From Barron's 6-11-07: "The markets tend to get very excited about the next big turnaround like the Gap which had a lot of excitement around it during the holidays. But the excitement was on Wall Street, not in the stores. Paying high multiples for the promise of a turnaround or dramatic growth becomes a riskier and riskier bet as the environment gets tougher. We are cautious. We are thinking in terms of consumer compression not consumer recession"

From Barron's 6-11-07: Question: What retailers do you like? Answer: "In our entire universe, we could find only one core retail holding, and that's Tween Brands (TWB)" "Tween Brands was formerly known as Too. But they changed their corporate name to reflect the fact that they have two very different distinctive retail banners today: the Limited Too, which is a great business, and a relatively new concept by the name of Justice. Limited Too is the most unheralded specialty-retailer growth story out there. They sell fantastically well-merchandised apparel and lifestyle products to girls between the ages of 7 and 14. It's a very steady, very profitable business with 570 stores and modest growth potential. Until recently, people thought they had no growth potential. But we think they'll increase square footage by 5% in the next couple of years. That's not immaterial. Combine that with 20%-plus operating profits at the store level, and this is just a great cash cow. Justice, on the other hand, is the strip-mall version of Limited Too: You get a similar feeling but lower prices and the strip-mall format, which doubles their opportunity for square footage and market penetration. They have opened 184 stores under the Justice concept, and same-store sales comparisons were up 22% in its most recent quarter"

From Barron's 6-11-07: "It's in many ways the most interesting because it is the messiest story. They offer a very special technology called String Ribbon wafer technology that allows manufacturers of solar products to use less silicon, which is important given the shortage of silicon. Evergreen's technology uses about five grams of polysilicon to produce one watt of power and we think that will go to about three grams a watt by 2010. The industry average is closer to nine grams per watt. Today, at a weighted average cost of $100 per kilogram of polysilicon, Evergreen spends about 50 cents a watt on silicon, and that's 30% of the total cost of the solar cell. A typical industry player would spend about a $1 per watt on silicon, representing almost 45% of the cost of goods sold. Evergreen has a 20% cost advantage. Even if silicon prices were cut in half, Evergreen would spend about 15 cents a watt on silicon, whereas an industry player would spend roughly twice that."

From Barron's 6-11-07: "They are the leader in global money transfer, providing people around the world with easy and trustworthy ways to send money. They have 300,000 agent locations in 200 countries. It provides an absolutely vital service. We think it is underappreciated because Wall Street has an Internet view of the world, which is, why not use PayPal and credit cards to send money? But a lot of Western Union customers don't have access to e-mail or a computer or a bank, for that matter. When they were preparing to be spun out by First Data, we were astonished at their returns on capital, their cash-flow generation, their 30% profit margins. Historically, this has been a very solid and a low-double-digit-growth business. Nice and boring and immensely profitable. But because of the costs of becoming a public company and some short-term disruption in the Mexican market, an important market for them, and some headline risk on immigration reform, we think the stock now presents an opportunity. Trading at just over 17 times 2008 earnings and just over 11 times 2008 Ebitda [earnings before interest, taxes, depreciation and amortization], it is just too cheap"