicon wafer into a finished part that's been
packaged and tested. That also shows why inventory is so important at
semiconductor companies -- particularly at those who specialize in analog
semiconductors.
The number of different analog parts is mind-boggling and most of those come
in an array of different package types. Consequently, you'll see that analog
semiconductor companies tend to run with inventories at levels significantly
higher than their pure digital cousins. More often than not, much of that
inventory will be retained as die banks, meaning the wafer has been processed,
cut into individual die, but not packaged and tested. This allows companies like
Texas Instruments a degree of flexibility when orders come in, particularly the
unanticipated upside that frequently happens at this point in the economic
cycle.
Texas Instruments' problem wasn't demand per se, but a miss-match between that
demand, its inventory, and its current configuration of assembly and test
equipment. Think of it like the mix-and-match game we all have to play with
batteries and toys on Christmas morning. But the difference here is that I can't
run out to 7-11 for the quick fix.
Let's say my customer needs more of part "X" in package "DD" than originally
expected. If I have it in finished-goods inventory, no problem. If I don't and
the two testers or assemblers that can handle package "DD" are busy, my customer
has to wait. I can order new equipment from Teradyne (TER) or Kulicke & Soffa
(KLIC) but the chance of them having what I need in "inventory" isn't very high.
Even if it was, I'll have to run the equipment through an acceptance process.
Texas Instruments brought in additional assembly and test equipment this
quarter and will again over the next two quarters. Obviously, it wasn't soon
enough.
The bottom line here is that this is an execution issue, not one of demand.
Texas Instruments perceived "disappointment" isn't a signal that the semi-cycle
is over before it's started. It's just a bump in the road and there will be many
more along the way. The devil is always in the details so study them to find out
what's really going on before you decide to jump either way.
AI
