—Election slogan in India, 2004

 

 

 

 

 

Roads probably constitute the earliest human demand for infrastructure, and the earliest known constructed roads have existed in Ur in modern-day Iraq since 4000 BC.1 Indeed, the pyramids couldn’t have been built without the roads on which the giant limestone blocks were dragged around between 2600 and 2200 BC. Civilizations advanced or declined around the quality of their road networks. The ancient Roman, Persian, Indian, and Chinese civilizations all built road networks that allowed them to rapidly transport military units while simultaneously encouraging commerce and trade. Trade routes joined the empires of China, India, Asia Minor, North Africa, and Rome. The next major infrastructure development came from the great discoveries of electricity and its applications by Benjamin Franklin, Nikola Tesla, André-Marie Ampère, Michael Faraday, Thomas Edison, and many others in the eighteenth and nineteenth centuries.

In the modern era, the term infrastructure refers to a wide array of industries with different characteristics. Traditionally, infrastructure refers to the following sectors of the economy: transportation, energy, telecommunications, water, and sanitation. Transportation refers to road, rail, airports, and ports. Energy refers to oil, gas, petrochemicals, and electricity generation, transmission, and distribution. Telecommunications refers to fixed lines and mobile telephony. Recently, information technology infrastructure that refers to the physical hardware used to connect computers and users is sometimes grouped with traditional infrastructure.a

Commentators have also grouped social infrastructure like schools, prisons, hospitals, and courts under the rubric of infrastructure.2 Grouping fundamentally different industries like mobile telephony with schools and prisons obscures the fact that the risk and return profile of these assets is very different. Consequently these assets demand different investment strategies. We therefore need consistent criteria for classifying infrastructure industries.

In this book, I use asset characteristics as classifying criteria. Using asset characteristics as an analytical scalpel yields insights into infrastructure asset risks that help us design optimal investment strategies. For example, electricity generation using coal, oil, and nuclear fuel differs in fundamental and important ways from alternative energy sources like solar, wind, and so on. Chapter 2 is entirely devoted to this topic. In this chapter I choose the traditional infrastructure industries which Chapter 2 shows possess distinct characteristics—electricity, transportation (road and rail), water and sanitation, fixed-line telecommunication, and pipelines—for analyzing demand across different countries.

Using two distinct methodologies for analyzing infrastructure demand, I make the case that demand is growing and requires large investments that private investors can provide. Demand analysis shows which sectors and geographical areas are likely to be attractive to investors. Furthermore, the returns from these investments match the needs of pension funds and insurance companies in particular and therefore offer attractive investment opportunities.

Infrastructure demand refers to the investment necessary to satisfy retail consumer demands as well as producer or industry demands based on projected GDP growth. This makes sense because it’s reasonable to assume that as disposable income rises, demand for a better quality of life in terms of electricity, water, sanitation, telecommunications, and transportation (air, sea, road, and rail) should increase. Industry, too, demands increased electricity, transportation, telecommunication, energy, commodities, and other inputs to keep pace with growth in its output. Demand, however, is unlikely to be uniform across industries and countries. This has been true historically and is likely to persist in the future. An analysis of sectoral and geographical demand is helpful to investors for targeting the appropriate industry and country. Exhibit 1.1 shows the change in worldwide composition of infrastructure stocks from 1960 to 2010. Water and sanitation is excluded because of lack of data.

EXHIBIT 1.1 Historical Change in Composition of Infrastructure Stocks

Source: M. Fay and T. Yepes, “Investing in Infrastructure: What Is Needed from 2000 to 2010?” World Bank Policy Research Working Paper 3102, 2003.

Exhibit 1.1 shows that the share of rail dropped from a third to a mere 5 percent and the share of electricity doubled while telecommunications quintupled, albeit from a low 2 percent to 10 percent, over the 50-year period. Therefore, any linear extrapolation into the future based on past trends even in mundane infrastructure must be interpreted with caution.

A complex interplay of factors influences demand in different infrastructure sectors. Some of these include technology, substitutability (mobile for fixed line, road for rail), complementarities (electricity generation to electric trains), sectoral structure (number and strength of established firms resistant to change), macroeconomic factors (GDP per capita), and so on. Technology has affected demand for fixed-line telecommunications with mobile telephony replacing fixed lines in high income countries and leapfrogging investments in fixed lines in low- and middle income countries. Technology also influences oil, gas, and electricity generation with the development of alternative energy sources, although these are not likely to catch up with traditional oil, gas, and coal even by 2030.3 Exhibit 1.2 displays the factors affecting demand for each sector and the interaction of the sectors.

In addition to the factors and interactions shown in Exhibit 1.2, investors must take into account cultural attitudes and government policy encouraging or discouraging particular infrastructure industries in order to forecast infrastructure demand with any degree of precision. The next section develops broad estimates of infrastructure investments up to 2030. Note that forecasts tend to lose accuracy as the forecasting period increases, and a forecasting period up to 2030 is still very long. The forecasts presented are therefore meant to be interpreted as providing broad estimates and are meant to provide insights into asset allocations across industries and geographies.

EXHIBIT 1.2 Infrastructure Demand Drivers Interactions

Source: Author analysis.

In this chapter I explore the methodology followed by the World Bank and the Organisation for Economic Co-operation and Development (OECD) to forecast investments for new assets as well as maintenance of existing infrastructure assets, using historical investment patterns as a percentage of gross domestic product (GDP). In addition, I use gross national income (GNI) to compare infrastructure stocks across countries with different levels of GNI. In the GNI method I extrapolate the future demand in low and middle income countries to match infrastructure in the high income countries.

In the percentage of GDP methodology, total spending on new infrastructure and maintenance varies with the GDP of countries. Developing countries need to spend a higher proportion of their GDP on infrastructure because of their lower infrastructure stocks and their greater growth rates. Estimates for new investments in developing countries range from 4 percent of GDP in the World Development Report4 to 3.2 percent in Fey and Yepes (2003). Middle income countries spend about 2.6 percent of GDP while high income economies spend about 0.4 percent of GDP on new infrastructure investment. The estimates for maintenance follow a similar pattern, with low income countries estimated to spend about 3.73 percent, middle income countries about 2.5 percent, and high income countries about 0.42 percent.

Apart from the demand for new infrastructure stocks, investments are also required for maintenance and replacement of existing stocks. Although total investments required in infrastructure should include funds for maintenance, maintenance funds essentially comprise costs required to keep the assets functioning smoothly. For the purposes of analyzing investment strategies, these costs determine working capital requirements and available free cash flows which in turn determine the return over investment. High maintenance costs naturally reduce free cash flow and returns. I do not consider maintenance funds further here since maintenance funds are not investments and do not translate into claims on assets. Suffice it to say that maintenance funds for electricity, road, and rail make up about 2 percent of the replacement cost of the capital stock, 3 percent for water and sanitation, and up to 8 percent for telecom fixed lines.5

Should we consider funds for replacement of existing infrastructure stocks? The amount of funds needed for the replacement of existing infrastructure depends on the level of existing infrastructure stock, the total life of the stock, and the rate of depreciation. For example, if we assume that the life of roads is 30 years and the rate of depreciation is spread out equally over the 30...