According to the World Economic Forum (WEF), the US ranks third globally out of 138 countries in competitiveness, behind Switzerland at first and Singapore at second. Infrastructure is one of the 12 primary factors (or pillars) that make up the WEF’s Global Competitiveness Index. In infrastructure, the US ranks eleventh in its total score globally, placing it between the UK and Korea at ninth and tenth and Spain and Taiwan at twelfth and thirteenth. With a total score of 5.94 out of 7.00, US infrastructure ranks better than the average advanced economy at 5.65.
Better than average quality of US infrastructure does not tell the full story
Eight sub-factors within the WEF total infrastructure score are shown in Exhibit 1. The total score of 5.94 considers 19 sub-factors within infrastructure, whereas the quality of overall infrastructure, where the US ranks twelth, is measured by the five quality sub-factors shown.
Exhibit 1. World Economic Forum: US infrastructure
The Global Competitiveness Report (2016-2017)
|US infrastructure quality||Value||Rank / 138|
|Quality of overall infrastructure||5.7||12|
|Quality of roads||5.6||13|
|Quality of railroad infrastructure||5.1||13|
|Quality of port infrastructure||5.7||10|
|Quality of air transport infrastructure||6.1||9|
|Available airline seat km/week, millions*||37.744||1|
|Electricity and telephony infrastructure|
|Quality of electricity supply||6.5||17|
|Mobile telephone subscriptions / 100 pop.*||117.6||66|
|Fixed telephone lines / 100 pop.*||37.5||25|
* Values are on a scope of 1-7 unless otherwise noted.
Although US infrastructure ranks better than the average advanced economy in its WEF total score and its overall quality score, many economists feel the age of US infrastructure due to low re-investment has not fully manifested itself. Further, 10 of the 19 sub-factors are quality, reliability or efficiency scores based on surveys. Hence they are perceived quality, not objective measures of quality. Although these are large surveys (485 respondents in the case of the US for 2016-2017) and the WEF has a very detailed methodology, only the remaining nine sub-factors are quantitative measures from third-party indices.
Infrastructure capability and performance matter more than perceived quality
Comparing infrastructure across countries and regions is difficult because of the unique characteristics of countries. Simple cross-country comparisons do not account for differences in current capital stock, demographics, population densities and transportation preferences across nations. Infrastructure scores also do not fully account for the composition of different economies, e.g., the quality of ports matters more to an export-driven economy such as Germany than to a service-oriented economy such as the UK, which scores relatively high on broadband connectivity. The real issue in measuring infrastructure is its capability and performance which is linked to the condition and utilization of the asset.
US infrastructure ranks relatively well even though it needs so much investment. This is because its performance elasticity of condition is in an elastic state.(1)
Low reinvestment in public capital stock
Another problem with assessing quality of infrastructure is that it is an output because it measures how people feel about the infrastructure they use. This does not translate into a higher input, i.e., how much a country is reinvesting in its infrastructure.
To analyze reinvestment rates, we looked at the change in public capital stock across the US and other advanced economies available on the IMF Fiscal Monitor Database. It is important to examine both the amount and the change in the amount of physical capital, not only the change in physical capital (i.e., the annual inflows of public investment into physical assets). It is the amount of physical assets (i.e., physical capital stock), such as economic and social infrastructure, that over time provides productive services.
Physical capital is well defined conceptually and follows a standardized approach across developed economies. Measuring public-private partnership (PPP) investment is difficult because of varying definitions of what constitutes infrastructure. We examined real net physical capital stock to adjust for inflation, depreciation and disposals. Capital stock includes a small percentage of R&D and education capital, whereas physical capital excludes these.
Exhibit 2. Real net public capital stock: US vs. other developed economies
Growth rate 1960-2016 (year-over-year % change)
As shown in Exhibit 2, developed economies including the US have invested at a declining rate for decades and especially since the financial crisis. The average growth rate for 1960-2016 in public capital stock in advanced economies (ex-US) was 3.3% per annum and for the US it was 2.2% per annum.
What is public capital stock?
Capital stock consists of public and private fixed assets (or physical assets), plus education capital and R&D capital. Physical capital consists of total fixed assets and is one of the three factors of production. The value of additions to fixed assets, net of or after disposals, is referred to as gross fixed capital formation. Public capital stock consists primarily of long-term government fixed assets such as infrastructure. Even though the concept of public capital stock is well developed internationally, the terms physical capital and fixed capital, and physical assets and fixed assets, are often used interchangeably. The real net stock of government fixed assets as produced by the BEA is a measure of US public infrastructure stock. This is a similar measure to public capital stock for developed economies as reported by the IMF.
