How Electric Vehicles and Autonomous Vehicles Are Destroying Our Roads, and Why Cities and States will have the Solution

Much ink has been spilled about the deterioration of America’s infrastructure in recent years: the American Society of Engineers grades our road system as a D, presidential hopefuls Hillary Clinton and Donald Trump debated infrastructure programs on the campaign trail, and The Economist asks “Is U.S. Infrastructure Destined to Crumble?

Even maintaining our current level of insufficient funding has proven to be a huge challenge. The $305 billion, five-year federal transportation bill signed into law in 2015 (after more than 20 short-term extensions over six years) maintains existing federal support to states at roughly the same level, $40 billion a year, and left the gas tax unchanged.

To make matters worse, modern transportation trends – namely electric vehicles and autonomous vehicles – conspire to make the traditional user fee approach to funding surface road infrastructure untenable.

As the transportation sector evolves, our funding mechanism for supporting roads must evolve with it. States and localities, as the agile testbeds for public policy, are well-positioned to scout the road ahead.

Many people think roads pay for themselves. They’re wrong.

The concept of a user fee to pay for roads is sensible enough: the more you drive on a road, the more you benefit from it and damage it through use, and the more responsible you should be for building new roads and maintaining existing ones.

GovEx calculation based on FHWA HF-10 data.
GovEx calculation based on FHWA HF-10 data.

And we do pay user fees when we use roads. (well kind of, but more on that in a moment). When you fill up your car with gas – assuming you have a gas-powered vehicle – you pay a tax on the fuel you consume to the federal government (since 1993 this has been 18.4 cents on gasoline and 24.4 cents on diesel) and to the state where you purchased gas (from 9 cents in Alaska to 50 cents in Washington State). When we drive over some bridges and through some tunnels, or even along some stretches of roadway, we pay a toll. That toll is often based on how many axles our vehicle has, which is a proxy for the damage we inflict on the road per mile.

All told, these user revenue streams now generate about $100 billion a year, about 42% of total revenue. Even at their peak share in 1967, user fees only accounted for at 75% of the share of revenue for roads.

Source: Federal Highways Administration, HF-10, 2014.
Source: Federal Highways Administration, HF-10, 2014.

The rest of the revenue comes primarily from general funds at all levels of government, especially at the local level. In 2014, $38 billion dollars came from local general funds, winning the battle for discretionary allocations that goes on between public safety, planning, community development, and other city and regional services. A significant portion also comes from bond proceeds, some of which may be thought of as a user fee, but that depends on how the bonds are paid off – based on road use or not.

Even if we assumed that all bonds are paid off through additional fees levied upon heavy road users and therefore constitute user fees, only about 55% of road revenue comes people who paid based on their road use. The rest of the bill is picked up by everyone: through income, sales, and property taxes, lottery ticket sales, and anything else that contributes to the general fund or investment income.

The set it and forget approach to funding transportation is a fallacy. Even maintaining our current roads system requires dogged and creative shuffling of revenue across government just to make ends meet.

An Equitable User Fee No Longer

While the gas tax and tolls were never sufficient to cover road costs, at least they were once equitably linked to use. When all vehicles were roughly the same, there was a linear relationship between how much we drove and how much gas we consumed. If everyone drives a car that gets 10 mpg and is powered by an internal combustion engine, then someone who drives 10,000 miles a year consumes 10 times the gasoline as someone who drives 1,000 miles a year, and pays 10 times as much into federal and state coffers through the gas tax.

Greater diversification of the vehicle fleet, increased fuel economy, and the failure of Congress to increase the gas tax lockstep, has broken the gas tax–user fee relationship. This is best evidenced through my own driving experience. I have a 2014 Chevy Volt hybrid that weighs about 500 lbs more than its exclusively gas-powered equivalent, the Chevy Cruze. My Volt gets about 75 mpg. That means that I pay less than half as much in gas taxes as a Chevy Cruze owner who drives the same distance getting 30 mpg, and I probably damage the roads marginally more because my Volt weighs more.  

