Menu
Home
Log in / Register
 
Home arrow Economics arrow Accelerated Economic Growth in West Africa
< Prev   CONTENTS   Next >

5 Juxtaposing the Relationship: Atleast for ECOWAS

The summary statistics of the main variables for ECOWAS countries are presented in Table 8.

As a first check, the validity of the internal instruments that was used in the SGMM, was considered. The checks are the Sargan and AR (2) test. Considering the statistics of the two tests [Sargan and AR (2)], in all the columns in Table 9, the instruments were valid and were not over-identified. This is following the p-values of the test results. This confirmation renders the result from this study relevant and reliable for inference.

To be concise, focus was only on the interactive variables (Control of Corruption x Infrastructure; Government Effectiveness x Infrastructure) that shows the juxtaposing between institutions, infrastructure and manufacturing export. This results was displayed in the last segments of Table 9. The signs and significant values of these interactive variables show that when considering the development of institutions (i.e. corruption), infrastructure and manufacturing export, we perceive that institutions and infrastructure play a substitutive role. As a matter of fact, infrastructural development will impact less on manufacturing export in countries where the control of corruption is improved. The coefficient of this variable is '-0.761' and it is significant at 1 % level of significance.

This result contradicts the prediction in this paper as we expected that institutions are complimentary factors in the infrastructure-manufacturing export nexus. Put different, infrastructure development was expected to improve manufacturing export in countries where the control of corruption is improved. Likewise, when considering government effectiveness and the role it plays in the infrastructuremanufacturing export nexus, we also perceive that infrastructure impacts less on manufacturing export in countries where the government effectiveness is improved. In essence, institutions play a substitutive role with infrastructure in influencing manufacturing export. The coefficient of this variable is '-0.552' and it is signif-icant at 10 % level of significance.

Faced with this somewhat contradiction, we raise a very important point: despite the fact that institutional development is supposed to enhance the influence of infrastructural development on manufacturing export, the case of ECOWAS is different. Possibly, mere institutional development is not enough to enhance

Table 8 Summary statistics of variables

Description of variables

Mean

Std. Dev

Min.

Max.

Mafex

Ratio of manufacturing export to total merchandise export

25.85

21.33

0.08

95.68

Magr

Manufacturing value added

8.16

4.73

2.24

21.68

Exch

Real exchange rate

793.09

1,161.88

0.54

6,658.03

Infra

Indicator of infrastructural provision

10.58

10.80

0.08

44.27

CC

Control of corruption (an indicator of institutions)

-0.60

0.48

-1.37

0.80

GE

Government effectiveness (an indicator of institution)

-0.79

0.49

-1.84

0.33

Source: Authors' computation

Table 9 SGMM results (dependent variable: manufacturing exports)

1

2

3

4

Manufacturing exports (-1)

0.381* (0.000)

0.364* (0.000)

0.373* (0.000)

0.350* (0.000)

Manufacturing value added

0.564**

0.453***

0.585**

0.544**

(0.025)

(0.073)

(0.018)

(0.032)

Real exchange rate

0.245

0.250

0.101

0.1458

(0.347)

(0.337)

(0.698)

(0.549)

Infrastructure

0.103

-0.291***

0.120

-0.190

(0.325)

(0.095)

(0.252)

(0.327)

Control of corruption

1.887

8.971**

(0.487)

(0.016)

Control of corruption x infrastructure

-0.761*

(0.005)

Government effectiveness

-4.124

1.340

(0.140)

(0.741)

Government effectiveness x infrastructure

-0.552***

(0.058)

Constant

9.152

14.465

5.460

9.968

(0.016)

(0.001)

(0.148)

(0.028)

AR (1)

0.005

0.003

0.001

0.001

AR (2)

0.657

0.718

0.639

0.647

Sargan test

0.393

0.568

0.435

0.473

Note: The values in parenthesis are the probability values *, **, *** are the significant levels of 1, 5 and 10 % levels of significance

infrastructural improvement for manufacturing export. As it is evidence from this study, there are other undertones that affect the expected result. We suspect that the available structures to drive institutions are not readily available and so, institutions are not able to achieve its objective of complementarity. In most developing countries—for which ECOWAS is no exception—institutions are measured by the policies that shows government's objective in reducing corruption and enhancing their effectiveness (see Henry and Miller 2008). Not to forget, the measures of institutions (control of corruption and Government effectiveness) are based on the perception of some groups, whose opinions are not distant from the public policies that support institutional growth. Like Acemoglu and Robinson (2012) observed in their blog on “Why Nations Fail: The Origins of Power, prosperity and Poverty”, institutions should go beyond policies and focus on structures that constraint, enhance and facilitate the application of policies. Most likely, ECOWAS countries are lacking in this regard and that's why the measures of institutions are not in sync with the improvement of infrastructure for enhanced manufacturing export.

Another important undertone is the 'power' of public officers who are supposed to enforce public policies. In African countries and ECOWAS, we find public officers who are very powerful and most times, they use their power to inform the dictates of public policies (Jo-Ansie 2007). In situations like this, it is expected that institutional development will most likely not achieve its objective. In this case, the submission of Acemoglu and Robinson (2012) that institutions should create a structure that drives incentives for the implementation of policies and in the case of Africa, create a cost for non-compliance. The situation in Africa is that: it is not as if cost and incentives are not embedded in the institutional structures, but such structures are not compelling to curtail excessiveness of public officers. Probably, the regional community can begin to play oversight to check public officers' compliance with institutional dictates.

