Estimating the End of Global Petroleum Exports Part 4 future global net export trends

The United States of America cannot afford to bet our long-term prosperity and security on a resource that will eventually run out.  Not anymore.  Not when the cost to our economy, our country, and our planet is so high.  Not when your generation needs us to get this right.

It is time to do what we can to secure our energy future.

So today, I'm setting a new goal: one that is reasonable, achievable, and necessary.  When I was elected to this office, America imported 11 million barrels of oil a day.  By a little more than a decade from now, we will have cut that by one-third.

I set this goal knowing that imported oil will remain an important part of our energy portfolio for quite some time.  And when it comes to the oil we import from other nations, we can partner with neighbors like Canada, Mexico, and Brazil, which recently discovered significant new oil reserves, and with whom we can share American technology and know-how.

Full text of Obama's energy speech: March 30

I think that it is every politician’s dream to know what will likely happen in the future, then set this as their goal, and when it happens, claim a great victory.  If the trend for declining exports continues, President Obama may get his wish of cutting exports by one-third without having to do anything at all.

Here in part four, I put together the results of my seven-region global analysis summarized in part 2 and part 3 of this series, and predict the future trends for global net exports. 

Global petroleum export trends

It is apparent from parts 2 and 3 that at present, of the seven regions considered, only four are net exporters: the Middle East (ME), Africa (AF), former Soviet Union (FS) and South America (SA).  The other three regions, Asia-Pacifc (AP), Europe (EU) and North America (NA), are all heavy petroleum importers and have been importers throughout 1965-2009. 

It follows therefore that the sum of regional net exports from ME, AF, FS and SA corresponds to global petroleum exports. 

Figure 8 shows the sum of reported regional net exports for ME, AF, FS and SA (open green circles), and, the sum of the predicted net exports based on my best fit NLLS analysis of these for regions, as described in parts 2 and 3 (solid and dashed curves). 

The predicted net export curves track the reported net exports fairly well throughout the 1965-2009 time range of the BP data set.  Based on the prediction curve, there were two peaks in global net exports: 11 bbs/yr in 1976 and 13 bbs/yr in 2007.  The peak value in total exports in 2007 is slightly less than the sum of net exports from all current (2009) exporters of 16,4 bbs/yr as estimated by George Lordos in Net Oil Exports Will Drop To Zero Long Before Oil Production Does (Table 1 of Lordos’s article indicates that the sum of net exports in 2009 equaled 44895 kbd or 16.4 bbs/yr).  I suspect that this 3.4 bbs/yr difference reflects the intra-regional exports that Lordos's analysis should include, while my present analysis only considers inter-regional exports, that is, net exports out of the ME, AF, FS and SA regions.

Figure 9 shows the predicted export curves (specifically, the "no sharing" scenario shown in Figure 8, as further discussed below) with the relative contributions from the four exporting regions.  I think that this figure nicely shows the dominant roles of the ME and FS as the number one and two export sources in the world today, and, going forward.  As I pointed out previously, Saudis Arabia and Russia, respectively, make up the majority shares of the net exports coming from these two regions.

ME in 2010 is predicted to account for 51% (6.5 bbs/yr) of the 12.7 total regional net export pool, with FS in second place at  28% (3.5 bbs/yr), AF third at 18% (2.2 bbs/yr) and SA at only 3% (0.44 bbs/yr). 

Figure 9 also nicely illustrates that AF and SA are both predicted to hit zero net exports in about 2018—only seven years from now.  So, by 2019, global exports are really down to ME and FS.

Sharing and no sharing scenarios for the ex-exporters AF and SA after 2018

The dashed line in Figure 8 shows the sum of reported exports for a scenario where, once a region becomes an ex-exporter, it just "goes away," and lives within its means, without attempting to import any oil from the remaining pool of global exports.  That is, the remaining exporting regions do not share their exportable oil with the ex-exporting regions.  For example, when AF and SA both hit zero-net regional exports in about 2018, the dashed line shows the remaining exports from ME and FS, assuming that AF and SA do not important any petroleum from ME and FS. Effectively, this means all the remaining exports continue to go to AP, EU and NA, until FS reaches zero net exports in 2033 and then ME reaches zero net exports in 2035. 