More telling is the change in the growth rate. For advanced economies (ex-US), it has declined 1.17% per annum since 1960 and for the US the growth rate has fallen 1.52% per annum. Between 1972 and 2001 the growth rate in US physical capital remained fairly flat. The steep drop in the late 1960s after most of the US Interstate highways were built, and again in 2003 after the prices of materials used to construct infrastructure increased rapidly, have contributed to a capital expenditure (“capex”) deficit for US infrastructure.
The conclusion of the WEF and public capital stock analyzes is that the relatively steady US infrastructure quality scores, as measured by the WEF surveys, do not yet reflect the declining reinvestment in physical assets, although they are likely to do so unless the US begins to invest. This recognition, along with positive movements in state budgets, federal financing programs and the P3 market, is why we feel US infrastructure is at an inflection point and poised for growth.
US infrastructure is aging
America’s infrastructure is aging, as declining funding has led to a lack of new investment, especially over the last 10-15 years. The 2016 US presidential campaign highlighted the issue, with both major-party candidates calling for additional investment. Reaching a political agreement to boost spending on infrastructure will not be easy, even though lower borrowing costs and reduced budget deficits may have softened opposition to it.
Infrastructure in the US faces three major issues. First, the average age of the public capital stock is rising (Exhibit 3) due to the low level of infrastructure re-investment. Airports and highways have aged the most in the last 10 years, and dams / levees and highways have aged the most over the last 50 years. This is particularly pronounced with certain types of infrastructure such as bridges, many of which will soon reach the end of their intended period of use, necessitating greater investment. The US Bureau of Economic Analysis (BEA) estimates that 67% of Interstate bridges are 26-50 years old and 15% are 51-75 years old, whereas their estimated design life is typically 50 years. About 30% of all bridges in the US are 51-100+ years old.
Only looking at the age of infrastructure can be misleading. The main reason these assets have aged is lower capital investment in new structures. Key to infrastructure efficiency and overall longevity, however, is operation and maintenance (O&M) spending, which increased 6% in real terms between 2003 and 2014, whereas new capital investment declined 23%. O&M spending has kept US infrastructure functioning and helps to account for its relatively good quality scores and average satisfaction scores (see Sequoia Research “Outlook for United States infrastructure: a market poised for growth,” Summer 2018, Appendix 2). Nonetheless, many assets are approaching their expected economic lifespans and need to be replaced.
Exhibit 3. Aging assets
Average age of public capital stock (years)
The second issue US infrastructure faces is a lack of capacity. Growth in the use of roads has outpaced new highway capital investment for most of the period since the 1970s, when much of the Interstate highway system was nearing completion (Exhibit 4). This has led to a gradual rise in congestion and travel delays on roads and other modes of transport. The hours delayed per 100 miles travelled has increased steadily from 0.30 in 1982 to 0.62 in 2015.
The term “highway” in this report includes all roads.(2)
Exhibit 4. Highway supply and demand
Annual growth in highway capacity and demand (year-over-year % change)
The third issue confronting US infrastructure is reliance on a declining revenue base. The federal gasoline tax of 18.4 cents per gallon and diesel tax of 24.4 cents per gallon are paid into the Highway Trust Fund. These taxes provide most of the funds for federal transportation spending but they have not been raised since 1993 and there seems to be little political appetite to raise them. Because of not raising the tax (even with inflation) and because of slower gasoline consumption, the federal highway program has been left with a funding deficit that Congress temporarily plugs every few years. Some states have raised gasoline taxes incrementally (e.g., California, recently) and a recovery in state budgets since the 2007-2009 recession is allowing a rebound in public investment, so the picture at the state level is brighter. Nevertheless, the current financing scheme is far from ideal.
The terms “spending” and “investment” with regard to infrastructure are used interchangeably in this report.
The US needs $1.5 to $2 trillion of additional infrastructure investment
The majority of infrastructure spending is at the state and local level. In 2014, the federal government spent $96 billion on transportation and water infrastructure and state and local governments spent $320 billion, for a total of $416 billion. This included $181 billion for new structures (referred to as capital spending) and $235 billion for O&M.
