CO2 per GDP

I presented the UN data for CO2 emissions equivalent in my environmental economics class on Friday. One student wondered (!*) what the ratio of CO2 equivalent to GDP looked like. Here is the chart with the Gg CO2 equivalents divided by real GDP (in millions) for the U.S. (click on the thumbnail for a bigger image):

 

Despite all the jokes and sarcasm and biases and cynicism, it appears that the CO2 intensity of GDP has been declining. The annual declines have averaged about 1.9%: -2.48 in 1998, -2.83 in 1999, -0.2 in 2000, -3.18 in 2001, -0.73 in 2002 and -1.98 in 2003.

Update: Could this be evidence of technological innovation?

*Note: I joke, actually, it has been a good class.

In response to some comments:

Update #2: The reduction in CO2/GDP could be due to the increase in the ratio of services to manufacturing in the U.S. economy.

Update #3: Given the current trends in GDP and CO2/GDP, CO2 would begin to fall in 2014 and fall back to 1990 levels in 2018. No doubt, current trends won't continue as all of the low hanging fruit is picked and diminishing efficiency returns set in.

Here is the data:

Year	GDP (millions)	CO2	CO2/GDP
1990	7112.5	5,046,059	709.46
1997	8703.5	5,747,512	660.37
1998	9066.9	5,838,752	643.96
1999	9470.3	5,926,125	625.76
2000	9817	6,130,768	624.51
2001	9890.7	5,980,065	604.61
2002	10048.8	6,031,568	600.23
2003	10320.6	6,072,181	588.36
2004	10650.47	6,121,192	574.73
2005	10908.88	6,143,224	563.14
2006	11167.29	6,159,263	551.54
2007	11425.7	6,169,310	539.95
2008	11684.11	6,173,365	528.36
2009	11942.53	6,171,427	516.76
2010	12200.94	6,163,497	505.17
2011	12459.35	6,149,574	493.57
2012	12717.76	6,129,659	481.98
2013	12976.17	6,103,751	470.38
2014	13234.58	6,071,851	458.79
2015	13492.99	6,033,959	447.19
2016	13751.4	5,990,074	435.60
2017	14009.81	5,940,197	424.00
2018	14268.22	5,884,328	412.41
2019	14526.63	5,822,466	400.81
2020	14785.04	5,754,611	389.22
2021	15043.45	5,680,764	377.62
2022	15301.86	5,600,925	366.03
2023	15560.28	5,515,094	354.43
2024	15818.69	5,423,269	342.84
2025	16077.1	5,325,453	331.24
2026	16335.51	5,221,644	319.65
2027	16593.92	5,111,843	308.06
2028	16852.33	4,996,049	296.46

Comments

[Mirko]

I guess that the problem here is that the assimilative capacity of the atmosphere has to do something with CO2 alone...I do not think atmosphere cares about our measure of energy efficiency, or does it?

[joshua corning]

this is a beautiful chart John.

I wonder what co2 per capita over time is?

[joshua corning]

Mirko,

"I do not think atmosphere cares about our measure of energy efficiency, or does it?"

well just imagine if that chart was flat but GDP still went up the way it did?*

anyway are you going to be the one who is going to tell everyone that they now have to be poorer? becouse i am not going to do it.

*(and yes I know GDP would not have gone up as much if that chart was flat becouse at least some of that growth is due to improvements in efficiancy)

[TFox]

Nice trend, let's hope it continues.

A chart along similar lines is the intercountry comparison, GDP per capita vs. energy consumption per capita. One version is here (see fig 2). The correlation is positive, and linear over two decades in wealth and power. The big question is how fast, and at what GHG cost, the developing world develops. Does anyone have data on the GHG cost of GDP increases in developing nations such as China and India?

[Dave Schuler]

Technology may be a factor but I'd bet that reductions in manufacturing here are a more significant influence.

[Robert Schwartz]

The incentives of the free market lead producers to eliminate waste in their operations. The inherent tendency of capitalism is to eliminate pollution.

[joshua corning]

"Technology may be a factor but I'd bet that reductions in manufacturing here are a more significant influence."

hmm i think you are confusing manufacturing jobs with manufacturing. i am not sure but total out put of goods in the US has not declined it just isn't the major sector of the economy as it once was...we still make as much if not more cars and shovels and crap as we always have we just do it for less money and with less workers then we did in the past.

Anyway I am not entirely positive so a confirmation or rebutle would be good.

[joshua corning]

"Technology may be a factor but I'd bet that reductions in manufacturing here are a more significant influence."

