Version classiqueVersion mobile

The Infrastructure Finance Challenge

Ingo Walter

15. Accelerating Infrastructure Finance

Texte intégral

1This study is intended as an up-to-date discussion of the global issue of infrastructure finance, starting with the fundamentals of infrastructure development and ending with concrete questions about tapping global pools of investable funds. We take as given the enormous financial requirements presented by the infrastructure sector in the years ahead, as well as the equally enormous pools of investable funds in perennial search of optimal portfolio allocation.

2The core question is how to better connect the requirements and the resources in a mutually advantageous way — the disciplined identification of viable and sustainable infrastructure initiatives on the one hand, and the equally disciplined risk, liquidity, and return attributes on the other. We suggest that this is decisively a positive-sum game, with impressive gains for both sides and, in most cases, equally impressive spillovers for society more generally.

15.1 Recap

3We began this study by defining the boundaries for the kinds of capital investments properly termed “infrastructure”. We continued with the attributes of infrastructure activities that distinguish them from other economic activities. These features include scale, capital-intensity, longevity, and generation of public benefits that can be difficult to value and internalize for purposes of creating viable financial structures.

4We next outlined key complexities in a section on the legal dimensions of infrastructure projects, including governing law and the use of eminent domain.

5All of these elements add to the underlying complexity of infrastructure finance. Each needs to be documented, priced, stress-tested across contingencies, and built into financial contracts that become the basis of financial instruments that fit investor portfolios.

6Perhaps even more problematic are the political, environmental, and social issues related to infrastructure development, whether or not anchored in legal documentation. These attributes and the associated political risks accompany all infrastructure projects, affect their cost and revenue characteristics, and sometimes block them altogether.

7Infrastructure finance investors try to address such issues by creating risk parameters reflected in their investment decision criteria as well as in the contractual design of infrastructure projects.

8Following the generic framework of infrastructure projects, we went on to consider the global infrastructure industry’s industrial organization — its structure, conduct, and performance — to demonstrate the importance of competition in the infrastructure-development industry, including the role of concessionary finance.

9Measuring the depth of globalization in this industry sector is complicated because of differences in how infrastructure is defined, different modes of participation (e.g., project financing, equipment supply, design, construction, and operation), and the fact that large projects often involve consortia with participants from diverse countries. The role of Chinese firms in the global infrastructure industry deserved special mention in this discussion.

10We turned next to the development, character, complexities, and risks of infrastructure finance — one of the most challenging activities in all of global finance.

11On the project side, we discussed the economic viability essentials for investors to review that incorporate a defensible assessment of risk attributes — both project risks themselves and the risk mitigation structures that surround them. We discussed project life cycles and the role of various forms of financing, from bank lending and sponsor equity to long-term fixed-income securities. On the investor side, we discussed how project finance fits into institutional portfolios managed by pension funds, insurance companies, hedge funds, and other large asset pools in terms of returns, risks, and liquidity.

12Liquidity was considered separately as a key issue, particularly with respect to gaps in current market microstructures that do not perform well in providing liquidity for the long maturities involved in infrastructure finance. Financial innovation may in the future create infrastructure-related obligations with improved liquidity properties, which could make institutional investors much more receptive to infrastructure bonds.

13As we emphasized in our empirical discussion of returns and risks associated with infrastructure finance as an asset class, the picture for investors is encouraging. This was true even during the extreme financial turbulence of the global financial crisis. We examined whether infrastructure-linked securities are more like bonds or more like stocks in the asset allocation of major institutional investors such as pension funds and life insurance companies — asset managers that seek to combine global stocks, bonds, real estate, and infrastructure in what they view as optimal portfolios.

14We then calibrated performance of the infrastructure asset classes from 2003 through 2015. We found that infrastructure financing is highly correlated with stocks and bonds but exhibits much higher returns than the relevant combination of stocks and bonds alone. These results suggest that infrastructure finance can indeed form a robust asset class in institutional investors’ portfolios worldwide. So what is next?

15.2 There Is No Alternative to Robust Debt Markets

15Project finance will inevitably be central to meeting the world’s infrastructure needs in both developing and developed countries, given the very long-term nature of infrastructure assets and the multiplicity of risks involved. We have emphasized in this study that project finance allows sponsors — whether corporations, governments, or financial sponsors such as private equity funds — to maximize the amount of debt in infrastructure initiatives’ capital structure, typically up to 70% to 80%. So the dynamics of the global debt markets are central to the whole discussion.

