Summary

It is clear in that most infrastructure sectors in the US need investment. We expect there will be priority sectors that receive this investment in different ways. The areas most likely to receive incremental private investment first are those that have clear user fees, are critical to the economy and are operating at a low standard. These include highways, airports and ports. We rank air traffic control first on the list because efforts are already underway for it to be privatized.

The electricity grid is facing many challenges from new sources and uses of power and has been identified by Congress as a priority. We expect continued near-record capital expenditures by the utilities and additional government incentives to invest in the smart grid. Renewables have a well-functioning incentive plan in place by the states and the federal government and we do not anticipate any material changes. There is little federal commitment for high speed rail.

Infrastructure sector analysis

Below is an analysis of the main economic infrastructure sectors in the US using three approaches:

Firstly, there is an analysis of sector performance, which suggests where additional investment is needed based on those sectors performing least well. Secondly, we examine changes in infrastructure quality over the last 10 years and identify what sectors have driven those changes to highlight areas that need investment. Thirdly, there is a sector-by-sector discussion highlighting specific characteristics of each sector and the outlook for investment.

In our sector conclusions, we combine these three approaches to identify areas that are most likely to attract the most incremental investment over the coming years.

To work out areas of demand that need spending, Goldman Sachs did an analysis in September 2016 in which they analyzed infrastructure capability and performance vs. population density and other measures. The analysis looked at four sectors in 15 developed countries:⁽¹⁾

• Transportation – road transport availability and congestion, rail transport availability, high-speed rail coverage and air transport availability
• Trade – ports and logistics availability and cost
• Utilities – electricity cost and electricity transmission & distribution losses
• Communication – broadband availability, speed and cost

Areas that need investment: high-speed rail, broadband, roads and ports

The analysis then identified areas where each of the 15 countries scored high or low (1 best, 15 worst). Below are the key take-aways which suggest areas for investment in the US. These are the areas that scored the worst:

• Road availability very low at twelfth out of 15
• High speed rail coverage the lowest at fifteenth out of 15
• Port cost relatively expensive at eleventh out of 15
• Port flow very low at twelfth out of 15
• Broadband cost the highest at fifteenth out of 15
• Broadband speed relatively slow at tenth out of 15

High-speed rail has challenges in most of the US

The US is unlikely to make high-speed rail a major priority because of the poor economics involved, which stem from a lack of population density such as in Japan, China and most of Europe. Further, America’s unique model of suburban development, the strength of property rights that makes securing land very expensive, and an existing rail network geared toward freight traffic make the economics of high-speed rail in the US challenging.

A notable exception is the California High-Speed Rail (CAHSR) that began construction in 2015. The Phase I portion will link Los Angeles and San Francisco and is being designed to reach speeds of 220 mph. The project has encountered political, financial and legal obstacles since federal funding was authorized as part of the American Recovery and Reinvestment Act of 2009 (ARRA). Current estimates are that Phase I will be completed by 2022.

Another exception is Amtrak’s Acela Express that runs along the Northeast Corridor (NEC) between Washington, DC, and Boston. The NEC is the most heavily traveled portion of the American passenger rail system. In 2016, Amtrak announced plans to upgrade the Acela Express with 28 new trains capable of reaching 186 mph. The contract was awarded to French manufacturer Alstom, which makes France’s famed TGV or train à grande vitesse (high-speed train).

Although US ports rank poorly considering only cost and flow, they rank average among the 15 countries using the World Bank’s Logistics Performance Index (LPI), in which the US ranks 4.0 vs. a mean of 3.9 and a range of 3.7 to 4.2. The LPI tracks six port-related variables.

Assessing infrastructure quality changes

Another way to identify areas of demand is to look at sectors where infrastructure quality has deteriorated. Most developed economies have fallen in quality of overall infrastructure, according to the WEF and as shown in Exhibit 1.

The largest declines over the past nine years (as far back as we could get data) in the 15 advanced economies analyzed have been in Germany, Belgium and Australia. The only increases were in the Netherlands, Japan and Spain. The US ranked sixth out of 15 in 2007-2008 and seventh out of 15 in 2016-2017. Its fall in the nine-year period was lower than average and in the thirty-third percentile, i.e., 67% of the countries fell more.

Roads and airports drop the most in quality within the US

Germany’s weaker quality scores were driven by a broad deterioration in roads, ports and air transport. This stands to reason because Germany has been at the forefront of promoting fiscal austerity in Europe. It was one of only four countries in the Eurozone to have a budget surplus in 2016. The surplus was €23.7 billion or +0.8% of GDP, a post-unification high, and the second largest only behind Luxembourg at +1.6%. Belgium’s largest quality decline was in roads, more so than in Germany and 3x the average drop in roads in the 15 countries. Australia’s deterioration was mostly in air transport, in which the decline was almost 3x the average decline. The US decline was mostly in roads and air transport.

