Catchup Countries Only

We have grouped the 131 countries according to several criteria and found that most countries in sub-groups 4 and 8 were catching up during the last few decades. However, not all countries in these groups were actually progressing. The regression results in Table 9.5 still reflect information from some countries that didn't progress vis-à-vis the US. If one assumes that all the information from countries that didn't make progress economically in the period we analyse is 'noise' and that only countries that really were catching up should be used to generate the parameters for our model, we should filter out the 'noise' and do regressions only for the remainder. All the catch-up countries in our samples, together with the US itself, are listed in Table 9.6. Most of them are in Group 8.

In 1971, the GDP in purchasing power parity (PPP) of the catch-up countries accounted for 66.34 percent of world GDP, including the US, and 43.1 percent of the non-US world GDP. It is fair to assume that all the countries that have been catching up during the past 30 plus years have had reasonably effective economic and political management during most of the period. We postulate that the energy-GDP relationship generated from these countries defines a theoretical trajectory that any country would follow, given reasonable economic management and in the absence of a 'growth stopper'.

Countries that did not catch up lagged behind or fell back for a variety of reasons. However, there are two groups of countries not on the list in Table 9.6, most of which were also not catching up with the US during

Table 9.5 a Regression results with different oil coefficients for groups 4 and 8 ( unweighted)

Oil Group 4 Group 8

Y = a + bx + cln (X) Y= a + bx + c (X)1'2 Y= a + bx + cln (X) Y= a+ bx + c (X)1'2

0 0.932 0.03 0.85 0.426 0.543 0.855 0.299 0.194 0.885 -0.57 1.783 0.893

0.1 0.624 0.03 0.867 0.257 0.483 0.875 0.343 0.19 0.901 -0.44 1.664 0.905

0.15 0.556 0.03 0.866 0.218 0.47 0.875 0.335 0.191 0.896 -0.43 1.651 0.899

0.2 0.51 0.03 0.865 0.191 0.46 0.875 0.319 0.193 0.889 -0.45 1.659 0.892

0.25 0.476 0.03 0.864 0.172 0.453 0.875 0.299 0.195 0.882 -0.48 1.678 0.885

Table 9.5b Regression results with different oil coefficients for groups 4 and 8 ( GDP weighted)

Oil Group 4 Group 8 coefficient

Y = a + bx + cln (X) Y = a + bx + c (X)1'2 Y = a + bx + cln (X) Y = a+bx + c (X)1'2

0

1.121

0

0.823

0.814

0.279

0.83

0.195

0.241

0.957

-0.63

1.934

0.959

0.1

0.662

0.02

0.862

0.324

0.419

0.878

0.245

0.224

0.965

-0.53

1.821

0.968

0.15

0.573

0.02

0.868

0.249

0.431

0.886

0.261

0.219

0.966

-0.5

1.779

0.969

0.2

0.515

0.03

0.871

0.205

0.435

0.89

0.272

0.214

0.966

-0.47

1.745

0.969

0.25

0.475

0.03

0.874

0.175

0.436

0.893

0.281

0.21

0.966

-0.45

1.716

0.968

Note: Countries that are far away from the main trend of each group were removed from the regressions. They are: in Group 4, Argentina, Jamaica, Netherlands, Antilles and Philippines; in Group 8, Luxembourg, South Africa and Turkey. Y is the dependent variable, GDP fraction, x is the independent variable, EP fraction.

EP fraction USA

Figure 9.4a Weighted regressions for group 4 (oil factor = 0.10)

EP fraction USA

Figure 9.4a Weighted regressions for group 4 (oil factor = 0.10)

EP fraction USA

Figure 9.4b Weighted regressions for group 8 (oil factor = 0.10)

Of. •-

9 sample

O fitted In

• fitted SQRT

Figure 9.4b Weighted regressions for group 8 (oil factor = 0.10)

the period. These are countries with high per capita GDP (Group A in the introduction) and countries with very low per capita GDP and development obstacles (Group C in the introduction). Figures 9.6a and 9.6b shows the trajectories of Group A and Group C countries, respectively. Instead

row --?*

^ sample O fitted ln • fitted SQRT

EP fraction USA

Figure 9.5 a Weighted regressions for group 4 (oil factor = 0.15)

Figure 9.5 a Weighted regressions for group 4 (oil factor = 0.15)

EP fraction USA

Figure 9.5b Weighted regressions for group 8 (oil factor = 0.15)

Q sample

O fitted ln

• fitted SQRT

EP fraction USA

Figure 9.5b Weighted regressions for group 8 (oil factor = 0.15)

of reducing their gaps with the US in per-capita GDP, most of them had been stagnant or even regressing.

In Figures 9.7a-e, the trajectories of catch-up countries are plotted for different values of the 'oil' coefficient in the EP, from 0.0 to 0.25. One can

Table 9.6 Catch-up countries

Country no.

