Get Into My Car: The Congested Future of Worldwide Auto Ownership


Given everything we hear about China’s economic miracle, you might think its vehicle ownership rate is about the same as that of the U.S. If so, you’d be partly right. China’s vehicle ownership rate is indeed the same as the U.S. rate — in 1916.

We’ve bought a lot of cars since then. So unless everything we know about the history of motorization is wrong, in the decades ahead China will too. And the impact of this will be large — not just there but here as well.

Just how many cars are we talking about? A recent paper by Joyce Dargay, Dermot Gately and Martin Sommer hazards some guesses.

You probably won’t fall out of your chair to learn there is a strong historical link between national income and national auto ownership levels. However, that pattern is not linear. Motorization proceeds at a pace that depends on the absolute level of a nation’s economic development.

Poor countries purchase few autos because very few people, even the relatively wealthy, can afford them. Up to national per capita GDP levels of about $4,000 (2010 dollars, GDP at PPP), the growth in car ownership has historically been quite low. For reference, $4,000 is about the per capita GDP of Honduras or the Philippines.

At about $4,000 a threshold gets passed. At this level nations begin to see the birth and development of a middle class (particularly, as Mark Kutzbach has pointed out in Access magazine, in nations with a fairly even distribution of income). So at $4,000 per capita GDP, auto ownership takes off, rising twice as fast as per capita income.

This rapid motorization slows when GDP/person hits around $14,000 (about the level of Mexico, Russia or Turkey); at that point, vehicle ownership continues to rise, but only at the rate of GDP/person growth.

Finally, at high national incomes (above perhaps $30,000, about the level of Italy, Spain or South Korea) auto ownership begins to saturate and level off. U.S. per capita vehicle levels have basically hit a plateau at about 800 per 1000 residents; since this is more vehicles than we have licensed drivers, we simply cannot handle more cars, except to the extent that our population grows.

Other rich nations have somewhat lower vehicle ownership levels than we do, but they are also starting to see the growth slow. The authors see European and other OECD nations peaking at about 650-800 vehicles, depending in part on their population densities and urbanization levels (less of each means more cars.)

And China? Currently its per capita GDP is about $7,500. That’s lower than that of Jamaica or the Dominican Republic (surprised?) and puts them right in the middle of the rapid motorization phase. Between 1960 and 2002 China rose from 0.38 to 16 vehicles owned per 1000 people.  Then, in the short period from 2002 to 2008, Chinese vehicle ownership more than doubled, to 37 vehicles per 1000. And that explosive growth rate seems set to continue into the distant future. The authors predict that by 2030, China will rise to an ownership rate of 269 vehicles per 1000 people.

It really wouldn’t be fair to talk about China without throwing India into the mix. Although the media sometimes makes it sound as if Indians are all a bunch of computer programming yuppies tooling around in sports cars thanks to the American jobs they’ve “stolen,” the sad reality is that despite India’s recent economic gains, as a group, the richest five percent of Indians have the approximate wealth level of the poorest five percent of Americans.

So with a per capita income of $3,300, India is just now beginning to reach the very first stages of mass motorization. It is lagging somewhat behind China, but still the authors predict its per capita vehicles level will rise from about 15 per 1000 people today, to 110 in 2030.

These per capita figures are large, and when multiplied by the gigantic populations of those two countries, the projected increases in total vehicles are colossal. The authors predict that by 2030 China and India will add about twice as many cars as the total number we currently have in the U.S.

By the way, we’re also going to see rapid growth in car ownership in other middle-income countries like Indonesia, Thailand, and Brazil. Those three countries alone are projected to add about half as many vehicles as we currently own in the U.S. between 2003 and 2030.

The bottom line is that vehicle ownership worldwide is expected to rise from about 800 million in 2002 to about 2 billion in 2030. And even then, Asia will only be at 15-45 percent of vehicle saturation levels, so rapid growth will continue into the distant future.

Prognostications of this sort are, of course, always dangerous. Rising congestion and fuel prices, or draconian government anti-auto policies, might slow the growth of motorization in the developing world. But past experience has almost universally proven that it’s impossible to put the auto back in the bag; even nations with excellent public transportation and hosts of anti-auto policies, like Japan or many places in Europe, have auto ownership levels that are fifteen to twenty times China’s today. Building a few light rail lines is not going to stem the tide.