Just looked up US consmption of steel. In 1998 and 2004 US consuption hit all time historic highs. (no graphs sorry if anyone has any better luck please post it)

I am begining to be more confident that "reductions in manufacturing" should read "increases in manufacturing"

[Dano]

Now if we can just get the actual amount of CO2 in the atm to go down.

Best,

D

[Anon]

I don't think anyone doubts that CO2 intensity has been declining. That's why Bush's promise to decrease CO2 intensity 18 percent in future years is considered a joke -- it's a promise to stick to the BAU (business as usual) path.

From what I've read, CO2e/GDP has been declining since 1990 because the US has shifted from a manufacturing economy to a service economy. I'd imagine that we still consume goods with as much embodied CO2e, it's just that the production of those goods (and the associated emissions) is now in China.

Also, as Dano pointed out, declining CO2e/GDP isn't going to stablize the emissions in the atmosphere. We need to cut absolute emissions in order to prevent severe climate change. The whole "carbon intensity" argument seems to be a canard to distract from real solutions and undercut Kyoto.

CO2e emissions per capita is interesting as well:

Australia: 26
Canada: 24
US: 23
Russia: 14
Japan: 11
EU: 10
Mexico: 6
China: 4
India: 2

No wonder China and India want the US to start reducing before they make any promises. And not only are we emitting many times more per capita now, but we're also responsible for the lion's share of the accumulate emissions that have already built up in the atmosphere.

[John Whitehead]

Joshua,

The number of manufacturing jobs have declined in the US. Much of our steel is imported.

John

[John Whitehead]

Anon,

It's always fun to discover that I'm the last person to discover something!

John

[John Whitehead]

Dano,

Given the current trends in GDP and CO2/GDP, CO2 would begin to fall in 2014 and fall back to 1990 levels in 2018. No doubt, current trends won't continue as all of the low hanging fruit is picked and diminishing efficiency returns set in.

John

[Anon]

"as all of the low hanging fruit is picked"

What low-hanging fruit is being picked? Some companies have made voluntary reductions, but I don't think any state or business has reduced CO2e emissions enough be driving this trend.

There was an interesting article in Environment magazine in October 2002. The author, David Gardiner, presented the following data from an analysis by World Resources Institute:

BAU, based on trend from 1990-2000: 1.5 percent annual increase in CO2e => 1,700 mmtCO2e/yr in 2012

BAU + 18 percent decrease in CO2e intensity: 1.3 percent annual increase in CO2e => 1,600 mmtCO2e/yr in 2012

Assuming that analysis is correct, then how will continuing the trend of decreasing CO2 intensity lead to 1990 levels (around 1,300) by 2018?

John, can you present your analysis? How do you get back down to 1,300 by continuing out current trends? How long to get to the Kyoto goal (7 percent below 1990)?

[ralph]

John and all, although it's not **directly** on point, Tim Harford's "The Undercover Economist" has a chapter on the benefits of globalization. One of his big points, taken mainly from Jagdish Bhagwati, is that economic growth reduces pollution relatively. His specific examples are CO2 levels vs. foreign investment; but his larger point is that newer technology, generally, pollutes a lot less than the older technology. Quote:

"Trade helps both indirectly, by boosting growth, and directly, because free trade in poorer countries has been associated with the end of subsidies to heavily polluting prestige industries such as petrochemicals and steel as well as the import of new, clearner technologies." (220)

[Anon]

Data issues (I just noticed your 3rd update):

1) I think the difference in the CO2 numbers I'm looking at vs. yours is either due to mine including CO2e, or the old ton vs. tonne problem. Shouldn't really affect the analysis, though, because it's internally consistent.

2) Where are your GDP projections from? About a year ago, the CBO was predict Real GDP (in $2000) of 14,430m, which is higher than what you're citing.

3) I'm assuming you're starting with the GDP projection, applying the CO2 intensity decline to future years of CO2/GDP, and then backing out CO2 -- correct?

[John Whitehead]

Anon,

2) The GDP numbers are from the BEA. The projections are a simple linear extrapolation based on the 1997-2003 trend.

3) yep!

John

[Anon]

John, it looks like you're projecting GDP forward by about 2.2 percent (is that right?). But when I calculate average annual change in GDP from 1997-2003 using the numbers you listed above, I get 2.9 percent. [Numbers I get are: 1997-98: 4.2%, 1998-99: 4.4%, 1999-00: 3.7%, 2000-01: 0.8% (ouch), 2001-02: 1.6%, 2002-03: 2.7%]

You also seem to be using a decrease in the carbon intensity of around 2.2 percent. I think you get a peak in 2014 because your rate of GDP increase is slightly below your rate of carbon intensity decrease.