5.1 Sensitivity Checks

The first sensitivity check is to ascertain the consistency of the result when excluding Nigeria from the sample of this study. The main reason for this is due to the economic size of Nigeria in the region. As it is, Nigeria's economic size is more than 57 % of the entire ECOWAS' economy (World Bank 2010) and this imply that their presence in the composition of the sample will likely influence the result. However, the result in Table 10 contradicts this expectation and it was obvious that—irrespective of the inclusion or exclusion of Nigeria, the interactive variable was signed similarly. In the last rows of the table, the behaviour of the interactive variables (Corruption x Infrastructure; Government Effective-ness x Infrastructure) was negative in all the columns. This further validates our earlier findings and submissions.

The second sensitivity check is to confirm whether the interactive variable still maintains its signs—as it is in the earlier Table 10—by checking the effect of alternative estimation technique and measures of investment. The alternative estimation technique is the use of OLS and FGLS; and instead of using the manufacturing value added, the gross fixed capital formation was used. The essence of the Feasible Generalised Least Square (FGLS) technique was applied because it allows for the presence of heteroscedasticity across the sampled countries and autocorrelation within the panels. This provides panel-corrected standard errors estimates.

Table 10 SGMM results excluding Nigeria

1

2

3

4

Manufacturing exports (-1)

0.376* (0.000)

0.363* (0.000)

0.369* (0.000)

0.343* (0.000)

Manufacturing value added

Exchange rate

Infrastructure

Corruption

Corruption x Infrastructure

-0.733***

(0.009)

Government effectiveness

Government effectiveness x infrastructure

-0.514***

(0.083)

Constant

10.274

(0.010)

14.647

(0.001)

6.988

(0.082)

11.102

(0.018)

AR (1)

0.004

0.004

0.011

0.002

AR (2)

0.655

0.714

0.644

0.649

Sargan test

0.471

0.633

0.494

0.555

Note: The values in parenthesis are the probability values. The sign '✓' imply that the variables were included in the estimated model. When this sign is not included, it imply that the variable was not included

*, **, *** are the significant levels of 1, 5 and 10 % levels of significance

The Ordinary Least Square (OLS) regression was included as a complementary estimation technique.

The result from this analysis presents the same sign for the interactive variables. As it is, column 1–4 of Table 11, where the alternative estimation technique was used for the baseline model that was estimated in Table 10, still presents negative signs. In essence, the stance that institutions in ECOWAS countries do not play a complimentary role in the infrastructure-manufacturing export nexus is valid and not informed by the estimation technique applied in reaching such conclusions.

In the same Table 11, the fifth to the eighth Column present a scenario where an alternative measure of investment and estimation technique was used. In these columns, the main variable of interest (Infrastructure x Institution) still maintains its negative sign and significant in all the columns. We can re-emphasise at this point that irrespective of the covariate applied (especially with regards to the measure of investment), the interactive variable still maintains its negative sign.

5.2 Conclusion

The main result from this study is that: contradictorily, institutions in ECOWAS countries do not play a complementary role to infrastructural development for improved manufacturing export. This result is robust, despite the alternative

Table 11 Sensitivity checks (dependent variable: manufacturing exports)

Control of corruption

Government effectiveness

control of corruption

Government effectiveness

OLS

FGLS

OLS

FGLS

OLS

FGLS

OLS

FGLS

Exchange rate

Manufacturing value added

Gross fixed capital formation (% of GDP)

Infrastructure

Institution

Infrastructure x Institution

-0.139

-0.139

-0.227***

-0.227**

-0.117*

-0.117*

-0.161*

-0.161*

(0.226)

(0.205)

(0.060)

(0.047)

(0.000)

(0.000)

(0.009)

(0.006)

Constant

2.203

2.203

2.178

2.178

2.805

2.805

2.547

2.547

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

R2

0.199

0.241

0.237

0.191

Wald

20.320

26.040

31.720

24.150

Prob. value

(0.004)

(0.001)

(0.001)

(0.000)

(0.001)

(0.000)

(0.001)

(0.002)


Note: The values in parenthesis are the probability values. The sign '✓' imply that the variables were included in the estimated model. When this sign is not included, it implies that the variable was not included

*, **, *** are the significant levels of 1, 5 and 10 % levels of significance

estimations. In the light of this finding, two main issues were identified as possible reasons for this contradictory result: the unavailability of structures that drive institutions may be a possible cause and the powerfulness of public officers who are supposed to enforce public policies.

Based on this finding, it is recommended that ECOWAS—as a regional economic community—can begin to play a supervisory role for countries in the community. By supervisory role, we imply that despite the 'beautiful' policies made by countries to forestall institutional development, there is the need for ECOWAS to ensure that related public officers adhere to the enforcement of such policies. Apart from this, the regional community can begin to develop frameworks that put public officers in member countries to check the applicability of member countries' policies. The reason for this policy recommendation is that; ECOWAS countries are not lacking in the development of policies—that shows institutional development—but the political will to put in place structures that ensures the accomplishment of the policies is probably not sufficient to enhance its effect. Therefore the regional community can act as a monitoring/enforcing body.

Just like it is obtainable in most empirical studies, we identified an area for future studies: that is, future empirical studies can focus on the consistency of our result when other measures of infrastructure are applied in our empirical model. The realisation of such study will be faced with data constraint in terms of macroeconomic data that reflects infrastructural development in African countries. If this constraint is mitigated, then a robust result will be necessary to check the consistency of the findings of this study.

Acknowledgement The authors are grateful to CREPOL and for the grant that made it possible to attend the annual conference on regional integration in Africa (ACGRIA5), organized by CREPOL, July 2014 at Praia, Cape Verde, where the first draft of this paper was presented. All comments from the conference participants are highly appreciated.

 
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Philosophy
Political science
Psychology
Religion
Sociology
Travel