Infrastructure: from French (1875), a collective term for the subordinate parts of an undertaking; substructure, foundation.
Oxford English Dictionary
As the etymology of the word suggests, infrastructure is the sub-structure of an undertaking on which other structures, systems, and activities are built. Its foundational feature lends it three essential characteristics: it is a public good at both local and national levels; it spawns a large number of diverse activities; and it is of long duration and embedded in long-term economic expansion.1
All major innovations in infrastructure, such as the introduction of electricity, gas, Internet, telephones, water, sewerage, public health, and education, display these characteristics. They have laid the foundations for economic activity around the world in every sector in most places over several centuries. But 140 years after the word was first coined, we still do not know how to provide and manage infrastructure. There is a chronic under-provision of it around the world, and where it is available it is often poorly delivered. The most advanced country in the world—the United States—is repeatedly criticized for its weak infrastructure. The European Union has launched a major infrastructure programme of €300 billion—the Juncker plan—to rectify deficiencies in European infrastructure. One of the fastest-growing economies in the world—India—has failed to provide the massive infrastructure programme the country desperately needs. The repeated inability to implement infrastructure programmes in Africa is one of the continent’s main impediments to economic growth. And China’s performance is in many respects no better than that of developed countries.2
The failure of infrastructure is not for the most part one of intention but of delivery and achievement. Despite the expenditures, UK and US infrastructures remain poor. Africa relies on China to builds its infrastructure. And the history of infrastructure in the European Union is one of repeated wasteful expenditure.
Why is it so difficult to get infrastructure right and what can be done to rectify the problems? The reasons are not hard to find: a failure to recognize the systems nature of infrastructure; its scale; its duration; incorrect accounting for infrastructure and consequent failures to fund it appropriately; and inadequate governance and management of infrastructure programmes. Together these have resulted in a failure to identify the purpose and benefits of infrastructure investments, to measure their impact on public and national accounts, to fund them appropriately, and to manage and implement them efficiently and effectively.
Until these are resolved, infrastructure will continue to display the same characteristics that it has to date, namely inadequate, wasteful, over-budget, over-time, and frequently corrupt expenditures. So long as this is the case then cynicism about government’s ability to manage public infrastructure budgets will prevail and excessive reliance will be placed on poorly specified methods of delivering the infrastructure programmes that the world desperately needs.
There are corresponding solutions. The first is to provide a proper framework for identifying what infrastructure is required and the costs and benefits associated with providing it in different ways. Though conventional cost–benefit analysis has a role to play, it fails to reflect and measure the full systems nature of infrastructure—its significance as a substructure of something much bigger and broader. Systems have to be conceived, designed, evaluated, and implemented not as increments considered in isolation but as integrated wholes.
Second, infrastructure needs to be financed. Whilst the private sector is familiar with provisioning for capital maintenance and enhancements, and has balance sheets to account for the state of the assets under their control, governments are driven by cash-based national accounts. Indeed governments typically have no accounting framework for addressing the creation of assets and liabilities for future generations, and no mechanism for closing the gap between future benefits and current liabilities.
Third, there needs to be a governance system that provides for political decision-making, without succumbing to corruption, short-termism, and what might be called ‘the trophy project syndrome’. The cynicism surrounding infrastructure expenditure is justified by its track record of failure. But if we can send spacecraft to Mars and Venus with stunning success, surely we can build roads, bridges, and electricity systems closer to home. An institutional, regulatory, and governance framework is essential for this and required to prevent the problems that afflict infrastructure.
Infrastructure investment is in general thought of as filling gaps in an economy. There is a deficiency in the provision of particular infrastructure in a specific location that needs to be fixed at minimum cost to the public accounts. It starts from the stock of existing assets and adds on bits here and there as tentacles entangling an ever-expanding economy and environment.
There are many advantages to this pragmatic approach, including avoidance of the converse risk of embracing large unworkable, unmanageable, and unaffordable programmes. However, it can easily descend into crisis management—waiting until the deficiency is so acute as to make new investment essential or in general long overdue—as in the case of US highways, European gas supplies, and UK airports—and systems have deteriorated to decrepit states.
