UPDATE MAY 21, 2008:

Interestingly, this post has generated inquiries from consultants at McKinsey, Boston Consulting Group, and Bain; a librarian at Deutsche Bank; a consultant at PWC; and also analysts at several infrastructure funds. They are all looking for more detailed numbers on world demand for infrastructure, by sector and by country, obviously all trying to come up with numbers to include in their PPMs and to try and figure out where to go looking for assets, which are few and far between. In response to this overwhelming level of interest, we have commissioned a detailed study on this topic. The findings are reserved for companies that are affiliates of the Collaboratory.

Here is a link to another study relevant to this topic, which appears promising but which I have not reviewed in detail: http://vector1media.com/top-stories/corporate-news/global-infrastructure-demand-through-2030-study-released-by-cg%10la-infrastructure/

ORIGINAL POST, MAY 26, 2006:

How much investment in infrastructure - roads, ports, airports, power, water, etc. - is necessary over the next 10 years? Here are a few statistics:

1. According to the UN, there will be approx. 1 billion new people living on the planet in the next decade, the large majority born in developing countries, and presumably they will all need homes and basic infrastructure;

2. According to Yun-Hwan Kim, Assistant Chief Economist in the Economics and Research Department at Asian Development Bank (ADB):

"Estimates on the investment level required by major countries in the [Asian] region to construct and improve physical infrastructure have varied from $20-30 trillion in the coming 10 years. The sheer magnitude of the region's financing requirements calls for urgent and diversified mobilization of financial resources. Asia needs to develop long-term financing and local currency markets that should channel the high level of savings in the region to infrastructure development. Deep and liquid bond markets, both local markets as well as regional markets, are key to addressing this challenge."

Unfortunately, Kim does not cite the source of the "$20 to $30 trillion" statistic.

Yet, Kim is not the only ADB employee to cite this statistic. In a 2003 speech, ADB Vice President John Lintjer made the following statement,

"The Asian region, as we are all aware, faces some major development challenges. To meet physical infrastructure needs alone, the investment level required by major countries in the region totals $20 trillion - $30 trillion over the next 10 years, according to ADB estimates. Asia's nonperforming loans amount to about $2 trillion, and need to be restructured and refinanced. This sheer magnitude of the region's financing requirements calls for urgent mobilization of funds."

3. In another ADB study -- "Connecting East Asia" infrastructure demand in the Asian region is projected more modestly at just over $1 trillion over the next five years, as follows:

"And so finally to the question of funding - that is, the scale of the resource requirement to address East Asia's infrastructure challenge, and how this requirement can be sourced. According to analysis undertaken for this study, to meet expected infrastructure service needs, East Asia would have to spend $165 billion a year over the next five years - or roughly 6.2 percent of its GDP annually - on electricity, telecommunications, water and sanitation, and major transport networks (see Figure 1.10).

These estimates take into account both investment and maintenance of assets (an equally and sometimes more cost-effective way of meeting service goals). In meeting these needs, it is estimated that 65 percent of expenditure would need to take the form of new investment, with the remaining 35 percent channeled toward maintenance of existing assets. In China alone, total needs account for almost 7 percent of GDP (and China's infrastructure needs account for 80 percent of the region’s total).

In low-income countries the needs are relatively greater than in middle-income ones. When other infrastructure needs are included (such as ports, airports, bridges, secondary roads, urban transport, and gas grids), the overall estimated need rises above $200 billion a year. Moreover, these estimates do not incorporate any strategic decisions to invest in infrastructure ahead of demand, or to increase access for the poor in line with the Millennium Development Goals (MDGs) or other targets." (Chapter 1, Pg. 29)

Within this study, the specific calculations are referenced back to:
Yepes, Tito. 2004. "Expenditure on Infrastructure in East Asia Region, 2006-2010".

4. It turns out that Tito Yepes is also the analyst at the World Bank who prepares official World Bank numbers for world infrastructure demand. In a 2003 study called "Investing in infrastructure: What is needed from 2000 to 2010?" he and a co-author conclude:

"New investment needs are estimated to be approximately US$370 Billion per annum for the period 2005-10, amounting to nearly 1% of worldwide GDP. Another $480 billion (1.2% of global GDP) are needed for maintenance. The total resources needed are therefore approximately 2.1% of GDP, excluding any expenditure on rehabilitation or upgrading. (page 11)

Thus, the study projects that actual global infrastructure investment will be 0.85 trillion per year (i.e. $370 million + $480 million = $850 million), which works out to a total of $8.5 trillion over the next 10 years.

Note, that $8.5 trillion is a far cry short from the $20 to $30 trillion that the Asian Development Bank has projected for South East Asia (recall item 2 above).

Why the huge difference? The difference probably lies in the calculation methods. For example, Yepes qualifies his predictions by stating, "It should be noted that predictions are based on estimated demand rather than on any absolute measure of "need" such as those developed in the Millenium Development Goals (MDG)."

It may be the case that the elusive $20 to $30 trillion stat from ADB is a "need" based statistic rather than an actual "demand" based statistic. It seems to me that if I were an analyst, I could define "need" in a great variety of ways -- it is such a subjective concept. For example, on one extreme, an analyst could assume that "need" encompasses only the provision of basic shelter, water, sanitation, etc.; whereas a more optimistic analyst might assume that all east Asians "need" infrastructure services at levels now common in the West. And given that ADB is in the business of providing infrastructure loans, their analysts may tend to be more optimistic in their assumptions.

5. Yepes comment about the MDG program inspired me to look at the MDG website, but I was unable to find a projection of world infrastructure demand anywhere within the materials published on the MDG website.

6. Finally, in 2005, the ASCE - a U.S. professional society of engineers - released the latest version of its Infrastructure Report Card, which pegs the 5 year "need" for investment in infrastructure within the U.S. at $1.6 trillion.

Posted by rjorr at May 26, 2006 11:25 AM