The next question to ask is whether the benefits of private finance are worth the premium apparently paid compared to government's borrowing cost. A whole industry - Public Sector Comparators, Value for Money guidance, and so on - has developed to answer this question project by project. But it rests on an important assumption: that government can sustainably fund projects from borrowing more cheaply than the private sector. Is this true?
It is certainly the case that government debt is cheaper than the debt provided to finance PFI projects, and cheaper still than the overall cost of finance for PFI projects, i.e. the weighted average cost of capital (WACC). But this is to compare apples with fruit salads. It ignores the position of taxpayers who play the role of equity in this financing structure.
Making a simple comparison between the government's cost of debt and the private sector WACC implies that the government can sustainably fund projects at a cost of finance equal to its risk free borrowing rate. But this would be true only if existing borrowing levels were below prudent limits. The constraints on public borrowing suggest, however, that borrowing levels are not currently too low. These constraints exist because government borrowing must ultimately be funded by the taxpayer. Prudent management of the public finances requires decisions on the sustainable level of debt that can be supported by taxpayers.
Take a corporate example. A company considering an investment should fund it entirely by debt only if its gearing is sub-optimal; in those circumstances that would be the cheapest thing to do. But if it were already optimally geared, 100% debt finance would be the wrong financing solution. Over-gearing would cause its credit quality to deteriorate and its cost of equity to rise. This would affect the business as a whole, not just the incremental investment.
Similarly, if government has already reached prudent borrowing limits, it would be wrong to favour government debt financing over PFI as a financing solution simply because the headline cost of the former was lower. This would be to ignore the taxpayers, who represent the equity in its financing structure.
Another way of looking at this is by considering the opportunity cost of government funding. The economic cost of providing funding for a risky investment should be equal to the returns foregone by not using it for an alternative equally risky investment. In other words, the cost of investing in one project is not necessarily the cash cost of that investment, but the opportunity which has been lost to invest that cash in other projects. The fact that governments can borrow more cheaply than the private sector does not mean that they should expect any lower return on their money.
Consider examples where Governments do invest large amounts of sovereign wealth. Examples of this include the Future Fund in Australia, GIC in Singapore and Dubai Holdings. These economies have large amounts of funding available for investment but do not use it as a vehicle for subsidising government projects. Why? Presumably this is simply because they can make better returns for bearing similar risks by investing in a balanced portfolio of stocks and bonds. Were these economies to invest instead in their own businesses or projects, the direct cost of funds might be low but they would be foregoing opportunities to invest funds in opportunities available on the market more generally. If, for example, the Australian Government chose to invest its Future Fund (which is aimed at funding long term government pension liabilities) in infrastructure projects or PPPs at a concessional rate which reflected its cost of funding advantage, it would be foregoing the opportunity to buy investments in the broader market where it could earn the same returns as any other investor (and therefore capitalise on its cost of funding advantage rather than concede it to an infrastructure project). This serves an important public policy outcome i.e. good stewardship of public resources. The Future Fund, in this case, would be in a position to pay more of the Government's pension liabilities in due course than it would have been able to do if it had simply conceded its cost of funding advantage to various projects. The opportunity cost of capital to the private and public sectors is the same in this sense.