Country

GDP of 2001 (PPP in billion 1995$)

Group no.

7

Austria

199.068

6

12

Belgium

256.049

8

21

Chile

130.826

6

28

Cyprus

14.629

4

32

Dominican Rep.

55.696

4

34

Egypt;

213.128

3

40

France

1 394.529

8

43

Germany

1 922.029

8

46

Greece

165.226

8

52

India

2707.164

4

53

Indonesia

560.887

3

55

Ireland

110.078

8

58

Italy

1 287.402

8

60

Japan

3 125.882

8

61

Jordan

18.731

4

64

Korea (S.)

674.911

8

73

Malaysia

181.962

4

74

Malta

4.731

4

91

P. R. China

4707.822

8

95

Portugal

166.752

8

103

Singapore

84.357

4

107

Spain

739.499

8

108

Sri Lanka

56.746

6

114

Thailand

356.876

4

124

USA

8977.8

8

Total GDP

2733.05

782.73

28112.8

World Total GDP

42374.34

% of World GDP

64.5%

2.85%

66.34%

see that, as the fraction of oil added to the energy proxy increases, several countries' imputed development tracks diverge from the major trend of the whole group. These countries are Jordan, South Korea, Malta and Singapore. Their oil consumption increases faster than their increase in electricity consumption and GDP. All four of these countries can be regarded as 'young': South Korea achieved formal independence in the mid-1950s after the very destructive Korean War. Singapore became independent of Malaysia in 1965. Malta became independent from Britain in 1964, while Jordan became independent of Palestine only after 1967, also after a war with Israel. It is possible that their abnormal behavior in regard

EP fraction USA

Figure 9.6a Development tracks of group A countries

Figure 9.6a Development tracks of group A countries

EP fraction USA

EP fraction USA

Figure 9.6b Development tracks of group C countries to oil consumption is due to having started from unusually low levels of motorization.

However, while the four outlier countries diverge, the others stay together. If we want to keep all the countries in our model, it is clear that electricity consumption alone is the best energy proxy. If we remove the

EP fraction USA

Figure 9.7a Development tracks of catch-up countries (oil factor = 0.00)

Figure 9.7a Development tracks of catch-up countries (oil factor = 0.00)

EP fraction USA

EP fraction USA

Figure 9.7b Development tracks of catch-up countries (oil factor = 0.10)

four countries from our simulation because of their too rapid increase in oil consumption, we still need to check whether including a fraction of oil consumption for the other countries in our proxy can improve the simulation. Regressions for the rest of the countries with different oil coefficients in the energy proxy were run and the results are given in Table 9.7. Table 9.7a

EP fraction USA

Figure 9.7c Development tracks of catch-up countries (oil factor = 0.15)

Figure 9.7c Development tracks of catch-up countries (oil factor = 0.15)

EP fraction USA

Figure 9.7d Development tracks of catch-up countries (oil factor = 0.20)

shows the regression results with no weights, and Table 9.7b shows regression results weighted by GDP. Considering the values of R2, the square root model is a little bit better than the natural log model. Even without weights, we get R2 values higher than 0.94 with only one independent variable expressed in two terms. The linear terms in the square root model for

EP fraction USA

Figure 9.7e Development tracks of catch-up countries (oil factor = 0.25)

EP fraction USA

Figure 9.7e Development tracks of catch-up countries (oil factor = 0.25)

some oil coefficients are not quite significant. However, in Table 9.7a all the other coefficients are significant. All the coefficients in Table 9.7b are also significant. The F-values are large.

By using GDP as weights, the R2 values improve to above 0.98 for the square root model. Moreover, the regression results are not sensitive to the oil coefficient in the EP, especially for weighted results. The plot of the samples used in the regressions and the simulation results are given in Figures 9.8a-e. The simulation lines fit the samples very well. There is not much difference among different models (either natural log or square root, whether weighted with GDP or not) for different values of the oil coefficient.

There might be questions as to why only one independent variable (EP) is included in the regressions in Tables 9.5 and 9.7, given that there are obviously other factors that can affect economic growth. The explanation is that we are not seeking a complete theory to explain growth (or its absence) in developing countries. Instead, we are asking how much of that growth can be explained by a single factor: energy (exergy) consumption as converted to useful work. In effect, we are treating economic growth as a physical process, analogous to heating water. In our case, energy consumption is the input of the economic growth system and the output is GDP (of course both inputs and output are expressed relative to US values in our model).