Nominal spending has been steady but real spending has dropped
Over the last 30 years, nominal public spending on infrastructure has hovered at about 2.4% of GDP. This is in line with the major economies in Europe, most of which spend 2% to 3% of their GDP on infrastructure. The outliers are Spain, which spent between 6% and 7.5% in the years 2000-2008 but has now normalized back to 3%, and Germany and the UK, which spent 1.75% and 1.5%, respectively, in the period 2010-2015. Australia, Japan and Canada have consistently spent more, averaging 4.7%, 4% and 3.5%, respectively, during 2008-2013.
Divergent trends in US public infrastructure spending
The US used to spend more than 2.4% of GDP on infrastructure. Real public spending on transportation and water infrastructure was 3% of GDP in 1959. More importantly, the composition of this spending has changed. From the 1950s to the mid-1970s, capital spending for new structures exceeded spending for operation and maintenance (O&M) by as much as 75%. From the mid-1970s to 2003, capital spending was comparable to O&M, but since 2003 real capital spending fell 23% through 2014, whereas O&M spending increased 6%. The higher capital spending in the 1950s and 1960s was due largely to the construction of dams and the Interstate Highway System and increases in federal grants to the states under the Clean Water Act in the 1970s.
According to the Congressional Budget Office (CBO), the nominal growth rate for 2003-2014 in public transportation and water infrastructure spending, which forms the bulk of public spending in US infrastructure, was 4.4% per annum. This is in line with nominal GDP. Adjusted for inflation, however, US public spending has slowed because the prices for materials and other inputs have rapidly increased. Adjusted by the simple GDP Price Index, real spending has increased only 1.5% per annum since 2003. If adjusted by infrastructure-specific price indices, however, it has fallen 2.1% per annum. This means the government is spending more on infrastructure but more than all of the nominal increase is paying for higher prices. This has exacerbated the US capex deficit.
Constraints at all levels of government have limited the ability to adequately reinvest in transportation, water and other critical infrastructure. Funding has been available for viable privately financed infrastructure projects such as P3s over the last several years, but P3s currently fund less than 5% of the nearly $400 billion of annual government investment in infrastructure.
What about future infrastructure spending?
The analysis so far has shown that the annual rate of real spending on US infrastructure has slowed by about 50% since the 1960s and it has contracted since 2002 when adjusted for infrastructure-related input costs. Another important metric is future needs and expected funding. The American Society of Civil Engineers (ASCE) does this as part of their score card on US infrastructure.
Exhibit 5 shows the total spending needed to keep US infrastructure in good condition, according to the ASCE. Estimated funding includes federal, state and local as well as private funding and it assumes that this funding stays flat over the 10-year period (i.e., no major infrastructure investment package from Washington and no major policy changes that better enable private investment and P3s). The funding gap is the difference between spending needs and estimated funding and illustrates the infrastructure capex deficit. These numbers are often used by analysts, economists and politicians to talk about the large need for US infrastructure spending.
It should be noted, however, that the gap of $2,064 billion in Exhibit 5 may be an overestimate. Some have argued that the ASCE will tend to overstate investment needs because their calculations are based, in part, on the physical condition or age of infrastructure, without linking it to system performance. This particularly applies to older and larger systems such as highways.
The term “funding” with regard to the ASCE numbers is synonymous with investment and spending as used in this report. Normally it means debt as opposed to equity capital but that distinction is not made here.
Exhibit 5. US infrastructure spending needs, funding and gap
|2016-2025 (10 years)
|Total needs||Estimated fundng||Funding gap|
|Water / wastewater infrastructure||$159||$45||$105|
|Inland waterways & marine ports||$37||$22||$15|
|Hazardous & solid waste||$7||$4||$3|
|Public parks & recreation||$114||$12||$102|
A caution on infrastructure numbers, scores and grades
ASCE is an infrastructure industry group that represents members of the civil engineering profession worldwide. Estimates of the US infrastructure funding gap often differ by organizations and should be interpreted with caution. Other organizations such as the EPA have highlighted a need for more infrastructure investment in certain sectors, such as drinking water, wastewater and storm water infrastructure. Each infrastructure sector has spawned supportive lobbies that push for higher spending. These groups often complain that their favored infrastructure is crumbling, congested or underfunded. Infrastructure scores can also be subject to biases. The annual WEF infrastructure scores are based in part on surveys and the ASCE grades are based on their internal grading using eight criteria.
