Lee Babiss left Roche, one of the world's largest pharmaceutical companies, for a contract research organization position in part because his days as president of drug research were filled less with driving research projects forward and more with containing his group's costs, a reflection of the current state of R&D productivity at major pharmaceutical firms.

Babiss speaks of the evolving relationships between pharmaceutical firms and CROs, including end-to-end virtualization of R&D.

Roche's rationale for buying Genentech must out-argue the one big reason not to do the deal: theirs has been the most successful relationship in pharmaceutical history. Instead, Roche is betting that this unique relationship has already borne its best fruit. In an emerging world of payor constraints, biological me-toos and growing oncology marketing expenses, the costs of keeping Genentech independent -- manufacturing transfer prices, up-front fees and royalties, and most importantly, no ability to leverage its investment in the US marketplace -- are simply too high. Moreover, Roche is clearly not convinced that Genentech's productivity would have continued at the rates it has in the last decade. Roche's vision of the pharma future looks a lot more like the cost-constrained world CEO Schwan knew at Roche Diagnostics - where innovation was rare and rarely paid for; where extraordinary business acumen counted for more than outsized research capabilities.