The solid line in Figure 8 shows the opposite scenario where, once a region becomes an ex-exporter, it is able import all of its needed oil from the remaining pool of global exports.  That is, the remaining exporters fungibly share their oil with these ex-exporters first before sharing with AP, EU and NA.  For example, when AF and SA both hit zero-net regional exports in about 2018, they are able to make up all of their predicted petroleum short fall (based on the best fit of the Hubbert equation to the consumption rate data) from ME and FS first, and then the remaining export pool of oil goes to AP, EU and NA.  According to this scenario zero net exports is reached globally in 2030.

Of course, I am not expecting either of these scenarios to actually occur.  Rather, something in-between is what I expect to happen.  The complete "sharing" scenario accelerates the time when global net exports end to 2030.  The "no sharing" scenario slightly delays the time when global net exports end to 2035.   Considering these two extreme scenarios therefore allows me to set the lower and upper time boundaries where global net exports are predicted to end—sometime between 2030 and 2035.

Despite the fact that I am combining the predicted export curves for four different regions (ME, FS, AF and SA) I find it remarkable how linear the declining sum of export curves looks through the 2010s and 2020s.  This also looks like a remarkably steep slope that I would not want to ski or even slide down on with a hard floor at the bottom, but I'm afraid that we won't have much choice in the matter.   

For instance, for either the "sharing" or "no sharing" scenarios, taking 2007 as the year of peak exports (100%), exports are predict to have declined by about 50% by about 2020.  That's an average annual decline rate of about -3.8 %.  For the "sharing" scenario the decline rate continues on at an even steeper pace hitting zero exports 10 years later in 2030—a -5%/yr decrease.  For the "no sharing” scenario, the decline rate is still steep enough to reach zero net exports 15 years later in 2035—a -3.3%/yr decrease

So, what about the prospects of President Obama reaching his “goal” of cutting foreign imports by 1/3 in a decade (although Canadian and Mexican oil are still okay it seems)?

Well, North American (basically USA) presently have net imports equal to about 44% of the oil it consumes (i.e., importing about 4 bbs/yr and consumes about 9.1 bbs/yr).  A 50% decline in the export pool (-2 bbs/yr) and assuming the consumption is down accordingly (7.1 bbs/yr) then only 28% of the oil will be from foreign sources (100 * 2bbs/yr / 7.1 bbs/yr) in 10 years.

What a victory! 

Only this victory is akin to saying that you have achieved a target of losing 50 pounds—by cutting off your right leg.  What would the economy of the USA look like if its consumption declines from 9.1 bbs/yr to 7.1 bbs/yr in 10 years?   A shambles I would say.  Certainly the prospects for growth look very dim.

What about Consumption?

Based on these export trends, I can also make an informed speculation about the future consumption trends for these seven regions.

Next time, I will discuss what the end of global exports could mean for the future consumption trends the seven different regions. 

Perhaps you will be surprised by which regions appear to be in the greatest jeopardy of having a total collapse in petroleum consumption.  Or, as the politicians will say, achieving the goal of energy independence!

Estimating the End of Global Petroleum Exports Part 3 Regional production, consumption and export trends for EU, FS, SA and NA

Here, in part three, I will finish my summary of analyzing the production, consumption and export trends for the remaining four regions in this study: Europe (EU), Former Soviet Union (FS), South America (SA) and North America (NA).

Europe

Figure 4 presents the reported production, consumption and net exports rates and the corresponding NLLS best curves for EU. 

Major producers in 2009 were Norway (52%) UK (32%) and Denmark (6%)—all of these being North Sea petroleum producers.  The top regional consumers were Germany (16%) followed by France, UK, Spain, and Italy, each of these responsible for about 10-12% of total consumption. 

