View Full Version : Peak Growth

08-12-09, 09:14 AM
In Spain we have our local "Dr. Doom", ala N.Roubini, that is warning of the incomming crisis at least since Summer 2007. (Santiago Niño Becerra. A colection of its writings - in spanish - here: http://www.burbuja.info/inmobiliaria/burbuja-inmobiliaria/51307-articulos-de-santiago-nino-becerra-post-oficial.html) (http://www.burbuja.info/inmobiliaria/burbuja-inmobiliaria/51307-articulos-de-santiago-nino-becerra-post-oficial.html). He expects a systemic crisis linked to Peak Oil to start about Summer 2010. He calls what we've been seeing since 2007 the "pre-crisis".

Not really very interesting for me, since he makes predictions, but doesn't explain why. At least not in his posts; he should explain better to his students, maybe. He's a teacher at a private university in Barcelona.

However, there is an article written by him and other two professors that I've found interesting. Original here: http://www.asepelt.org/ficheros/File/Anales/2008%20-%20Barcelona/ANALES_2008.pdf, starting at page 365.

The basic idea of the article is modelling world GDP growth by a Hubbert logistic curve. And his conclusion, using an analogy with the 70's oil crisis as seen in this model, is that another crisis should start in 2010, and after that, annual world GDP growth will average 1% for the following decades in a range of -0.5% to 2.6%. And that is really a doom scenario :(.

I've included the article in pdf form with graphics, and then I post its google translation in pseudo-english. I'm interesting in comments about if the work is sound, and how it compares to itulip theories. Feel free to move to junk economics if you think it deserves it. My first feeling is that it is an elegant theory - not that I'm any kind of expert, of course.


This paper describes the evolution of global GDP and its apparent association to the evolution of oil
consumption. This relation prompts us to propose a model for the evolution of world GDP in the coming
years following the logistic curve used by Hubbert to fit the evolution of oil stocks. Adjusted data to the
model of global GDP in Geary-Khamis dollars yields a turning point in 2003, thus starting this year to
decline in growth rate. The forecast shows an average growth of 1% for a period starting in 2010.

Key Words: World GPD, economic growth, oil consumption.

Thematic Area: International Economics and Development.

1. Aim
Describes the evolution of global GDP and as it appears associated with the evolution
oil consumption, which is the main source of energy that is used and source
a huge number of raw materials. This partnership makes it possible to propose a model
for the evolution of world GDP in the coming years based on the model of Hubbert
for oil stocks.

2. The capitalist system and oil
Energy sources have characterized the different stages of evolution of the system
capitalist (Table 1). The burning of wood was replaced by the power of steam
water obtained by combustion of coal, resulting in the first
industrial revolution to the late eighteenth century, almost one hundred years later
began to market two new sources of energy: electricity and oil.
After the Second World War (1950) oil was the source of
massive use of cheap energy. Its consumption grew exponentially. Oil
became a source of a large number of materials involved in the
manufacture of many consumer goods that surround us daily. A
use no less important, but often ignored, to the point that it is very
difficult to find a good or service around us who may not have needed the
oil or any of its derivatives to put at our disposal, both for their
preparation for transport.


Currently the main sources of energy in our society are those arising from
fossil fuels (oil, natural gas and coal) and uranium, all subjects
with finite global reserves. Of which crude oil accounts for 35% of energy
primary world (IEA, 2007). The possibility that in a medium-term future to draw
Oil is not profitable and the demand clearly exceeds the capacity of
provision of supply, makes you talk of the end of oil and increasing concern
the future world energy (Rosell, 2007)

2.1. World GDP.

To measure the economic activity of a country and most significant is the use
Gross Domestic Product, one can study the evolution of world GDP as the sum of
the national GDPs, in which case we will have a global economic activity.
A recognized source of estimates of global GDP is the book of Maddison
(2003), in which values are collected for years (since 1500) and countries expressed in
Geary-Khamis dollar, currency unit that validates the value of coins
various countries (UNSD, 2005), a more detailed explanation can be found at
a document of the Maddison (2003).
From the data published by Maddison (2007) and estimates made by
the authors based on the relationship between dollars and gdk USD 1990 for the
period 2002-2005 has been prepared Figure 1 shows the evolution of GDP
worldwide since 1820, at the beginning of the current economic system. There is a growing
very fast since 1950.

