Biz Dev in NYC by Rob Tsai

I read Mish’s posting on Minyanville yesterday, which paints a pretty tough picture heading into 2008:

• I have received many emails recently in regards to yield and inflation adjusted returns. The most common complaint goes something like this: “Prices are rising at x%, the CPI is a crock, so if I get less return than x% then I am losing money.”
• Yes, that is likely true. It has also fueled bubbles in many equity markets (and perhaps even commodity markets) by those looking to do something about it.
• However, no one is entitled to positive returns. It is entirely possible that for a period of time the best thing one can do is minimize “real” (CPI adjusted) losses.

Mish goes on to press his case for a continued “assymetric unwind of the credit bubble”, based on the following assumptions:

• Housing is going to continue to be weak
• Commercial real estate is going to be weak
• Capital impairments at banks will be an issue
• Unemployment is going to rise
• Consumer spending will be weak
• Credit card defaults will rise
• Foreclosures will rise
• Corporate earnings will be weak
• The Yen carry trade will unwind

The question arises whether it’s too late to make money being short the homebuilders, commercial REITs, financials, consumer discretionary (restaurants, retail), or long the Japanese Yen (FXY).   I personally think there’s still room in these trades, and I have a few positions that reflect these themes.

Mish concludes:

• The credit conditions that fostered those so called “profits” are not returning no matter what the Fed does. S&P earnings are now estimated at 23 not 16. A PE of 23 is not cheap. In fact it is more closely associated with market peaks. If more writedowns than expected come, even 23 will be too low an estimate.
• Please remember that for the last few years nearly every investment class rose in union. It is entirely possible if indeed not likely, that the reverse happens for a few years. What was correlated on the way up can certainly remain correlated on the way down, even if there is an asymmetrical skew to the unwind.
• In a muddle through economy with weak corporate earnings, mistakes will be punished quickly and severely. Thus safety should be the primary concern. Furthermore, risk in equities and risk in the economy are both heavily skewed to the downside.
• Long-short strategies that can play relative strength within sectors and or stronger sectors vs. weaker sectors may be the best equity strategy going forward. Otherwise there is simply nothing wrong with building a CD or treasury ladder while one waits for better opportunities down the road.

Last night on Fast Money, Guy Adami was recommending a long position in Intel, after the stock got crushed on a Bank Of America analyst’s downgrade.  SeekingAlpha has a good posting supporting this pullback as a buying opportunity.

Jon Najarian also made the case of shorting KLA-Tencor, the semiconductor equipment supplier.

Thematically this makes sense to me.  Intel has a dominant position in the microprocessor space, powering everything from laptops, desktops, servers and even Apple computers.   This article pegs Intel’s market share at 78.7% compared to AMD’s 13.9% in Q3 2007, according to estimates from iSuppli.   So certainly, a global slowdown in tech spending would hurt Intel, but you have to love their market dominance, and it’s not clear to me that people won’t keep spending money on computers.  I’d look to Intel to continue innovating, pushing the limits of Moore’s law to improve pricing and performance, and there’s no doubt of Intel’s ability to maintain good gross margins compared to commoditized chips like memory (DRAM, flash).  

• For example, Intel’s gross margins in Q3 2007 was 51.2%, compared to 46.9% in Q2 2007, and 49.1% in Q3 2006.   So margins improved both sequentially and year over year.

I thought I’d use Doug Lehrman’s tutorial on pairs trading at Traders Log to analyze this pairs trade - long INTC/short KLAC.

• Creating a pairs trade involves the following four steps:

1. Perform correlation analysis to match stocks.

2. Apply technical and statistical indicators.

3. Perform fundamental and technical tests to confirm relationship.

4. Manage and exit the trade.

I downloaded the daily closing prices for the latest year (251 data points) from Google finance, and ran the correlation analysis, and found that there was virtually no correlation.   The correlation coefficient is 0.059.

If you look at the 1 year price comparison charts, you see that Intel returned close to 25% compared to KLA-Tencor’s -5% return.  Intel underperformed KLA largely through June.  From June to September, it looks like they traded in step with one another, before diverging substantially in September.   What was explains KLA’s downward move?

• October 4 - CIBC analyst Gary Hsuehdowngraded the stock citing market share loss at a large DRAM customer, Rexchip
• October 26 - KLA reports Q1 profits down 35 percent, and Citi analyst Timothy Arcuri downgrades the stock

So the question becomes whether you think KLA’s downward trend persists or not, whether Intel will recover from this recent downgrade, and whether the spread between the two stocks will widen or narrow. 

So the trade at the end is not really a statistical mean-reversion trade by any means, but rather a bet on whether a market dominant chipmaker like Intel can perform simultaneously when semiconductor equipment capital expenditures are reduced.  Probably the next level analysis would go into utilization of fabs. 

One final interesting data set to look at is the semiconductor industry’s book to bill ratio.  The analysis uses trailing three months of bookings data compared to billings to smooth out volatility.  

I downloaded the historical book to bill ratio’s from SEMI, an industry organization, and then charted the ratio from Jan 91 to Nov 07 (p) below:

You’ll notice first that the book to bill ratio, and the semiconductor industry, is highly cyclical - reflecting (over)investment in capital expenditures followed by slowdowns.   The latest book-to-bill ratio is at 0.82, signifying that chip makers received $82 in new orders for every $100 in billings.  The last peak was at 1.14 in June 2006, so this downtrend in book-to-bill has been about a year and a half.   Interestingly enough, KLAC stock was in an uptrend from June 2006 through August 2007 - even while the industry book-to-bill was moving from peak to trough.

The author is long INTC and short KLAC