Hacker News Reader

> 1) Increase efficiency. Jevon's paradox makes this a fail, if you make consumption more efficient then you effectively make it less expensive and so people do it more.

I'm not sure I'd call this a "fail"; U.S. domestic energy usage per capita has actually fallen slightly since 1990, despite a massive explosion of new electronics and personal devices. (The EU has experienced a 1% per capita growth in energy consumption over the same period.)

People are certainly doing it more, but -- so far -- the recent emphasis on increased efficiency has been keeping this demand from translating into an increased demand for energy.

If improving efficiency alone can entirely halt growing demands for energy, then that's half the battle right there in reducing consumption.

Could this not be caused by increased income disparity, rather than energy use efficiencies? I would be interested in looking at people throughout the years with some kind of comparable inflation-adjusted income to see how their energy usage has changed over the years.

There is also the matter of rising energy costs. When people having less money + oil being more expensive does nothing better than keep consumption at approximately the same level, it doesn't leave much left to claim as the benefit of efficiency.

Mind you efficiency is still great because it allows us to derive more benefit for the same price (or the same level of resources), but it does nothing for climate change.

Domestic energy usage includes electricity, which is not significantly produced by petroleum (about 1% in the U.S.), so I don't see evidence that the rising cost of oil has had a significant impact on energy consumption.

The 2007-2008 financial crisis and subsequent economic recession were specifically mentioned as factors in the reduction of energy demand in the U.S. (and EU, if I remember right), but growth in energy demand had already started to decline before then.

The University of Michigan in 2007 started trying to follow the average MPG of cars sold in the U.S.; although I'm not familiar with their methodology, they've found a 4.5 MPG increase from 2007 to 2013, from 20.8 MPG to 24.6, for all cars, light trucks, minivans, SUVs, etc.

So, people would have to be driving significantly more now for those gains in efficiency to not contribute to reduced oil consumption, and driving significantly more doesn't really work well with an economic recession.

Have a look at the chart at http://www.investingdaily.com/13503/oil-demand-emerging-mark...: U.S. per capita oil consumption has been flat since about the 80s, a trend that transcends economic recessions and political conflicts.

I see no evidence at all for "Jevon's Paradox". Everything I can find about it suggests that it's a good-sounding economic theory that hasn't yet been comprehensively checked against available statistics.

> Domestic energy usage includes electricity, which is not significantly produced by petroleum (about 1% in the U.S.), so I don't see evidence that the rising cost of oil has had a significant impact on energy consumption.

And what about the rising cost of electricity?

> I see no evidence at all for "Jevon's Paradox".

Jevon's Paradox is a straight forward corollary of supply and demand. If you make the cost per mile go down, all else equal, the "demand for" miles driven will go up, which will erode if not exceed the savings from higher efficiency. The last decade is not representative because the "all else equal" bit is totally off: Gas prices have doubled from ~$1.80 in 2004 to ~$3.60 today and we've had a global financial crisis that suppresses demand for everything.

> And what about the rising cost of electricity?

See, this is why I prefer to argue from data in policy discussions.

The cost of electricity has been flat or declining since 1990.

I used the data from http://www.eia.gov/electricity/data.cfm, specifically http://www.eia.gov/electricity/data/state/avgprice_annual.xl.... I found that the EIA publishes prices in nominal amounts, i.e., not adjusted for inflation. So, I used the inflation calculator at http://www.bls.gov/data/inflation_calculator.htm, and then I plotted electricity prices from 1990 to 2012. That chart is at https://infogr.am/electricity-prices-are-flat. The yellow line is the raw data from the EIA: the average retail price, in cents per kilowatt-hour, of electricity for all regions and markets. The red line is the inflation-adjusted retail price in cents of the same. (Interesting aside: at a glance, it looks like it's almost entirely the unregulated service providers which are skewing the price of electricity upward.)

> Jevon's Paradox is a straight forward corollary of supply and demand.

A corollary is not data! This is exactly why arguing from ideology is bad: you found something that sounds good, and you believed it, and you never checked to see if it was true.

I insist that you find some data to support your argument, because so far, every single bit of data I can find suggests that Jevon's Paradox is complete bunk.