 The reported production rates and consumption rates both have two distinct stages.  To model this, I divided the production rates into two stages corresponding to 1965-1989 and 1990-2009, and, I divided consumption rates into two stages corresponding to 1965-1982 and 1983-2009.  The best fit parameters of Qo, Q∞ and the rate constant "a," obtained from fits to these time ranges, are summarized in Table 4 below:

Table 4 summary of best fit parameter for production and consumption for EU
	Qo (bbs)	Q∞ (bbs)	a (yr-1)
Production 1965-89	0.54	37	0.18
Production 1990-2009	13	68	0.15
Consumption 1965-1982	24	166	0.15
Consumption 1983-2009	199	662	0.036

For production, the best fit parameters for the 1965-89 and 1990-2009 time ranges are consistent with 18 and 15 percent per year increases, respectively,  until the peak production at 2.5 bbs/yr in 1999 when production peaked.   Consequently, from 2000 onwards, the production rate has been declining at about this same rate of 15 percent per year.

For consumption, the best fit parameters are consistent with a high rate of increasing consumption from 1965-89 at 15 percent per year, followed by a much slower rate of yearly increase of 3.6 percent per year.  The all-time peak reported consumption rate was way back in 1979 at 6.2 bbs/yr, but by 2006, consumption was almost back to this level at 6 bbs/yr.  The the Hubbert equation best fit to the 1983-2009 time range suggests that consumption peaked in 2006 and now is slowly declining at 3.6 percent per year. 

Like AP, EU has not been a regional net exporter since 1965 and probably longer (if ever).  The reported net import rate follows a complex time course, as the multiple production and consumption rates peaked at different times.  Despite the recent trend for declining consumption, the need for imports is predicted to increase because production is declining (i.e., -15%/yr) at a much faster rate than the declining rate of consumption (i.e., -3.5 %/yr).   The bet fit export trend predicts peak net imports occurring in 2021 at -5.1 bbs/yr. 

Former Soviet Union

Figure 5 presents the reported production, consumption and net exports rates and the corresponding NLLS best curves for FS. 

Regional production in 2009 is dominated by Russia (76%) with Kazakhstan a distant second (13%).  Likewise, consumption is predominantly from Russia (68%), distantly followed by Ukraine (8%) as the second highest petroleum consumer. 

Once again, we see that both the reported production rates and consumption rates have two stages. To model this, both the production rates and consumption rates were analyzed as two separate stages corresponding to 1965-1996 and 1997-2009. The best fit parameters of Qo, Q∞ and the rate constant "a," obtained from fits to these time ranges, are summarized in Table 5 below:

Table 5 summary of best fit parameter for production and consumption for FS
	Qo (bbs)	Q∞ (bbs)	a (yr-1)
Production 1965-1996	14	151	0.12
Production 1997-2009	22	167	0.12
Consumption 1965-1996	9.4	104	0.12
Consumption 1997-2009	58	165	0.035

For production, the best fit parameters for the two time ranges are consistent with the about same increasing rate of production, 12 percent per year, and, about the same peak production rates of 4.6 and 5.0 bbs/yr, respectively, for the 1965-1996 and 1997-2009 time ranges.  Production in the latter time range is predicted to peak in 2012 at about 5 bbs/yr. 

For consumption, in the 1965-1996 time range, the consumption rate also increased at about 12 percent per year, peaking at 3.2 bbs/yr in 1984, which is the same year as the first production peak.  However, after declining by about 60% from the peak in the mid 80s, throughout the second time range, 1997-2009, consumption rates increased much more slowly at 3.5% per year, looks pretty flat in a range of 1.4-1.5 bbs/yr over this time span, with a slight peak predicted to occur in 2014.

Mostly because of its continued trend for flat consumption, FS has increased as a net exporter in the last decade (i.e., from 1997-2009, compared to 1965-1996).  The best fit export trend predicts peak net exports occurring in 2012 at 3.6 bbs/yr and zero net exports by 2033. 

South America

Figure 6 presents the reported production, consumption and net exports rates and the corresponding NLLS best curves for SA. 

