Building a Case for Infrastructure
Institutional Investors Evaluate a Multitude of Opportunities
By Bruce Pflaum and Bob Moreland | The Institutional Real Estate Letter
The recent collapse of the I-35 bridge in Minneapolis was both a shock and a tragedy. The only positive outcome has been the national attention the disaster focused on the 73,533 bridges rated "structurally deficient" by the Federal Highway Administration. Repairing all of these bridges would require $188 billion.
The infrastructure demand is broader than just bridges. By some estimates, the United States will need 35 percent more electricity capacity by 2030. The American Society of Civil Engineers estimates there is a $40 billion annual infrastructure deficit in the United States; estimates in Canada run as high as $60 billion annually, while Australia is estimated to have only a $19 billion annual deficit.
In recognition of this growing need for new and improved infrastructure, as well as more efficiently managed infrastructure assets, there has been a growing movement for institutional investors to invest in assets characterized as "infrastructure."
At present, numerous investment managers are seeking to raise billions (in a multitude of currencies) from institutional investors who are wrestling with evaluating and explaining these new opportunities to their respective investment committees. The key questions on the minds of investment committee members are: what is infrastructure investing, what are the risk/return profiles, and where should infrastructure fit into a larger investment strategy?
Institutional investors will need to evaluate the multitude of different infrastructure investment options they are likely to see in the near future. Topics for discussion should include:
What constitutes infrastructure investments?
What are some commonalities between infrastructure and real estate?
What returns can one expect, and what is the expected return continuum?
How much risk is involved in infrastructure investments?
What investment vehicles are available to investors, and what fees can one expect to pay to garner an exposure to infrastructure?
INFRASTRUCTURE PRESENTS DIVERSE OPPORTUNITIES
No single definition can serve as an all-encompassing standard for what constitutes infrastructure investments, especially in light of the fact that infrastructure actually encompasses investments and assets from a number of industries or sectors of the economy. The table "Infrastructure Investment Sectors" on page 38 shows five segments: energy and power, water, transportation, communications, and social services.
Other ways of categorizing infrastructure investments consider a risk/return element. These include: greenfield versus brownfield projects, emerging markets versus developed markets projects, and regulated versus unregulated projects. In addition, the risk profile of projects changes as infrastructure assets mature.
Many infrastructure investments have been ones for which the social externalities were the primary driver for their undertaking. Many of these assets were publicly funded, directly or indirectly, and — consistent with the social benefit mindset — many were not revenue generating in a traditional fee-for-service model (the I-35 bridge for example). Subsidies or taxes supported their construction and maintenance, and their social value was seldom questioned.
INFRASTRUCTURE AND REAL ESTATE COMMONALITIES
Physical assets and capital intensity are common threads between infrastructure and real estate. Nowhere is that physical land-based analogy more evident than in the case of seaports, airports, toll roads and toll bridges. The capital-intensive nature of large-scale industrial real estate and office space, as well as the need for long-term tenants, is also closely akin to the capital-intensive nature of communications networks, power and fuel distribution systems, and assorted water distribution and waste treatment facilities.
The difference between the real estate model and the infrastructure asset has more to do with identifying specific long-term users/tenants and long-term user populations where, actuarially speaking, demand for the services will persist.
Another oft-cited similarity is the high cash flow nature of infrastructure assets. Many of the first privatized infrastructure assets did have very high cash yields and payouts, exceeding that of core real estate. Note, however, that some real estate strategies, such as value-added and opportunistic, do not offer significant early positive cash flow, and there again we can draw analogies to infrastructure.
In addition, greenfield projects can be likened to development real estate, with a subtle but very meaningful difference: Speculative building does not translate well into infrastructure asset investing. Most, if not all, greenfield infrastructure projects to date have been undertaken with end users effectively contracted to take the service when it was completed, and some of those have been partnerships with public entities. The contractual "take out" is more like build-to-suit real estate development than "spec" development.
THE RISK-RETURN SPECTRUM
The core to opportunistic risk/return profile of conventional real estate investing provides a useful framework to map infrastructure investments. Core real estate properties generally entail less leverage, lower vacancies and, therefore, more predictable cash flows and debt service. Opportunistic real estate investments frequently require major capital expenditure, repositioning and operational improvements, and they are more highly leveraged. As a result, opportunistic real estate investments require more active management and have more risk. Core-plus and value-added strategies fall in between. So it is with infrastructure investments.
