Tuesday, October 24, 2006

How many taxes will it take?

Here's a practical question for Pigouvians: How many Pigou taxes will it take to achieve the objective of relieving traffic congestion in cities? The reason for asking: A comment posted by anonymous the other day came to the defense of London's congestion charge.

I had said that London, which already has $4US a gallon gas taxes, also imposes a $16US daily congestion charge on vehicles entering the central core of the city. "And that doesn't seem to work much either!" Anonymous came back with a quote from a paper that claimed the charge, originally set at about $10US, had produced a significant response, reducing traffic congestion by 30%.

That's the claim, although a recent visit to London leads me to believe congestion is still an issue. If it was such a success, why did they have to raise the charge, and why is the city now talking about raising it again to about $20US? The tax is certainly a cash cow; maybe that's one reason. Another: Maybe it doesns't work all that well.

The latest report from the London authority actually says it's too soon to reach final conclusions. Also, the 30% reduction in congestion is a bit of a statistical game. It compares the amount of time it takes to travel one kilometre within the city before and after the charge was imposed. Buat it doesn't.

The total driving time for one kilometre pre-congestion charge was 4 minutes and 30 seconds. The first 2 minutes are considered to be normal. The congestion element is listed at 2 minutes and 30 seconds. The claim is that the congestion portion was reduced by 30%. But that's 30% of the 150 seconds per kilometre that make up the congestion portion, which means that the total congestion reduction is about 45 seconds off the time it takes to travel one kilometre. That's success?

The net effect: Before congestion charge the time to travel one kilometre was 4 mins and 30 secs. Time to travel one kilometre after charge: 3mins 45 secs. Actual reduction, or total time saved. equal to about 15%. That's based on the report after the first year of operation. The latest report reduces the numbers a bit. Bottom line analysis: If you were driving, say, 5 kilometres across central London in 2002, pre-charge, it would have taken 22 minutes. To travel the same distance today would take almost 19 minutes. That three minute reduction in travel time today cost $16US.

That's Pigovian tax policy in action. So Brits now pay two Pigovian taxes: $4US a gallon in gas tax and another $16US in congestion charge. But London is still bogged down in traffic.

So the fundamental question is: How many Pigovian taxes will it take to meet the objective of reducing traffic congestion?


Anonymous TheTsunami said...

how about proposing an alternative to the congestion problem, eh? if you reduce the cost of driving (by eliminating the taxes) the problem will get worse, even if the demand for driving in London is very inelastic.

10:54 PM  
Anonymous AB said...

Doesn't this show that the Pigouvian taxes are working fine, and if anything should be utilized further? If such a substantial increase in the costs of driving has had so little effect on the actual number of people traveling through the city center, it seems that people are able to handle this tax extremely well. If this is true, then there is little reason to call for an elimination of the tax, but perhaps a reason to increase the tax. An increase would bring one of two results: it could significantly lower congestion - the aim of the tax, or it could significantly raise state coffers if few people changed their behavior - thus not negatively effecting them.

8:03 PM  

Post a Comment

Links to this post:

Create a Link

<< Home