By Steve Coulter, LSE
The takeover tussle between rival drug giants Pfizer and AstraZeneca has uncovered some interesting truths about government and corporate policies on taxes and R&D funding. Of course, the shadow of Pfizer still looms over UK-based AstraZeneca, with continuing speculation that the US-headquartered Viagra manufacturer will revive its bid later this year. Both firms face imminent ‘patent-cliffs’ as profitable older drugs near the end of their patent protection. Pfizer’s strategy, backed by the Economist, is to acquire new drug products through acquisition. AstraZeneca is a particularly attractive target because the UK’s corporation tax rate is 20% compared to 40% in the US, allowing Pfizer to shift its headquarters to the lower-tax jurisdiction. But some UK politicians worry about job losses among UK research staff if the merger goes ahead – although AstraZeneca’s research activities are already highly de-centralised.
However, the real issue, as outlined by Mariana Mazzucato, is twofold. First, the knowledge behind many of Pfizer’s products has been bankrolled by the American taxpayer through generous government support for R&D. The vertically-integrated drug companies of old have been replaced by flatter organisations which increasingly farm out their R&D to government-backed research institutes. And the money saved is increasingly spent on share-buybacks rather than being ploughed back in to drug development. Second, these ‘innovative’ firms also seek constant handouts through R&D tax credits, fuelling wasteful tax competition between countries. Mazzucato cites the example of the UK’s ‘patent-box’, an ill-conceived tax break that will do little to increase the amount of R&D done in the country, according to the Institute for Fiscal Studies.
Mazzucato calls for an end to the current, parasitical, innovation eco-system which puts shareholder’s returns ahead of new product development. Until that happens, it’s going to be a long wait for that new generation of antibiotics which doctors are calling for…