As an ecotax, gas taxes may create an incentive to buy more efficient vehicles, after all who would want to drive the Chevy Cruze in that scenario? But as a user fee, the gas tax has become antiquated and it’s going to get worse over the next 50 years with further uptake of electric vehicles and autonomous cars.

The Electric Slide towards Delinquency

2016 was a landmark year for the electric vehicle. The plug-in electric vehicle (PEV) share of new autos sold in the U.S. nearly doubled from 1.4% in January to 2.9% in September, no small feat given that PEVs accounted for only 0.1% of new car sales back in January 2011. All told 111,000 PEVs were sold in 2016 YTD and over 500,000 PEVs are now quietly cruising American roads.

Sources: Argonne National Laboratory “Light Duty Electric Drive Vehicles Monthly Sales Updates” website, September 2016, and U.S. Department of Energy, Vehicle Technology Office.
Sources: Argonne National Laboratory “Light Duty Electric Drive Vehicles Monthly Sales Updates” website, September 2016, and U.S. Department of Energy, Vehicle Technology Office.

And the future looks increasingly bright for PEVs too.  Chevy promises delivery of the 238-mile Chevy Bolt this fall and over 400,000 Tesla Model 3 were pre-ordered when it debuted over the summer. As of September 2016, 17 automakers now have PEVs in their showrooms from BMW (maker of the i3 and i8) to VW (e-Golf).

Despite low oil prices, which reduce the incentive to purchase electric vehicles, the share of PEVs is expected to climb dramatically over the next 50 years driven by improvements in battery technology. Less expensive and more efficient batteries mean that electric vehicles will have even longer range and be more competitively priced with gas-powered cars in the years to come. Consistent with other projections, Bloomberg New Energy Finance suggests that the PEV share of new vehicles sold will climb from 3% now to 35% by 2040.

Visualization from: http://www.bloomberg.com/features/2016-ev-oil-crisis/
Source: http://www.bloomberg.com/features/2016-ev-oil-crisis/

Set against the backdrop of dramatically improving fuel economy standards, rapid electrification of the U.S. vehicle fleet creates a huge challenge for maintaining our nation’s roads.

As a back of the envelope exercise, let’s imagine for a second that we could snap our fingers and tomorrow 35% of vehicles on the road are electric vehicles (probably this will occur sometime around 2050 because it takes about 15 years for the U.S. fleet to fully cycle). Today, the sales-weighted average fuel economy of new vehicles is 25.3 mpg. Plug-in electric vehicles  have an average sticker MPGe rating of around 100. Tomorrow, when everything is the same except 35% of vehicles are PEVs, the average fuel economy will double to 50 mpg.

That’s half the fuel consumed to go the same distance and half the gas tax revenue received by federal and state governments. And this back of the envelope calculation ignores that the fuel economy of the non-electric fleet will continue to rise thanks to improved CAFE standards.

Gas tax revenue, once a $92 billion a year revenue stream paid by those who use our roads, drops to something closer to $50 billion. It’s as if the entire federal Highway Trust Fund, and its $40 billion annual contribution to states, disappeared.

Autonomous Vehicles Accelerate the Pace

In the transportation world, if 2016 is not deemed the “Year of the Electric Vehicle” then it should be monikered the “Year of the Autonomous Vehicle.” Tesla introduced autopilot, the Google Car topped two million miles traveled in autonomous vehicle (AV) mode, surveys were conducted to explore ethical questions through the lens of decision algorithms, Uber and Lyft developed partnerships to test AVs, and the Obama Administration released a Federal Autonomous Vehicles Policy with a model state policy to pave the way for broader adoption.

With dramatically evolving technology and new research partnerships popping up nearly every day, any projection on how AVs will affect our national vehicle fleet must be taken with a grain of salt. One estimate has about 4.5 million AVs on U.S. roads by 2020; another estimates that autonomous vehicle operation will be a standard feature on most new vehicles by the 2050s, with AVs accounting for potentially 50% of the fleet.