Air-traffic control and airports heading toward privatization

The US air transport sector ranks first in available seat miles per capita and fifth in the number of airports per capita, but the quality of US airports has declined due to a slowdown in real O&M spending since 2008 and a steep drop in real capital spending since 2004. The US air traffic control (ATC) system is in the process of being upgraded from a 40-year old radar system called Host to a GPS system called NextGen, which is expected to be spun-off into a private, non-profit, newly formed corporation similar to Nav Canada. Host and all US ATC operations by the Federal Aviation Administration (FAA) are funded with aviation gasoline taxes of 19.4 cents per gallon and jet fuel taxes of 21.9 cents per gallon. Civil aviation is a critical enabler of US growth. It accounts for about 5% of GDP, compared to about 3% for the UK and 3.5% for the rest of the world. It supports nearly 12 million jobs. Americans take more trips annually than any other nation, ranking seventh globally in air trips per capita.

US airports and airlines lag in performance

US aviation infrastructure moves more people than any other nation’s system – more than twice as many as China, with the world’s second largest number of passengers carried in 2016. Yet the US lags on performance indicators compared to other leading industrial economies. No US airport has made the top 25 of the latest World Airport Awards, an annual international customer rating of best airports. Only Delta Air Lines made the top 10 airlines for on-time service among the international carriers, according to FlightStats, a flight data services company. Four US airlines also ranked in the world’s top 10 for the oldest aircraft fleet.

Lagging performance by the US aviation system can partially be explained by lower investment in the last decade. Every two years the FAA submits a consolidated five-year plan to Congress on airport improvement, comprising project ideas for over three thousand US airports. The 2015 edition showed a 21% decrease in development projects. This is due to a variety of factors, including lower growth expectations and less traffic as passengers occupy fuller and fewer flights.

While US public aviation infrastructure spending on O&M has been growing slowly in the last 10 years, capital spending has declined markedly, from $21 billion in 2004 to $13 billion in 2014, as shown in Exhibit 2. Reasons for the fall in capital spending include a decline in the federal share of spending, project delays due to budgetary constraints and the recession of 2007-2009. Annual development costs at US airports are expected to plateau at $14 billion. Delayed infrastructure investment can be costly in terms of time lost due to airport delays. In 2010, FAA/Nextor completed a comprehensive study on the costs and impacts of flight delays in the US and estimated the annual costs to be $31 billion.⁽²⁾

The ASCE estimates total spending needs of $157 billion through 2025 for airports and air traffic control.

Prices for air travel have fallen 50% in the last 30 years

In the 30-year period after deregulation began in 1978, US airlines lost nearly $60 billion that included bankruptcy of all the legacy carriers. However, US airlines now rank among the world’s most profitable thanks to a boost from market consolidation and tight capacity. Airlines have focused on increasing fares, filling available seats and cutting costs while minimizing capacity growth. For example, load factors averaged 72-74% in the early 2000s but for the last several years have consistently been in the low-80% range. Travellers have also benefited as real prices for airline travel have fallen by about 50% in the last 30 years.

Paying for highways and bridges in need of repair

Highways account for the largest portion of the US public budget on transportation and water infrastructure at $165 billion in 2014, or 40% of the total. Real spending on highways accounted for 48% of the federal budget in 2014, virtually unchanged from about 45% in 1956 but up from 27% in 1980. In 2014, 66% of new federal capital spending went to highways, the largest share of capital spending. In 2003, real capital spending started to decline because prices of infrastructure-related inputs started to increase rapidly. This has resulted in a drop in real spending over the last 13 years.

Exhibit 3 shows the National Highway Construction Cost Index (NHCCI) increased 80% between March 2003 and June 2008, almost doubling the Producer Price Index growth of 42% for the same time period. Although it dropped post 2008, it is still up 68% through 2016. The index measures the cost of domestic highway construction and maintenance. It includes inputs such as asphalt, concrete and metal, which rose 107%, 61% and 45%, respectively, between 2003 and 2016.

State and local real government spending on highways, as a percentage of total transportation and water infrastructure spending, has also been in a steady decline for nearly 60 years. In 1956, it was 66% but in 2014 it was only 37%. One of the major problems is that revenues for highways come mostly from gasoline taxes but they have not been raised for nearly 25 years and there is little talk of increasing them. The solution is to change the way users pay for driving on US roads. One answer is broader tolling and a more recent suggestion is a mileage-based road user charge.

Tolling and highway funding

In the Northeast and mid-Atlantic region, many of today’s highways were originally toll roads, often built and operated by private investors. While tolling often made it possible to build or improve roads at minimal cost to taxpayers, many of these roads failed due to overly optimistic revenue expectations and an inability to attract sufficient investment to pay for improvements and competing capacity.

