Vehicle miles travelled grow in the US, but not fuel consumption

Arthur Berman on the site oilprice.com reports on an interesting statistic indicating two key trends that should be of interest to transport policy makers in the US.

This figure indicates that vehicle miles travelled are at a record high.  For some time, green transport advocates have claimed that what they called "peak car" had been reached, implying that there had been a generational switch away from growing car trips and mileage:

Gasoline sales and Vehicle miles travelled

The claim was generally an assertion, whereby it was thought that younger generations preferred to use public transport, were more environmentally conscious and so the age of the private car was starting to wane.  This appeared to be an overly simplistic interpretation of data which seemed to reflect that with the economic slowdown associated with the so-called "Global Financial Crisis" (GFC) and high oil prices, that this reflected a cultural trend (one which may seem apparent in some relatively affluent urbanised centres like Berkeley, but which was much more questionable outside the geographic and cultural locations of those who made the claim).   That doesn't appear to hold true.

It would be simplistic to say that the drop in oil prices has helped promote demand, but Berman says that the relationship doesn't appear to be that direct, although it undoubtedly helped.

US gasoline sales related to prices

This graph indicates that as prices declined, demand did not respond immediately, no doubt because for many the cut in that price was a saving that could be used for other expenditure, not necessarily more transport.    Further analysis in the report indicates a lack of price sensitivity around demand for gasoline, but this appears to be a bigger issue when it looks like gasoline is increasingly a less important element in road transport costs.

For those of us in the world of funding and charging roads, the most interesting statistic is how the rise in vehicle miles travelled is not matched by fuel consumption.  There was a 3% increase in total vehicle mileage over a year in 2015, but a 2% increase in gasoline sales.  Does that mean that increased VMT are resulting in only a 50% increase in fuel consumption?  It may be too soon to conclude quite that, but there is definitely a declining correlation between fuel consumption and distance travelled.

Inflation adjusting fuel tax wont be enough

What this means is clear.  Even inflation adjustment of fuel tax wont be enough to offset ever declining yields.   Higher traffic levels don't necessarily mean proportionate increases in road maintenance costs, unless the increase is from heavy vehicles.  Around 40-60% of road maintenance costs may be fixed, and unrelated to traffic volumes, but increases in traffic do tend to mean increased capital spending on improving capacity at bottlenecks.   So continuing to use fuel taxes to recover road capital and maintenance costs is not going to be sustainable (and inflation adjustment is likely to only to delay the inevitable by a few years).

Alternatives will have to be found, unless it is deemed politically and economically desirable to simply keep increasing fuel tax, with the costs of road infrastructure falling on a reducing proportion of road users.  Taxation of vehicle ownership has its own limitations, as it imposes deadweight economic costs that distort wider economic activities.   The most economically advantageous approach would be to move towards charging for road use.

Some jurisdictions are using tolls as a way of achieving this, although tolls may be viable for major highways and crossings, they will not enable efficient charging of all roads.  Only charges based on distance or time will do this.

So it will be the likes of Oregon and California, pioneering such options in the US, that will be ahead of this. Charging by distance or time (and by time I mean actual time on the network, not so much prepaying like a vignette in Europe), with factors for vehicle size and weight, will far more accurately charge for road infrastructure costs than a proxy such as fuel consumption.  There is also more potential to charge varying by location, to reflect infrastructure costs and time to reflect congestion factors, but these are neither necessary, nor always desirable.

The issue of fuel use vs. road use is one of sustainable revenues at the moment, it is increasingly going to be an issue of equity, for it is difficult to see why the motorist who buys a Tesla and pays nothing to use the roads, should be subsidised by the low income motorist with a twenty year old 6-cylinder car.

Long term trend towards less oil use - what does this mean for fuel tax?

An excellent article in Slate describes the long term trend in the use of transport fuels in the United States:

According to the EIA, the U.S. transport system required about 6 percent fewer BTUs of energy to function in 2014 than it did in 2007. And it used nearly 10 percent less oil than it did that year. In fact, oil consumption was lower in 2014 than it was in 2000. And as a proportion of transportation fuel, petroleum hasn’t been this low since 1954, when coal was still a significant transportation fuel. Petroleum’s market share has fallen from 96.5 percent in 2004 to 91.5 percent in 2014.

