Description

CCMP is a short because its core product is becoming a commodity and recent entry by Dupont and Air Products is speeding up the process. These factors - along with the need for high R&D/capital spending in order to keep up with rapid changes in chip making technology - are steadily driving down CCMP’s returns. Yet Mr. Market treats the company with “darling” status (24x EV/EBIT 2003e) in expectation of a high peak in the semiconductor cycle. There are several reasons, however, to question whether CCMP will benefit the way the market expects.

Negative Catalysts:

· Deteriorating business fundamentals, commoditization of core
product
· Market entry by Dupont, Praxair, Air Products and others
· Recent Level-4 CFRA warning on q303 earnings release
· Potential earnings misses and disappointments (q3 spike in DSO,
DOI)
· Eventual reversion of ROE to 10-12% commodity-type level
· Lower FCF due to resumption of normal capex levels
· Potential product substitution (some day)

Background:

Chemical mechanical planarization (CMP) is a process in integrated circuit (IC) manufacturing. CMP allows IC makers to get more layers of material on a silicon wafer by using chemicals to polish and flatten the layers. The Company makes the chemicals used in this process (CMP Slurries) and sells them to IC manufacturers including Intel, UMC and TSMC. CCMP was first in its space and established an 80% market share in slurries used to polish tungsten (80% of sales). In the last four years this share has declined to 65%.

Industry Characteristics:

CCMP’s only source of competitive advantage is a kind of limited customer captivity arising from a) the fact that slurries must be qualified by customers (an expensive, 6 month process) before they can be used in production and b) once a slurry has been selected and qualified for a product line the risks and costs of switching outweigh the incremental cost savings in most circumstances. The problems are that a) chip lifecycles are short b) manufacturing processes are also subject to rapid change and substitution and c) Intel demands constant product upgrades or price reductions even in qualified slurries.

And the industry is in constant flux. For example, IC makers are increasingly using copper instead of tungsten, because it has better conductivity. CCMP’s “limited franchise” is in tungsten slurries. The copper slurry market is much more competitive. A slew of new entrants - including Dupont, Air Products, Praxair and Fujimi – are making forays into the copper slurry market, which may cause shifts in market share and will in any case expand supply and keep negative pressure on pricing, even in a cyclical peak for semis.

While tungsten will probably continue to be used for many years in the making of lower grade, commodity-type chips, its growth potential is limited. Not surprisingly, prices for tungsten slurries have fallen such that CCMP’s revenue growth has slowed from annual highs of 84% to –6% yoy in q303.

This collapse is not solely due to the downturn in the semicon cycle, but also to increased sources of supply and strong competition. In short what was once a “first-mover advantage in a growing market” story, is now characterized by fierce competitive rivalry, high buyer power, high threat of entry and high threat of substitution.

In addition, makers of manufacturing technology (e.g. Applied Materials) are focusing on ways to use less slurry in the production process. In fact, Applied Materials has published papers about a new CMP process that does not use slurries at all. I have no idea about the potential adoption and use of this process, but it highlights the threat of substitution

CCMP has high customer concentration: 63% of 2002 sales came from five customers. CCMP also relies on a take or pay contract, with annual 2% fixed price increases, with one supplier (parent company Cabot) for fumed oxides – its key feedstock.

CCMP is trying to keep up with change through high R&D spending (15% of sales for CCMP) and capital expenditures ($35M vs. Depreciation of $12M in 2002) such that ((EBIT + D) – maintenance capital expenditures) is a better measure of economic earnings. On this basis, the company trades at EV/Economic Earnings 2002 of 37x.

In short, my thesis is that CCMP’s rich valuation grossly overstates its prospects and fierce competitive rivalry should create negative catalysts for the stock.


Financial Analysis

1999 2000 2001 2002 q103 q203 q303
Net Revenue 98,690 181,156 227,192 235,165 57,273 62,201 64,288
% gmgn 51.5% 52.4% 52.3% 51.9% 51.7% 48.9% 51.2%
EBIT 19,076 46,818 62,439 59,960 13,942 13,572 14,634
% of rev 19.3% 25.8% 27.5% 25.5% 24.3% 21.8% 22.8%
DA 2,777 4,891 7,787 12,009 3,745 3,767 3,976
Capex (9,313)(38,923)(35,328)(35,259)(2,286)(2,428)(3,347)
EBITDA - Capex 12,540 12,786 34,898 36,710 15,401 14,911 15,263
Sales/Assets 3.78x 3.07x 2.36x 1.91x 1.69x 1.71x 1.61x
Fin Lvg 0.42x 0.55x 0.58x 0.58x 0.60x 0.61x 0.64x NI/Sales 12.4% 16.8% 18.4% 17.3% 16.2% 14.6% 11.7%
ROE 19.8% 28.4% 25.2% 19.1% 16.3% 15.3% 12.1%

The financials confirm the story. Revenue growth has decelerated dramatically. After growing 68% and 84% in 1999 and 2000, CCMP’s topline grew 25% in 2001 and 4% in 2002. In CCMP’s shady q303 (more details below), revenues fell yoy by –6%.