Over time toll roads came to be regarded as obstacles to the free flow of commerce. In 1916 Congress established the forerunner of today’s federal-aid highway program. It emphasized the principle that roads should be free and that “all roads constructed under the provision of this Act be free from tolls of all kinds.” Subsequent acts allowed some provisions for tolls, but after WWII states built toll roads without federal aid. The federal tolling prohibition was reiterated in 1956 with the Federal-Highway Act that authorized the Interstate highway system. It increased gasoline taxes to pay for the construction and established the Highway Trust Fund where the taxes were sent.

Highway toll revenue nationwide came to $14.35 billion in 2014, according to the Federal Highway Administration (FHWA). While the amount of toll revenue has grown significantly in recent years, toll revenue as a share of total spending on highways has been relatively steady for more than half a century, in the range of roughly 5% to 6%. On average, facility operators (or concessionaires) collected $2.36 million per mile of toll road or bridge in 2014. As of 2015, only 6,000 miles of roads, bridges and tunnels had tolls compared to just over 1 million miles (0.60%) of public roads eligible for federal highway aid and 227,000 miles (2.7%) comprising the National Highway System. All revenues from tolls flow to the state or local agencies or to private entities that operate the toll roads.

One way of estimating the revenue that could be raised by tolling the Interstate highways is to assume that the public would pay the same average annual amount per mile, $2.35 million, as is raised on existing US toll roads and bridges. In this case, tolling all Interstate Highways would be expected to raise roughly $112 billion per annum. Of this, approximately $8 billion is already captured by existing toll facilities on Interstates, leaving approximately $104 billion of new revenue. This would be more than enough to maintain and operate the proposed toll network. However, it is doubtful that this average could be supported over the entire length of the Interstate system because a large proportion of current toll revenue is collected on heavily traveled roads and bridges in urban areas. The rural Interstates that account for a majority of Interstate highway mileage carry far less traffic.

A less ambitious alternative would be to convert only the urban Interstates. Approximately 8.4% of the roughly 18,500 miles of urban Interstate highways are tolled already, leaving over 17,000 miles of roads available for conversion to toll roads. By assuming tolls would be imposed at rates that generate the current average of $2.35 million per mile, tolling the currently free urban Interstates could produce nearly $40 billion in annual revenue, nearly as much as the Highway Trust Fund now receives from motor fuel taxes.

Mileage-based road user charges as an alternative to tolls

A mileage-based road user charge would be a toll-like charge on each mile driven. It has been advanced as an alternative to the gasoline tax, which is a form of pay-as-you go user charge. In future years, the revenue-generating capacity of the motor fuel tax will be at best problematic as cars become more fuel efficient and as a new generation of electric cars and hydrogen-powered cars come online.

Most existing highway tolls are based on weight of the vehicle and distance traveled, and road user charges could be structured in a similar manner. Both electronic tolls and mileage-based road user charges could be used to implement congestion pricing, in which drivers are charged more for using a road at a busy time. Many widely criticized aspects of mileage-based charges have been less prominent than in discussions of tolling. Such criticisms have included the difficulty of accommodating drivers who lack credit card accounts to which the charges could be billed and concerns that the vehicle tracking system would invade drivers’ privacy.

Cost of $162 billion annually to improve roads

The ASCE’s 2017 Infrastructure Report Card estimates that congestion on US roads costs motorists $160 billion annually from lost time and wasted fuel. Forty-five percent of this cost is attributable to recurrent congestion that includes inadequate capacity and poor signalling design.

The National Transportation Operation Coalition (NTOC) gave the US a D+ grade in its 2012 national traffic signal scorecard. In 2015, the American Association of State Highway and Transportation Officials estimated an annual cost of $162 billion through 2020 to improve American roads to an acceptable level, as opposed to the ASCE that calculates the spending need to a good level. Of that figure, $85 billion annually is required merely to maintain the physical condition of existing roads.

Cost for bridges in need of repair $123 billion

The degradation of America’s bridge inventory is reaching a crisis point, with 9% of US bridges structurally deficient and nearly one in seven functionally obsolete. The average bridge is 43 years old, with a 50-year service life. Delays in rebuilding bridges are dangerous and further deterioration drives up costs. The federal government estimates the total cost for bridges in need of repair is $123 billion.

Inland waterways and ports need modernization

There are 25,000 miles of inland waterways and 239 locks that form the “water highway” in the US. This intricate system, most of which is operated and maintained by the US Army Corps of Engineers (USACE), supports more than half a million jobs and delivers over 600 million tons of cargo each year, about 15% of all domestic freight. Many of the commercially important waterways consist of the Mississippi River System – the Mississippi River and connecting waterways which includes the Gulf Intercostal Waterway. The Columbia River is one of the few rivers on the West Coast that is navigable for a significant length.