  • 1 [n.a.], “A Long and Winding Road”. The Economist. 22 March, 2014.

16Institutional investment funds in the pension and insurance sectors alone are estimated to have managed more than $50 trillion in 2015, but only 0.8% of these assets were invested in infrastructure projects worldwide.1 Given infrastructure debt’s built-in inflation protection, stable returns, high recovery rates, and low correlations with other asset classes, the global market for it should be much larger among the world’s growing long-term investment pools than it is today. Whereas the construction phase, particularly in greenfield projects, will likely continue to rely on banking expertise and bank lending as the form of finance, it seems logical that securitized bank debt and bonds covering the operating phase of infrastructure projects should be well suited for the key institutional investors.

17For a high-leverage approach to work in this context, the debt emanating from infrastructure projects must be available in ample supply, and the average debt cost must be kept low — in today’s market, usually in the mid-single-digit coupon range. Low debt costs imply the need for investment-grade credit ratings for most project finance debt.

15.3 Banking pressures

18The first impediment to improved bridging of sources and uses of infrastructure debt comes from commercial banks. As noted, these banks dominate the project finance debt market and are particularly important in the pre-completion phase. Overall, commercial banks provide roughly 90% to 95% of total project debt in any given year.

19Because commercial banks earn significant up-front fees for structuring and advising on projects, they are often willing to lend at fairly thin margins, sometimes in the low single-digit (2% to 3% yield) range. Attractive debt cost is compelling for borrowers, as is the ability to negotiate with a small group of relationship lenders, particularly during the critical project construction period. Consequently, most project and infrastructure finance sponsors have little incentive to refinance bank debt with project bonds, even when a project begins generating cash flow, unless the lending banks face balance sheet capacity issues.

20The progressive implementation of the Basel III banking regulations under the auspices of the Bank for International Settlements (BIS) affected the conditions in the project finance debt market. The thinking was that the new minimum capital and liquidity ratios would prevent commercial banks (particularly European banks) from holding illiquid long-term project finance loans on their balance sheets. Indeed, almost 10 years after the financial crisis, key European banks continue to wrestle with capital adequacy resulting from their failure to raise capital soon after the financial turbulence. So these banks presumably would be sensitive today to holding risky long-term assets on their balance sheets and less willing to finance up to 90% of project debt, forcing sponsors to inject more equity.

21Perhaps surprisingly, the European project finance banks seem to have adjusted to the Basel III regulations fairly smoothly and without materially throttling infrastructure project lending.

22Despite some reallocation of market share among the major project lending banks — notably, the Japanese banks have become more aggressive — there has been no noticeable decline in overall new project lending volumes, no forced secondary sales of on-balance-sheet loans, and no meaningful increase in project bond refinancings during the past few years. Moreover, because project loans have relatively high average credit quality, low default rates, and high recovery rates compared with loans in most other sectors, project loans arguably remain a better use of bank balance sheet capacity than some alternatives.

23At the same time, the investment-grade bond market is very well suited to project and infrastructure financing.

  • The market is very large ($5.4 trillion outstanding at market prices as of December 31, 2015).2

  • It is long-dated (10.4 years average maturity) and global in nature.

  • It is well versed in analyzing cross-border sovereign risk.

  • It is low cost (3.00% to 4.00% average yield-to-worst for all issues).

  • It is dominated by institutional investors such as life insurance companies and pension funds, which typically prefer long-dated bonds to match their long-term liabilities.

24In short, the global high-grade bond market seems to be an ideal arena to generate the large amounts of debt capital required to finance projected infrastructure needs. So far, however, it has been significantly underutilized.

25The investment-grade bond market appears optimal for most project finance bonds, particularly in a continued environment of extremely low interest rates and scarcity of new public bond issuance following a period of heavy corporate refinancing. However, the market has issued only $25 billion to $50 billion per year from 2003 through 2015. The run-rate of investment grade project and infrastructure bonds appears minimal, and it should be possible to materially expand it.

15.4 Infrastructure Finance Expertise

26A second infrastructure finance bottleneck lies with the specific skill-sets involved in analyzing and pricing project finance bonds. The complex financial and nonfinancial risks discussed earlier, the document-intensive nature of project finance structures, and the variety of legal contracts and bond covenants built into most such financings demand a high level of analytical skill.