While the rate of adoption may be uncertain, there is a growing consensus that AVs will change the way our vehicle fleet looks and operates. Autonomous vehicles are likely to be primarily electric cars that will travel many more miles per vehicle, resulting in a smaller and more heavily utilized fleet, particularly in urban areas. One study found that, under certain conditions, AVs could reduce average vehicle ownership rates by 43% and would result in a 75% increase in the annual miles per vehicle.

The autonomous vehicles revolution will concentrate road use on a smaller number of vehicles. As mostly electric vehicles, they won’t pay gas taxes, despite using the road even more.  Electric vehicles and autonomous vehicles are like a one-two punch combination on user fees. Together, they severely limit the ability of the government to generate equitable user fee revenue using the existing framework.

New Roads Forward

States and localities are already taking action. Over the last three years, 20 states either increased gas taxes or pegged them to inflation. More bandaid than solution, gas tax increases address the need for more revenue, but do not tackle the fundamental problems outlined above. States and localities have the opportunity to explore more common sense approaches that both generate revenue and recognize the changing landscape of road use in the 21st century.

Vehicle Miles Traveled Tax

AKA: shouldn’t we pay based on how much we use the roads?

Just because the gas tax is an ineffective user fee, doesn’t mean user fees don’t make sense. Modern technology and changing norms have cleared the way for experimenting with vehicle miles traveled (VMT) taxes. It’s as simple as it sounds, VMT taxes levy fees on road users based on how many miles they drive. Oregon’s voluntary VMT tax, called OReGO, is the most well-developed program of this ilk. Up to 5,000 participants can opt into being charged 1.5 cents per mile driven instead of the state fuel tax of 30 cents per gallon (they receive a gas tax rebate).

In 2016, Tennessee has also created a VMT tax levied specifically on autonomous vehicles. As part of legislation establishing a certification program for testing autonomous vehicles, Tennessee created a mandatory 1 cent per mile tax on autonomous vehicles. The main challenges to a VMT tax are related to implementation. These programs require tracking every vehicle’s mileage using an onboard device or an enormous amount of trust in self-reported mileage, but these examples from Oregon and Tennessee suggest there might be something replicable here.

Transportation Taxes on Electricity

AKA: if I drive an electric vehicle, shouldn’t I contribute to the use fee too?

If the gas tax is meant to be a user fee, then electric vehicles aren’t paying their fair share. One option is to levy road use taxes on electric vehicles through the electricity they consume. As a fuel, electricity consumed by an EV is almost perfectly analogous to gasoline consumed by a gas-powered car. Of course, improved efficiency for electric vehicles will create the same problems that improved fuel economy has for gas-powered ones, so the tax would need to be pegged to some efficiency measure to be effective in the long term.

Rethinking Freeriding with Transportation Network Companies

AKA: if I’m a passenger for 20 miles in an autonomous Uber, shouldn’t I contribute too?

The advent of the transportation network companies (TNC), like Uber, Lyft, and Sidecar, that operate electric, autonomous vehicles will have a dramatic impact on the way Americans get from point A to point B. Right now, because TNC vehicles are operated by a driver who typically purchases gas, the vehicle owner contributes gas taxes as anyone else might. But as these companies progress towards electric AVs, there will no longer be a gas tax levied. The most efficient way to internalize the impact TNC vehicles operating on roads may be to charge riders a fee based on the distance they travel.

States and Localities Hold the Keys

It is clear that eventually our current user fee based system will fail us beyond recovery and we will need to conceptualize anew how our nation’s roads should be funded in the 21st century. While overarching trends towards electric and autonomous vehicles loom in the distance, there will still be funding battles to be fought over annual budgets in states, cities, and regional governments across the country just to make ends meet using the current revenue streams.

States and localities hold the keys to test new equitable approaches to raising revenue for roads, from authorizations in statehouses to test autonomous vehicles, to oversight on the operation of TNCs in their jurisdictions, to determining the rates imposed on electricity customers through public services commissions. Consequently, the task will likely fall on states and localities, our nation’s test bed for policy innovation, to rise to the challenge of developing a sustainable and equitable model for funding our road system.