US transport sector energy consumption by fuel (Source: Slate)

So in proportionate AND absolute terms, there is less oil being used for transport in the US.  This is not because there is less tonnage being shipped (quite the opposite) or fewer people travelling, no it is because of fuel efficiency and the emergence of alternative fuels and motive power for road vehicles. Given that in 1954 the reason petroleum was a lower proportion was because railroads still used coal for steam locomotives (they virtually all use diesel now), the transformation today is considerable.  

4.7% of consumption is now alternative fuels including ethanol and biodiesel, 3.5% is now natural gas.  In 7.5 years the average fuel consumption of cars sold in the US has improved by 25%.

Although oil prices have dropped, there is little indication that this trend is about to be reversed, which poses an obvious question.  What is the future of fuel tax (or "gas tax" in the US)?

In the United States and a few other countries, fuel taxation is a direct source of revenue that is dedicated to government spending on roads and public transport subsidies.  However, in many other countries it is simply another form of taxation, although it tends to be acknowledged as a way of charging bluntly for the negative externalities generated by burning petroleum products.

Of course, lower consumption does mean some obvious benefits, from lower noxious emissions and contributions to CO2 emissions.  These benefits can be monetised, and there is an argument for maintaining this momentum, but it is unlikely that fuel taxes are the key driver for this, not least because they are so low in the United States compared to other markets (in some cases less than a tenth of the taxes in Europe).  

The growth of alternative fuels and greater fuel efficiency creates the obvious issue that revenue per vehicle mile is dropping, so what should be the response to this?  There are three broad paths that jurisdictions can go down:

1. Do nothing: This is, in effect, the default position of the US Federal Government, but also all those jurisdictions that do not increase such taxes.  The implication of this is that either funding for roads gets subsidised from non-usage based taxes (which is what the US Federal Government does), or funding gets cut, or those dependent on funding find alternatives (which in a federal-state relationship can mean greater use of tolls).  As easy as this option is, it lacks any strategic focus and is essentially an adhoc approach to charging road vehicles and raising revenue for highway infrastructure.  The mere fact that this has been the primary approach of most US state governments, as well as Federal, is a damning indictment on the political process to think strategically about the highway sector.  In Europe, given such revenues are rarely even partially dedicated to transport spending, it may be easier, although this simply means as a source of revenue, fuel taxes diminish over time.  It doesn't necessarily mean transport funding has to, but it does beg the long term question about reform of taxation.

2. Increase fuel taxes:  By increase, I mean go beyond inflation (which a few jurisdictions already so, such as New Zealand), and increase to take into account the real loss of revenue due to efficiency. This may have some initial appeal, as it is likely to encourage a more rapid shift to alternative fuels and more fuel efficient vehicles.  Yet the distributional impact of this is going to be more mixed.  It will impact those who are least able to afford new vehicles the most, and also those with fewer alternatives in terms of mode or trip consolidation (e.g. rural motorists).  On the other hand, it also means those driving alternatively fueled vehicles or highly fuel efficient vehicles are paying much less to use the roads, so the vague relationship between what is paid and the costs imposed upon the highway system (or capital consumed in using roads) reduces even further.

Of course, in many jurisdictions fuel tax isn't even nominally intended to be a way of recovering the infrastructure, let alone external costs, of highway use, although the economic case for doing so is clear.  When there are clear policy goals to avoid subsidising overuse of highway networks (because of congestion, pollution and lack of public funding for maintenance, renewals and network enhancements), it seems obvious that moving towards a user pays approach makes sense.  Fuel taxes may be a first, blunt, but low administrative cost way of doing this.  Yet, if the relationship between fuel consumption and road use widens more and more across the vehicle fleet, it simply means that some road users are paying much more than others, going above and beyond the pollution costs that may be fair to recover from those users.   Even if there were a hypothetical solar powered car, it consumes road space and road capital tied up in the infrastructure.  It should pay for using that infrastructure, even if the case to charge it for negative externalities is zero.

So, whilst increasing fuel taxes may be a short term palliative, it feeds the cycle of ever decreasing consumption of petroleum and inequities between those who pay more through such taxes, and those who avoid such taxes because they can afford to buy more fuel efficient vehicles.