Mature infrastructure investments (typically brownfield assets) can be compared to core real estate investments. Core infrastructure assets are often viewed as a buy-and-hold investment. These assets possess long-term contracts (up to 99 years), providing the basis for stable cash flow. There is often little operational improvement or additional leverage that can be added to amplify returns. Some investors regard these investments as a proxy for fixed-income investments and may place them within that allocation; a subset of these investors view infrastructure as an inflation hedge due to price inelasticity of the asset allowing for unit pricing to be increased ahead of inflation.
By contrast, development-stage infrastructure projects (often greenfield projects) require much more active management, analogous to a real estate development project or an opportunistic property investment. Active infrastructure management includes operational improvements, rate structuring and refinancing. Often the greenfield stage infrastructure investor shoulders risk similar to that of financing land development. The focus of the project is securing entitlements (contractual concessions) and concluding site preparation, thereby reducing risk before bringing in additional outside investors or selling the project outright. For shouldering this risk, investors earn opportunistic returns.
In between core and opportunistic lie core-plus and value-added strategies. According to Ross Israel at Queensland Investment Corp., "Core-plus investments are defined as existing infrastructure assets that seek capital for expansion [e.g., adding a lane to a toll road or terminal to an airport]. Also core-plus/opportunistic deals have an opportunity for asset turnaround through better management."
The leverage levels differ between core, core-plus and opportunistic infrastructure projects. Opportunistic greenfield development projects that have not been stabilized can support limited debt service, and thus have lower leverage levels (typically 20 percent to 40 percent LTV). Mature core infrastructure assets with stabilized cash flows can be leveraged up to 90 percent LTV between senior and mezzanine financing. Much like a highly leveraged, fully leased class A office building, most of the attributable value for the equity holder in the core infrastructure transaction is derived from attractive acquisition pricing and from the leverage.
RETURNS MAY VARY
Return parameters generally match the core, core-plus, value-added and opportunistic strategies in real estate. Core infrastructure can produce 9 percent to 11 percent gross IRRs at the project level. Core-plus or value-added infrastructure can produce 12 percent to 15 percent gross IRRs. Opportunistic infrastructure can produce 15 percent to 25 percent IRRs.
Of course, these are general parameters and vary considerably by vintage year and by project sector. For example, historically, some of the longer standing developed market brown-field funds have exceeded those return numbers, with long-term returns (during the past 10 years) above 15 percent per year. As one might expect, this sometimes has the effect of raising investor expectations. Investors should not assume this history will repeat itself, as the environment — rising P/E ratios and falling interest rates — has been favorable over the past decade, and the infrastructure market has become more efficient and competitive with additional capital and managers entering the space.
Little ex ante data exists for infrastructure investing in less developed economies. The early to mid- 1990s saw some investment activity in both Asia and Latin America, but significant economic downturns in those regions and/or currency crises negatively impacted those returns.
Targeted returns on funds currently in the market (focusing on developed country, brownfield assets) are centering in the high single digits to low teens range. Greenfield return targets (albeit from a small sample size) seem to average roughly 3 percent above brownfield return targets. The returns targeted by funds in the market addressing less developed countries exceed those of their developed brethren by roughly 3 percent per year, in both brownfield and greenfield segments (low teen returns for brownfield investments, mid- to high teens or low twenties for greenfield investments). As with greenfield data, sample sizes of less developed country data are small and should be judged accordingly.
VOLATILITY OF LISTED INFRASTRUCTURE
A possible surprise to the investor new to infrastructure investing is the risk or volatility in return associated with different investment vehicles. Some infrastructure funds are listed securities, while a greater percentage are unlisted. With listed funds, shares in the fund can be bought or sold in a secondary market (much like a listed REIT). Unlisted funds, often in the form of partnership interests for which there is no secondary market, yield liquidity only when the underlying portfolio holdings are sold, which is infrequently.
Intuitively, an investor might believe a listed security is less risky because liquidity can be more readily obtained. However, when examining volatility in annualized returns, a far different risk story emerges. Because these listed funds can be traded in the market at values that are not necessarily the stated NAV of the underlying securities, large premiums can be built into these share prices at times, similar to publicly listed REITs. While one can argue whether this market pricing correctly predicts the future prospects of a specific pool, what seems indisputable is the added volatility that is introduced in measured returns.
Historical evidence from the mid-1990s through 2005 indicates listed funds showed volatility factors almost three times that of similar unlisted funds (16 percent to ~6 percent). RREEF and UBS estimate that the 10-year performance of global and U.S.-only listed infrastructure securities has been 14.2 percent and 10.4 percent per year, respectively.