This is certainly good news if you own stock in Toyota, but what does it mean for the rest of us? Next time, some thoughts.

Jill Blevins

If we were to design a transportation system, this wouldn't be it.

Joshua Northey

Some thoughts:

1) Population density is important! Auto ownership is not the same in very dense areas, China is very dense.

2) Owning a car is going to get relatively more expensive so the costs of using other modes of transit are decreasing comparatively.

3) There is a small but growing countervailing trend of people aware that we are consuming resources in a way that isn't going to work in the long run. I bike to work 6 miles one way every day, and at times have biked to work daily as much as 14 miles. This is still very rare. BUT I am reasonably sure I would not have been doing this if you say plopped me back 25 years but left the rest of the variables the same, and I am positive I wouldn't have been doing it 50 years ago. It also seems to be more common now than it was 11 years ago when I started.


One issue that doesn't seem to be addressed is that of capacity. Cars are not used in a vacuum...they require the existence of roads to drive on in order to maintain their status as a valuable tool...after all, a car that sits in a garage all day doesn't have much utility.

One of the places you are seeing this is in urban China, where population density is so high that despite a program of almost continuous road building over the past 10-15 years, the supply of roads is not keeping up with demand for road space. In major cities like Beijing and Shanghai traffic is usually bumper-to-bumper a significant portion of the day, and travelling by car is often the slowest way to get from point A to point B rather than the fastest. This despite an equally continuous program of subway building as well (Beijing will have gone from two subway lines to fifteen in the space of ten years), as well as the fact that the authorities are also placing restrictions on new vehicle registrations, thus placing an artificial cap on automobile demand.

This raises the point that in densely populated locales, it is simply not possible to build enought roads to keep up with automobile demand if it is left unconstrained. Personally, I don't see how it's going to be possible to build enough roads to facilitiate a rise in vehicle ownership from 800 million to 2 billion.



Regarding the "sudden" boost in chinese car ownership between 2002-2008 as compared to 1960-2002 a back of the envelope calculation reveals that while in the 6 years between 2002-2008 the car ownership doubled in the 42 years between 1960 and 2002 the car ownership in fact increased over 42 fold. Now, by my calculations that means that car ownership in china doubled 5 and a bit times between 1960 and 2002.
Basically what i mean is that while in recent times, in 6 years ownership doubled, in the previous 42 years ownership had doubled every roughly 8 years or a bit less.

Wouldn´t say thats such a drastic difference as the numbers make it seem.

Jaimeap least not as far as rate of icrease is concerned.
(when coupled with rising population it still huge)


I would have substituted the word "enlightened" for "draconian".

And let's hope that, like in Holland, high rates of ownership don't translate into high rates of use. Unfortunately in Shanghai there have been a few instances now of roads or lanes previously reserved for cyclists given over to cars.


If we were to design a transportation system, this wouldn’t be it. hoganspaccio


I certainly hope that China can take a lesson from its neighbors of Hong Kong and Taipei - comprehensive public transport is the way to go!

Hong Kong has an incredibly interconnected transportation system, which is admittedly aided by its density as a city. Meanwhile, Taipei, known for its roads jammed with mopeds, has had dedicated bus lanes within the city for quite some time and has embarked on a major subway building push in the past decade.


What is the Easiest way to spread the gospel of the late merge to non-believers?


It's surprising how we sermonize on mass transit systems to developing countries when it's almost non-existent here in the US!

Just because they happen to be in a position to buy cars now, doesn't make them any less (or more) deserving than us to the comforts of automobiles! So stop blaming others. We're also equally at fault. It's everyone's planet after all.

saverio manzo

no question there are many assumptions made here but there is one certainty: there will be more cars purchased as time ticks. As an investment, this certainly bodes well for the car manufacturers who produce and sell to China and the BRIC, but what about insuring or servicing a car? does this make for a good investment in the car or auto insurance companies, knowing that there will be millions more to insure over time?
saverio manzo