Why Infrastructure isn't a Bubble

Kit Roane has a weirdly bearish piece on infrastructure investment today, looking at it from the point of view of pension funds and bemoaning the fact that they can no longer get predictably high returns on any deal they like.

Roane makes the good point that infrastructure returns are largely bear-proof: the income streams are long-term, predictable, and often guaranteed or at least legislated by the government. In an environment where most other asset classes are much more correlated with each other and have much more downside, it's easy to see why infrastructure is attractive.

But Roane also says that "prices have already been driven too high" and worries that infrastructure is "fated to become the next asset bubble". I don't buy it, for a number of reasons.

For one thing, you can't have a bubble without speculators: people who buy with the intention of flipping, at a profit, to a greater fool. So far, the secondary market in infrastructure investments is slim to nonexistent. Everybody who's buying is buying to hold, not to flip.

And for all that the amount of money in infrastructure funds is rising, it's still minuscule by comparison with total infrastructure-investment needs worldwide. Roane puts a figure of $160 billion in infrastructure investment funds raised over the past two years; that compares to $53 trillion in needed infrastructure investment over the next 25 years. Annualize both numbers, and you have $80 billion a year in private funds chasing $2.1 trillion a year in opportunities. No matter how much leverage you add to those private funds - and all that leverage counts as infrastructure investment as well, remember - you're not even getting close to the point at which too much money is chasing too few opportunities.

Where most observers would see a vibrant marketplace, Roane sees unhealthy competition: financial sponsors are "often competing madly against each other," he says, noting that one investor will "sometimes walk away from a deal". But that's how markets work: investors look at lots of opportunities, and pick the most attractive ones. If you're not walking away from potential investments, you're not investing in a remotely mature market.

But the most important thing to bear in mind with infrastructure investment is that you can almost never have too much of it. The world is crying out for new roads and railways, power plants and ports - the developing world in particular. And there are really only two ways of funding those projects. The old-fashioned way is that governments would pay for them; the new-fashioned way is that a large amount of the money comes from the private sector. And private-sector involvement is a good thing, because it reduces costs, increases efficiency, and stops the government from having to needlessly run up enormous deficits.

A lot of the first generation of infrastructure investment was in the form of privatizations: previously public assets were sold off to the private sector. That's good too, since all of the proceeds flowed directly into the public fisc. But the "greenfield developments" which Roane worries so much about are actually even better: the public gets all of the benefit of the new infrastructure, without having to bear any of the up-front cost.

Yes, bond and stock investors in Eurotunnel, to use one of Roane's examples, were both hit financially when the chunnel didn't prove as profitable as had initially been expected. But the hit was taken by institutional investors who were putting risk capital to work. The alternative would have been for the losses to have been borne much more by UK and French taxpayers. And I really don't see why that would have been preferable.