The CBO shows that the government spent $181 billion in new capital in 2014. The Office of Management and Budget (OMB) estimates similar federal spending for 2017. The ASCE spending needs are mostly for new capital as well.
To compare the ASCE and CBO numbers, we isolated public transportation and water from other sectors in the ASCE numbers, e.g., public parks and schools. This lowers the total needs to $3,606 billion. Assets that are largely privately financed, e.g., power and freight rail, also need to be excluded. This brings the total needs down to $1,418 billion for public transportation and water infrastructure.(3) This is about 70% of the estimated ASCE funding gap for transportation and water. Our estimate is that the US needs $1.5 trillion over 10 years to bring its infrastructure up to a state of good repair, which would be a B grade by the ASCE.(4)
A plan for plugging the infrastructure funding gap
We have identified the problem: low re-investment and the resulting funding gap. The question now is how does the US plug this gap. Exhibit 6 illustrates a specific roadmap for achieving this.
There are five components to finding the $4,590 billion of total infrastructure spending needs. The first is the public construction budget, which Goldman Sachs estimates could grow at the expected rate of headline CPI over the next 10 years. To this add $411 billion in additional P3 investment that would occur over 10 years if the US spent the median of other countries that use P3s. Add further the $193 billion that is possible from privatization of airports and ports and divestiture of state toll roads through an asset recycling program. This brings total public and private potential spending to $3,889 billion. With another $700 billion in investment, the US can plug the gap and fund its total infrastructure need of $4,590 billion as estimated by the ASCE. This means the US needs $1.3 trillion in additional public and private spending (the items in light blue). If the administration and Congress are successful in implementing plans for investment of $1 trillion over the next 10 years, this will plug about 75% of the gap and give the US the best infrastructure it has had in the last 50 years.
Exhibit 6. How P3s can change the infrastructure investment landscape
USD billions over 10 years
Both parties have a plan to invest $1 trillion in infrastructure over 10 years
The administration and the Democratic members of Congress have put forward initiatives that would see $1 trillion of new spending in the US over the next 10 years.(5) Such an increase would be a favorable development for the infrastructure sector as it would facilitate funding for new projects and help the US maintain its aging infrastructure assets. Either party’s 10-year plan would represent a meaningful down payment on required spending and a suggest a 25% average annual increase above the current $416 billion in annual public infrastructure spending, if the spending is completely incremental to current spending.
A breakdown of US infrastructure spending
Almost all spending on transportation, drinking water and wastewater treatment infrastructure is done by the public sector. These broad sectors make up about 75% of all public infrastructure spending with the remainder going mostly to K-12 public schools, followed by public parks.
The largest amount of public infrastructure spending goes to highways, which includes Interstate and local roads, followed by water utilities, mass transit, air transport, water resources, water transportation and rail (Exhibit 7). Water utilities consist of water supply and wastewater treatment. Water resources consist of water containment systems (dams, levees, reservoirs and watersheds) and sources of freshwater (lakes and rivers). Spending on air transport includes state and local spending on airports, federal spending on air traffic control and federal grants to airports. Water transportation includes inland waterways and rail includes Amtrak and other short-line commuter rail systems.
Exhibit 7. States bear most of the infrastructure financing burden
Real spending on capital and O&M in 2014. USD billions (LHS) and % of GDP (RHS)
State and local governments pay for most of the facilities that schools require. Municipalities primarily administer and fund water and wastewater infrastructure. The public sector has little direct role in funding power or other components of the grid, communication infrastructure or freight rail.
Taxes and user fees ultimately pay for infrastructure
Funds for any infrastructure project ultimately come from taxation, user fees or a mixture of the two. This simple, fundamental point can sometimes be obscured by the complexity of the financing programs and arrangements surrounding a project. This is ultimately one of the main reasons nominal US infrastructure spending has not grown to required levels.
On the public side, gasoline taxes have not been raised since 1993 and these taxes fund highways, the largest single sector in the public infrastructure budget. On the private side, many projects do not have clear user fees (e.g., most of the Interstate highways) and these fees are an important component of investor returns. Reform such as liberalizing tolling restrictions has an important role to play in attracting private capital to help build out US infrastructure.