An economic approach should start from a more considered position of what is the desired and needed infrastructure system. What type of airport system, energy transmission and distribution networks, broadband, water supply, and rail and road network does an economy need over the next ten, twenty, and fifty years? These systems are public goods, and are almost always underprovided by markets, except during periods of ‘irrational exuberance’ as, for example, in the early railway mania in mid-nineteenth-century Britain and the dotcom mania across the developed world in the late 1990s. It is not a question of whether governments have a role to play but what it should be.
Technology and economies change. There have been some infrastructure networks that have remained roughly the same over long periods. The water and sewerage systems in many major urban areas—for example in London, Paris, and Rome—were put in place in the nineteenth century, and many of the assets still function more than a century later. The canals were built in the eighteenth century, the railway networks in the nineteenth century, and some roads have been in existence for a millennium. Other networks are more recent. Broadband is a new infrastructure less than a couple of decades old.
In some systems the speed of technical change creates a paralysis of decision-making, but a notable feature of current programmes is that many are very long term. New nuclear power stations are planned with a life of sixty years; rail programmes and new airport capacity are likely to be in use in the twenty-second century.
There are therefore three key features of infrastructure programmes—their scale, their breadth, and their duration. The magnitude of expenditures dwarf those that single private entities on their own can finance or manage. The programmes, if not universal, encompass at least a broad segment of a country’s population in the provision of a diverse range of services. And the relevant timescale of the programmes exceeds that of conventional private-sector investment programmes and capital investment appraisals.
All of these point to the problems involved in relying on private capital markets to elicit the right scale or form of infrastructure provision without public-sector coordination and planning. But they are also the source of public-sector failure. The conventional tool with which to evaluate infrastructure programmes is cost–benefit analysis. The total social benefits of infrastructure are evaluated, set against the total social costs, and discounted back to the present at a social discount rate. If the resulting number is positive then the programme progresses, and, if there is a funding restriction, those projects with the highest positive net present value per unit of expenditure are chosen until the available budget is exhausted.
There are numerous problems with this. The first and most obvious is that given the timescale of the programmes all three components of a cost–benefit analysis—costs, benefits, and discounting—are subject to substantial uncertainties. Second, the objective of large infrastructure programmes is to change the nature of the economy in terms of both its overall activity and spatial distribution across people, goods, and services, whereas cost–benefit analysis is a partial analysis that takes all other components of the infrastructure and the economy as given. The appropriate question is therefore not so much whether a particular piece of infrastructure should be built but rather what infrastructure system does an economy need to underpin its long-term development. Thus the specific investment decision should be viewed in a wider context as part of a broader set of decisions, combining the infrastructure for the core services of water, sewerage, rail, road, air transport, broadband, and energy.
Recognizing the system characteristics and the interrelationships between systems provides the basic architecture of infrastructure policy. But it is not just the scale and breadth of infrastructure programmes that mark them out from other investments. It is also their duration. They span not just one but multiple generations and have to reflect the interests of the unborn as well as the current voting population.
When the Victorians built the London sewers, it is unlikely that they had in mind the benefits Londoners a century and a half later would reap. They were concerned with the ‘Great Stink’ in London, and the very immediate needs to clean up the mess. The project was economic in the shorter term, without worrying about the long run, but there are many projects where time horizons make a considerable difference.
The combination of the interconnected-systems nature of infrastructure, the vast scale of required expenditures, and their long duration combine to make underinvestment a ubiquitous problem that afflicts the developing world even more than the developed. Together they render infrastructure particularly prone to the ‘market failures’ that make the private sector unable to provide it unaided by government. In particular, in the absence of public-sector funding, the private sector is prone to discount the future benefits of infrastructure more rapidly than society as a whole would wish.
However, the framework of this book provides a more insightful view of the problem and its potential solution. It points to the importance of commitment to the ability of both the private and public sectors to contribute to the provision of a country’s infrastructure and the particular difficulties that both parties face in committing to programmes of the breadth, scale, and duration of infrastructure systems. It suggests a solution by emphasizing the similarities between private and public sectors rather than their differences. The conventional dichotomy between the state and the firm sees the same type of division between public rule-makers and private profit-makers in infrastructure as prevails elsewhere. This has erected damaging barriers between the two that have undermined the effective functioning of either. The removal of these barriers provides the means to address the global chronic underinvestment in infrastructure.