In short, we believe that the evidence compiled in this chapter demonstrates that the EP, discussed above, is indeed an important factor of production, at least in situations where growth is not distorted or restricted

Table 9.7a Regression results with different coefficients for catch-up countries only ( unweighted)

Ul1 Y = a+ b-x + c-ln(x) Y = a + b-x + c-(x)1/2 coefficients--

b

c

R2

b

c

R2

0

0.588 (t =

23)

0.108 (t =

20)

0.92

-0.211 (t =

-6.8)

1.359 (t =

40)

0.94

0.1

0.600 (t =

25)

0.117 (t =

25)

0.93

-0.115 (t =

-3.5)

1.298 (t =

37)

0.95

0.15

0.612 (t =

25)

0.116 (t =

24)

0.93

-0.081 (t =

-2.4)**

1.266 (t =

35)

0.94

0.2

0.620 (t =

26)

0.114 (t =

23)

0.93

-0.057 (t =

-1.6)*

1.241 (t =

33)

0.94

0.25

0.625 (t =

26)

0.113 (t =

23)

0.92

-0.039 (t =

-1.1)*

1.221 (t =

32)

0.93

Table 9.7b Regression results with different coefficients for catch-up countries only ( GDP weighted)

Table 9.7b Regression results with different coefficients for catch-up countries only ( GDP weighted)

Ul1 Y = a+ b-x + c-ln(x) Y = a + b-x + c-(x)1/2 coefficients--

b

c

R2

b

c

R2

0

0.342 (t =

19)

0.169 (t =

31)

0.964

-0.465 (t =

-15)

1.675 (t =

42)

0.976

0.1

0.345 (t =

23)

0.173 (t =

33)

0.973

-0.417 (t =

-16)

1.632 (t =

45)

0.982

0.15

0.353 (t =

23)

0.172 (t =

33)

0.975

-0.397 (t =

-15)

1.610 (t =

46)

0.983

0.2

0.360 (t =

24)

0.170 (t =

32)

0.975

-0.379 (t =

-15)

1.590 (t =

47)

0.983

0.25

0.366 (t =

25)

0.168 (t =

32)

0.975

-0.365 (t =

-15)

1.574 (t =

47)

0.983

Notes: 61 (Jordan), 64 (Korea), 74 (Malta) and 103 (Singapore) were dropped. * Not significant at 5% level. ** Not significant at 1% level. Y is the dependent variable, GDP fraction, x is the independent variable, EP fraction.

Notes: 61 (Jordan), 64 (Korea), 74 (Malta) and 103 (Singapore) were dropped. * Not significant at 5% level. ** Not significant at 1% level. Y is the dependent variable, GDP fraction, x is the independent variable, EP fraction.

EP fraction USA

Figure 9.8a Simulation results for catch-up countries (oil = 0.00)

o

o

Sample

Unweighted logarithm

Weighted logarithm

—*

Unweighted square root

♦ • -

Weighted square root

EP fraction USA

Figure 9.8a Simulation results for catch-up countries (oil = 0.00)

EP fraction USA

o o

Sample

Unweighted logarithm

Weighted logarithm

-------*

Unweighted square root

Weighted square root

EP fraction USA

Figure 9.8b Simulation results for catch-up countries (oil = 0.10)

EP fraction USA

o

o

Sample

Unweighted logarithm

Weighted logarithm

-

—*

Unweighted square root

♦ • -

■ -♦

Weighted square root

EP fraction USA

Figure 9.8c Simulation results for catch-up countries (oil = 0.15)

EP fraction USA

Figure 9.8d Simulation results for catch-up countries (oil = 0.20)

o

o

Sample

Unweighted logarithm

Weighted logarithm

—*

Unweighted square root

• -♦

Weighted square root

EP fraction USA

Figure 9.8d Simulation results for catch-up countries (oil = 0.20)

EP fraction USA

Figure 9.8e Simulation results for catch-up countries (oil = 0.25)

o o

Sample

Unweighted logarithm

-—•

Weighted logarithm

-------*

Unweighted square root

Weighted square root

EP fraction USA

Figure 9.8e Simulation results for catch-up countries (oil = 0.25)

by exogenous constraints. Factors such as political system, institutional situation and so on are beyond the consideration of our model. They constitute another dimension of the economic growth issue. However, there are factors, such as a country's location, its hydro-electricity fraction and so on, that can certainly affect the energy/work relationships.

Then what is the story these regression results can tell us? Mainly, the story is as follows: for countries whose per-capita GDP is very low (or which are at an early stage of development), energy consumption (as converted to useful work) can generate rapid (more than proportional) catchup in terms of GDP. However, as countries' GDP approaches the US level, the catch-up rate slows down. Or, more accurately, the economic catch-up attributable to energy consumption decelerates.

Trash Cash Machine

Trash Cash Machine

How recyclable trash can save the world and bank us huge profits! Get All The Support And Guidance You Need To Be A Success At Recycling! This Book Is One Of The Most Valuable Resources In The World When It Comes To How To Make Profits With Trash!

Get My Free Ebook


Post a comment