Barges convey large volumes of bulk commodities for long distances in a fuel-efficient manner

A principal value of the inland waterways is their ability to convey large volumes of bulk commodities for long distances in an environmental and fuel-efficient manner. Towboats push barges lashed together to form a tow. A tow may consist of four or six barges on smaller waterways and over 40 barges on the Lower Mississippi River.

A 15-barge tow is common on large rivers with locks, such as the Ohio, Upper Mississippi, Illinois and Tennessee rivers. Such tows are an extremely efficient mode of transportation, moving about 22,500 tons of cargo as a single unit. A single 15-barge tow is equivalent to about 225 railroad cars or 870 tractor-trailer trucks. If the cargo transported on the inland waterways each year had to be moved by another mode, it would take an additional 6.3 million rail cars or 25.2 million trucks to carry the load.

More than 50 percent of the locks and dams operated by the USACE are well beyond their 50-year design life. Many of the medium-sized 600-foot locks in the system (which comprise 60% of all locks) were built in the 1930s or earlier, including those on the Ohio, Upper Mississippi, Illinois and Tennessee rivers. Many of today’s tows operate with 12 or more barges. Passing through a 600-foot lock requires the tow to be cut into two or more sections. Such multiple cuts are time consuming and cause long lines.

Investment in the waterways system has increased in recent years but upgrades still take decades to complete. America’s outdated system of inland freight shipping networks causes expensive delays. Between 2000 and 2014, the average delay per lock nearly doubled from 64 minutes to 121 minutes and 49% of vessels experienced delays in 2014.

Delays cause far more losses than the costs of new and rehabilitated infrastructure. A 2012 National Waterways Foundation study concluded that each year of construction delay for new projects results in lost efficiencies amounting to 37 cents for every dollar in ultimate investment. Six years of delays more than doubles the cost of the investment.

The ASCE estimates total investment needs for US inland waterways over the next 20 years at about $5 billion.

Ports are responsible for the bulk of $4.6 trillion in exports and imports

Major ports in the US are owned by public port authorities but there is significant private sector involvement. In addition to providing many port services, such as pilotage, towage, trucking and rail, some public ports lease the entire operation of their terminals to private companies. The waterside of ports, e.g., deep-draft navigation channels and harbors, are owned by the federal government because they have navigable waters and support interstate commerce which is federally regulated. The landside of ports, e.g., terminals, storage and the connections to transport corridors for inland distribution, are generally owned by state and local governments.

The US has 926 ports that are essential to the nation’s competitiveness, serving as the gateway through which almost all overseas trade passes. Ports are responsible for the bulk of exports and imports, which totalled $4.6 trillion in 2016. Six of the nine largest ports are operating at near record highs for traffic. US international trade has been growing steadily as both exports and imports have strengthened. Exhibit 4 shows US port volume and measures the flow of containers from land to sea transport and vice versa.

As ships get larger, congestion at landside connections to other components of the freight network increasingly hinders port productivity. Similarly, on the water side, larger ships need deeper navigation channels – typically 45 feet or more – to be able to access a port. The Panama Canal Expansion allows ships that can carry 13,000 twenty-foot equivalent units (TEUs) to reach East Coast ports but only a few of the nation’s existing ports are currently able to accommodate ships this large. As bigger ships are built, the majority of existing port infrastructure will not be able to accommodate these larger vessels. (A TEU is a unit of cargo capacity based on a 20-foot-long intermodal container.)

To remain competitive globally and with one another, ports have been investing in their facilities and plan to spend $155 billion from 2016 to 2020 on repair, expansion and modernization. However, connections to these ports are also in need of modernization. These include roads, rail and inland waterways on the landside of ports and deep-draft navigation channels on the waterside. Landside connections are scheduled to receive $11 billion in new federal funding for freight improvements through 2020, which includes $2 billion for FEMA’s Port Security Grant Program, but baseline projected needs are $32 billion through 2025, according to the American Association of Port Authorities. The total estimated cost to fully modernize the waterside of ports through 2025 is a further $34 billion.

Electricity transmission grid faces a multitude of shifts and identified as high priority

The US electric power grid scores well, ranking fourth out of 15 developed countries in low power prices and average in transmission and distribution losses. Nevertheless, developed markets, including the US, have a very large need for power infrastructure to manage greater complexity coming on-line soon.

National power systems are facing a multitude of shifts, including penetration of solar and wind causing more volatility in the power supply, increased use of energy-efficient alternatives weighing on demand, a steady rise in electric vehicles increasing demand and the rising use of distributed generation. These shifts necessitate more dynamic grids. Mckinsey & Co. estimates that the power sector needs the most investment globally.⁽³⁾ (Distributed generation refers to power generation at the point of consumption, such as rooftop solar and micro-grids. Historically, this was performed by combustion generators.)