In a study tracking 19 unlisted Australian infrastructure funds from third quarter 1995 to second quarter 2006, conducted by Hsu Wen Peng and Graeme Newell from the University of Western Sydney using data from Mercer and other sources, the unlisted funds produced an average annual return of 14.1 percent with an annual volatility of 5.8 percent and a Sharpe Index of 1.47. In contrast, listed infrastructure funds tracked by Peng and Newell produced a 22.4 percent average annual return during the same period with a much higher volatility of 16 percent and corresponding lower Sharpe Index of 1.05.
The listed fund volatility is more in line with public equity markets generally, while the unlisted fund volatility is more in line with core real estate. Equally important to institutional investors is the impact this volatility increase has on such risk/return efficiency measures as Sharpe ratios. Unlisted funds from the period cited enjoyed Sharpe ratios of ~1.5, whereas listed funds saw their Sharpe ratios fall to just above 1.0. The magnitude of the difference is significant, given that the primary difference in the respective samples is the listing aspect. Regardless of the listing aspect, what seems clear is that infrastructure investing is not a highly volatile investment.
Investors also should understand the return and volatility distinction among and between various infrastructure investment types. According to Cynthia Steer with Rogers Casey, "There are very definite and meaningful [return] distinctions between asset types." Anecdotal evidence from Australian, European and Canadian markets suggests that this is true. The table titled "Infrastructure Risk-Return Profiles" on page 40 provides illustrative broad return parameters for different types of infrastructure assets based on anecdotal data points from market participants. (The table is illustrative only; actual returns in individual transactions depend on a variety of factors including risk reduction through contract or concession structure.)
RELATIVE VALUATIONS
The characteristics that draw investors to infrastructure include monopoly-like barriers to entry, a captive customer base, predictable earnings due to long-term contracts, high operating margins, low volatility of cash flows and inelastic demand enabling episodic price increases. From a portfolio construction standpoint, infrastructure's attractive features include the large investment scale, long duration, hybrid fixed-income and real estate profile, inflation hedging aspects, and low correlation with other investable assets. These features have not been lost on public pension plans with long-dated liabilities; infrastructure's long-dated assets and stable earnings have been an enticing characteristic.
More recently, Goldman Sachs has argued that infrastructure assets are undervalued relative to most real estate asset classes in the United States and Europe, perhaps due to a flood of capital into real estate in recent years. Goldman Sachs estimates suggest that infrastructure assets are valued 5x to 9x less on an enterprise value to EBITDA multiple basis.
According to Rogers Casey's Steer, whereas "core real estate cap rates globally are generally hovering in the 5 percent to 6 percent range, infrastructure cap rates are closer to the 10 percent to 14 percent range." Lower real estate cap rates are a function of the amount of capital that has poured into real estate from other asset classes. Over time, as more institutions embrace infrastructure and commit capital to the asset class, real estate and infrastructure valuations may begin to converge.
STRUCTURES AND FEES
For those institutional investment programs that have determined that infrastructure is a suitable investment, it becomes necessary to find an appropriate vehicle. As with real estate, larger programs have more options than smaller programs. Investment options include public securities, separate accounts, and open- and closed-end funds.
If investments deliver at the higher end of return expectations, then the gross-to-net fee differential of a "2 & 20" opportunity fee structure may make sense. However, according to Dragana Timotijevic from Mercer Consulting, "Core infrastructure returns with a private equity fee structure don't work for institutional investors." This sentiment was echoed by Rogers Casey's Steer, Queensland Investment's Israel and a representative from a large Midwestern public plan. All felt that, given the returns of corefocused strategies, core fund sponsors should evolve to an open-end fund/separate account fee model. As this occurs, there will be greater adoption of core infrastructure strategies in the marketplace among institutional investors. At present there are only a handful of institutional funds with this open-end structure.
INSTITUTIONAL INVESTOR ADOPTION
Investor adoption of infrastructure programs comes from two distinct directions. Outside of the United States, investors appear to have migrated into infrastructure from a fixed-income, liability-driven investing perspective. These groups seek out long-dated assets to match longdated liabilities. By contrast, U.S. institutions have viewed infrastructure from an equity-oriented mindset. Early adopters seem to be large public plans, though many consultants are receiving infrastructure investment inquiries from all types of institutions.
Australia has been an early adopter of infrastructure investing, with many of the superannuation programs incorporating infrastructure into their asset allocation models. Some Canadian programs also have been moving into infrastructure investing. In fact, some Canadian and Australian pension plans have infrastructure asset allocations of as much as 10 percent to 15 percent. Australia-based investment consultants Access Economics, Sovereign Investment Research and Mercer Investment Consulting have supported infrastructure recommendations to their superannuation clients. For example, Australia's Queensland Investment Corp. is a big believer in infrastructure. The team, headed by Ross Israel, has a mandate to do fund investing, co-investing and direct investing in infrastructure.
