You can understand why big-pharmaceutical industries would be interested in studying the placebo effect. In order to fabricate good medicines, the effects have to be real… in the sense that the healing effects are not simply due to psychological residues.
Robin Hason, on Overcoming Bias suggests that their are gaming the system; seeing how, when and where they should conduct research on certain drugs to maximize the placebo effect. In other words, placebo effects for different drugs is widely variant based on different cultures, geographies and the like. Since no drug trial is truly randomized (despite appearances), some studies will report different effects when tested in different areas or subgroups. Pharmaceutical companies would like to test their drugs in ways that allows them to maximize their apparent effectiveness against the placebo ‘background.’ Worse, the NIH and (probably) FDA are staying mute on the subject because they don’t want to lose their corporate funding. Talk about the importance of financial signals!
In a related post, Mind Hacks has an excellent post about how the placebo effect isn’t just the effect of “mind over body” healing, but it is also statistical variations in the way different people respond to disease and medicine. If the subjects of a study are not properly randomized into groups (or a poor sample population is chosen in the first place) then the placebo effect will change the “background” signal of disease progression sans drugs, which will change the apparent effectiveness of the drug. (HT2 TGGP in the comments at OB)