Monday, February 27, 2012

Why are Gas Prices High?


Gas prices are going up again, resulting in a lot of discussion by people who don't normally think about the oil markets, and therefore aren't necessarily that well informed about the subject.  As a certified oil-obsessive these last seven years, I thought I'd put up a "cheat sheet" with just the key graphs that would allow you to understand the major forces that affect the behavior of gas prices over time.

First, I've got this graph of gas prices adjusted for inflation:1


As you can see, gas prices are not as high as they were in 2008 or even last year.  However, they are going up and they are certainly much higher than they've been at most points in the last five years, and way higher than the happy days of the nineties.  Plus most people probably think in terms of the nominal prices (ie without adjusting for inflation) and that makes it seem worse:


It seems like a fair likelihood that gas prices will top $4 again soon, and of course regionally they already are:


No fair! the exasperated consumer cries.  So why is this?

The first thing to understand is that the price of gas over time is fairly closely related to the price of oil.  This next graph shows the weekly retail price of gasoline versus the price of oil (per gallon, vs the usual per barrel) for the same 1995-2012 period as above:


You can see that there's a very strong relationship over time. Technically, 97% of the variance of the price of gas is explained by the price of oil.  Less technically, the blue 45 degree line is what would happen if gasoline cost exactly the same as the bulk price of the oil in it.  The blue dots are what it actually cost.  The difference is what refiners and distributers and retailers get (what in oil industry terms is called "downstream").  You can see that by and large the downstream industry gets a minimum of about 75c a gallon - occasionally they are able to tack on an extra 50c or so, but not consistently.  The vast bulk of the price of a gallon of gas is going upstream - to the folks who own and extract the oil from the ground.

So that's the big evil oil companies, right?  Yes and no.  More no than yes.  This next graph shows the top ten countries as a source of oil:


(Data from BP).  You can see that the top two by far are Russia and Saudi Arabia, and OPEC countries dominate the rest of the list too.  Note particularly the prominence of unfriendly countries like Iran and Venezuela.  With the partial exception of the US, western countries are bit-players as oil sources at this point,  So most of the oil revenue is going to semi-friendly or actively hostile countries - and by and large all those countries nationalized their oil reserves long ago and either don't use western oil companies at all, or pay them a moderate fee per barrel and keep most of the profit.  So while it's certainly true that the big western oil companies (Exxon, Chevron, Shell, etc) make a lot more money when gas prices are high than when they low, those companies aren't the main cause of the problem.

The main cause of the problem is staring you right in the face in that graph above - roughly speaking oil production stopped increasing in about 2004.  Historically, most of the time2 as economies grow, more houses, offices, and factories get built on the edge of towns everywhere, people buy more and bigger cars, and oil demand increases.  Generally, when the economy increases in size by about 10% (as measured by GDP) oil demand increases by around 6% or 7%.  Usually in the past, oil production has increased as a result.  However, lately, it's become harder and harder to find ways to increase oil production.  In particular, you can see that both Russia and Saudia Arabia were increasing production in the late 1990s through 2004 and that was a lot of the increase in total global production.  But since 2004 they are roughly flat.

This means that the world is engaged in a bidding war for oil - prior to about 2004, there was quite a bit of excess capacity in the global oil production system and it was essentially a "buyer's market" in oil.  Since then it's been very much a seller's market.  You can see this very clearly in the price graph we started with:


Prior to 2004 or 2005, gas prices were relatively low and stable (under about $2/gallon).  Since then they've been high and volatile.  With flat global oil production and a still growing economy, someone has to be squeezed out and persuaded not to use as much oil, and it takes high prices to do that.  The "growing economy" part is particularly important in the developing world - a little oil makes a big difference to productivity.  The productivity enhancement coming from getting around on a small motorbike vs a bicycle is much bigger than the productivity enhancement that comes from switching from a car to an SUV, though the latter change requires much more oil.  Thus we find developing regions of the world are not slowing down much in their increase in oil consumption:


Comfortably over half of global oil is now going to these regions.  So the US and Europe, who traditionally used the lion's share of the planet's oil production, find themselves squeezed between flat global production and rising developing country consumption.

Under these circumstances, gas prices will continue to be high and volatile - and indeed will very likely go higher yet before we are through.  US and European consumers are going to have to conserve more and more oil - there's just no two ways about it.  There's a reason that just about every car company is experimenting with electric cars.

Finally, it's worth one last graph to explaining the detailed price movements over the last few years:


Footnotes:

1.  Technically, this was done using the CPI less energy from the BLS and adjusted to Jan 2012 price level.  Gas prices are monthly retail all grades national average from the EIA. February is extrapolated from the weekly data.

2.  The seventies were an important exception when major supply disruptions caused economic disruption, conservation efforts, and a reduction in oil usage.

10 comments:

Chris Reynolds said...

Thanks for the post Stuart,

It's good to see you point out the apparently counterintuitive issue of utility and it's impact on demand from growing economies.

RMD said...

Yes, nice post. Do have a question though: notice that the slope of the gas vs. oil price line is not 1 but 1.2. Why is this?
I can understand the offset - as you explain that is the profit for the gas refiners - but why should that "profit" go up as oil prices increase?

Stuart Staniford said...

RMD - the slope is 1.1 (1.1025 from the equation in the chart above) not 1.2. There's probably some inflation entering into it - that particular chart is not inflation adjusted (since it's not easy to inflation-adjust weekly data) and the low end is dominated by older data while the high end is dominated by more recent data.

Oale said...

You. do. not. have. high. gas. prices.

Alexander Ac said...

Yes, Americans do have high prices, but only relatively. And they are not as efficient as we (Europeans) are...

Alex

Mr. Sunshine said...

"If you think $4.93 per gallon is expensive, go outside and push your car 20 miles."

Stuart Staniford said...

Mr Sunshine - LOL - exactly.

Stephen B. said...

Mr. Sunshine, that comment deserves to be reproduced on a tee shirt!

dr2chase said...

Hah, I pushed my vehicle 20 miles yesterday, including up and over a 300 foot hill in the morning.

It's pretty crazy how much energy we're willing to piss away for the privilege of riding in giant armored boxes (cars, busses, or trains, a bicycle or an electric scooter beats all of them handily, and as it happens, yesterday on the ride home the bicycle was equally fast on a 10-mile commute -- a colleague left the building to catch a bus at the same time I left to hop on my bike, and even with a stop to take a phone call on the way, I still met him as he walked from the bus stop towards his home at the other end of our commutes.)

Idiolect said...

What is the effect of oil speculation on prices? People (Matt Taibbi etc..) who dismiss peak oil say the speculators cause the prices to rise. We need to deal with that argument.