Saturday, August 24, 2013

The coming and increasing inconvenience of combustion engine vehicles

(NOTE: Unlike most of my posts, I want to keep returning and updating this one, so content will change over time. One critique I'll make of my original post is that I focused too much on range anxiety. The range anxiety issue is a subset of a broader issue of relative convenience of EVs versus gas engines. A virtuous cycle is just starting to happen, where EVs become more convenient and eat into the gas engine market, making them less convenient, spurring even greater EV adoption. More updates follow the original post below.)

The big challenge for electric vehicles is that we have the societal infrastructure to support gas engines but not electric. Despite that fact, in California the hybrid and EV plug-ins are now 1.8% of the market. Globally and nationally, estimates range from one to three percent of all vehicles in 2019-2020 will be the two types of plug-ins (see page 4).

The infrastructure challenge is real even if "range anxiety" is overhyped. To the extent that challenge to plug-ins is met and overcome, though, then the shoe is on the other foot. Every sale of a plug-in vehicle decreases support for the infrastructure supporting gas vehicles. When plugins are taking only one percent of the market, the decrease is insignificant. When they take three percent, it's starting to be significant, and we can expect the effect to increase and to be concentrated in some areas. If nearly two percent of California car sales are electric now, then what will be the percent in a decade, and what will be the situation in the San Francisco and LA metropolitan areas?

There are going to be fewer gas stations and fewer gas engine car servicing businesses. They'll try to adapt but plugins and especially EVs won't need charging stations in the same places and they won't need as much maintenance.

I think we may be about a decade away (or less) from hearing the first complaints of that gas stations aren't as easy to find as they use to be. What I'd call the "relative convenience factor" between ICE and EVs change not just as EVs get more convenient with longer range and more charging options. The convenience of ICE vehicles decrease as they lose market share and start losing parts of their support infrastructure.

The important thing about relative convenience is the process could be a virtuous cycle - ICE vehicles become less convenient as their infrastructure disappears with decreasing sales, so people buy fewer ICE vehicles and more EVs, and the ICE infrastructure thins out still more. The process may accelerate as investors hesitate to put money in the gas vehicle infrastructure, knowing they'll need a number of years to make a profit while their market is shrinking.

I know I'm pretty good at counting my chickens before they hatch, and I recognize that cars have a 10-year lifespan so it takes a while for the existing car population to reflect changes in sales (UPDATE:  per the comments, more like 20 years, although older cars are driven much less). Still, I can see this happening in a decade in some markets, especially where I live.

UPDATE:  from September 2013 -  Norway, especially Oslo, should be the first test of the virtuous cycle. Too bad I'm not an expert on Norwegian gas station businesses or we might already know the  answer.

UPDATE:  from a June 2016 post - Others noticing gas stations disappearing, and that EVs will have an effect. Norway and other countries discuss future gas engine bans, with varying levels of seriousness.

UPDATE:  from July 2017 blog post -

  • I think there's already a good argument that in many two-car households, having one vehicle as an EV is already more convenient than two gas engines.
  •  I agree that increased fuel economy has reduced gas demand and gas station numbers, and that in turn reduces the convenience of gas engines. By itself though, increased mileage will only go so far in knocking down station numbers. Absent further improvement, there should be a stabilized point where the number of stations balance with the new, lower demand.
  • I also agree that factors making land more valuable for uses other than gas stations are the primary motivators so far in reducing station numbers, particularly in dense urban cores. While extraneous, this also reduces gas engine convenience. Like increased gas mileage, it only reduces convenience up to a point as opposed to being a virtuous cycle.
  • Gas stations will (and have) reacted to find non-gas ways to boost sales. As they go further in that direction though, they'll have fewer pumps or no pumps, and the inconvenience will still increase. 
  • Gas engine repair and maintenance will also become less convenient - EVs need less maintenance and different equipment, so fewer mechanics will train on gas engines. Those repair bays will get replaced with expanded coffee shops and (possibly) electric charging stations.
  • While EVs have only a small impact so far on gas demand, they can affect what's happening right now based on people's expectations of the future. Take for example a family business that owns a dozen gas stations, with the parents nearing retirement and kids deciding whether to take over versus having a very different career. Those kids may well be concerned about what EVs will do to the business in 10-15 years (and should be) and tell the parents to sell instead. Anyone else thinking of a 10-year investment knows there's a risk that EVs will significantly hurt the resale value. These EV effects on gas stations are happening now.
  • Not an expert on this, but I'm guessing that ultimately there's not much long-term future for stand-alone EV charging stations replacing gas pumps, except on interstate highways. As EV range gets further and further above 200 miles, and as fast chargers become ubiquitous at work, shopping, and home, there just won't be a need except on highways where lots of people are traveling long distances. This could affect gas station viability over time.
  • And finally, the above is mostly predictive rather than policy-related, but if the virtuous cycle is real then it does have policy implications. Aggressive long-term EV targets are achievable and should be pursued because there's a virtuous cycle effect we have barely experienced yet that will make them work. Outright bans on gas engine sales like those proposed in the future for Norway and elsewhere will be politically feasible because the writing will be on the wall concerning EV superiority.