Exhibit 5 shows the estimated grid investment through 2019 and Exhibit 6 shows the breakdown of the grid investment, in which renewables are included in generation. The estimates are by the Edison Electric Institute (EEI), a primary industry trade group. The administration identified the US electrical grid as one of several high priority sectors in its 24 January 2017 executive order “Expediting Environmental Reviews and Approvals for High Priority Infrastructure Projects.”

Transmission investment is expected to remain elevated as the renewables build out continues. The EEI expects $21.5 billion in investment in 2017 and approximately $22 billion per annum in investment through 2020 at investor-owned utilities. This is nearly an all-time high and more than double the levels from 10 years ago.

The electric power transmission grid of the contiguous US consists of 120,000 miles of lines operated by 500 companies. The electrical grid that powers mainland North America is divided into multiple regions. The Eastern Interconnection and the Western Interconnection are the largest. Three other regions include the Texas Interconnection, the Quebec Interconnection and the Alaska Interconnection. Each region delivers 60 Hz electrical power.

The majority of assets in the grid are investor-owned by publicly listed corporations or privately-owned power generation companies and they finance their investment needs in the capital markets. In 2017, however, 37 states plus Washington, DC, took action to modernize their power grids to make them more resilient and interactive. The most common action that states took was in advanced metering infrastructure (AMI). Smart meters are a critical building block of smart grid deployment, which is a key component of the government’s push to modernize the US grid.⁽⁴⁾ The American Recovery and Reinvestment Act (ARRA) of 2009 appropriated $4.5 billion in matching funds and made them available to the states to invest in the smart grid.

Renewables to emerge as the major driver of spending in power

We expect a continuation of the federal tax equity incentives currently in place to finance sustainable energy projects and a continuation of incentives at the state level. Components of new tax reform bill have increased uncertainty in the tax equity market, however (see tax equity below). It is likely that the tax equity market would get a significant boost if US corporations repatriate their overseas earnings.

Investment of $25 billion annually expected through 2030 being driven by the states

Goldman Sachs expects US renewables to emerge as the major driver of infrastructure spending in the power sector. They estimate the US will add over 200 GWs of combined utility-scale wind and solar and distributed solar capacity through 2030, resulting in $25 billion per annum in spending and a 7% CAGR. This is up from their earlier estimate in September 2016 of 150 GWs of incremental capacity by 2030. By sector, Goldman’s estimates are that 90 GWs of wind (60 GWs prior), 60 GWs of utility-scale solar (50 GWs prior), and about 55 GWs of distributed solar (50 GWs prior) will be added by 2030. This increase is being driven by 30 states plus Washington, DC, now having mandated Renewable Portfolio Standards (RPS), and these standards are increasing.

In 2013, the highest RPS was 40% in two states. By 2016 six states had minimum standards of 40% or greater and five had 50% or greater. California recently raised its RPS to 60% by 2030 vs. 50% today and 100% by 2045. RPS require in-state utilities to procure a certain amount of generation from renewable resources by a given year or incur penalties. Eight states also now have voluntary renewable portfolio goals. Renewable Portfolio Standards are also referred to as Renewable Electricity Standards (RES).

The renewables build out is also being driven by improved economics from continued cost declines, albeit at a slower pace going forward vs. the 11% and 6% declines per annum for solar and wind for the period 2011-16 (see Exhibit 7). Goldman recently decreased its cost assumptions for utility-scale solar and wind from $1,500/kW and $1,800/kW, respectively, to $1,400/kW and $1,600/kW, respectively.

Increased usage of corporate purchase power agreements (PPAs) gives a third tailwind to renewable development driven by increased appetite for price certainty. The administration is expected to have little impact on the trajectory of the US renewable sector given that RPS drivers are at state-level. The positive economics of the tax equity market would only be disrupted if federal tax credits were repealed or new taxes levied.

Tax equity market on-hold pending clarity on the effect of tax reform

The tax equity market has been very strong over the last few years. This financing reached $12 billion in 2017 and accounted for 21% of the $58.5 billion in total renewable investment in 2016. There is concern that the lower corporate tax rate of 21% will decrease the present value of the tax credits. This would make it harder for developers/sponsors to earn their required rate of return because tax-equity investors, usually cash-rich banks and insurance companies, will require more project economics to make up the difference. A corporate tax cut was already being discounted by investors, however, who have been making tax reform assumptions in their models and the economics have still been strong (see inset).

The main concern now is the new Base Erosion and Anti-Abuse Tax (BEAT) designed to insure a minimum tax is paid by multi-national corporations similar to the Alternative Minimum Tax (AMT) for individuals. When a corporation’s tax liability reaches a threshold of 10%, then tax equity investments will not be able to further decrease its effective tax rate. Many investors in tax equity are not multi-national corporations, however. It is estimated about one-half of the investors in the market may be affected by BEAT.