27The analysis required is more typical of the high-yield bond market than of the investment-grade bond arena. Nevertheless, a small community of investment-grade bond investors has developed expertise in this area. Most project finance bond offerings to date have been bought by a handful of large insurance companies with private placement groups, such as MetLife, John Hancock, Travelers, and AIG in the US.

28A number of global asset management firms including BlackRock, JPMorgan Asset Management, and Allianz SE have built investment teams with project finance expertise and raised funds to invest in project and infrastructure loans and bonds. The bulk of these funds come from second-tier and smaller insurance companies as well as pension funds that lack the in-house capacity to analyze such debt.

29Moreover, some of the aforementioned insurance companies have also raised third-party funds to invest in project debt, marketed on the back of their private placement franchises.

30Such capital-raising moves were geared to take advantage of expected changes in the project finance bank debt market resulting from the application of Basel III rules. As noted earlier, these changes did not play out quite as expected. The end result is that much of this newly raised capital has not yet been deployed because of a worldwide scarcity of viable project and infrastructure finance deals, while pricing on closed project debt financings has tightened in response to increased competition.

15.5 Liquidity issues

31The third major obstacle to project and infrastructure financing growth through the global investment-grade bond markets is liquidity.

32Most project finance bonds issued to date have been in traditional private placement format, with limited secondary market trading once a new issue is placed with institutional investors. Typically, most insurance companies would not opt to sell a project finance bond unless a credit event resulted in a downgrade of the issuer below investment grade.

33Compared to classic private placements, using the more-liquid US Rule 144A private placement format, which allows reselling bonds to “qualified institutional buyers” and opens project and infrastructure finance bonds to a wider universe of potential buyers, assuming the investing institutions have the requisite analytical skills.

34So far, most 144A project finance issues have been dominated by traditional private placement life insurance-industry investors. If the volume of 144A project finance bonds were to increase significantly, it is safe to assume that other institutional investors would opt to build their in-house capacity to research and trade these securities as opposed to outsourcing this function for a management fee — this would effectively help disintermediate the aforementioned asset management firms.

  • 3 Including 144A bonds as well as Section 4 (a)(2) and Regulation D private placements.
  • 4 The US National Association of Securities Dealers (NASD) introduced TRACE (Trade Reporting and Comp (...)

35Improved secondary market liquidity would also help facilitate broader trading interest in investment-grade project and infrastructure finance bonds. At present, only three or four Wall Street firms actively trade project finance debt,3 although the number of market makers arguably will increase in line with greater new-issuance volumes. Increasing secondary-market price transparency would also help facilitate more of a two-way aftermarket.4

36Another step to expand the universe of project finance bonds is accessing the upper end of the US high-yield (“junk”) bond market, although this market is limited in tenor (7 to 10 years) and provincial in its credit focus (only domestic issuers, including US independent power producers and liquefied natural gas project developers). Moreover, in a rising interest rate environment, the economics of issuing high-yield project finance bonds, even those rated BB, may be problematic.

37Lastly, discussions around developing a liquid, index-type project finance bond structure using CDO-style architecture have not yet led anywhere. Nonetheless, the concept of using a portfolio approach to project finance bond issuance clearly has merit, because it would diversify overall cash flows and improve market receptivity for infrastructure-related bonds.

38We have noted the chronic lack of infrastructure finance data in terms of correctly identified statistics, investment patterns, debt ratings, debt and equity performance, liquidity, and other attributes that help to define markets and reduce the risk of engaging with them. A 2016 joint initiative on the part of the Monetary Authority of Singapore (MAS), the French business school EDHEC, and the French investment bank Natixis aimed to launch a $14 million research unit staffed by 10 economists and statisticians mandated to create a comprehensive infrastructure financing dataset by 2021. This dataset could then be used for benchmarking infrastructure financings around the world, improve institutional investors’ understanding of the asset class, and improve the management of risks.

39This initiative’s ultimate objective is creating tradable products such as a new generation of infrastructure-based collateralized debt obligations (CDOs). Unlike the last wave of subprime mortgage-backed CDOs in the mid-2000s, these assets would be transparent, tradable, and fully understood by institutional investors. This venture could put Singapore at the center of the action in terms of origination and trading of a new global asset class with acceptable liquidity properties.