[FIGURE 1: Development of world GDP. ]

2.2. Relationship between GDP and oil consumption.
Taking the data of world oil consumption published by BP (BP, 2006) and
graphically compared with the global GDP can be described as the building
Figure 2, it shows the growth of final consumption cheap price (1965 --
1973), the crisis between 1973 and 1979, a transition period that began in 1979 with a
decline in consumption until 1989 that the consumer recover the same level of 1979.
From the 90 consumption grows more slowly than before because of the 73
productivity growth in the use of oil, with some minor adjustments
in the slope. Furthermore, the world GDP growth seems insensitive to these
vagaries of oil consumption, but their behavior is not independent of
consumption as shown in Figure 3, it shows a strong association
linear relationship between two variables (over 99% of the variability of GDP growth
appears associated with increased consumption), structural changes are observed
giving rise to two trends, one between 1965 and 1979 and another
pronounced between 1985 and 2005 in the first period shows the effect of the first
crisis period of 1973-1979 and in the second period are slight changes in
1993 and 2000.

[FIGURE 2: Oil consumption (thousand barrels per day) and the evolution of world GDP (GK billion U.S. dollars) between 1965 and 2004. ]
[FIGURE 3: Relationship between oil consumption (thousand barrels per day) and GDP global ($ millions GK). ]

An interesting observation is that in the period 1985-2005 has increased
energy efficiency (more GDP per barrel consumed), compared with the previous period to
1973, but consumption continues to increase at a rate of 1.65% per annum on average in
past 20 years, which in 2005 was 1.36 times that consumed in

3. Models.
Here is the model of oil reserves and described by Hubbert
Based on the association between oil consumption and GDP is the same type of
equation to predict a possible behavior of the global GDP.

3.1. Estimates of oil reserves.
Laherrère and Campbell (1998) explain that in 1956 Hubbert proposed a model for
U.S. oil reserves based on a logistic curve and concluded that the maximum
production is to reach the 70s as it was. The model curve
Hubbert correctly explains global production. Hubbert (1956) also concluded
that by the year 2000 would be the best of global production. Forecasts
recently (Guse, 2007) put the maximum in 2007 and show that 2019 will be
spent 90% of the reserves.

3.2. Anticipating the evolution of world GDP.
Since the evolution of GDP is linked to the consumption of oil as shown in
Figure 3, and Hubbert's logistic model, although simple, has been described
acceptable form of the evolution of reserves, this type of model can be
used to project the evolution of world GDP. To do this, using GDP data
in Geary-Khamis dollars between 1950 and 2005 was adjusted by logistic model
described by equation 1

[EQUATION 1: PIB = PIBm x ( 1 / ( 1 + exp( b0 + b1 * t ) )]

t is the year minus 1950
GDP is the value of GDP for year t
PIBm is the upper asymptote, the maximum value that can reach as GDP
b0 and b1 are the coefficients of the exponential function of the model, whose value is
must estimate

The results of the estimation by nonlinear least squares, using
Statgraphics 5.1, are shown in Table 2.


Figure 4 shows the fitted model and the position of the asymptote.
The turning point of the curve is calculated

[EQUATION 2: t1 = -b0 / b1]

Is obtained by replacing the turning point corresponds to the year 2003, with a
confidence interval 95% between 1998 and 2008. The turning point of the curve
set up is similar to Hubbert's model, the "peak oil" which
Hubbert had estimated for 2000. The "peak oil" is a milestone in reserves
oil and that from that moment began to decline, so that for
years aims to study and debate (Bentley 2002), which would identify that
year as the time when the growth trend of the economic system
pass through a turning point, beginning with a new trend identified
possibility of a much lower growth.

[FIGURE 4: Setting a logistic model to describe the evolution of world GDP]

In Figure 5 shows the percentage change in GDP (growth rate) between
1950 and 2005 along with the trend of this change and would be expected from
logistic model adjusted. The data indicate two levels of change and notes that
the model is adjusted away from the first level (5% between 1950 and 1973) is about
the second level (3% between 1973 and 2005), crosses (1993) and leaves falling.

[FIGURE 5. Data and trend of variation of world GDP between 1950 and 2005 based GKD a dollar and its relationship to the adjusted logistic model.]

The model suggests that by 2010 the difference between the level of growth
period (3.0%) and predicted by the model (2%) is very similar to that
to 1973 (5% growth on average since 1960 compared to 4%
predicted model); from that year the average level of growth for the
period 1975-2005 was 3%, if the pattern is repeated at an average level
world GDP growth would be reduced to half in value between 1.0% and 1.3% for
a period beginning in about 2010. If the variability of annual growth
remains, 90% of annual growth rates would be between -0.5% and
2.6%. (Figure 6).
Some believe that the Hubbert curve is going to have a global decrease
rapid climb the slope before reaching the maximum (Bardi, 2005 and Mohr,
2008), which would make the former model was "optimistic."

[FIGURE 6: Variation in the growth of global GDP levels and prognosis after 2010.]

