##### Document Text Contents

Page 1

Economic Transformation and Sectoral

Growth Options in Ethiopia:

Considerations for Achieving Middle Income Country

Status

Zerayehu Sime Eshete

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE AWARD OF DOCTOR OF

PHILOSOPHY IN ECONOMICS OF THE UNIVERSITY OF

NAIROBI

November 2013

Page 2

i

DECLARATION

This thesis is my original work, which has not been presented for a degree in any

university.

Signature Date

Zerayehu Sime Eshete

This thesis has been submitted for examination for PhD in economics with our approval

as University supervisors.

Signature Date

Prof. Peter Kimuyu

School of Economics

University of Nairobi

Signature Date

Dr. John Matovu

School of Economics

University of Makerere

Page 97

85

Figure 13:-The Dynamics of Sectoral TFP growth by regression-based approach

Source:- wn calculation, estimated using the regression-based approach

Comparing the estimated sectoral TFP growth rates from both approaches, the study

chooses the growth accounting approach in order to calibrate the induced sectoral TFP

growth. This is mainly because of 1) maintaining a consistent assumption of constant

return to scale across the paper including the CGE model 2) The results from the growth

accounting approach show that the agriculture uses labour intensive technology while

the industry and service use capital intensive with negative TFP growth rate across

sectors, on average. This result relatively reflects the actual economic performance of

5.1.2 Analyzing the Determinants of Sectoral TFP Growths

Using the estimates of the TFP growths for the agriculture, industry and services, the

paper specifies, estimates, and analyzes the determinants of sectoral TFP growth using

VARX model. On the basis of the econometric results, the study uses statistically

significant explanatory variables in order to generate the induced sectoral TFP growths.

Hence, the empirical results of descriptive analysis, optimal lag length, stationarity test,

regression outcomes, diagnostic test, impulse response function, and variance

decomposition are presented as follows.

-0.300

-0.200

-0.100

0.000

0.100

0.200

0.300

1970 1980 1990 2000 2010 2020

ATFPG

ITFPG

STFPG

Page 98

86

Descriptive Analysis

Descriptive analysis gives the general features of the data and emphasizes how it looks

like the distribution of time series. Table 18 below presents both central tendency and

variation measures of the distribution of the data in 1972-2011. The average value or the

mean value of the sectoral TFP growth for the three sectors is negative, indicating how

the sectoral TFP growths can be the daunting challenge for economic transformation and

sustainable growth. All variables in the VARX model have a wide range of dispersion

around the mean with a mix of positive and negative skewness and peakedness.

Table 18:-Summary of Descriptive Statistics

Variable Obs Mean Std. Dev. Variance Skewness Kurtosis Min Max

TFPGA 40 -.005 .081 0.006 -0.81 4.73 -.274 .182

TFPGI 40 -.116 .487 0.23 -1.91 7.55 -2.04 .64

TFPGS 40 -.037 .246 0.06 -1.73 6.12 -.884 .317

GARD 40 .170 .470 0.22 1.69 6.61 -.627 1.77

GRNW 40 .046 .0479 0.002 0.72 3.94 -.0502 .172

GIMC 40 .058 .366 0.13 -0.37 3.26 -.99 .724

GPEP 40 .083 .104 0.10 -0.04 2.79 -.163 .315

OPP 40 28.35 10.62 112.95 0.56 2.21 11.73 51.53

IR 40 15.73 5.61 31.51 0.14 1.56 6.15 24.65

INF 40 9.25 11.29 127.53 1.11 4.64 -9.81 44.4

Source:-

TFPGA stands for TFP growth rate for agriculture; TFPGI stands for TFP growth rate

for industry; TFPGS for TFP growth rate for service; GARD denotes growth rate of

government expenditure for agricultural R&D, GIMC denotes growth rate of imported

capital goods ,GPEP denotes growth rate of pupil in primary school; GRNW stands for

growth rate of road net works in kilometers ; stands for openness of the economy;

LR denotes service trade liberalization index and INF for inflation.

