Westminster Blog

George Osborne wants to channel billions of pounds of pension fund money into big infrastructure projects such as rail, road and energy projects.

This is the premise of Monday’s “infrastructure review”, in which the chancellor will update us on progress on getting this big plan off the ground – along with a list of 40 or 50 “top priority” infrastructure schemes which we should see happening quite soon.

The curious thing, however, is that pension funds already pay for infrastructure projects, and not only through PFI schemes. This is because they regularly buy tens of billions of pounds of government bonds (gilts) which are then used to finance general government stuff: including capital investment.

Investors in infrastructure schemes usually expect more income given that risk is perceived as higher than supposedly risk-free gilts (although the Eurozone crisis has put a dent in that concept).

So why would Osborne want to borrow in this more expensive way? Could it be a rather less visible way of accumulating government debt than simply going to the capital markets? That was of course one of the true rationales for the now controversial PFI system.

Here is the government spiel:  The World Economic Forum ranked the UK at just 33rd in the world for the quality of its infrastructure, down from 9th in 2005. The government anticipates that some £200bn needs to be invested in UK economic infrastructure over the next five years, with most of that spent in energy and transport. The coalition believes that previous governments failed to produce a coherent view of the long-term needs for British infrastructure.

Ministers are determined to lever in private sector investment and reduce the cost of capital for projects and programmes – and officials have been working for months on “efficient and effective funding models“. Key to this is encouraging participation by pension funds also overseas sources such as sovereign wealth funds.

In the UK the level of infrastructure investment is estimated to be under 1 per cent of pension fund assets compared to 8-15 per cent in Australia or Canada. Officials and ministers want to change this.

So far, so straightforward.

The key is to lower the cost of capital, which increases as “risk transfer” increases. For example building a toll road is high-risk because both demand and operational risk are transferred away from government.

Sorry to get technical; but one way to solve this problem and cut the cost of capital would be to extend the regulatory asset base (RAB) concept to new sectors, a move urged by the British Chambers of Commerce in a report this summer.

“RABS” are in effect contractual agreements between regulators, investors and infrastructure providers which allow a fixed rate of return to investors. It is an approach already in place for electricity, gas, water and rail networks but it could be extended into other fields, proponents believe.

This was one of the main suggestions in a similar infrastructure report published a year ago by Infrastructure UK, the relevant quango.

It will form part of a review of PFI which was launched last week by chancellor George Osborne and is expected to take 10 weeks to complete.

Mr Osborne said he wanted to draw on private sector innovation at a lower cost to the taxpayer.

A recent Commons Treasury Committee report put the cost of capital for a typical PFI project at 8 per cent, which is double the long-term government gilt rate of around 4 per cent. The committee said paying off a PFI debt of £1bn could cost taxpayers the same as paying off a direct government debt of £1.7bn.

Mr Osborne said back then that the government wants “a new delivery model which draws on private-sector innovation, but at a lower cost to the taxpayer and with better value for public services”. The model needs to be cheaper than PFI, access a wider range of financing sources and carry a better balance of risk between private and public sectors.

I’ve heard that Monday’s scheme would involve RPI-related increases. Because most state-owned infrastructure schemes do not have visible revenue streams, however (apart from social housing and toll roads) the cashflow would have to be synthetic. This is financial engineering that may remind some of the world of mortgage-backed securities.

Which takes us back to my original question; why not just issue more gilts to raise the money? Is this all about – like many believe PFI was – keeping the debts off the public balance sheet?

Richard Threlfall, head of infrastructure at KPMG, tells me that with the government determined to protect its credit rating it has no choice but to turn to the private sector for more financing.

“The government is caught in the grip between that austerity and wanting to pump money into the economy to get growth; hence one solution is bringing in private capital in this way,” he said.

There are a few glitches to iron out in the favoured solution, apparently. The Treasury wants its mechanism to enable pension funds to invest “from the word go” into big capital projects.

Within the industry there is some scepticism, however, with some experts believing that the government will still have to use conventional debt for the risky construction period of such schemes. Only once they are up and running will many pension funds want to invest in the projects, they say.

Nick Prior, head of infrastructure and capital programmes at Deloitte, told me that the debate on a successor model to PFI was important but remained “substantially academic” in respect of immediate economic growth.

He was also sceptical over whether pension funds would want to invest in infrastructure schemes at their early, risky stage.

“The business models of pension funds gear them towards low-risk, stable, long-term yields,” Mr Prior told me. “Introducing them to the riskier build and development phases of infrastructure projects doesn’t currently fit with this outlook. One has to be sceptical that any model transferring these risks to pensions funds can be delivered, particularly in the current economic climate.”