3. Implement road usage charging:  In my view, the only solid case for fuel tax is as a carbon tax.  If it were set at an appropriate price to reflect this, it would easily be the most defensible way of recovering that cost.  It is, not, a good way of recovering the imputed costs of noxious pollution, because it isn't a tax on particulates or NOx or other emissions (as diesel vehicles emit more of these by-products that petrol vehicles, but emit less CO2 and use less fuel). Beyond that, it has long been established that the relationship between fuel consumption and infrastucture costs is very weak. One reason being that the greater stresses imposed on road infrastructure due to heavier vehicles is not reflected in proportionately higher fuel consumption to recover such costs.  Furthermore, it is clear that there is no relationship at all between the long run amortised capital costs of roads and the consumption of fuel of vehicles as they use them.

However, it is not the argument that charging vehicles through fuel taxes is inefficient and badly targeted that is driving investigation of options for distance based road usage charging in the USA, but rather concern about the sustainability and equity of continuing to do so.  Oregon is already well on the way to implementing such a system for the most fuel efficient cars, as an option. and other states are investigating how to move in that direction.  What it raises are a whole host of questions, not least how to transition from the current taxation of fuel to charging by road usage, and also whether there is a role for continuing to tax fuel (if only because of the environmental externalities).

 Conclusions

Many of the problems of highway management and infrastructure in the US today are due to a lack of funding, but also poor pricing and recovery of the costs of maintaining such infrastructure.  Looking at charging directly for use, rather than through the proxy of fuel taxes would help answer one element of this, and opens up wider questions of thinking of the links between using roads, paying for roads, spending money on roads and their management.  It isn't just an issue for the US, but in all jurisdictions where fuel taxes are a significant contribution to public funds.

Doing nothing may remain the easy option, particularly if the politics of raising fuel taxes are difficult, (and the politics of considering direct charging even more so), but what it ought to do is open up a dialogue and discussion about what roads are for, who should pay for them, how funding for them should be allocated, and where policy on highways should be heading.  I advocate an approach that more closely linked road users to decisions on funding, and how and by how much they are charged, because doing so has delivered greater benefits in other economic sectors.   It requires some strategic thinking that goes beyond how vehicles are taxed.

A small step for Oregon, a giant leap for the United States

The Oregon State Legislature has now passed the first legislation to allow for the implementation of a distance based road user charging system for cars on existing roads, in the United States.

Extensive background about the Oregon system was written and linked to in these articles:

Advocates of distance based road pricing look to Oregon in the coming months

Oregon progresses pilot of vehicle mileage tax

The SB810 Road Usage Charging Bill passed the Oregon State Senate by 24 to 6.

It passed the Oregon State House of Representatives by 48 to 12.

It now awaits the signature of Governor John Kitzhaber, who is understood to be fully supportive.

What it means is that Oregon will be implementing distance based charging for ultra fuel-efficient vehicles (those with greater than 55 MPG fuel consumption).

This provides a benchmark approach that some other states are likely to follow, whereby a small number of road users - those which pay little to no fuel tax - will be paying to use the roads directly.   It offers a long term path to shift vehicles from taxing energy use, to taxing road use.   Most Oregonians will notice nothing, unless they purchase such fuel efficient vehicles, and over the next decade or two, more and more will.

I don't have to explain to readers what a significant evolutionary step this is, and how long term the policy is.  A stark contrast to the approach of some states and countries.

I will be open, that I played a very small part in this project a couple of years ago, assisting the team at D'Artagnan Consulting LLP, which has been one of the key firms advising and developing the system architecture for the Oregon Department of Transportation.

However, those who know this project understand that it is, in particular, through the efforts of Jim Whitty - Office of Innovative Partnerships and Alternative Funding Manager at Oregon Department of Transportation - who over some years, with his team, have taken the best experience from across the US and across the world (dare I say that sometimes Americans can be accused of minimising the relevance of experience from other countries, Whitty did not), to develop a solution for Oregon.

I congratulate Jim his team, both within ODOT and his advisors and consultants, on this landmark success.

ODOT Road User Charging website

Florida advisory panel proposes distance based road pricing

Columnist Kingsley Guy in the South Florida Sun Sentinel has commented on the proposal from the Florida Metropolitan Planning Organization Advisory Council (take a breath) that the state introduce a 2/c mile tax on road use.