The level of investment in this market is highly correlated with the level of US corporate profits. For example, in 2008 and 2009 when net aggregate profits fell to $4,243 billion and $4,812 billion, respectively, the tax-equity market slowed dramatically. (The recession of 2007-2009 affected the market as well.) In 2017, profits were very strong at $3,585 billion for 1H2017 and the consensus estimate compiled by Trading Economics is for $3,195 billion in 2H2017, for a total of $6,780 billion. This would be the second-best year on record, only behind 2014 at $6,973 billion. This level of profits bodes well for the tax-equity market, which is already very strong. If the $1 trillion in estimated cash profits held offshore is repatriated, it could give the market a further boost.

A noteworthy trend affecting tax equity deals in the energy market is a significant increase in commercial and industrial power purchase agreements that are showing up in the wind market. Traditionally, there have been commercial and industrial PPAs in solar distributed generation, but there are a lot more of these now in the wind space and they are expected to provide a further boost to the market.

Production tax credits may lose their inflation link

Another concern is the effect of tax reform on the production tax credit (PTC) that developers currently earn. The Tax Cuts and Jobs Act recently passed by Congress kept the planned phase-out of PTCs for the period 2016 through 2019 but eliminated the inflation link. This would lower the credit from $0.023/kWh to $0.015/kWh and cost the industry $12.3 billion between 2018 and 2027, or about $1.3 billion per year. This is relatively insignificant compared to the $13.6 billion in average annual investment in wind alone since 2006.

Natural gas investment expected to see a modest pick up

Natural gas is expected to see a modest pick-up in investment over the next four years, up from the trough levels of 2006-2016. The very low levels of investment over the last 11 years resulted from a surplus of capacity due to a wave of gas-fired power generation that came online in the early 2000s after the US power markets were deregulated. This pick-up in investment will be driven by inexpensive gas prices, a need for more quick-starting generation capacity to manage the intermittency of renewables, and the 40-50 GWs of coal plant retirements that occurred over the last few years. Gas-fired power generation overtook coal in 2016 and new combined cycle gas turbine (CCGT) capacity is expected to increase through 2025. The US Energy Information Administration (EIA) estimates that new CCGT plants are 25% more efficient than older coal plants.

In 2016, the US generated 12% of its power from renewables, up from 9% in 2005. It is estimated that this will increase to 26% by 2030. CO₂ emissions from power generation were down 25% in 2016 vs. their peak in 2005, largely due to the shift from coal to natural gas. This also marked the first time electricity contributed less to total CO₂ emissions than transportation. Emissions from power are expected to drop another 9% by 2030 driven mostly by renewables growth.

Strong capacity growth in liquefied natural gas expected over the next five years

Away from power generation, the natural gas market is expected to see significant growth in liquified natural gas (LNG) in the years ahead. This will be the second wave of the natural gas revolution in the US after shale, which began a steady climb in 2010 after the Renz 1 well was drilled in Pennsylvania and Marcellus production began to soar. In 2016, the US was the world’s largest natural gas producer at 750 billion cubic meters (bcm), with Russia second at 628 bcm. Other producers, from Australia to Qatar, were in the 71-190 bcm range.

There were only 15 countries consuming LNG 12 years ago and today that number has grown to nearly 40. This increased consumption pushed LNG global demand growth to 6.2% per annum between 2000 and 2014, outstripping the demand growth of 2.9% per annum in dry gas, according to the International Energy Agency.

Fifty percent of global planned capacity is set to come from the US

Five countries – Qatar, Australia, Malaysia, Nigeria and Indonesia – account for more than 70% of the world’s natural gas liquefaction capacity. Qatar alone holds nearly one quarter of the total. Current US capacity is very small at 2.1% of world capacity. This may be set to change. According to the International Gas Union (IGU), 50% of the 115 metric tonnes per annum (mtpa) of global capacity that is under-construction is in the US (58 mtpa) and 27% is in Australia (31 mtpa). This build-out is being driven by very favorable economics in the LNG markets, particularly in the US where dry gas is so cheap. In its 2017 World LNG Report, the IGU said it expects that the US will be the primary source of incremental liquefaction capacity growth over the next five years. It is less expensive for large importers, such as Japan, South Korea, China and India, to buy LNG in the US and import it rather than to buy it closer to home.

As shown in Exhibit 8, the current cost of transportation from the US to China is $1.20/million British thermal units (mmBtu) and the current US LNG spot price is circa $3/mmBtu, so the cost is $4.20/mmBtu vs. the current NE Asian LNG spot price of $7-9/mmBtu. Meanwhile, average LNG spot charter rates fell from about $150,000 per day in January 2012 to $30,000 per day in January 2015 but have since levelled off. These positive economics are expected to drive strong growth in the US LNG market.

