Should pension funds invest in infrastructure?

Investors new to the infrastructure asset class need to understand the risks as well as the return potential. Some lessons from the financial crisis are worth studying, writes Duncan Hale.

During the past five years there has been conspicuous growth of infrastructure as an asset class for European investors – long a favourite of Canadian and Australian pension funds – because of the expectation of equity-like returns for volatility akin to a bond portfolio. While the financial crisis has helped correct these expectations, we believe that infrastructure remains an attractive investment, mainly because the natural durability of infrastructure assets means that it provides diversification from other more cyclical assets within a pension fund portfolio.  However, as with most asset classes, recent market events have given infrastructure investors a chance to reflect on the way the asset class performed, and a number of lessons have been learnt, particularly with respect to how investors’ exposure is achieved and manager products are structured.

One of the first challenges for a pension fund initiating an infrastructure allocation is deciding what it is attempting to achieve.  Infrastructure is a broadly defined asset class which encompasses a large and varied number of investments, and thus a wide spectrum of risk and returns.  However, generally infrastructure refers to any investment in a project with the following characteristics:

  • provides essential services to businesses and/or communities
  • capital investment
  • long asset life
  • monopolistic in nature, with relatively high barriers to entry or competition.

As such, the main benefit to investors is the robust revenue streams and associated cash flows that can be derived from these essentially monopolistic services.  These provide diversification benefits as infrastructure is, in aggregate, less economically sensitive than other asset classes like public or private equity.

Infrastructure is also often seen as an asset class that offers a real return.  Some infrastructure assets provide a natural inflation hedge because their revenues are contractually related or linked to inflation (CPI or RPI). Also, the capital appreciation of assets offers an inflation hedge similar to that offered by equities.

In addition to these two factors there is an abundance of potential investments in the asset class due to the need for major capital works in countries all around the world, which explains why infrastructure is a mainstay of pension fund investment in some parts of the world and is now growing significantly in Europe and the UK.  Allocations by UK pension funds should be a balance between having a large enough allocation to make a meaningful impact at the total scheme level and to achieve diversification, tempered by the very illiquid nature of the asset class. Consideration also needs to be given to allocations to other illiquid asset classes in the portfolio, such as real estate and private equity.  Given these parameters, many UK institutions are allocating between 2 percent and 5 percent to infrastructure, while some have targeted allocations of up to 10 percent; a figure which is dwarfed by a number of the large industry funds in Australia which have made allocations of between 20 percent and 25 percent.

However, the characteristics outlined above do not mean that the asset class is without risk. The stability of cash flows enables a capital structure to be employed which includes significant amounts of debt.  Debt levels of 50 percent to 70 percent of the value of an asset are not uncommon for economic infrastructure (such as toll roads and pipelines), while for social infrastructure (such as schools and prisons) where the counterparty is the government, debt levels of over 90 percent are not uncommon.  These levels of debt are not necessarily imprudent, but they do end up magnifying any deviations in revenues for equity holders. 

The use of leverage in the asset class does have the potential to undermine returns for equity investors. This is particularly true when the level of debt is too great, or is not structured well.  In the benign credit environment of 2006-2007, a combination of easily available debt and aggressively bid infrastructure projects resulted in a number of these projects taking on more and increasingly complicated levels of debt.  With the financial crisis, some of these projects have begun to struggle, and the stable, robust asset class that some investors thought they were accessing has not materialised.

While the travails of a few high-profile assets have made headlines, it would be unfair to tar the entire asset class with the same brush. During the recent financial crisis, infrastructure provided some protection from falls in the equity market, albeit not as much as many investors had hoped for.  Many of the best structured assets have long-term, sustainable debt funding in place and debt for new, high-quality infrastructure projects remains available, although the price has increased. Further, despite the issues facing a number of the strategic infrastructure owners – such as constructors and investment banks – we see this as a period that provides good opportunities for skilful investors who can drive financial gains through greater operational improvements and can access deals without having to participate in auctions.  As such, while many of our clients made few commitments during the past 12 months, we still see an appetite to invest in the asset class.

Recent events highlight the key challenge facing a pension fund investor: designing an appropriate manager structure at the right price.  Given the relative immaturity of global infrastructure as an institutional asset class, the diversity of manager structures (independent, investment bank sponsored etc) and the illiquidity of the asset class, selection of managers is arguably more challenging than in other asset classes.  The past 18 months have crystallised some traits infrastructure managers will require in future:

  • a stable business and team which will be around over the life of a fund
  • a robust philosophy, such that investors can be confident with the product they are purchasing
  • strong discipline, both in terms of ensuring what is bought is appropriately bid, but also that the monitoring and operational improvements of the asset are undertaken in a diligent fashion
  • identifiable and differentiated deal flow, as models that rely on participation in widespread auctions have been largely discredited.

Recent market events have highlighted the need for diversification in an infrastructure allocation.  Infrastructure funds are often very concentrated; whether that is in number of projects, geography, drivers of returns or a combination of all three.  A structure that provides diversification across manager, geography, return driver and vintage year is vitally important.

The last 12 months have also provided an opportunity to reflect on how infrastructure funds are structured.  In 2006 and 2007, there was spectacular growth in demand for infrastructure funds, which meant that it was largely a seller’s market.  Most infrastructure managers structured their funds as private equity vehicles and, given the differences between private equity and infrastructure, it is arguable that this was not in investors’ best interests.  However given the current scarcity of capital, it has largely become a buyer’s market and we believe it is a good time to make the structures of infrastructure funds more attractive for pension fund investors. While this should include negotiation of more equitable terms in fund documentation, such as appropriate key-man provisions and no-fault divorces, the main area that we believe needs addressing is fees. 

The present fee model for infrastructure borrows heavily from private equity with many funds charging 1.5 percent pa plus on commitments and investments, as well as 20 percent of outperformance once an 8 percent return target has been reached with catch-up. We believe this is not appropriate.  The level of fees currently paid by the investor is too high; both directly through fund fees and indirectly through hidden fees paid by the portfolio company to the manager, particularly given the lower return expectations for many infrastructure funds relative to private equity funds.  We would like to see this addressed in three ways:

  • a reduction in the headline fee
  • introduction of a true hurdle (potentially based on a real return) for the performance fee, with no catch up, so investors only pay for true outperformance
  • elimination of some of the hidden fees (particularly acquisition fees) and greater transparency on all other areas of manager revenue generated from its infrastructure investments.

We have been discussing these issues with managers for a number of years, however it is only now that managers are willing to discuss a fairer fee deal and that we have had success in negotiating better fee structures.

Overall, we believe that infrastructure remains an attractive asset class for those clients that want to diversify away from other, more traditional asset classes, subject to fee arrangements.  We expect the majority of European pension funds to continue to access the asset class through infrastructure funds, while a number of large, sophisticated Northern European investors will invest directly in individual deals. Direct investing offers benefits to these investors through a headline reduction in fees and greater control over the assets in a portfolio. 

The financial crisis has highlighted many of the risks associated with the asset class, but it has also provided an opportunity for the asset class to mature, in terms of return and risk expectations; experience of both investment teams and investors; and fund structures, thereby making it more attractive to pension funds.

Duncan Hale is head of infrastructure manager research at Watson Wyatt in London.