The November 2006 Bulletin paper by Martin McKee et al., Publicāprivate partnerships for hospitals (84(11):890ā96), makes several valid and useful points about the Private Finance Initiative (PFI) in the building of new United Kingdom of Great Britain and Northern Ireland (UK) hospitals. In particular, the emphasis on the failure of the UK government to undertake a rigorous evaluation of the PFI is welcome, as is the coverage of ministersā attempts to discredit the work of scholars whose research has disproved many of the claims made in favour of it.1,2 At the same time, the paper has some serious weaknesses. First, at several points it treats the key issues raised by the PFI as if they were still open, when from a scientific viewpoint this is unjustified. The paper fails to draw attention to an extensive and detailed empirical literature dealing with the affordability problems created by PFI and their impact on public expenditure and the scope of service provision. Given the quantity of detailed research devoted to this and related issues over a period of more than ten years, not least by my own research team (previously at University College London and now at the University of Edinburgh)3ā9 and the failure of any PFI supporter to refute the findings, it is surprising to see the debate as a whole described by fellow scientific researchers as āideologicalā. The paperās concluding statement that āit is impossible to say whether the model underlying publicāprivate partnerships is flawed or whether the difficulties ā¦ are the result of mistakes in its executionā is not even consistent with the authorsā own apparent conclusions in the body of the paper. Second, the authors seriously understate some of the problems they acknowledge that the PFI presents. There is no reference to the systematic failures of risk transfer, which underpins the governmentās justification for using private financing, rather than less costly public financing. The paper claims that PFI investments are high-risk and near ājunk bondā status, suggesting extensive risk transfer. The paper cites the Office of Government Commerce (OGC) in this context. But the OGCās claim does not relate to the risks borne by investors, which are generally triple A, and only refers to bonds (www.hm-treasury.gov.uk/media/6066B/ppp_GuidanceonCertainFinancing.pdf). It is true that the underlying credit strength of PFI projects is usually in the range of BBBā to BBB+. However, most PFI bonds to date have been āwrappedā by a monoline insurer, allowing the bonds to benefit from an AAA rating. The cost of this insurance is borne by the public. Therefore, risk transfer in bond financing is paid by the public through the costs of private finance and the insurance costs. The paper acknowledges that the very low risk actually transferred to the private consortia has enabled the latter to enjoy āsignificant benefitsā. But the scale of the profits made through refinancing ā in the case of the Norfolk and Norwich University Hospital project giving returns of 70% on the contractorsā original investment ā is surely more than a āsignificant benefitā: it is a misuse of public funding that fundamentally undermines the risk transfer argument. Third, the authors overlook further research that has demolished āthe one positive findingā they claim has been established concerning the PFI: the governmentās often-repeated assertion that it reduces the cost and time overruns of hospital procurement relative to the traditional system. A study of the evidence for this claim shows that it rests on a single erroneous report by a consultancy with a major interest in PFI projects (http://www.health.ed.ac.uk/CIPHP/publications/unison_2005_pfi_a_policy_built_on_sand_pollock.pdf, http://society.guardian.co.uk/societyguardian/story/0,1600183,00.html).10 The focus of this report was on price certainty after contracts are signed ā a method that is bound to favour the PFI. A more valid comparison between PFI and traditional procurement pricing would involve increases from the Outline Business Case (OBC) stage, not post-contract increases. The PFI process from OBC to Final Business Case (FBC) is relatively longer than that in traditional procurement, and there are bigger differences between OBC and FBC figures for price and delivery time. Department of Health figures show that the average cost increase between OBC and financial close for āprioritisedā PFI projects is 74.5%. Therefore, the paperās statement that ācompared with the traditional system, PFI facilities are more likely to be built on time and within budgetā is unfounded. ā