I reran your numbers out to 2028 using a GDP increase of 2.9 percent, and I don't get any peak in CO2, it just keeps going up. I admit I'm half-watching tv in the background, so maybe I've missed something, but the numbers seem screwy.

[Dano]

Ralph,

if you look at footprints or appropriation of Net Primary Productivity, the higher GDP countries appropriate more resources. That is, wealthy countries exploit more resources than poor countries do [as the comment above says.

You can do accounting tricks and move beans around, but not NPP or footprints.

It is certainly true that poor countries do some dirty things. But it's all about scale, which is what the implicit argument in your quote neglects.

Best,

D

[joshua corning]

"I'd imagine that we still consume goods with as much embodied CO2e, it's just that the production of those goods (and the associated emissions) is now in China."

yup i saw that coming 100 miles away.

you really should look at how much the US produces and then look at how much is imported from China before you say that.

anyway there seem to be a couple things going on here that REALLY need to cleared up.

1. Just becouse the service economy has surpased the manufacturing economy does not automaticaly mean that manufacturing has declined...i can also mean that (and is probably the case that) the service economy just grew faster.

2. less manufacturing jobs does not mean less manufacturing...it does not take a genious to point out it now takes less workers to make a car then it did 15 years ago. And it is even possible to increase manufacturing and reduce the number of jobs AT THE SAME TIME.

3. before I pointed out that steel consuption in the US is at an all time historical high..this leads me to belive that steal consumption being a good indicator of manufacturing and steal consumption being up that manufacturing has also increased. Ie the US produces more cars shovels and crap then it did 15 years ago.

4. If manufactuing is UP and the ratio of sevices to manufacturing is up could mean that that the reduction is a phantom...but it by no means indicates that the US is exporting it co2 emissions. It only means that we are making a but load of cash with services.

[joshua corning]

which is why i asked like at the start:

"I wonder what co2 per capita over time is?"

Which i think is a far more useful number

Anyway I also want to point out that in Johns GDP numbers: GDP will double starting in 1990 from ~7 trillion to ~14 trillion in 2018.
It will reach 16 trillion by 2026....happy 250 years A'murica.

[John Whitehead]

Anon:

I used MS Excel to make a linear forecast in GDP based on the 1997-2003 linear trend. The annual GDP growth rate is 1.6% = ((GDP2-GDP1)/GDP1).

Different assumptions about future GDP growth and the CO2/GDP trend will get you different results.

I'm not trying to say that I think this is what will happen. It is all speculative based on only a few years worth of data. The point I'm trying to make is that the CO2/GDP trend over the past few years might indicate some real progress.

I'll try to deal with the "low hanging fruit" assertion later this week.

Joshua: You are making good points about the mix of services and manufacturing. Ultimately it is an empirical question ... that would make an interesting research topic. What accounts for the decline in CO2/GDP? It is complicated enough that we won't find an answer here.

Thanks for the great comments everybody!

John

[Anon]

John, that makes sense, thanks for explaining it -- I've never used the "forecast" function in Excel, will have to check it out.

I was pushing so hard on the numbers because I think it's important to know what the assumptions are in this sort of analysis, even if it's back of the envelope. The idea that we can fix climate change by doing nothing, because of trends in CO2/GDP, is pernicious. Even if you're just throwing it out there for discussion... well, it needs to be discussed.

By the way, if I remember Okun's law correctly, annual GDP growth of 1.6 percent would suggest massive unemployment and economic decline. We'd probably have enough "hot air" emissions credits (a la Russia) that we'd be clamoring to join Kyoto or its successor.

[tc]

This is an interesting discussion.

You may be interested in checking out the proposal by the National Commission on Energy Policy - a bipartisan proposal that formed the basis of the Bingaman-Domenici climate change discussions in the Senate during the Energy Bill debate.

The proposal is to create an emissions cap that is based on an energy intensity reduction target. Here is an excerpt from the summary:

"Under the Commission’s proposal, the U.S. government in 2010 would begin issuing permits for greenhouse gas emissions based on an annual emissions target that reflects a 2.4 percent per year reduction in the average greenhouse gas emissions intensity of the economy (where intensity is measured in tons of emissions per dollar of GDP)."

They include a cap of $7 per ton in their proposal to prevent CO2 prices from getting out of control (i.e., to ease the economic risk).

There is a cool graph of what this means for emissions on page 22 of the final report (page 41 in the pdf).

Their business as usual case shows a constant growth in emissions. With the intensity reduction target, they have a leveling off of emissions around 2020. I'm not sure where their numbers diverge from the numbers you show in update #3 but the website has a bunch of documentation for those who want to dig deeper.