The first obvious symptom of which is the pattern of 1973 is the current rise
price of oil, but unlike then it is not expected to re-price
down to average levels of 90 years but if declines will occur due to
reduction in demand caused by the current crisis, to eventually become a
increasingly rare view of political and economic strategies (to be published
comments which indicated that would lead to double or triple the current price). This
lead to an increase in the price of all goods, which will not be matched
with an increase in wages (to avoid an increase in inflation, which would
increase competitiveness and reduce unemployment even more), therefore,
predicts an overall reduction and significant purchasing power and thus of
consumption, this in turn influences the decline in economic growth.

4. Conclusions
Has studied the association between the oil and the evolution of GDP
world, noting that both magnitudes are strongly associated. In
result has been adjusted to the data of world GDP between 1950 and 2005, the same
type of model, a logistic curve that Hubbert used in predicting the evolution of
oil stocks, the fitted model correctly describes the historical data
and allows one to predict average global GDP growth by just over
1.0% during a period beginning in 2010.


IEA (International Energy Agency): World Energy Outlook (WEO 2007: in
English.pdf http://.www.worldenergyoutlook.org/weo2007/WEO 2007: last

Bardi, U. (2005), "The mineral economy: a model for the shape of oil production
curves. Energy Policy 33 (1), 53-61

Guse, R., Dalla Valle A. Guidolin, M. (2007), "World Oil Depletion Models: Price
effects compared with strategic or technological interventions, "Technological
Forecasting and Social Change, 74 (4), 452-469

Hubbert, M.K. (1956). "Nuclear energy and the fossil fuels." Drilling and Production

Campbell C.J. and Laherrère, J.H. (1998), "The end of cheap oil", Scientific American
(March 1998), 60-65

Maddison, A. (2003), "The World Economy: Historical Statistics, Development Center
Studies. OECD, Paris

Maddison, A. (2007), "World Population, GDP and per capita GDP, 1-2003AD"
(available in http://www.ggdc.net/maddison/Historical_Statistics/horizontalfile_
03_2007.xls, last accessed 21/4/2008)

Mohr, S.H. and Evans, G.M. (2008), "Peak Oil: Testing Hubbert's Curve via Theoretical
Modeling ", Natural Resources Research 17 (1), 1-11

Rosell, J., (2007), "And then the oil, what? Lights and shadows of the future
Global energy, Ediciones Deusto, Barcelona
UNSD (United Nations Statistical Division) (2007) (available at
http://unstats.un.org/unsd/methods/icp/icp7_htm.htm, last accessed 21/4/2008)

08-12-09, 01:36 PM
thanks for your summary. relating gdp and oil consumption makes really good sense under current technological conditions.

This lead to an increase in the price of all goods, which will not be matched with an increase in wages (to avoid an increase in inflation, which would increase competitiveness and reduce unemployment even more), therefore, predicts an overall reduction and significant purchasing power and thus of consumption, this in turn influences the decline in economic growth.this is identical to the itulip prediction: commodity inflation causing a shrinkage in purchasing power, with wage inflation lagging. from that point becerra focuses on the stagflationary global [anemic] growth environment, while ej looks at the currency/inflationary and financial market effects.

is there any analysis of differential effects around the globe? and, if so, is it related to energy development per se? to say 1% [-0.5-+2.6%] global growth is bad, indeed, but we can expect that within this overall outcome, some regions will be doing better than others.

08-12-09, 02:50 PM
is there any analysis of differential effects around the globe?Becerra says often that we well get out of this crisis via increasing the productivity. He thinks that the current tecnology allows enourmous increases in productivity that had not been applied yet. So he's more positive on the countries that have the know-how (USA, Germany, Japan) and is particularly negative for Spain, that makes products with low value added and intensive in human labor. He's also positive on Britain, but I don't remember well the context.

Productivity is necessary to produce the maximum of really necessary products with minimum resource consumption. What makes Becerra so pessimistic is his conclusion about anemic growth and high productivity: If production is limited by some resource (oil or whatever), productivity doesn't imply more production, but more unemployment. He thinks that that will mean that most of the work force will be unnecessary, and will need to live from subsidies. He thinks that a survival subsidy will be implanted in most countries and sees the extensions of unemployment benefits being applied in the USA and Spain as the first steps in that direction.

08-15-09, 04:53 AM
Wealth = cheap energy supply.

Becerra makes sense to me.

08-17-09, 06:59 AM
Wealth = cheap energy supply.

Becerra makes sense to me.

Yes, it's a good description of the current status of the economy. It used to be

Wealth = work

since in a world were there are plenty of resources, the limiting factor is human work. In an overpopulated world, the different "peaking" resources are the limiting factor. Now is cheap energy. If we found a good replacement for oil, it could be the next commodity under stress: food, drinking water, whatever.

Could it be possible that work becomes to the economy like air is to life: requisite but valueless, because you can have as much as you want?