In Canada, the Canada Pension Plan Investment Board and Ontario Municipal Employees Retirement System were early adopters of infrastructure investing. In the United States, institutions that reportedly have incorporated or are actively considering incorporating infrastructure into their asset allocation include the Alaska Permanent Fund Corp., Baylor University, Illinois State Investment Board, New Jersey Investment Council, North Dakota State Investment Board, Missouri State Employees' Retirement System and San Bernardino County (Calif.) Employees Retirement Association.
Many institutional investors, however, are taking a wait-and-see attitude. One large Midwestern public plan that is active in private equity and real estate has not moved on infrastructure. "We won't rule it out, but we are concerned that the returns may not pan out and that there is a little too much 'flavor-of-the- month' with infrastructure at present," says an investment officer with the pension plan.
Still, infrastructure is winning many converts, some from nontraditional sources. One industry consultant reported that his firm was receiving inquiries about infrastructure "from programs with no private equity exposure at all. The appeal of infrastructure appears to be quite broad, possibly broader than private equity, but not as broad as real estate."
As infrastructure investing is gaining acceptance in institutional portfolios, investors are displaying a willingness to expand outside of markets such as Australia, the United States, and Europe, and outside of metropolitan areas within those developed markets.
The Alaska Permanent Fund is a recent convert to infrastructure, recently creating a 2 percent allocation to infrastructure and making commitments to two funds. According to David Stuart, formerly of the Alaska Permanent Fund and now at TIAA-CREF, "Alaska pursued a 'barbell strategy' with the dual recommendation of the two funds. The first has more of a core focus in development markets, whereas the latter has more of an emerging market, greenfield strategy."
FINDING THE RIGHT FIT
Many believe over time more specialization in offerings will occur. Some believe that larger plans will migrate toward a multi-manager separate account model because of specific investor requirements. For example, some programs have heightened sensitivity to specific risk types (e.g., labor risk), and certain infrastructure assets (e.g., airports) have modest exposure to these specific types of risk. Given this, some investors may seek custom-tailored separate account solutions that will enable infrastructure exposure generally while emphasizing assets without these types of risk. At the same time, endowments and foundations that have an equity risk, illiquidity premium mindset may seek out smaller, niche managers in specific regions to optimize their portfolio.
At the core end, infrastructure investing does have some qualities of real estate, but it is a distinct investment. Evidence suggests that infrastructure is more correlated with inflation than core real estate, giving rise to the belief that core infrastructure investing may be a stronger inflation hedge. Equally important, core real estate has an established going-in cap rate with immediate yield. Infrastructure investing, by contrast, takes longer to deploy because assets are harder to procure and produce yield; therefore, there is a distinct "J curve" element to infrastructure investing, which is not present in core real estate investing. At the same time, the asset life is long, as much as 15 to 99 years in some cases.
A feature of these long-dated assets is their underlying contract concessions; in many cases these stipulate rate increases that translate into cash-flow/equity-distribution kickers. The contractual nature of these increases provides for a high degree of visibility into the timing of the stepped up cash flows/distributions, which is not present with core real estate. However, these concessions often run longer than the life of the typical closed-end fund, which creates a possible exit dilemma at the termination of a fund: Will there be a ready market into which the sponsor can sell the asset in order to liquidate the fund and distribute proceeds to the limited partners?
Infrastructure investing is not appropriate for every institutional investment program. Those charged with generating high absolute returns may find the core aspects of infrastructure unattractive and may feel that the development risk of greenfield projects is inappropriate for their program. Others with shorter liability time horizons may not take comfort in the extended J curve, requiring a more timely return of principle investment capital. Still others may find the long-term asset management fees by fund sponsors to be an overall drag on otherwise attractive asset performance, especially for core fund performance. The data on performance remains thin, and there have been few true vintages of infrastructure funds to benchmark, so for investment committees needing to see longer data sets, the timing may be a bit early.
As institutional investors continue to ponder the asset class, the demand for infrastructure construction and repair continues to climb. The Minneapolis bridge replacement project has been fast-tracked and is scheduled to be rebuilt in two years or less. As more investment opportunities present themselves close to home, more institutional investors will begin actively evaluating whether infrastructure investing is right for them.
Bruce Pflaum, based in San Francisco, and Bob Moreland, based in Chicago, are fund placement agents at BerchWood Partners.