The proposal comes simply from concerns that fuel taxation is becoming increasingly less reliable as a source of revenue to fund the state's transportation expenditure.

The state has 12 different fuel/vehicle ownership taxes (plus two Federal fuel taxes), two of these taxes haven't been increased since 1941 and 1943 respectively (meaning they are now worth 7-8% of what they originally were set at).  3 are inflation adjusted.

Sources of Florida state transportation trust fund revenues

Black line shows where revenues have to be to remain constant in real terms

FMPOAC concludes that six options should be looked at.  These are:

- Index all fuel taxes to inflation;
- 1c optional Municipal sales tax;
- Increase State Highways fuel tax by 2c a year, above inflation;
- Vehicle Mileage Tax;
- 5c local Diesel tax for counties;
- Redirect part of fees that were shifted from transport funding to the State Transportation Trust Fund.

On VMT it said:


This recommendation is to have the Legislature commission and fund an extensive effort to deal with the systemic issues of fuel taxes becoming less sustainable as a primary surrogate for a transportation user fee. While fuel taxes served as an adequate substitute for a true user fee for decades, significant increases in mandated vehicle fuel efficiency and the introduction of all electric and plug‐in hybrid vehicles are eroding transportation revenues. It is recognized that there are significant concerns over the concept of charging users of the highway system based on each mile traveled. These include privacy of citizens, the cost of implementing such a system, and institutional issues associated with revenue sharing. This effort is intended to address these issues at a minimum, deploy a demonstration of the concept and develop a business plan and implementation roadmap to move Florida to a VMT‐based system.

Guy rightly rejects Virginia's bizarre hypothecation of a new sales tax to pay for transport, as if people who rarely drive should pay more for potholes (through their shopping) than a trucking firm does.  May as well pay for electricity through general taxation, or telecommunications.   

He notes that tolling in Florida has had a distance dimension to it, and that is widely accepted.

However, he also notes some important concerns:

- Privacy.  Suggesting an odometer only option.  Which is fair enough, as long as it cannot be defrauded.  However, privacy does need to be respected, not dismissed.  Too many wingnuts think distance based road pricing is part of a conspiracy  (one that would be far better implemented if done in cahoots with mobile phone operators!);

- Fraud.  He expresses concern about people cheating a distance based system, which is fair enough, but this can be addressed and has been elsewhere.  Enforcement is a key part to any such system;

- Visitors.  A big issue for any state only system is covering those from out of state quickly and efficiently.  I'd suggest some sort of vignette (time based charge), but there are other options too.

- Heavy vehicles.  He notes that fuel tax charges heavier vehicles more, and that there needs to be a reflection of weight as well.  He's right, and fuel tax does this very poorly, as it tends to charge the smaller trucks too much, and the heaviest ones not enough.  Notice that the trucking sector can often be very unhappy about better pricing of roads, partly for this reason.

- Transition.  He wonders if there may still be a fuel tax as well as distance charging.

He also notes that there remains an issue of poorly spent money, which is separate from that.

I'm encouraged that Florida is starting to look at this issue more seriously.  The state has done well to use tolls to a wide extent, which has helped it a great deal and will help it with any transition to more direct user charging.  However, it could do worse than look at progress in Oregon, and the lessons learned from distance charging in other countries as well.

Reduce oil dependency by introducing congestion pricing?

The Mobility Choice Coalition has released a report (PDF) that focuses on improving choice and competition in transport, but which also claims that congestion pricing and HOT Lanes could have a dramatic impact on oil consumption.   By 2020 a combination of HOT lanes and congestion pricing in the US is claimed to save 80 million barrels of oil p.a., and nearly double that by 2030 if pricing is even more comprehensive.  The options outlined in the report are basic (cordons, HOT lanes, distance charging and truck toll lanes), but the point is raising the idea in the first place.    It is ranked as the second best option in reducing oil consumption in transport (after a punitive fuel tax) and this neglects the wider economic and environmental benefits of pricing roads more efficiently.

The organisation itself appears to be a blend of free-market advocates, environmentalists and those interested in reducing the US transport sector's dependence on oil.

However, it is notable that more efficient pricing of roads is finally being acknowledged as a way of reducing oil consumption, with the reduction coming both from reduced demand, but also - critically - from more efficient and steady flows of traffic.