Value investing in infrastructure

Infrastructure investors have historically focused on the operational stability of infrastructure assets when underwriting investments. The last four years have shown that, in many cases, operational resilience can be overwhelmed by the investment risks inherent in high entry valuations and debt levels. In this article, we offer several observations on the importance of valuation in the infrastructure context, and how value might be assessed to secure a margin of safety on infrastructure investments.

Returns observed in listed infrastructure suggest private market fund return benchmarks may have been set too low.

The Dow Jones Utilities Index provides the best benchmark of long-term, US-listed infrastructure returns. 

The Index has delivered 9.5 percent compounded returns (including dividends) per annum since 1970, slightly below the 10.0 percent returned by the S&P 500 over the same period. 

By simply levering the Index to customary infrastructure fund leverage levels, one would increase the return to 12 percent. That is, on a leverage-equivalent basis, private market funds must provide net returns of 12 percent simply to match the historical performance of the Dow Jones Utilities Index.

The table below depicts the gross returns that an infrastructure fund must achieve to provide a net 12 percent return to its limited partners, assuming various fee and carry terms.

 

The analysis demonstrates clearly that a low double-digit return hurdle, which is somewhat common among private infrastructure funds, is too low to generate any meaningful outperformance against publicly-listed infrastructure, and provides investors no compensation for illiquidity.

Entry price may be the most important determinant of success for an infrastructure investment.

Historically, the key to achieving mid-teens returns in infrastructure has been to invest at a disciplined entry value.  This should come as no surprise. It is difficult to grow one’s way out of an excessive purchase price multiple for developed-market infrastructure, which often does not benefit from strong secular growth.

This conclusion is demonstrated in the table below which shows ten-year total investment returns versus entry multiple for the Dow Jones Utilities Index. We have divided the data into quintiles based on the monthly price-to-earnings (PE) ratio for each month since 1980.

 

The data show that lower entry multiples are strongly correlated with a substantial improvement in investment performance. Indeed, the listed index provided mid-teens returns on low leverage when investing at lower entry multiples.

Valuation cyclicality is the major source of volatility in infrastructure investing.

Infrastructure multiples are cyclical over time, as the chart below shows for multiples of EBITDA, or earnings before interest, tax, depreciation and amortisation; an examination of rate-base multiples yields much the same results.

 

Furthermore, as Exhibit 2 above demonstrates, there is no escaping the impact of valuation risk, even for long-term hold returns. We believe cyclicality of valuation multiples is the primary source of investment volatility in infrastructure investing.

Good infrastructure assets tend to have relatively stable cash flows. Even GDP-related assets, such as airports and toll-roads, tend to have predictable cash flows over the medium-to-long term, looking past the effects of short-term cyclicality. As a result, operational volatility is not usually a major source of investment volatility in infrastructure.

Consequently, valuation risk deserves at least as much attention as operational risk in assessing infrastructure investments.

Valuation should be measured against the distribution of expected returns, not just the “base case”.

Infrastructure returns should be considered in terms of a distribution of possible returns, rather than a single-point, or single base case, return. The following chart shows return distributions for two hypothetical utility investments.

 

The red line (Investment A) depicts a utility that earns a highly predictable regulated return on equity (“ROE”), making the base case (the mode) relatively easy to pinpoint. Here we portray a situation in which an investor has paid a high multiple of rate-base (and EBITDA), and utilized high leverage to achieve a 13% base case return. Downside risk exists due to the potential of an exit at a lower multiple and the possibility of a refinancing under unfavorable circumstances. At the same time, there is minimal opportunity for out-performance on valuation, cash flow growth, or financing. Hence, returns are skewed to the downside, with a greater probability of an outcome below the base case than above it.

The blue line (Investment B) depicts a utility with a major upcoming rate case that would increase rates substantially, though the amount of the increase is uncertain. This utility has been acquired at a lower purchase price multiple with more conservative leverage, leaving more potential for upside. As can be seen, Investment B has a greater spread of possible outcomes (i.e. it is not as predictable as Investment A), but the downside is better-protected and returns are skewed to the upside. Investment B is clearly the better, and less risky, investment.

Moreover, in a managed fund context, a portfolio of investments that have left-skewed return distributions like those of Investment A will have an expected return substantially lower than the average of the individual base cases.

Long-term levered IRR is a poor tool for assessing return distribution.

Long-term hold IRR models continue to be the predominant analytical tool in infrastructure investing. It is appropriate to run such a model given the longer term holds of infrastructure investments. 

However, it is also critical to run a five- to ten-year model, regardless of intended hold period, with exit multiples set to long-term averages (and sensitivities showing returns on even lower multiples). A shorter-term model highlights valuation risk in a way that long-term or hold-to-maturity models cannot, and forces investors to account for left-tail valuation scenarios in their investment decisions.

This is most easily demonstrated with an example of a mature toll road, acquired with the intention of holding it for twenty years.  The purchase price is 14x EBITDA, and EBITDA is expected to increase steadily at 3.5 percent per annum from a combination of toll and traffic increases. Assume that the long-term average EBITDA multiple for similar toll roads is 11x EBITDA, and the transaction is capitalized with net debt of 7x and equity of 7x:

 

A five-year hold model makes it clear that paying 14x for an asset that has historically traded at 11x is a dangerous proposition. The multiple for this investment will likely compress from 14x to 11x during the holding period; by definition, it should be below 11x for around half of any extended period.

A twenty-year hold model disguises this valuation risk. Even using the 11x exit multiple, a twenty-year model could deceive an investor into believing that the value of the investment will compound at a steady 12.2 percent annually for twenty years.

In reality, the IRR will likely turn significantly negative in the early years, as industry multiples revert to the mean and cause a significant decrease in equity value. If this coincides with a large debt maturity or other need for liquidity, which is impossible to predict with certainty, the result could be a significant realized loss on the investment. If not, the IRR will take several years to climb back to 0 percent (par value) and a full twenty years to reach 12.2 percent. By over-paying for the investment, the investor has assumed substantial valuation risk and incurred a significant opportunity cost in terms of foregone return and potential for upside.

Conclusion

The touchstone for all good infrastructure assets is that they benefit from lower operational volatility and an enduring franchise value created through the provision of an essential service and high barriers to entry. These characteristics fulfill a necessary, but insufficient, condition for success in infrastructure investing. Only by combining operational defensibility with disciplined valuation techniques can infrastructure investors achieve lower-risk returns in the private market that exceed listed benchmarks.

Michael Dorrell, David Tolley, and Trent Vichie are the founding partners of Blackstone Infrastructure Partners.