The starting point is the observation that there is no proper accounting for infrastructure. Chapter 6 recorded how we fail to account properly for natural assets. Natural capital is a form of infrastructure providing a lifeline for the preservation of our existence, but it is just one of a broader form of infrastructure for which we fail to account adequately.
National accounts are statements of current incomes and expenditures, originally designed to facilitate the management of the macroeconomy. Governments record liabilities as part of public-sector deficits but only keep incomplete records of their public assets. This has several effects. First, it creates an impression of burgeoning deficits and profligacy when governments engage in investments for the benefit of future generations. Second, as in the case of natural capital, it means that there is an incomplete record of the expenditures required to maintain infrastructure assets and, as a consequence, a natural tendency, as has occurred in many developed countries, to allow them to deteriorate.
Third, since the private sector does have balance sheets, when infrastructure assets are transferred from the public to the private sector, there is a spurious increase in measured assets. The private sector pays for the assets that were not recorded on government balance sheets thereby augmenting the public-sector accounts and records the cost of purchasing the asset on its own balance sheet to offset the expenditure, thereby resulting in no deterioration in its accounts. There is therefore an apparent magical creation of economic value from the transfer of ownership, thereby providing an impetus to the programmes of privatization of utilities and infrastructure that have swept around the world.
The allocation of infrastructure between public and private sector should reflect the relative merits of the two parties in defining, implementing, and managing their assets, not arbitrary distinctions in their accounting conventions. The absence of proper public-sector accounting has created a distortion in the ownership and operation of infrastructure. But failure of accounting is just one manifestation of a much larger public-sector deficiency.
The systems nature of infrastructure makes the design of infrastructure inherently a public policy matter. The size, shape, and form of a country’s rail, road, electricity, water, and telecommunications systems are naturally matters of public interest. The specification of the systems is therefore the role of governments, albeit assisted and advised by private-sector parties.
Since they set plans, governments also revise and update them. This makes private-sector providers inherently exposed to changes in policies and reallocation of resources. For example, the success of a particular toll road is dependent on connecting roads and competing motorways. It is undermined by the building of neighbouring high-speed railways or improved air transport.
To the extent that the prices of the provision of infrastructure and utility services are regulated then private-sector companies are exposed to changes in the terms on which services can be offered. A tightening of the cap on the prices that can be charged diminishes the returns on the investments that have been made.
Some of these changes may be benign and undertaken for sound economic reasons but there is equally a political incentive to engage in exploitation of private-sector investments. Expenditures on infrastructure are particularly exposed because of their ‘sunk’ nature. They are not readily recoverable or transferable to other locations and so, once made, they are vulnerable to systematic redistribution of benefits from companies to users, and governments may feel strong political imperatives to undertake such redistributions.
Political risks are particularly great in developing-country contexts, and the prevalence of corruption makes private infrastructure providers subject to both costly and illegal activities. The consequence is underinvestment and the avoidance of locations where political risks are especially significant.
Correcting these failures requires the identification of public-sector commitment devices—ways in which governments can commit to abstaining from engaging in opportunistic changes in policy and exploitation of private-sector providers. International enforcement is one such commitment mechanism through, for example, multilateral insurance agencies like the Multilateral Investment Guarantee Agency (MIGA). These agencies offer private-sector providers insurance against political risk backed up by the threat of withdrawal of support by international organizations such as the World Bank in the event of political violations. Where developing countries uphold the interests of their investors then they can obtain larger amounts of funding on more favourable terms with the insurance that MIGA offers than those countries that renege and are excluded from its assistance.
Such commitment devices can be powerful deterrents to political interference. However, there are alternatives. Where infrastructure programmes are organized by public-sector corporations then the potential exists for joint ownership and control of such organizations. Instead of granting governments exclusive control over the design and operation of infrastructure, they can be organized as collaborative arrangements between the public and private sector.
In particular, coming back to the importance of nomination committees discussed in Chapter 5, large-scale private-sector providers as well as governments can be involved in the process of selecting and removing the directors of public infrastructure companies. They can both delegate decisions regarding the nature and funding of infrastructure programmes to jointly nominated directors. Viewed from the perspective of the private sector, this provides a way of committing the public sector but equally it addresses the private-sector commitment problem that the public sector legitimately perceives.
