Infrastructure investing: will it take off as an asset class?

I attended the EFFAS-EBC conference in London on the 14^th and 15^th of October (link here).

The overall theme was long term investing but there was a large proportion of the talks on infrastructure as a long term asset class.

The Economist published a small article here with the following statistic: only 0.9% of UK pension assets are invested in infrastructure while there are colossal investment needs ahead.

The EU commission mentions between €1.5-2trn of infrastructure investments needed by 2020 in the EU in the fields of transport, energy and broadband. This prediction was made in 2010 which means that the target will be missed.

At the same time as infrastructures are needed, pension funds are looking at long dated fixed income investment to match their long-term liabilities. Once those securities are offered, the pension managers and insurers are rushing to buy them. The utility sector has been for instance able to issue long dated bonds at attractive rates. For instance, Fortum in Finland has issued 10yr bonds at 2.2% last year!

However, debt capital markets only represent 10% of current infrastructure investment as the public purse remains the main payer. The key question is therefore how to develop those capital markets in order to get private money to fund public infrastructure.

The EU and the US (through a programme called TIFIA: The Transportation Infrastructure Finance and Innovation Act) have come up with a solution: they will guarantee part of the senior secured bonds (up to 20% of them in a given project) such that the interest costs can remain favourable and/or that the bonds/loans be favourably rated and attract insurers and pension funds (most of them cannot invest more than 5% of assets in non-investment grade securities). The EU commission has earmarked €29bn for the next 7 years for this purpose.

What is impeding investments? Five reasons can be mentioned:

1)      The EU commission is complaining that there is a dearth of viable projects available. I find this quite surprising if they also believe that €2trn of investments are needed in the next 10 years

2)      Pension fund managers complain that the regulatory framework is too fuzzy, which deters them from investing into the asset class. Why hire a specialised infrastructure team when a government can change the rules a few years down the road (e.g. retroactive solar subsidies) or cancel capital expenditures once in power (e.g. the UK conservatives in 2010)

3)      Prudential rules assign higher risk weights to long dated loans by banks. Therefore, they are not in a rush to underwrite long term securities as they will use too much capital

4)      Infrastructure investments are hard to model, value and compare. A single project is so singular that a historical example may not be relevant. Therefore, assigning hurdle rates is still a guesstimate exercise.Once there are enough transactions in the history database such as an index can be created, then infrastructures could be truly considered within a portfolio context

5)      My understanding: most private investors do not earn the positive externalities of an infrastructure project, contrary to the national governments and therefore need a higher hurdle rate than can be supported by a project. A metro line investor will not earn the added business taxes of the local council seeing activity boosted at the surface for instance.To me this is the main problem: private capital seeks a clear risk/return arbitrage. Most infrastructure investments do not offer positive economics.

To sum up, I do not think that the private sector will pick up most of the slack anytime soon.

The prospects of the EU supporting the credit rating of private infrastructure investment is also frightening to me: it reminds me of Ambac, MBIA and the other credit insurers which failed spectacularly in the last crisis.