Appendix: Compelling economics of infrastructure
Numerous studies have shown infrastructure investment can raise economic growth, productivity and land values, while also providing positive and significant second-order benefits in economic development, energy efficiency, manufacturing and public health. For example, investing in newly paved roads raises housing values and living standards. The rise in housing values on affected streets significantly exceeds the cost of paving.
Infrastructure affects the real economy, not just the financial economy
Whereas QE had its largest effect on the financial economy, there are good reasons to believe that increased infrastructure spending will affect the real economy. Infrastructure is used by a large percentage of the population in one way or the other and while interest rates affect most people, financial asset prices do not.
Studies have shown a multiplier effect of 2-3x from public-sector investment, i.e., for every $1 spent on infrastructure by the government, GDP increases by $2-3 over the medium term (typically four years). This effect is especially large when the investments are debt financed during recessions, the output gap is large and nominal short-term interest rates are at their natural floor (called the zero lower bound or ZLB). During these times, the fiscal multiplier can be 4-5x. The current US output gap is almost zero as the economy is operating near full employment, but nominal interest rates are still very low by historical standards so the multiplier would still likely be positive.(6)
Higher returns from operation and maintenance spending rather than new capital spending
The CBO found that additional investment in existing infrastructure is one of the most efficient means for raising output and employment. For example, research has shown that the greatest return on investment (ROI) can be achieved by spending on the maintenance of existing highways to bring them up to their minimum state of good repair. Spending beyond this level, or even on new roads in many cases, produces a much lower ROI. Investing in transportation infrastructure also creates middle class jobs. Studies have shown that 80% of the jobs created would be in the construction, manufacturing and retail sectors.
Acknowledgment, sources and endnotes
Sequoia would like to thank Michael Verhoeven for his assistance in preparing this report.
Primary sources: American Society of Civil Engineers, Bipartisan Policy Center, Bureau of Economic Analysis, Cato Institute, Common Good, Congressional Budget Office, Congressional Research Service, Council of Economic Advisors, Department of the Treasury, Federal Highway Administration, Goldman Sachs Global Investment Research, IMF Fiscal Monitor Database, Moody’s Investors Service, National Council on Public Works Improvement, Preqin, US Department of Transportation and World Economic Forum.
(1) A Gallup poll in 2011 found that 69% of Americans are satisfied with highways and 61% are satisfied with public transport. This places the US in the middle of the 32 OECD countries surveyed for infrastructure satisfaction.
(2) This report uses a broad definition of highways that includes all roads. In the context of federal spending, the term refers only to those roads that are eligible for assistance under the Federal-Aid Highway program – the 164,000-mile network of the National Highway System (of which the Interstate Highway System is a subset) and 1 million additional miles of urban and rural roads. The official name of the Interstate Highway System is the Dwight D. Eisenhower National System of Interstate and Defense Highways. It includes 46,876 of roadway mileage and is spilt roughly 35% urban Interstate and 65% rural Interstate.
(3) Amtrak demands 20% of the rail ASCE needs. The remaining 80% is privately-owned freight rail, which is excluded from public needs. About 80% of US dams are privately owned hydro projects and the remaining 20% are large federal dams owned by the Bureau of Reclamation. They take up an estimated 50% of spending needs for dams.
(4) An ASCE grade of B is considered good-to-excellent and has minimal capacity constraints but shows some general deterioration. The grades are based on eight criteria. The ASCE has consistently scored US infrastructure overall at a D or D+ for the period 1998 to 2017. In 2017, rail scored the highest at B and transit the lowest at D-. The grading concept was first introduced by the National Council on Public Works Improvement (NCPWI) in 1988, where it assigned an overall grade of C for US infrastructure. The first ASCE grades were then published in 1998 and have been done every 3-4 years since then. Rail includes freight rail and intercity passenger rail but not commuter rail, which is part of transit.
(5) The Democratic members of the Senate announced in January 2017 “A Blueprint to Rebuild America’s Infrastructure.” It called for $1 trillion of investment.
(6) The zero-lower bound (ZLB) occurs when the short-term nominal interest rate is at or near zero and causes a liquidity trap thereby limiting the capacity of the central bank to stimulate growth through monetary policy. This is when fiscal spending, especially public investment financed with debt, has the largest multiplier effect. The effective federal funds rate has averaged 6.25% over the last 50 years. The output gap is potential GDP less actual GDP as a percentage of potential GDP. It was -0.43% in 2016 and is expected to flat in 2017. In 2009 it was -4.1%.