Problems of commitment are at least as pervasive in the private as the public sector. Privatization has been successful at promoting operational efficiencies but not capital expenditures. Private sector providers seek to deliver as little as possible at as low a cost as they can over as long a period as feasible. Competitive tendering is used to elicit the best value-for-money construction and delivery of infrastructure projects and services, and contracts are specified as precisely as feasible to avoid gaming. But contracts are at best incomplete and prone to abuse when they involve the provision of subjective qualities of service over long periods of time.
The problem is that the interests of government and regulators, on the one hand, and private-sector companies, on the other, are in direct conflict. Governments and regulators are (at least in principle though unfortunately not nearly always in practice) concerned with the public interest. Private companies are interested in maximizing their returns. Governments and regulators want maximum quality at lowest prices for the largest number of, in particular disadvantaged, consumers. Companies want the highest revenues from the provision of the lowest-cost projects and services. Even if competitive tendering, well-designed contracts, oversight by international agencies, and independent public bodies are put in place, they do not resolve fundamental problems of commitment between private and public sectors.
The current institutional solutions to the inherent conflict between the private and public sector in the delivery of infrastructure services are becoming increasingly fragile. And that is before we have even begun to think about our potentially most important infrastructure asset and that is our natural capital and environment. We simply cannot afford to retain a system that is degrading both our material and natural infrastructure.
There is a straightforward resolution to the governance as well as the measurement problem. Regulated utilities that provide infrastructure services operate under a licence to supply. However, those very same companies’ terms of employment of the investments under the corporate laws that define their existence state something very different. The licences specify the amounts, qualities, and duration of the services they provide; corporate and company laws relate to the financial interests of their investors. The focus of infrastructure licences is on quality and value for money; the fiduciary responsibilities of directors are to maximize the returns on their companies’ investments.
It is no wonder that the executives of such organizations experience a degree of schizophrenia in their professional lives. It is as if they were telling their children on the one hand to be good, honest, and upright citizens and on the other to do whatever it takes to get rich quick. It is both unsustainable and unnecessary.
This problem can be readily resolved by making the licence condition of their infrastructure operations part of the charters or articles of association of their companies, or at least the subsidiaries that provide the relevant services. That does not extinguish their obligations to their shareholders but it puts their licence conditions to their infrastructure companies on an equal footing and it aligns the private purposes of infrastructure providers with the public purposes of governments and regulators.
That is precisely how infrastructure companies were for centuries structured across the world. The railroads and canals were built by public companies under royal charters or licences issued by acts of parliament. The public obligations were a pre-requisite to the granting of licences to operate. We lost that association with freedom of incorporation in the nineteenth century, and, while liberalization may or may not have been appropriate for private companies that were not supplying public goods, it is most certainly not right for the provision of infrastructure.
Incorporation of public licences in private charters and articles of association eliminates the conflict and contractual abuse that is endemic in private–public partnerships, private finance initiatives, and privatizations. It goes beyond what was described as being needed in private companies in general, namely the incorporation of their purposes in their articles of association and a demonstration of how their corporate structures assist with the fulfilment of them, to a requirement that the stated purposes of regulated companies include their licence conditions to operate. It makes the private provision of public-sector services a natural function of the private sector and it extinguishes the divide that exists between the two sectors. It does so by converting the conflict that is intrinsic to regulated corporations into cooperation, and avoids having to trade off the inefficiencies of the public sector against the social injustices of the private sector.
Where the inclusion of such licence conditions in corporate charters and articles of association is permitted under existing corporate law then this should be a pre-requisite for the granting of an infrastructure licence. Otherwise, a legal form analogous to the Benefit Corporation in the United States, which explicitly allows companies to specify a public purpose beyond their commercial interests, should be adopted with the licence condition being the stated public purpose. In both cases, the role of the regulator then becomes essentially a governance one of ensuring that the company abides by its public duties under its charter and publishes relevant measures of performance that demonstrate this. Conversely public–private corporations can result from the injection of private capital into state-owned corporations, as occurred in China from the mid-1990s onwards, in which case their purposes should reflect the interests of private as well as public-sector investors.