(aside: "annoyed" is my username there not because of this discussion, but because of the annoying way that infogr.am handles account setup.)

edit: after thinking about it a bit, I think residential electricity usage for lighting should be a great test case for Jevon's Paradox. Although residential lighting is only responsible for something like 1% of domestic electricity usage, there has still been a remarkable change in the efficiency of residential lighting since 1970. If Jevon's Paradox is true, we should see a flat or increased demand for residential electricity for lighting over the same period; if it is not true, as I suspect, we should see an overall decreased demand for electricity for residential lighting. That is: if you're right, then more efficient bulbs should mean that people are leaving the lights on more often or are adding more lighting to their homes, roughly in proportion to the increase in efficiency.

I chose residential lighting because the annual costs are relatively low, compared to heating and cooling (so trends should be resistant to changes in the economy), because bulb efficiency has changed quite a lot over the last several decades, and because more efficient bulbs have been accessible to people in most economic tiers.

Unfortunately, although I can find current data on residential electricity usage for lighting, I haven't been able to find any historic figures.

> I found that the EIA publishes prices in nominal amounts, i.e., not adjusted for inflation. So, I used the inflation calculator at http://www.bls.gov/data/inflation_calculator.htm, and then I plotted electricity prices from 1990 to 2012.

See, this is the problem with data. "Lies, damn lies and statistics." The period from 1990 to 2000 can't help you because electrical consumption significantly increased over that period.[1] From 2000 to 2012 you take an upward sloping graph and try to flatten it by adjusting for inflation. The problem is, it still shows an inflation-adjusted increase of ~10% from 2000 to 2012, and over the same period the inflation-adjusted median income was on the decline.[2] And that's before you account for the loss of personal wealth resulting from the dotcom bust, 9/11, the housing market crash and the consequent lack of credit availability. And even then, all you can say is that electrical consumption failed to increase significantly. There is no reduction over time, the number of MWh generated even increased slightly.

[1] http://www.eia.gov/electricity/data/state/ [2] http://www.davemanuel.com/median-household-income.php

> after thinking about it a bit, I think residential electricity usage for lighting should be a great test case for Jevon's Paradox.

> Unfortunately, although I can find current data on residential electricity usage for lighting, I haven't been able to find any historic figures.

Test cases are meaningless anyway. It's a mathematical equation. You can test the degree of the effect, but all you're really testing is the price elasticity of demand in a particular market. It won't be the same number for light bulbs as for air conditioners or for transportation.

Agreed, the only way to truly get rid of oil is to make it outdated by replacing it with better (where better equals cheaper mostly) energy technology.

>Subsidize research...

At some point (IMO already reached) this becomes just "Research". Properly directed R&D pays for itself many times over.

>...into alternative energy production...

And utilisation! Electric cars FTW!!

> At some point (IMO already reached) this becomes just "Research". Properly directed R&D pays for itself many times over.

The "problem" with research is that the costs are centralized while the benefits are diffuse. There is no way for a private entity to capture 100% of the social benefit of an invention, which means that solely private research will inherently be funded at a level below the optimal level for its social utility.

There are also some things that are so expensive to research that private capital is not willing to take the risk that they don't pan out. Making hot fusion commercially viable for power generation is the holy grail of energy research, but you can't exactly do it in your garage.

if you make consumption more efficient then you effectively make it less expensive and so people do it more

Will that really be a big factor in petroleum use? My driving is not dictated by the cost of Petrol (not yet anyway). I am not going to start driving twice as much if my car suddenly only uses 50% as much fuel.

You need to be thinking more macro: energy production via oil rather than, say, coal...China switching from bicycle s to gas scooters...etc

You won't drive twice as much, but that's not the only use case for petroleum. Consider how much more people would fly if airline prices went down due to cheap gas. Or think about how many more gas engines will be used to power other tasks if gas is cheap and plentiful.

> Will that really be a big factor in petroleum use? My driving is not dictated by the cost of Petrol (not yet anyway). I am not going to start driving twice as much if my car suddenly only uses 50% as much fuel.

Plenty of people might buy a large hybrid SUV if its fuel consumption was closer to that of a conventional compact car. Fuel cost goes into the calculation of whether to drive instead of taking the train, or carpool, or how much you're willing to pay in rent to live closer to where you work (or to mass transit), or how much you're willing to sacrifice in fuel to get home sooner, or whether to take a road trip, etc. etc.

It's also not all about cars. Coal and petroleum are used to generate electricity and heat homes. Make it cheaper to move the thermostat five degrees in the direction of comfortable and people will.