UPDATE March 2019: we have data! Gas stations and gas consumption in SF Bay Area, for future comparison.

UPDATE July 2019: still easier for a tourist to use an ICE vehicle rental than an EV in rural Norway.

UPDATE September 2019: updated gas station and consumption data for the SF Bay Area show significant decreases for both, despite the strong economy. 

UPDATE May 2023: Gas consumption in California shows a decrease, not yet for the US as a whole.


Anonymous said...

Tom Murphy has a related article:

John Puma

David B. Benson said...

It would be a good idea to somehow promote more use of trolleys, other light rail and even buses.


Meanwhile, back in the Bight Of Biafra, there being too few cars to create much gasoline deand, the locals are burning crude by the megabarrel in open pits to produce homebrew bunker oil they can sell on the waterfront.

Brian said...

John P - Murphy seems more skeptical than me. I'm somewhat skeptical of non-experts doing their own analysis, except as an exercise for its own sake. I assume his math is okay, but his data and especially his assumptions may not be.

David - it would be interesting to look at data from a century ago to see how the infrastructure supporting widespread use of horses in cities got replaced by trolleys and cars.

Russell - that sounds healthy.

EFS_Junior said...

Of course it would be useful to see if your utterly absurd specious argument holds ANY water? (PDF for the 2nd quarter of 2013 released in August of this year)

So in the form of an IF THEN statement;

IF the total number of vehicles sold in CA (year-over-year) exceeds the total number of 100% electric vehicles sold in CA (year-over-year) THEN the total number of gas stations in CA will increase or decrease?

Marketshare does not mean squat WITHOUT looking at the entire dataset of vehicles sold year-over-year. Go figure (e. g. Do the math!). :(

EFS_Junior said...

Just to make sure about my last post, the total number of vehicles sold year-over-year MINUS the total number of 100% electric vehicles sold year-over-year compared to the total number of 100% electric vehicles sold year-over-year. Got that?

Turboblocke said...

I bought a Nissan LEAF: I love it. Max torque from zero revs and no feeling that a vacuum cleaner is sucking my wallet empty when I floor it. I only visit a petrol station when I need to make a 1500 km trip every 5-6 weeks. I live in rural France, so recharging points are few and far between, but at least I'm using relatively low carbon electricity.

Brian said...

EFS - hybrid plugins create very little demand for gas or gas stations. I disagree with your initial assumption about 100% electric being the only ones that matter.

And while there may be some increase in cars due to population increases, but it won't keep up with a shift in market preferences to plugins.

Also, more people create their own barriers - driving five miles to a station in a dense urban environment is much more of an obstacle than it is in a low density suburb.

James Wimberley said...

Cars have a much longer lifespan than 10 years. The mean age of all light vehicles in the US is 11 yeas, implying a mean lifespan closer to 20. The switch will therefore be gradual.
Incidentally, this means that low estimates of carbon savings from EVs based on today´s electricity generation are nonsense. You need to guess what the generating park will be like 10 years from now - certainly much les carbon-intensive.

EFS_Junior said...


You know what? I really appreciate a well thought out argument, however yours is at about this level;,_Star_Bright

So let's begin again, shall we?

Let's start with your "they'll have to drive five miles or more out of their way each time they want to gas up" argument.

I have a square diamond shaped grid (cardinal street directions ENWS 1st, 2nd, 3rd and home bases). I can drive 5 miles E,N,W or S or 5/sqrt(2) NE,NW, SW, or SE. Regardless, the area turns out to be 50 mi^2. Capisce?

Now we should also add to that area the user's typical driving patterns from their ground zero per tank of gas.

Unless you think people are truly as dumb as rocks, like they go out to start their car from a full tank of gas and go nowhere until the tank runs dry, I believe that most people think about gassing up during their normal driving patterns.