The corporation is a powerful vehicle for binding public and private sectors together in the delivery of infrastructure programmes. The particular form that it takes depends on the nature and phase of an infrastructure programme. There are typically three phases to a programme: design, build, and operate. These have specific financing requirements and risks associated with them and consequently require different types of arrangements between private- and public-sector providers to make them viable.
In combination the three phases typically generate a long horizon. Design may take anything from two to eight years; building may take three to seven years; and the operation phase is likely to extend for at least two decades. No government or company can credibly commit for the totality of a programme; instead, they should be disaggregated into their three phases.
The design phase sets the parameters that determine whether the subsequent stages can attract private finance. Since it sets key parameters, design is in many respects both the most critical and the most complex phase. It is critical because it defines the purposes and objectives of the project and needs to demonstrate credibly an ability to deliver in the subsequent phases. It is complex because it involves coordination with several different players and a commitment on their part to participate in the subsequent activities.
Currently insufficient resources are often attracted into the design phase, which, for a significant project, are likely to amount to tens of millions of dollars spread over several years. If this phase is to attract private finance, then from the start the prospects of commercial returns from investing in design must be sufficiently attractive. At present this is often not the case.
At the completion of the design stage, building the project needs to be a bankable proposition, recognized as advantageous by the government. Hence, the most crucial element of design is to define the governance of the programme. Investors feel exposed to risks of expropriation by governments, and governments are vulnerable to exploitation by private-sector contractors. There needs to be a common purpose and understanding from the outset, means by which the different parties can credibly commit, and ways of resolving disputes in a low-cost and effective manner.
The most effective way of achieving this is through joint ownership. The different parties—private-sector funders, contractors, and operators need to be brought together with public-sector providers (donors and government) to define the roles and responsibilities of the different parties. Projects should be ring-fenced as self-contained activities with the governance of each project delineated from that of other projects but within the context of a broader framework or architecture for the sector as a whole, set by the government after consultation with investors and other stakeholders. In other words there should be a hierarchy of collaborative arrangements between private and public parties from high-level sector-wide oversight to specific infrastructure projects.
The importance of the high-level oversight is that while the management and financing of individual infrastructure projects can be contained and ring-fenced, the impact of other activities in the sector or indeed from other sectors cannot. So for example, the success of a road-building programme might be seriously affected by the initiation of railway programmes. The potential for infrastructure assets to be stranded is therefore immense even in the absence of specific expropriation of returns from a particular project. The way to manage this is for the infrastructure programme of a country to be designed in close collaboration with investors.
The build phase is when major finance has to be raised. It is also currently a high-risk phase. There are two critical components to effective management at the build stage. The first is commitment mechanisms and the second is collateral so there are costs and compensation for reneging on contracts. Building large infrastructure projects commonly involves substantial cost overruns and delays. There is an established psychological bias towards optimism among planners since best-case scenarios tend to become adopted as the benchmark for outcomes. At the build stage these risks can potentially inflict high costs on the private contractor, which are consequently priced into the project. Hence, these risks are borne by the private investor who undertakes upfront finance for building the project ahead of recuperating returns on the investment. To address this there need to be mechanisms for imposing penalties on defaulting governments and ways in which private-sector contractors can recover their costs.
Underwriting by international agencies is one way in which governments can commit to private-sector providers. MIGA has offered an effective method of discouraging default by acting as a collection device backed by the threat of sanctions from the World Bank. The question is whether this can be scaled to a greater level than to date. This is a matter of the amounts of capital that can be raised to provide underwriting facilities. With its strong record in enforcing contracts, MIGA has a good base on which to build its capital; however, it requires a global initiative on the scale of the European Bank for Reconstruction and Development (EBRD) for this to have real impact.
Capitalizing MIGA and similar organizations should be viewed as an effective form of aid in which, for the most part, the aid does not need to be spent because the underwriting never has to be provided. Governments should recognize their collective power to promote economic development by creating agencies such as MIGA, backed by the World Bank, that act as commitment vehicles and allow developing country governments to commit in a way in which they wish to but cannot credibly do at present. It is therefore substantially welfare-enhancing at low cost.
At the operate stage the risk is a government one that the private-sector operators fail to deliver their promised level of services. This requires a commitment on the part of companies to meet obligations in terms of quality and price of services and to desist from exploiting their monopoly positions to the detriment of customers. This is traditionally achieved through regulation.