Six LNG projects are under construction in the Gulf of Mexico and the US East Coast. Three of the six projects, both expansions and newbuilds and sanctioned in 2015, are expected to be online by 2019. Elba Island LNG began onsite construction in 2016. Apart from Corpus Christi LNG, the projects under construction are brownfield developments associated with existing regasification terminals. The US currently exports from only two trains at Cheniere Energy’s Sabine Pass export terminal in Louisiana on the Gulf Coast. Sabine Pass is a deep-water port and natural outlet from Sabine Lake into the Gulf of Mexico.

The US is in a strong position to dominate incremental LNG capacity growth in the years ahead because it has the most capacity globally under construction and it is the world’s largest natural gas producer. The economics in the LNG market are favorable and we do not expect any meaningful public spending. Investment in LNG will be led by the private sector.

Freight rail bottlenecks and insufficient passenger rail capacity

For more than 150 years the rail network has been a critical component of the US transportation system. Today it carries approximately one-third of US exports and delivers five million tons of freight and approximately 85,000 passengers each day. The private freight rail industry owns the vast majority of the nation’s rail infrastructure and continues to make significant capital investments – $27.1 billion in 2015 – to insure the network’s good condition. US rail still has clear challenges, especially in passenger rail that faces the dual problems of aging infrastructure and insufficient public funding.

Defining the rail network system

The total rail network is comprised of nearly 140,000 miles of track and over 100,000 bridges. The system is divided into two categories: private freight railroads and intercity passenger rail. Freight railroads consist of seven long-haul Class I railroads, 21 regional Class II railroads and 547 short-line Class III railroads. Intercity passenger rail is operated almost exclusively by Amtrak. More than 90% of Amtrak’s network runs on tracks owned by freight railroads, and to a lesser extent commuter railroads, and it pays the owner for its use. As a result, Amtrak relies predominately on freight railroad maintenance and system support to deliver quality, timely service. Commuter rail is classified with public transit by the CBO for spending purposes, even though it provides a different service. Public transit includes light rail operated in urban corridors (e.g., trams), plus buses, subways and metro systems. Commuter rail offers scheduled, short-distance travel on a non-reservation basis between central business districts and adjacent suburbs. Subways and metro systems are also referred to as rapid transit.

Declining system mileage but increasing freight ton miles causes bottlenecks and delays

Freight rail bottlenecks resulting from insufficient rail capacity cost the economy over $200 billion per year, according to the ASCE. These costs will only increase as more freight is moved over US railways, and the US Department of Transportation (DoT) projects a 32% increase in rail-based freight tonnage by 2040. Inadequate infrastructure compounds these problems in denser areas such as Chicago, where the average freight train requires 30 hours to traverse the city.

The cause of bottlenecks can be seen in Exhibit 9. System mileage (or track miles owned) for Class I railroads has been on a steady decline since the early 20th century due to increased competition from trucking. In the 1970s, system mileage fell sharply because of a lack of investment, due in part to heavy regulations by the Interstate Commerce Commission (ICC) and several railways went through bankruptcy. An example is Penn Central in 1970, which was the largest bankruptcy in US history at the time.

Ton-miles of freight moved, on the other hand, have increased steadily since the early 20th century, except in the last 10 years. As effective measures of supply and demand in the freight rail, these metrics illustrate the dilemma the industry faces: annual ton-miles have grown 200% from 1960 to 2015 but annual system mileage has dropped 52% during the same period. The result is bottlenecks and delays.

Growth, maturity, decline, rationalization and possible resurgence

US railroads have gone through a classic industry life cycle. The industry was in an extraordinary growth phase from 1860-1915. It was a national priority to build a transcontinental line running between the East and West coasts, and the connecting branches and trunks formed an interconnected rail system. System mileage peaked in 1916 at 254,000 miles of track.

The period of 1915-1950 was the age of rail transportation dominance and marked its mature phase. In 1930, the rail network accounted for 65% of all freight tonnage carried in the US and accounted for nearly all long-distance passenger transport. After WWII, competition to the railroads was beginning to take hold: automobiles and airlines weighed on passenger traffic and trucks negatively affected freight traffic. System mileage began to decline and unprofitable lines were abandoned.

The years 1950-2000 were a period of decline and rationalization for railroads as the core of economic activity shifted from manufacturing to services. The most important factor causing the industry to move from mature to decline was road competition. As a result, railroads got into financial difficulties during the 1970s and Congress deregulated the industry through the Staggers Act of 1980, which allowed railroads to set their own rates and service levels and sell or lease unprofitable rail segments. A total of 12,300 miles of track was abandoned and between 1950-1990 the number of scheduled passenger trains fell from 2,000 per day to 200 per day, a 90% drop. Rail transportation then became predominantly freight oriented.