Now, I happen to think that most people think about these things WHILE actually driving, you apparently not so much. Now suppose a typical driver on one tank of gas drives 10 miles in the eight cardinal directions (E, NE, N, NW, W, SW, S and SE), that is 9X the area of 50 mi^2 or 450 mi^2. Maybe you think that's too high, so let's make it a nice round 100 mi^2. Capisce?

Since it is rather obvious that "last gas station for the next 100 miles" only occurs in very sparsely populated areas (e. g. rural) and you have indicated densely populated urban areas ("dense urban environment"), let's look at this list;

So now, you being you, you'd probably pick Sitka or Wrangell AK, while I'd pick some real urban cities like NYC or OKC at ~300 mi^2 and 600 mi^2, respectively.

To believe your "they'll have to drive five miles or more out of their way each time they want to gas up" argument, NYC or OKC would only have ~3 and ~6 gas stations, respectively. ROTFLMFAO!

I can keep this up forever you know, at least when you try to make a bad argument, cite some literature, any literature, you know with at least some projections, seriously your "I think we may be about a decade away from hearing the first complaints of range anxiety and range irritation coming from gas vehicle owners in some markets" seat of the pants argument is so bad even Cartman says; LAME! WEAK!

It would appear as if in this specific case, your think tank ran dry along time ago. :(

Brian said...

I appreciate your willingness to switch to a completely different argument, EFS.

Here's an interesting post:

"'While rural communities struggle to fill empty stations, New York City has a different problem. Property values are so high that stations are being converted to more profitable uses, like high-rise buildings, giving drivers fewer places to fill their tanks. The city had 809 gas stations in 2011, down from 872 in 2006, according to the Department of Consumer Affairs. Of the remaining gas stations, only 44 are in Manhattan.'

Decreasing profits on gas, rising property values, and lower demand due to fewer and more fuel efficient cars are putting pressure on big city gas stations. No one is pretending that this is the beginning of the end, but it does seem like yet another negative feedback loop for cars in large cities like New York. Fewer cars mean fewer gas stations, which makes gas harder to find, which makes owning a car less convenient. "

EFS_Junior said...


Apparently you can't read your own post or even your 1st reply to my 1st post.

In both cases you directly mentioned the "driving five miles to a station in a dense urban environment" and "driving five miles to a station in a dense urban environment" or did you already forget that weak/lame part of your own argument already?

Meanwhile, back at the ranch, no real counterargument, it really does show your complete lack of critical thinking skills!

Oh, look, a squirrel!

NYC being too expensive (real estate-wise)for gas stations has NOTHING to do with electric cars or plug-in hybrids.

But well 44 gas stations in just Manhatten at a whopping 23 mi^2, not too shabby, that's ~2 gas stations/mi^2, where your really bad math would have it at ~0.01 gas stations/mi^2, so your off by over two orders of magnitude, even for your weak/lame attempt at a "Oh, look, a squirrel!" comeback.

Change the subject much? Meaning look in the mirror, Brian, before you accuse others of doing the very thing you just did. Project much?

BTW, I did say "I can keep this up forever ..." so here's a clue, I'm taking baby steps just for you Brian, because, well, you really are just that ...

Anonymous said...

Brian says "I'm somewhat skeptical of non-experts doing their own analysis, except as an exercise for its own sake."

The irony is thick as pea soup at this site.

But self-skepticism is healthy, as is exercise, so by all means, carry on.

Anonymous said...

It's a big jump from "fewer gas stations" to "range anxiety."

The reason range anxiety is an issue for EV drivers is that they don't have much range to begin with. Also, EVs are much more vulnerable to changing weather conditions.

When the low fuel light comes on in my vehicle I've got ~50 miles of range at my disposal (more if I'm careful); gasoline cars can carry that kind of reserve. Having to drive 5 miles to fuel up would be an inconvenience, but not really an emergency. Even if an ICE driver manages to run the tank dry, it's not too hard to get them running again.

Will some neighborhoods probably lose a few gas stations? Yes, but it will only be an issue in very specific circumstances, like people who commute within a big city every day but rarely take their car out of downtown (that's already madness on many levels, and fewer gas stations would merely reinforce the point).

Worth noting that many gas stations aren't making much money on fuel anyway; they rely on the pumps to bring traffic into the stores where they can sell cigarettes and coffee.


Brian said...

Anon - agreed.

Unknown said...
This comment has been removed by a blog administrator.