Hi - It has been awhile since I posted on SI, due to other commitments. However, I did write a piece for my newsletter that might be of interest here. I am providing it below for your review and comments.
Repost of Message from Yahoo Message Board (4/05/02) - WFR Valuation Now Crazy
Long Term Sentiment: Strong Sell - For the record, I just shorted this stock today at $7.15.
The stock seems to have been going up on short covering and a very small float. Did anybody read the 10-K? TPG now owns 185 million shares (fully diluted) of WFR, about 90% of the company. WFR has over 200 million shares outstanding. That's a $1.4 billion equity valuation, plus $220 million of debt = $1.6 billion enterprise value. For $1.6 billion what do you get? You get a company with about $500 million in sales that is losing money in an extremely competitive, cyclical business.
Even in the go-go days of '99 and '00, WFR generated operating losses, BEFORE interest expense. They just announced a further decline in sales in the fourth quarter and projected operating profit (before interest) "could approach break-even." There is good reason to believe that this business will never be as good as it was in '99 and '00. And they couldn't make money then. Even with the restructuring, WFR still needs to make a 6% operating margin just to cover interest expense. They need about a 10% margin to cover interest and capex. They've never even come close to that kind of operating profit. Even if they achieve a never-before-seen 10% operating profit, they'd be breaking even, generating zero cash flow. So, $1.6 billion for losses and, if things go really well, maybe WFR breaks even? Not a good deal to me.
Great deal for TPG. TPG paid nothing for WFR. They did have to loan WFR money, on a senior basis. They only have to pay if WFR's EBITDA for '02 is greater than $100M. They have a scale up depending on WFR's EBITDA. The maximum they would have to pay is $150 million if EBITDA exceeds $300 million. That's the greatest deal in the world. Their 90% stake in WFR is now worth $1.3 BILLION. AND their stake is in convertible preferred, so it is senior to the common, meaning $1.3 billion for nothing.
What if WFR somehow achieves a record shattering $300M in EBITDA? How could they do that? WFR would have to grow sales from the current run-rate of $500M to over $1.5 billion. And then they would have to achieve a 20% EBITDA margin. Neither of these things is remotely possible. WFR's highest sales level was $871M in '00. Their EBITDA margin then was 18%. I don't believe WFR even has capacity for over $1 billion in sales.
So, TPG and E.ON thought WFR was worth nothing at the current sales and profit levels. And was at most worth $150mm, if WFR achieved mind-blowing results. Yet, the market is valuing WFR at $1.4 billion. Something is wrong.
Radarview Response to the Message - First of all, it is difficult, if not close to impossible to short WFR stock in any great quantities. Most of the shares are held by TPG, which probably is not allowing the shares to be shorted. Second, what this person has to say makes a good deal of sense, but WFR is the premier provider of 12" (300mm) wafers, which is ramping up. WFR should be benefiting from the new round of contract prices for wafers that started in January of this year. Remember, its contracts run from January to January for the most part.
Our average cost basis in WFR is ridiculously low and I would hazard a guess that I might know where another 1.5% of the shares reside, with these shares unavailable for shorting. Even though TPG got WFR for practically nothing, we know these are sharp people. We expect these guys to eventually do a secondary offering to recoup some of their investment. At that time we will be looking carefully at whether we want to sell or not. We currently are considering an LLT at $6-$6.50, but this might be the wrong time to bail out. With the recovery underway, higher margins for 12' (300mm) wafers, more 300mm fabs ramping up, writing off all of the bad past businesses, shuttering some of the older facilities, etc., WFR might very well be in the midst of a turnaround by a group of investors who are letting management do what is necessary.
WFR, in our opinion, fell from glory for a number of reasons, one of which was the VEBA (E.ON) takeover where the mother company did not know how to run the business. WFR went down the crapper when the IC fabs reduced capacity and when fabs shut down. Now that capacity is ramping and new fabs are coming on line, our guess is that we should see some improvements.
What goes through our mind is that a 40,000 wafer per month fab uses more than 55,000 wafers per month (yield and test wafer usage). A fab running at 50% utilization runs 27,000 wafers at roughly $80 per wafer (8" or 200mm). This translates to $2.16M revenue per fab that WFR sells to. When ramped back to 90% (100% for this example), it is an additional $2.15M per customer. All it takes is 100 customers (an there are many more than that) to generate $215M. Of course my numbers are very conservative since 10,000 wafers per week is a drop in the bucket to some IC fabs. Also, raising the current run rate of $500M to $1.5B does seem very difficult, but when you consider the current fab capacity utilization, a jump from $500M to $1.0B seems like a walk in the park. Add to this the ramping and new $300mm fabs, and it is not totally outrageous to see a figure like $1.5B.
One thing we have been remiss in doing, is finding out the revenue per 300mm wafer. Based on prior costs and revenues, it appears that revenues and costs are closely aligned to square centimeters (area of wafers). A 6" wafer generated ~$30 in revenue. With an 8" wafer being 1.8 times larger area, one would assume the revenue from an 8" wafer would be close to $55, but it is closer to the $80-$90 range. If we do a direct proportion, we could raise this factor to 2.7, due to the increased complexity of creating larger diameter wafers as well as the number of wafers that can be made at 8" (remember that as wafers get larger, so does the thickness). Therefore, given the number of wafers that yield and the thickness, a 2.7x multiplier seems reasonable. If we use the 2.7 on the upcoming ramp of 12" (300mm) wafers, we can expect a minimum of $225 per wafer in revenue. And if we stretch it one step further and make the not so absurd assumption that each reduction in the number of 6" wafers sold (obsolescing technology) is directly tied to the number of 12" (300mm) wafers sold, we come up with the following observation.
There is a 4X area difference (ratio of the squares of the radius), so we assume there is a 4 to 1 replacement ratio of 6" to 12" and there is no incremental increase in usage (that is more conservative than reality). Four 6" wafers generates $120 in revenue, while one 12" wafers generates $225, for a $105 delta or a factor of 1.88. Let's assume that 6" wafers make up 30% of the revenues (30% of $500M is $150M). This would mean that this revenue would be replaced with $280M in 12" revenues. Taking this one step further:
$150M becomes $280M, which at 100% capacity utilization becomes $560M.
$350M (rest of business), at 100% capacity utilization becomes $700M.
This conservatively totals $1.25B and does not factor in any NEW fabs being built for expansion and not for replacing obsolete fabs or retrofitting existing fabs. This does not include any business WFR might get in China either.
While it might be a stretch to get to $1.5B, it might not be impossible. And of course, the final nail in the coffin of the Yahoo author is not recognizing the economies of scale. WFR is being crippled by depreciation of 12" equipment that was never run at full capacity.
Running the operation at severely reduced capacity utilization has eviscerated WFR. The business has a high level of fixed costs, which do not go away when you are not running full. The variable costs are not as dramatic. Therefore, as they ramp up their factory, the cost of goods sold will drop dramatically. This is where the author might be in error. We do not believe the cost efficiencies of running full have been factored into his numbers. NOTE: This is also why I really like the IC manufacturers on our list of companies. As they ramp up fab utilization and capacity, wafer costs drop significantly, and they will be making more wafers at higher gross margins. |