The top regional contributions to production in 2009 were from Venezuela (36%), Brazil (30%) and Columbia (10).   The top regional consumers were Brazil (42%) Venezuela (11%) and Argentina (8%).

The reported production rate time course has two distinct stages, corresponding to 1965-78 and 1979-2009, which I modeled separately.  Like ME and AF, consumption rates in the SA are well modeled using a single Hubbert equation fit to the entire 1965-2009 time span.  The best fit parameters of Qo, Q∞ and the rate constant "a," obtained from fits to these time ranges, are summarized in Table 6 below:

Table 6 summary of best fit parameter for production and consumption for SA
	Qo (bbs)	Q∞ (bbs)	a (yr-1)
Production 1965-1978	16	48	0.14
Production 1979-2009	18	151	0.066
Consumption 1965-2009	22	277	0.035

For production, the best fit parameters for the two time ranges are consistent with the production rates rapidly increasing in the first stage from 1965-78 at 14 percent per year, followed by a slower rate of increase, 6.6 percent per year, from 1979 to present.  The best fit to the second time range predicts the peak in production occurring in 2008 at 2,5 bbs/yr.

For consumption, a steady increasing rate of consumption at about 3.5 per cent per year is predicted to peak in 2034 at 2.4 bbs/yr, well past the predicted peak in production.

Exports in SA nearly dipped to zero in about 1980 as production declined and consumption increased.  Since then, the export rate has gradually increased, but not nearly to the same peak of  1 bbs/yr in 1968.  The best fit to the present trend in exports predicts that exports peaked in 2002 at 0.6 bbs/yr, although the reported peak export year was 1998 at 0.75 bbs/yr.  The decline in production and the increase in consumption cross in 2018, signifying the year of zero net exports.

North America

Figure 7 presents the reported production, consumption and net exports rates and the corresponding NLLS best curves for NA. 

In 2009, the USA was the major petroleum producer (54 %) followed by Canada (24%) and Mexico (22%).  Previously, I have discussed the trend for a decline in Mexican production so these proportions may not hold going forward.  Consumption was dominated by USA (82%) with Mexico and Canada roughly evenly sharing the remaining consumption. 

The reported production rate time course has three distinct stages corresponding to 1965-76, 1977-90 and 1991-2009, which I modeled separately.  The reported consumption rate has two distinct stages corresponding to 1965-82 and 1983-2009, which I modeled separately. 

The best fit parameters of Qo, Q∞ and the rate constant "a," obtained from fits to these time ranges, are summarized in Table 7 below:

Table 7 summary of best fit parameter for production and consumption for NA
	Qo (bbs)	Q∞ (bbs)	a (yr-1)
Production 1965-1976	31	127	0.15
Production 1977-1990	55	193	0.11
Production 1991-2009	251	567	0.036
Consumption 1965-1982	45	252	0.12
Consumption 1983-2009	211	908	0.040

For production, the best fit parameters for the time ranges are consistent with two earlier periods 1965-76 and 1977-90, with rapidly increasing production rates (15%/yr and 11%/yr, respectively), followed by the present trend of decreasing production at gradual rate of about 3.6 percent per year.  The best fit to the second range predicts that peak production for NA occurred in 1985 at 5.5 bbs/yr. 

For consumption, after the rapid increase in consumption rates at 12 percent per year, peaking in 1978 at nearly 8 bbs/yr, consumption declined until 1984, and then started to increase again at a slower rate of 4 per cent per year.  From the best fit to the second time period, peak consumption is predicted to occur in 2013 at about 9 bbs/yr. 

The closest NA got to in the last half-century having no net imports was in about 1983, with net import of only -1.1 bbs/yr, down from the first maximum in imports of -3.1 bbs/yr in 1977.  Since then, regional imports have increased and are not predicted to peak until 2022 at -4.5 bbs/yr. 

Okay, that finishes my summary of the data analysis done on the seven regions.  Next time, I will discuss the implications these data and the best fit results has for global exports going forward.