The regulatory process should be seen as a commitment device on the part of companies. It forces them to avoid abusing their monopoly power in the charges that they levy on customers and the quality of services that they provide. However, regulation has to balance the benefits it confers on customers through lower charges and higher-quality services with the disincentives it imposes on providers to invest and operate in infrastructure markets. In particular, regulators can act in as arbitrary a fashion as governments in seeking to impose populist agendas. Alternatively they are subject to capture and fail to promote the interests of customers.
There is an inherent conflict between shareholder driven corporations and the public interest and in general the regulator comes off worse from the conflict. This is particularly serious in a developing-country context because of the reliance that has to be placed on overseas operators. There is, therefore, a real risk of the process of private-sector engagement in infrastructure being rapidly discredited as utility companies are perceived to be exploiting vulnerable economies.
To avoid this it will be necessary to ring-fence the utility activities of foreign operators from the rest of their activities. It is a process that has been successfully implemented in utilities in the United Kingdom where, in particular in the water industry, the regulator requires a clear delineation between the utility and the non-utility parts of businesses. Assets cannot be transferred between the two, they are incorporated as separate subsidiaries with their own boards of directors, and the payment of dividends from the subsidiary to the parent is limited. The ring-fenced activity should be structured as a benefit company with its stated public purpose being its licence condition to operate.
In sum, the three phases of an infrastructure programme are delivered by different types of corporations: publicly owned corporations with private-sector participation in the first, design phase; private corporations in the second, build phase, with public-sector engagement constrained where necessary by international organizations; and private corporations with public as well as private purposes defined by their licence conditions in the third, operate phase.
The solution to the public-and private-sector commitment problems are mirror images of each other. In both cases, the corporate form provides a powerful vehicle for aligning interests that are currently highly divergent. Regulation does not solve the problem because conflicting interests encourage private companies to do what they can to circumvent the regulation and public organizations to renege on the terms of the regulation.
Instead we should use the corporate vehicle as a powerful instrument for promoting partnership where conflict currently exists. It does so through conferring a degree of engagement of private providers in public-service functions, such as the design and regulation of infrastructure services, and through including public licence obligations in the articles of associations of private providers of infrastructure services.
The corporate form also addresses the accounting problem of the absence of balance sheets in national accounts described above. By organizing infrastructure in public as well as private corporations, infrastructure assets are all automatically reported in corporate accounts.
A combination of proper accounting for infrastructure across the private and public sectors together with the reformulation of the purpose of private-sector providers offers the prospect of transforming infrastructure around the world. With correct measurement and diminution of the inherent conflict between public-sector purposes and private-sector interests, there is at least a chance of the world obtaining the infrastructure it needs for its future prosperity and survival; without it, there is little or none.
Infrastructure has brought us full circle back to the Roman origins of the corporation described in Chapter 3 as the provider of public works. This stands in contrast to the way in which the book opened in Chapter 1 by noting that the nature of the corporation is fundamentally altering from a tangible to an intangible entity in the guise of the mindful corporation. The most significant manifestations of this are the social networks of Facebook, LinkedIn, and Whatsapp and the Internet providers, such as Google.
By virtue of being full of minds rather than matter, the twenty-first-century infrastructure companies are low- not high-capital-intensive businesses. But far from implying the irrelevance of the principles of infrastructure described in this chapter, they magnify its significance. It is not financial capital that is ultimately the problem—it is social capital. The systems features of infrastructure with which this chapter opened are statements about the social as against the private relevance of infrastructure. It binds us as one community, society, and human race. And as such, the private provider is operating in a social space. It cannot ignore its public place or if it does then it risks the wrath of the politicians, competition authorities, and regulators bearing down upon it, as they have done increasingly over the past decade, with the ultimate threat being nationalization if it exploits its dominant market position.
This brings us back to the consciousness and values of Chapter 2. The corporation is a conscious entity that has values. But when its sphere of operation is public not private, when it interacts with others in fulfilling its functions, and when it is collectively part of a bigger whole, its consciousness has to embrace its environment, not just itself. That is the challenge of the twenty-first-century corporation, government, and world, and it is what will make the subject of the corporation one of the most fascinating for many years to come. We await the coming of the seventh age of the corporation as the trusted corporation.