Net margins have fallen too, 570 bps from 18.4% in 2001 to 11.7% in q303. As prices for slurry have fallen, so have gross margins and asset turnover. Asset turns have fallen in half since the heady days of 1999.

All in all, CCMP is a deteriorating business with shrinking margins, lower ROE, lower asset turnover and higher financial leverage.

Accounting & Further Financial Analysis

A look at q303 highlights this deterioration:

1. Capex had been fairly stable at $35 - $39m per year from 2000 – 2002, but through the first three quarters of 2003 the Company spent only $8m, despite guiding $22m for the year on the q203 call. This does not at all appear sustainable and perhaps was done to cover up cash flow deterioration.
2. The company puffed earnings in q303 by manipulating its inventory and warranty provisions. CFRA issued a level 4 warning on 9/25/2003 highlighting this issue, adjusting net income down by $0.04 and gross margins down by 240bps.
3. DOI are up 23% yoy in q303, finished goods as a % of total inventory is up 1%
4. DSO’s are up 15% yoy in q303

Bull Case

The bull case for CCMP is that a) the slurry market will explode and b) CCMP can sustain or increase its market share. It’s non-scientific, but among the studies I’ve looked at, a bullish estimate for slurry market growth is 15% per year through 2006, implying a $579m market in 2006.

$ ‘000 2003 2004 2005 2006
Market 381,000 438,150 503,873 579,453
Share 65.00% 65.00% 65.00% 65.00%
Revs 247,650 284,798 327,517 376,645
Op Mgn 22.90% 27.50% 27.50% 27.50%
Net Mgn 14.40% 18.40% 18.40% 18.40%
NI 35,662 52,403 60,263 69,303
EPS $1.45 $2.13 $2.45 $2.81
Price $58.07
Mkt Cap $1,427,941
EV $1,357,830
P/E 40.0x 27.3x 23.7x 20.7x
EBIT 56,712 78,319 90,067 103,577
EV/EBIT 23.9x 17.3x 15.1x 13.1x

We’re already three quarters into FY 2003, so I left margins for this year as they are. In 2004-2006 I used the best NI and EBIT margins CCMP has ever had. Even in this scenario, CCMP seems over-valued.

Bear Case

Given the rapid change and increased competition mentioned above, I think it is very unlikely that CCMP will maintain a 65% market share or return to 18.4% net margins. Though I think margins can decrease from here, I assumed that they stabilize at current levels and that CCMP loses market share in a market that grows 12% per year. 22x EV/EBIT 2006 for a company with flat earnings and diminishing ROIC offers a lot of room to make money on the short side. A guess at fair value is 20x P/E 2004 for a price target of $30 - $35/share.

$ ‘000 2003 2004 2005 2006
Market 381,000 426,720 477,926 535,278
Share 65.0% 60.0% 55.0% 50.0%
Revs 247,650 256,032 262,860 267,639
Op Mgn 22.9% 22.9% 22.9% 22.9%
Net Mgn 14.4% 14.4% 14.4% 14.4%
NI 35,662 36,869 37,852 38,540
EPS $1.45 $1.50 $1.54 $1.56
P/E 40.0x 38.7x 37.7x 37.1x
EBIT 56,712 58,631 60,195 61,289
EV/EBIT 23.7x 22.9x 22.3x 21.9x


Conclusion

Full disclosure: I am short CCMP. The primary risk to this story is that a new peak in the IC cycle is greater than expected and the negative impact of competition manifests slowly. I am well aware of the conventional wisdom not to short a growing market and really have no answer other than to point out the risk. Perhaps this is better as a hedge in a portfolio that is long IC, or a better bet once the bad news is more evident.

On the other hand I think time is not on CCMP’s side and the scuttlebutt work I’ve done in speaking to competitors and industry sources leads me to believe they are in a very tough business. They purchase feedstock from one supplier, sell to Intel, compete against Dupont and face rapid changes in chip making technology. To me this doesn’t add up to sustaining a 20% + ROE.

Catalyst

· Deteriorating business fundamentals, commoditization of core
product
· Market entry by Dupont, Praxair, Air Products and others
· Recent Level-4 CFRA warning on q303 earnings release
· Potential earnings misses and disappointments (q3 spike in DSO,
DOI)
· Eventual reversion of ROE to 10-12% commodity-type level
· Lower FCF due to resumption of normal capex levels
· Potential product substitution (some day)