Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : DELL Bear Thread -- Ignore unavailable to you. Want to Upgrade?

To: Geoff Nunn who wrote (1873)	9/7/1998 11:27:00 AM
From: The Phoenix	Read Replies (2) \| Respond to of 2578

Geoff, If I'm taking rudedogs position it's pure coincidence. I'm arguing from a common sense position. 1. Whether Dell is in the picture or not, Intel must carry a finished-goods inventory to meet the needs of other boxmakers. Now, what does Dell's presence do? The question is whether Intel would want to carry less inventory if Dell moved away from JIT. Alternatively, would Intel want to carry additional inventory if CPQ somehow converted to JIT? I submit the answer to both of these questions is no. Now you're smokin somthin. If Intel has 5 customers for which it inventories and each customer typically take 100,000 Pentiums a month then they will inventory for that expectation. If however Intel has 6 customers each of which take 100,000 a month then they will inventory for that expecation.. You're argument is that they will not change their inventory or build practices depending on the inventory, purchasing habits, and volumes of their customers is seriously flawed. 2. Occasionally you will hear someone claim that the main idea behind JIT is to force the upstream suppliers of a firm to bear its inventory burden - and absorb the costs. I think this is totally wrong. It looks at inventory control as a zero sum game. It assumes that if firm A carries less inventory, then firm B - a supplier - must carry more. The truth may in fact be just the opposite: if A moves closer to JIT, it may also allow B to move closer too. Is that what Intel does? I don't think the fab practice allows for that as well as the assembly practice. Do you have information that indicates that Intel is running JIT? If it is and DELL needs goods that aren't produced (because Intel can not predict DELL's run rate) does DELL delay shipping to customers? 3. What would JIT mean for Intel? JIT in finished goods would mean that a chip coming off the production line would arrive in inventory precisely at the moment an order for it is received - not one moment sooner! THanks for the definition. I think we all understand the premise behind JIT. 4. Dell has devoted a great deal of attention to increasing its component-inventory flow or "velocity." In reality it is not a JIT system, which implies infinite velociy, but a 7-8 days system. This means that components in inventory on a value weighted basis need to be replaced on average every 7-8 days. Dell's stated goal is to accelerate the flow to a 5-day system. If Dell accelerates how would this affect Intel? Let's perform a conceptual experiment. Suppose Dell were somehow able to raise velocity to a one day system. Let's say Dell on the morning before each daily production run acquires only enough chips for that day. How would this affect Intel? The Rudedog- Gary hypothesis is that Intel would need to carry more inventory, but how can this possibly be? Dell's inventory needs on one day system are more predictable, not less. Intel's task of smoothing its own inventory would be made easier, not more difficult, because fluctuations generated by Dell's order flow would be less disruptive. This suggests to me that Intel would, if anything, carry less inventory not more. First, they are not on a one day system. Second you're assumption that each day is more predicatable than each week is flawed. If each day is so predictable then each week should be equally as predicatable. Third, the fabrication process is too costly to reduce flows for daily runs. Fourth, shipment charges would become a greater influence on the total cost of business. And fifth, management and support of a daily process such as this will also add costs. I think these last two points are fundamental to keeping JIT to 5-8 days turns. Geoff, You have also ignored the fact that daily turns give Intel no window onto their future business. If DELL is purchasing 2000 chips a day today what will they be purchasing 6 months from now? 3000? 1500? How does Intel plan for that given that they can not fab chips in a day? They must purchase inventory with an expecation of what DELL will be purchasing...so although they may not be holding finished goods they will be holding inventory simply to meet DELL's unknown delivery cycle. If they don't they risk losing a customer. OG

To: Geoff Nunn who wrote (1873)	9/7/1998 5:18:00 PM
From: rudedog	Read Replies (2) \| Respond to of 2578

Geoff - Intel's task of smoothing its own inventory would be made easier, not more difficult The problem with this is the huge difference in manufacturing cycles. Intel's has a very long time constant, so the ability to know a day or a week in advance what demand will be, is of absolutely no value. They need to know months in advance. I agree that Dell is only a portion of their manufacturing load. CPQ is a much larger fraction. At one point CPQ took nearly 30% of Intel's output. Although that is no longer the case they are still Intel's largest customer. Intel is willing to provide aggressive pricing to keep at least that section of their output stable. This is a complex issue, and I think your point about the linkage between supplier and customer in a JIT world is a very interesting one. I devoted several years of my life to this issue in the late 70's when General Motors was shifting to JIT. At the end of the day the suppliers took it in the shorts. The issues were similar - GM had a short time constant, the suppliers in most cases (the steel makers in particular) a much longer one. Many of the supliers passed their costs through, and it ended up costing GM for a while until the other big players also went to JIT and leveled the playing field. I don't see that happening here, since almost by definition Dell has a unique requirement because of its manufacturing model. That's the very reason their model is so competitive on the other side (their customers). I believe not only that this is one of the few areas where Dell is at a potential disadvantage, but that Dell senior management is well aware of this. There's just no easy answer since Intel (in particular) is so hard to push around.