Fun With Research: Higher Fuel Prices Increase Economic Productivity

Todd Litman's picture

Last week I posted a blog, "Win-Win Transportation Emission Reduction Strategies: Good News for Copenhagen" which described emission reduction strategies that also help achieve economic and social objectives. I've continued doing research on the subject and made some additonal discoveries that I can report on now.

I've been investigating the relationships between various transportation policies and economic productivity, generally measured as per capita Gross Domestic Product (GDP). I had already found that GDP tends to increase with transit ridership and density, and decline with vehicle miles traveled (VMT) and roadway supply. This essentially validates general analysis on the economic benefits of agglomeration (the value of locating economic activities close together, which is why cities exist) and of a more diverse and efficient transport system (indicated by higher transit ridership). Then it occurred to me to test the effect of fuel prices on productivity.

Fortunately, data on fuel price is readily available at, and per capita GDP for individual countries is available from various sources, including some listed in a Wikipedia page. I started with the OECD countries (about two dozen wealthy countries), and found a relatively strong positive relationship between fuel price and GDP, particularly if the U.S. and Canada are excluded.  I decided to expand the dataset to include more countries. There are about 170 in total.

Fortunately, I found that I could copy country-GDP data directly from the website into a spreadsheet, but I still needed to hand copy the gasoline price data. That took about an hour. The results were not very impressive. Considering all countries there is only a weak relationship between fuel price and GDP (R2 about 0.1). So I tried a few variations, such as analyzing continents individually, which did not prove very helpful.

Then I tried a different approach. I divided countries into those that produce oil and those that do not. This made a huge different. Among countries that do not produce petroleum, there is a very strong positive relationship between GDP and fuel price, particularly if I excluded those with very low-income (GDP averaging less than $2,000 annual per capita), showing a high R2 value of 0.3633.

I think that the results, illustrated below, are facinating (for a better view and addional discussion see the full report, Evaluating Transportation Economic Development Impacts).

Fuel Price and GDP For Various Countries


Gross Domestic Product (GDP) tends to increase with fuel prices for countries that do not produce petroleum.  

These results are consistent with my current research which shows that excessive motor vehicle travel can reduce economic productivity, and mobility management strategies that improve transportation options and price transportation tend to increase productivity. This implies that relatively high fuel prices, and other vehicle user charges, can be applied without threatening economic development.

This is important for current debates about the economic costs of emission reductions. Critics claim that, because mobility is essential for economic activites, any effort to reduce vehicle travel is economically harmful. This research indicates otherwise. It suggeests that, beyond an optimal level, increased vehicle travel reduces economic growth because it is used for less productive purposes (a form of declining marginal benefit) and increases costs such as congestion, facility costs, accident damages and sprawl. As a result, in high VMT regions, mobility management strategies that improve travel options and more efficiently price roads, parking, insurance and fuel tend to support economic development. We can have wealth and a clean environment, if we focus on efficiency!


For more information see:

Win-Win Emission Reduction Strategies (

Smart Emission Reduction Strategies ( )

Are Vehicle Travel Reduction Targets Justified? Evaluating Mobility Management Policy Objectives Such As Targets To Reduce VMT And Increase Use Of Alternative Modes ( )

Evaluating Transportation Economic Development Impacts ( )



Todd Litman is the executive director of the Victoria Transport Policy Institute.



This makes me think

What would happen if the real value of gasoline was somehow inputed into your equation? I'm at the very least pretty sure that, if applied retrogressively, the true costs of our ravening sprawl M.O. maybe could be better quantified. Below are the results of some very quick off-the-top-of-my-head research:

( "If we look at the average annual Inflation adjusted gasoline prices for each of the following years (1958, 1968, 1978, 1988, 1998, and 2008 )we see the following Inflation Adjusted Gasoline Prices

"Year Price
1958 $2.24
1968 $2.11
1978 $2.16
1988 $1.75
1998 $1.35
2008 $3.23
2009 $2.28

"So if the long term average price is $2.37 then in 1988 gas was very cheap and in 1978 it was only slightly below average but in 1981 and in 2008 it was extremely expensive on a historical basis. In 1998 gas had gotten really cheap by historical standards allowing people to buy gas guzzlers like SUV's and Hummers. But that reversed in 2008 as prices rose above the long term average. As of this writing, the monthly average price of gasoline in November of 2009 was $2.61 just slightly above the long term average price of $2.37. With the Annual Average for 2009 at $2.28 being extremely close to the long term average."

Among other URLs, here's a pretty concise summary of gasoline's real cost:

( "Estimates on the "true" or "real" cost of gasoline vary by study and by year -- I've seen numbers ranging from $5 per gallon to $10 per gallon to $14 per gallon and higher. Over at the liberal opinion site AlterNet, Jason Mark notes that it is a conservative think tank whose research put real gas prices above $5 -- and that was a couple years ago. Presumably that number would only have risen since. A top-notch researcher here passed along some information about a graphic found in the June 2004 issue of National Geographic, alongside this story. I have not seen the graphic, but I'm told this is what it said (note: I have that issue and this is indeed what the graphic says):

"Congestion costs society about $1/gallon
Local pollution costs society about $0.80/gallon.
Global pollution costs society about $0.40/gallon.
Business cycle instability costs society about $0.12/gallon.
Leaks cost society about $0.12/gallon"

(As a quick sidenote: Simon Kuznets, the inventor of GNP/GDP, had this to say: "The welfare of a nation can scarcely be inferred from a measure of national income.")

David Parvo
Most Senior Fellow
The Placemaking Institute

What about oil wars?

I think that the cost of our ongoing, middle eastern oil war---hundreds of billions of dollars per year---should be figured into the cost of gas, but I won't hold my breath.


And remember what happened in the mid-1970s when the U.S. reached its peak production of oil (which is essentially when we began going abroad in order to satisfy our sprawl-driven GDP economies ravening thirst for oil) - what happens when we reach peak production at a global scale (which some argue has already occurred)? At that point we as a society will have no other choice but to decrease VMT and increase density, no?

David Parvo
Most Senior Fellow
The Placemaking Institute

A "truer" cost of sprawl?

Have begun endeavoring down the path this here post opened the door to...For those interested:

David Parvo
Most Senior Fellow
The Placemaking Institute

Brand new! Urban Grid City Collection

Each city has its own unique story. Commemorate where you came from or where you want to go.
Grids and Guide Red book cover

Grids & Guides

A notebook for visual thinkers. Available in red and black.

City Map Posters are in!

Available in 9 different cities.
Planetizen Courses image ad

Looking for the perfect last-minute gift?

A one-year subscription to Planetizen Courses is the gift that will continue giving all year long.