With these descriptive statistics, the variables show some fashion of statistical

relationship among themselves as captured by the Pearson product-moment correlation

coefficient (Table 19). This coefficient is symmetric and gives some clue about the

Page 194

182

Savings - Investment Balance:

.(1 ). . . .

i i i c c c c

i INSDNG c C c C

MPS tins YI GSAV EXR FSAV PQ QINV PQ qdst WALRAS

Non government saving + government savings + foreign savings = fixed investment + stock change

i

MPS = Marginal propensity to save for

domestic non-government institution

(exogenous variable)

i

tins = Exogenous direct tax rate for domestic

institution i

i

YI = Income of domestic non-government

institution

GSAV = Government savings

EXR = Exchange rate (LCU per unit of FCU)

FSAV = Foreign savings (FCU)

c

PQ = Composite commodity price

QINV = Quantity of investment demand for

commodity

c

qdst = Quantity of stock change

WALRAS= Saving- Investment Gap

.(1 )

i i

MPS mps MPSADJ

i

MPS = Marginal propensity to save for

domestic non-government institution

(exogenous variable)

i

mps = Base savings rate for domestic

institution i

MPSADJ = Savings rate scaling factor (= 0 for

base)

Government balance: -

YG EG GSAV

Government revenue =government expenditure + government savings

YG = Government revenue, EG = Government

expenditures

GSAV = Government savings

Factor Market Balance:-

fa f

a A

QF QFS

: Demand for factor = Supply for factor

fa

QF = Quantity demanded of factor f fQFS = Quantity supplied of factor

Page 195

183

Total absorption: -

. . . . .

c ch ac ach c c c c c c

h H c C a A Ac C h h c C c C c C

TABS PQ QH PXAC QHA PQ QG PQ QINV PQ qdst

Total absorption=household market consumption + household home consumption + government

consumption + fixed investment + stock change

TABS = total nominal absorption

c

PQ

= Composite commodity price

ch

QH

= quantity consumed of commodity c by

household h

c

qdst = quantity of stock change

ach

QHA

= quantity of household home

consumption of commodity c from activity a for

household h

c

QG

= government consumption demand for

commodity

ac

PXAC

= producer price of commodity c for

activity a

Ratio of investment to absorption: - . . .

c c c c

c C c C

INVSHRTABS PQ QINV PQ qdst

(Total absorption ratio) ( total absorption) = fixed investment + stock change

INVSHR = Investment share

TABS = total nominal absorption

= Composite commodity price

=quantity of investment demand for

commodity

c

qdst = quantity of stock change

Ratio of government consumption to absorption: -

: (Government consumption absorption ratio) (total absorption )=

government consumption

= government consumption share in

nominal absorption

TABS = total nominal absorption

= Composite commodity price

= government consumption demand for

commodity

Economic Transformation and Sectoral

Growth Options in Ethiopia:

Considerations for Achieving Middle Income Country

Status

Zerayehu Sime Eshete

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE AWARD OF DOCTOR OF

PHILOSOPHY IN ECONOMICS OF THE UNIVERSITY OF

NAIROBI

November 2013

Page 2

i

DECLARATION

This thesis is my original work, which has not been presented for a degree in any

university.

Signature Date

Zerayehu Sime Eshete

This thesis has been submitted for examination for PhD in economics with our approval

as University supervisors.

Signature Date

Prof. Peter Kimuyu

School of Economics

University of Nairobi

Signature Date

Dr. John Matovu

School of Economics

University of Makerere

Page 97

85

Figure 13:-The Dynamics of Sectoral TFP growth by regression-based approach

Source:- wn calculation, estimated using the regression-based approach

Comparing the estimated sectoral TFP growth rates from both approaches, the study

chooses the growth accounting approach in order to calibrate the induced sectoral TFP

growth. This is mainly because of 1) maintaining a consistent assumption of constant

return to scale across the paper including the CGE model 2) The results from the growth

accounting approach show that the agriculture uses labour intensive technology while

the industry and service use capital intensive with negative TFP growth rate across

sectors, on average. This result relatively reflects the actual economic performance of

5.1.2 Analyzing the Determinants of Sectoral TFP Growths

Using the estimates of the TFP growths for the agriculture, industry and services, the

paper specifies, estimates, and analyzes the determinants of sectoral TFP growth using

VARX model. On the basis of the econometric results, the study uses statistically

significant explanatory variables in order to generate the induced sectoral TFP growths.

Hence, the empirical results of descriptive analysis, optimal lag length, stationarity test,

regression outcomes, diagnostic test, impulse response function, and variance

decomposition are presented as follows.

-0.300

-0.200

-0.100

0.000

0.100

0.200

0.300

1970 1980 1990 2000 2010 2020

ATFPG

ITFPG

STFPG

Page 98

86

Descriptive Analysis

Descriptive analysis gives the general features of the data and emphasizes how it looks

like the distribution of time series. Table 18 below presents both central tendency and

variation measures of the distribution of the data in 1972-2011. The average value or the

mean value of the sectoral TFP growth for the three sectors is negative, indicating how

the sectoral TFP growths can be the daunting challenge for economic transformation and

sustainable growth. All variables in the VARX model have a wide range of dispersion

around the mean with a mix of positive and negative skewness and peakedness.