As of the early 2000s, rationalization appeared to have been completed and transportation analysts began to expect a resurgence. A more efficient rail system emerged that was based on high-capacity, long distance corridors connecting maritime gateways and inland terminals. Freight rail began to see a surge in traffic linked to globalization and a rise in energy prices made it more competitive vs. trucking, as more locomotives switched from diesel to diesel-electric. But this all changed beginning in 2008.

Even though the DoT expects an increase in freight rail tonnage over the long term, it has contracted over the last 10 years. The largest contributor to this has been a steep decline in coal shipments, which fell 42% between 2008 and 2016. Even though shipments have fallen, coal is still the largest category of freight tonnage at 30% of the total. Intermodal also declined for the first time in 2015 due to heavy competition from trucking. Despite the drop in freight volume, rail still moves 40% of all freight in the US. As a comparison, freight rail accounts for 8% of total freight volume in the European Union.

Emerging opportunities for rail

Crude oil has become an emerging opportunity for freight rail as US domestic oil production has nearly doubled in the last 10 years. In 2008, Class I railroads originated 9,500 carloads of crude oil. By 2016, it had climbed to 211,985 carloads because many parts of the US and Canada lack sufficient pipeline infrastructure. Railroads have transported over 60% of North Dakota’s crude oil production, although this will decline now that oil is flowing through the North Dakota Access Pipeline. North Dakota has the second largest amount of proven oil reserves and production in the US, behind only Texas.

There are new proposals to electrify railroads and run them on renewable energy. Rail corridors could also be used as electricity superhighways to carry power from remote solar and wind installations to population centers. Rail electrification is common in other parts of the world. Around the globe, electricity serves nearly 25% of railroad track miles and supplies over 30% of the energy that powers trains. In the US, under 1% of tracks are electrified due to high upfront capital costs, an obstacle that publicly owned railroads in other nations do not face.

Railroads are well-positioned for this opportunity because trains can be more easily electrified than other long-haul transportation vehicles such as ships, planes and trucks. Rail is already one of the most efficient forms of ground transportation and it has an unparalleled capacity to provide clean freight and passenger mobility. Electrification would be accomplished in conjunction with track modernization and transmission lines would be run through rail corridors. Financing could be accomplished through P3 concession agreements.

Future transportation demand will grow and new rail capacity will be needed. Railroads are preparing for this now. Unlike trucks, barges and airlines, which travel mainly on infrastructure that the government provides, America’s privately-owned freight railroads operate almost exclusively on infrastructure that they own, build and maintain. In recent years, freight railroads have been investing their operating cash flow into their networks more than ever before. From 1980 through 2016, they spent more than $635 billion of their own funds on renewal, maintenance and expansion of their infrastructure and equipment. That is more than 40 cents out of every rail revenue dollar going to capital expenditures. Much of the emphasis has been on corridor and inland terminal development and efficiency gains.

The ASCE estimates the total cost of upgrading America’s rail infrastructure at $154 billion through 2025 but the industry has been spending more than this. Freight rail makes up about 80% or $125 billion of this amount. Amtrak has a state-of-good-repair investment backlog of $28 billion through 2025.

Sector analysis conclusions: air traffic control, highways, airports and ports expected to receive the most incremental private investment

It is clear that most infrastructure sectors in the US need investment. We expect there will be priority sectors that receive this investment in different ways. The areas most likely to receive incremental private investment first are those that have clear user fees, are critical to the economy and are operating at a low standard. These include highways, airports and ports. We rank air traffic control first on the list because efforts are already underway for it to be privatized.

The electricity grid is facing many challenges from new sources and uses of power and has been identified by Congress as a priority. We expect continued near-record capital expenditures by the utilities and additional government incentives to invest in the smart grid. Renewables have a well-functioning incentive plan in place by the states and the federal government and we do not anticipate any material changes. There is little federal commitment for high speed rail.

We include bridges and tunnels with highways and inland waterways with ports.

US highways and airports present the most significant long-term growth opportunity

According to Goldman Sachs, the most significant long-term growth opportunity in transportation for developed markets may be the US, where highways and airports are underinvested, still mostly in public hands and often unprofitable. US infrastructure is ripe to be opened up to private capital because of government budget constraints, underinvestment and federal politics. These impediments are diminishing as currently 70% of the states have P3-enabling legislation and two-thirds of those are now considering expanding the eligibility of assets that qualify for P3.

Less than one-half of one percent of roads in the Federal Highway System are tolled. There would be a strong incentive for private investment in US highways if tolling restrictions were liberalized. US airports could attract significant private capital if they moved toward European-style dual-till regulation in which retail activities are unregulated.

Acknowledgement, sources and endnotes

Sequoia would like to thank Michael Verhoeven for his assistance in preparing this report.