Paradox of the day: Jevons’ Paradox

Posted on Updated on

New EU rules are going to make transport greener according to a report in Business Green today. Maybe, or maybe not.

But I was struck by the following sentence:

“The Commission said that for a long distance lorry covering 100,000km a year the changes would cut fuel costs by €5,000 a year and reduce greenhouse gas emissions by 7.8 tonnes.”

Can anyone spot the problem? If not you should acquaint yourself with the views of this man, Stanley Jevons:

As Jevons remarked “It is a confusion of ideas to suppose that the economical use of fuel is equivalent to diminished consumption. The very contrary is the truth.”


5 thoughts on “Paradox of the day: Jevons’ Paradox

    Jamie said:
    April 16, 2014 at 10:59 am

    Are you suggesting that a reduction in fuel costs will lead to an equivalent rise in activity so as to offset the saving?

    I reckon this is unlikely as there are many other factors which affect freight vehicle operations such as drivers’ hours, availability of orders etc.

    Freight vehicles will only be operated when they have stuff to transport and improved fuel economy of a vehicle won’t generate more orders (or indeed increase the availability of the vehicle to operate). Increasing economic activity might generate more orders but that is nothing to do with Jevons.

    Doubtless there will be some rebound seen but this should have been taken into account in the savings calculation.


    Dan Olner (@DanOlner) said:
    April 16, 2014 at 4:59 pm

    Jamie – have they taken it into account? In what way? You’re right that the intricacies of freight and logistics make it more complex, but Robert’s point stands: increased efficiency will probably lead to increased demand for transport even given other factors. E.g. there’s a tendency for third party logistics firms (and in-house for that matter) to shift more towards keeping goods on the move – reducing warehousing costs. Just-in-time etc etc. Increased efficiency / lower fuel costs makes that more possible – just one way transport intensity might increase as efficiency increases.

    There aren’t many goods whose consumption drops when the price drops (giffen goods, technically)…


    Jamie said:
    April 17, 2014 at 7:37 am

    I have no idea if they have or not because I don’t know where the numbers come from but they should have. Any GHG savings calculated for the European Commission should take rebound into account because it’s standard practice.

    I don’t agree that the point stands. The Wikipedia definition of Jevons states: “the increase in efficiency with which a resource is used tends to increase (rather than decrease) the rate of consumption of that resource.” This is an extreme version of the rebound effect known as backfire. It’s a total efficiency fail as the end demand is greater than before the intervention (i.e. rebound >100%).

    I contend that improving vehicle efficiency by X% will lead to a saving in total freight vehicle energy demand of X% * (1 – R%) where R% is the rebound effect and R < 100% (and most likely substantially < 100%).

    I don't see how JIT would lead to a large increase in activity. It's the same quantity of freight and that freight isn't moved around randomly just for the sake of keeping it out of the warehouse. It still goes from producer to distribution to end user but they just do it with less time spent in warehouses. To achieve this the logistics operator might transport smaller quantities of goods more frequently but they'll use a smaller vehicle for that (and presumably that vehicle will also have benefited from improvements in fuel efficiency). Smaller quantities of freight transported in smaller vehicles operating more frequently won't be as efficient as fewer trips in larger vehicles but it won't double the energy demand.


    Dan Olner (@DanOlner) said:
    April 17, 2014 at 1:08 pm

    Actually Jamie, yes, I think you’re right / I was wrong, though I’d think about it in terms of elasticities. Demand for transport is inelastic. Any efficiency gains (putting aside their development cost) means cheaper per-tonne-km good movement. Per-pound demand will increase as it becomes cheaper as competition passes that onto the buyers – but if price elasticity of demand is <1 (which I think it is), per-unit demand won't rise back up to the same level (it'd need to be unit elastic for that). Given there's a relatively simple one-to-one fuel/co2 relationship (and the fixed costs haven't changed), co2 output will drop.

    That's even more the case since fuel costs as a percentage of total freight costs look like this: so the effect of fuel efficiency savings are damped by overall costs.

    It was new to me that Jevon’s paradox was describing situations where price elasticity of demand > 1. These exist, of course, but inelastic demand is the norm in the energy sector AFAIK?

    An aside, though: “freight isn’t moved around randomly just for the sake of keeping it out of the warehouse. It still goes from producer to distribution to end user but they just do it with less time spent in warehouses.” It is moved around just for the sake of keeping it out of warehouses. Storage and stockpiling is expensive. Ref from Zotero: Rushton, A., Croucher, P., Baker, P., 2014. The Handbook of Logistics and Distribution Management: Understanding the Supply Chain. Has a load of great detail on that.

    But that doesn’t change your point, though, I agree: efficiency increases will cause a drop in price and an increase in per-pound demand, but that won’t use the same or more fuel than before the efficiency gain. Jevons doesn’t apply.


    Dan Olner (@DanOlner) said:
    April 17, 2014 at 1:41 pm

    Sorry, meant to stick toy example in. Completely arbitrary numbers, assuming fuel is the only cost. The thing being bought is transport – say moving a widget of set weight over one mile:

    Before efficiency gain: 1 mile, 2 gallon of fuel, $2.
    After: 1 mile, 1 gallon of fuel, $1…

    Demand for transport increases as price drops to a dollar per mile. If unit elastic, demand rises and two miles are bought – taking fuel demand back to its pre-efficiency level. In reality, demand is inelastic: demand increases to between 1 and 2 miles, but there’s still a fuel saving (now between 1 and 2 gallons not the previous 2) and thus a CO2 saving.

    Adding in the other variable and fixed costs makes the efficiency gains even stronger, as the overall price change is smaller, effecting demand less. But handy to see even if fuel were the only cost there’s still a CO2 gain.

    If Jevons paradox really is only about PED>1 situations, it doesn’t apply. But is that actually correct? I always thought it was just trying to point out the efficiency gains mean price drops mean per-unit demand increases, got that wrong too it turns out…


Comments are closed.