Governance, Quality of Institutions and Economic Growth: Empirical Evidence from a Cross-National Analysis

4. Empirical analysis of the relation between governance and economic growth

On this level of analysis we seek to study the impact of various measurements of the governance on the long run global economic performances in 96 countries. Basing our study on the model of growth of Mankiw & al. (1992), Knight & al. (1993) and Ghra

& Hadjmichael (1996) [quoted by Demetriades and Law, 2006, p. 5)]. Our starting point is the following Cobb-Douglas production function:

Y t K _t H _t ( A _t L _t )

á â - á -

â

(1)

1

=

Where Y is real output, K is the stock of physical capital, H is the stock of human capital, L is the raw labour, A is a labour-augmenting factor reflecting the level of technology and efficiency in the economy and the subscript t indicates time.

It is assumed that á + â < 1, i.e. decreasing returns to all capital (physical capital and human capital). Raw labour and labour-augmenting technology are assumed to grow according to the following functions:

L t = L ₀e nt (2)

t A 0

gt +

pè

(3)

A = e

where n is the exogenous rate of growth of the labour force, g is the exogenous rate of technological progress, P is a vector of financial development, institutions and other factors that can affect the level of technology and efficiency in the economy, and è is a vector of coefficients related to these variables.

In this model, the variable A depends on the exogenous technological improvements, the degree of commercial opening and the level of other variables. It is obvious that A in our study differs from that employed by Mankiw and al. (1992). This modification is particularly appropriate to the empirical validation of the relations between Institutions quality and economic growth. The technological improvements are encouraged by effective institutions (North 1990) and by healthy institutional environment (World Bank).

In equilibrium condition, the output per capita increases at a constant rate G (the exogenous component of the growth rate of the variable reflecting the level of technology and efficiency of an economy). These results can be obtained directly from the definition of output per effective worker (Average Labour Productivity):

Y =
t

AL

t t

;

( ) ^á ( )â

k . h

t t

Y

t A . k . h

= ( ) ^á ( )^â (4)⁴

t t t

t

L

With

? ? * t = Y y ? ^t L ? ? t ?

*

As we apply the logarithm on the two sides for the equation (4) and to simplify calculation we eliminate the time index, we have then:

á â

- -

Y L

^t A . = t t

( ) ^á ( )â k . h t t

1

Y K H A L

á â ( )

4 t t t t t

=

t

AL

t

( ) á á â â

1 - + - +

A L

t t

á â á â

1 - -

Y ? K ? ? H ? ? A L ?

; t t t t t

= ? ? . ? ? . ? ? = ( ) ^á ( )â

k . h

t t

ALA L

t t ? t t ?A L A L

? t t ? ? t t ?

Y ( ) ( ) ( ) ( ä)

á â á â

+

*

Ln( = + + +

) Ln A g.t .P

è - + + (5)⁵

L 1 - -

á â 1 - -

á â 1 - -

á â

0 k h

Ln s + Ln s Ln n g

The equation (5) determines the output per capita in the equilibrium state; also in this equation the vector p gather the institutional variables which will be defined in the fallowing paragraph.

Because of a data limitation, one can suppose in this study that s^h and g.t do not change through time (Demetriades and Law (2006)), whereas s^k and n vary Indeed, Ln (A 0), g.t and s^h can be gathered in a constant á0 in the equation (6). Therefore, output per capita (also called Average Labour Productivity) is given by:

Y á ( ) ( ä

? á â

+ ?

Ln( *

) = + +

á è

0 . .P k

Ln s + -

? ? .Ln n g

+ + ) (6)

L 1 - -

á â ? 1 - -

á â ?

With P: the vector gathering institutional variables.

After simplification of the equation (6), we obtain an equation of evaluation for the relation between the quality of institutions and output per capita:

Lny 0 1 . INS 2 . Ink 3 . In n g

= + + + + +

á á á á ( ä ) (7)

With, y Gross domestic product per capita (GDP/capita); INS a vector gathering the institutional variables, k is the stock of capital investment or physical capital accumulation; n is the rate of labour force; g is the rate of technology growth or technological progress and ä is the rate of depreciation. g and ä are assumed to be constant across countries and over time and their sum equals 0.05, following Mankiw and al. (1992, p.413). á0, á1, á2 and á3 are the parameters to be estimated.

1 1

and

1 â â á â

? s . s

- - -

? 1

⁵ At equilibrium one has; k * ??

h

k

= ?? ä + +

g n

1 - á á

1

á â

- -

? s . s ?

h * ??

k

h

= ?? ä + +

g n

4.1. Econometric approach

The equation (7) is considered as the base for the empirical models which will be estimated by the cross-section method:

For the econometric approach the equation (7) will be modified as follows:

y= á0 + á' X+ î(8)

ii i

i = 1, 2, ..., 96

With, y is real GDP per capita in logarithm, á ₀ constant, á^' = (á₁,á₂,á₃) a vector of dimension (3;1), X i = (X1 _,i ;X2 ,i ;X3 ,i ) vector of explanatory variables, and îi innovations supposed to be independently identical of null mean and variance ó î 2.