Cheap Big Pharma stocks?

Analysing pharmaceutical companies proved to be much more challenging than I initially thought. I started screening best stocks in this sector with my usual parameters (ROE, ROA, gross margin etc.) and first focused on the ones paying over 3% dividend and having the lowest P/E ratios.

As the efficient market hypotesis (EMH) suggests, all information available should be already reflected on the stock price. Therefore, there should be a good reason why some Big Pharmas trade at P/E over 14 and some below 10. While I don't fully agree with EMH in it's broadest form which suggest basically that people like Warren Buffett should not exist, I do acknowledge that Big Pharmas are probably priced correctly as they are probably being followed by hundreds of professional analysts and there is no free lunch to be found. Nevertheless, I am curious as to why such a big differences in P/E.

So far, I have looked at two companies: AstraZeneca and Merck that have P/E below 10. While Merck has some fairly big one-time items in profit and loss statement explaining low P/E calculated from 2009 result, I found out something much more interesting.

I have been aware of the significance of patent protection for drugs, but I imagined the big companies had large enough existing portfolio and a lot of new drugs in the R&D pipeline to offset the impact of patent expiry. I may have been wrong. Although, it is immensely difficult to assess the drugs in the R&D pipeline for their eventual selling potential.

Three of the most revenue generating drugs for AstraZeneca are Nexium, Crestor and Seroquel. Each of these drugs generated between 4.5-5.0 billion USD of revenue 2009. That's a lot compared to sales of 32,8 billion for the whole company. Overall, four out of 10 best selling drugs will be without patent protection by 2012:

Patent expiration will likely take away most of that sales as generic drug companies will come into play and the prices will be slashed. AstraZeneca better have some pretty good stuff in the R&D pipeline as many of their key selling products will face competition in five years.

For Merck the situation seems to be pretty much the same, although they seem to have their sales spread over their product portfolio more evenly:

I will be going through some more annual statements in near future to look for similar information as I want to understand if these patent expirations explain the differences in P/E of big pharma companies.