Table 18:-Summary of Descriptive Statistics

Variable Obs Mean Std. Dev. Variance Skewness Kurtosis Min Max

TFPGA 40 -.005 .081 0.006 -0.81 4.73 -.274 .182

TFPGI 40 -.116 .487 0.23 -1.91 7.55 -2.04 .64

TFPGS 40 -.037 .246 0.06 -1.73 6.12 -.884 .317

GARD 40 .170 .470 0.22 1.69 6.61 -.627 1.77

GRNW 40 .046 .0479 0.002 0.72 3.94 -.0502 .172

GIMC 40 .058 .366 0.13 -0.37 3.26 -.99 .724

GPEP 40 .083 .104 0.10 -0.04 2.79 -.163 .315

OPP 40 28.35 10.62 112.95 0.56 2.21 11.73 51.53

IR 40 15.73 5.61 31.51 0.14 1.56 6.15 24.65

INF 40 9.25 11.29 127.53 1.11 4.64 -9.81 44.4

Source:-

TFPGA stands for TFP growth rate for agriculture; TFPGI stands for TFP growth rate

for industry; TFPGS for TFP growth rate for service; GARD denotes growth rate of

government expenditure for agricultural R&D, GIMC denotes growth rate of imported

capital goods ,GPEP denotes growth rate of pupil in primary school; GRNW stands for

growth rate of road net works in kilometers ; stands for openness of the economy;

LR denotes service trade liberalization index and INF for inflation.

With these descriptive statistics, the variables show some fashion of statistical

relationship among themselves as captured by the Pearson product-moment correlation

coefficient (Table 19). This coefficient is symmetric and gives some clue about the

Page 194

182

Savings - Investment Balance:

.(1 ). . . .

i i i c c c c

i INSDNG c C c C

MPS tins YI GSAV EXR FSAV PQ QINV PQ qdst WALRAS

Non government saving + government savings + foreign savings = fixed investment + stock change

i

MPS = Marginal propensity to save for

domestic non-government institution

(exogenous variable)

i

tins = Exogenous direct tax rate for domestic

institution i

i

YI = Income of domestic non-government

institution

GSAV = Government savings

EXR = Exchange rate (LCU per unit of FCU)

FSAV = Foreign savings (FCU)

c

PQ = Composite commodity price

QINV = Quantity of investment demand for

commodity

c

qdst = Quantity of stock change

WALRAS= Saving- Investment Gap

.(1 )

i i

MPS mps MPSADJ

i

MPS = Marginal propensity to save for

domestic non-government institution

(exogenous variable)

i

mps = Base savings rate for domestic

institution i

MPSADJ = Savings rate scaling factor (= 0 for

base)

Government balance: -

YG EG GSAV

Government revenue =government expenditure + government savings

YG = Government revenue, EG = Government

expenditures

GSAV = Government savings

Factor Market Balance:-

fa f

a A

QF QFS

: Demand for factor = Supply for factor

fa

QF = Quantity demanded of factor f fQFS = Quantity supplied of factor

Page 195

183

Total absorption: -

. . . . .

c ch ac ach c c c c c c

h H c C a A Ac C h h c C c C c C

TABS PQ QH PXAC QHA PQ QG PQ QINV PQ qdst

Total absorption=household market consumption + household home consumption + government

consumption + fixed investment + stock change

TABS = total nominal absorption

c

PQ

= Composite commodity price

ch

QH

= quantity consumed of commodity c by

household h

c

qdst = quantity of stock change

ach

QHA

= quantity of household home

consumption of commodity c from activity a for

household h

c

QG

= government consumption demand for

commodity

ac

PXAC

= producer price of commodity c for

activity a

Ratio of investment to absorption: - . . .

c c c c

c C c C

INVSHRTABS PQ QINV PQ qdst

(Total absorption ratio) ( total absorption) = fixed investment + stock change

INVSHR = Investment share

TABS = total nominal absorption

= Composite commodity price

=quantity of investment demand for

commodity

c

qdst = quantity of stock change

Ratio of government consumption to absorption: -

: (Government consumption absorption ratio) (total absorption )=

government consumption

= government consumption share in

nominal absorption

TABS = total nominal absorption

= Composite commodity price

= government consumption demand for

commodity