40The first benchmark, focused only on OECD infrastructure financing, was due to be published at the end of 2016. Without trading data, the initiative’s success will depend entirely on cooperation from institutional investors, banks, and other investors. The dataset reportedly contains 500 infrastructure debt and equity issues over 25 years, with plans to eventually expand to 2,000 entities.5

15.6 Lack of Viable Infrastructure Projects

41So far, all of the impediments to increasing the flow of infrastructure debt capital from the investment-grade bond market have not presented a serious blockage to market expansion because of the dearth of viable projects in need of financing. Moreover, a wave of refinancing activity since the 2007–2008 global financial crisis has further tempered any financial supply issues because interest rates have plummeted. From 2010 through 2015, about one-fourth of total new project finance debt issuance involved refinancing of existing on-line projects.

42Already-operating or brownfield projects located in OECD countries with strong equity sponsorship (especially projects that are government-related) and long-term offtake contracts with high-quality counterparties currently have no trouble finding financing. In fact, most such deals are oversubscribed.

  • 6 An interesting exception is a bond issue for the 2015 $522 million “Rutas de Lima” toll road infras (...)

43At the other end of the spectrum, projects that struggle to find economically-priced debt and are effectively closed out of the global financial markets are located mainly in lower-rated, non-investment-grade countries where sovereign risk looms large, predominantly in Latin America and Africa. This situation is highly unfortunate because the properly executed and financed development of power, water, and telecommunications infrastructure could dramatically accelerate economic growth and living standards in such countries.6

44For perspective, based on the Standard & Poor’s rated sovereign universe as of December 2015, roughly two thirds of the 193 UN member countries were rated below investment grade or not rated at all (and thus implicitly non-investment grade). Approximately 40% of the world’s current population lives in these 124 countries, which in the aggregate generated only an estimated one quarter of world GDP in 2015.

45No amount of bond market innovation or contractual/legal structuring, however, can effectively mitigate or compensate for the poor governance, rampant corruption, and expropriation risk that afflicts such countries. Even off-shore cash lock-boxes and international arbitration panels provide little help, because the underlying weaknesses are mainly political and social in nature. Investor memories of the Argentine and Venezuelan project bond defaults some 10 to 15 years ago have yet to fade. At worst, poorly conceived, governed, and executed infrastructure projects that fail to generate the economic capacity to service the contractual financing can leave countries and their citizens worse off even if the financing providers are made whole.

46Much more subtle are local rules governing infrastructure investments that effectively discriminate against foreign investments which compete with local interests.

47A good example is India’s Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest Act of 2002 (Sarfaesi), which provides access to non-performing secured assets without the intervention of Indian courts by following a streamlined alternative procedure. Given India’s inefficient legal system, Sarfaesi was designed to encourage investments, including in infrastructure, by cutting legal risk in the event of default. Foreign creditors, however, are not classed as secured lenders under the Act and, unlike their Indian competitors and co-financiers, could not avail themselves of Sarfaesi’s streamlined provisions (notwithstanding pari passu and inter-creditor agreements). Under such circumstances, foreign lenders end up at the mercy of lengthy and uncertain Indian court proceedings.

48Such rules, which seem clearly protectionist and provide a potentially decisive advantage to domestic lenders, are unlikely to promote the Indian market for project investment. Given India’s size and importance, this regulation could extract a high cost in terms of the country’s future infrastructure development.


1 [n.a.], “A Long and Winding Road”. The Economist. 22 March, 2014.

2 Barclays Bank at

3 Including 144A bonds as well as Section 4 (a)(2) and Regulation D private placements.

4 The US National Association of Securities Dealers (NASD) introduced TRACE (Trade Reporting and Compliance Engine) in July 2002 in an effort to increase price transparency in the US corporate debt market. The system captures and disseminates consolidated information on secondary market transactions in publicly traded TRACE-eligible securities (investment grade, high yield, and convertible corporate debt), representing all over-the-counter market activity in these bonds. See

5 investment-database

6 An interesting exception is a bond issue for the 2015 $522 million “Rutas de Lima” toll road infrastructure project in Peru. This example shows that a well-structured PPP initiative in a developing country with impressive public policy credentials can open up new, cost-effective channels of financing. For background, see


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search