Linkages Between Formal Institutions, ICT Adoption, and Inclusive Human Development in Sub-Saharan Africa

Abstract

This study empirically assesses the effects of formal institutions on ICT adoption in 49 African countries over the years 2000–2012. It deploys 2SLS and FE regression models (a) to estimate the determinants of ICT adoption and (b) to trace how ICT adoption affects inclusive development. The results show that formal institutions affect ICT adoption in this group of countries, with government effectiveness having the largest positive effects and regulations the largest negative effects. However, while formal institutions generally affect ICT adoption positively, population and economic growth tend to constrain ICT adoption more in low-income countries than middle-income countries. The results further demonstrate that the effects of ICT adoption on development are comparable to those of domestic credit and foreign direct investment. Ceteris paribus, one may conclude that external factors like foreign aid are more limiting to inclusive development than internal factors. This suggests that developing countries, African countries in this specific case, can enhance their ICT adoption for development by improving formal institutions and by strengthening domestic determinants of ICT adoption. Both represent opportunities for further research.

Supplementary Material for the Editor

1.1 Supplement 1: The Theory Behind the Model

We assume a basic Schumpeterian model in which the economic activity is described as:

$$ {Y}_i=\left({A}_i^{\alpha_i}{S}_i^{\beta_i\ }{X}_i^{\gamma_i}\right) \exp \left({\mu}_i\right) $$

(10.3)

where, Y _i is the real GDP of the ith economy, in Schumpeter’s terminology A _i (technology, including ICT) and S _i (socioeconomic setting, including institutions) are “evolution components”, X _i are “growth components”, including conventional factors of production, and all variables are dated ([53]; cf. [54, 55]). Central to growth among X _i is capital accumulation, which over time depends on investment (I) equal to savings in a steady state. Savings come from profit (π) made possible by technological change and the socioeconomic setting surrounding it. The evolution of the socioeconomic environment is a function of resources, technology, and level of development, i.e.:

$$ \frac{d{ K}_{i t}}{ d t}= k\left[\frac{d{ I}_{i t}}{ d t}= f\left({\pi}_i\ \left({A}_i,{S}_i\right)\right)\right],\frac{d{ S}_{i t}}{ d t}= s\left({X}_i,,{A}_i,,{S}_i\right),\pi =\mathrm{profit}. $$

(10.4)

A Schumpeterian technological change is discontinuous due to five initiators: (a) introduction of new ideas, requiring technological know-how; (b) introduction of new production techniques for which funds (credit) are essential; (c) discovery of new sources of supply; (d) discovery of new markets; and (e) change in the structure and organization of the industry involved. Thus, in dynamic form Eq. (10.1) is characterized by the Schumpeter-Kondratiev waves (cycles), such that A _i over time is sinusoid, i.e.:

$$ {A}_i(t)={A}_0 \exp \left(\varphi t+ \cos \left( bt+\psi \right)\right) $$

and $\partial A/\partial t={A}_0\left(\varphi - b\right) \sin\ \left( bt+\psi \right) \exp\ \left(\varphi t+ \cos\ \left( bt+\psi \right)\right),$ which is consistent with [15] Eqs. (10.3 and 10.4 (p. 6)), but we do not pursue this line of thought further. Instead, from Eq. (10.1) we solve for A _i as:

$$ {A}_i={Y}_i^{1/{\alpha}_i}{S}_i^{-{\beta}_i/{\alpha}_i}{X}_i^{-{\gamma}_{i/{\alpha}_i}}. $$

(10.5)

Dividing both sides of Eq. (10.3) Equation 10.5 by some specific ${X}_i={X}_i^{\ast }$ such as population or labor (worker), and taking the natural logs on both sides, we get a per capita (per labor, per worker, per head) indicator of adoption with diffusion as follows:

$$ {\dot{\mathrm{A}}}_i={\alpha}_i^{\ast }{y}_i+{\beta}_i^{\ast }{\dot{s}}_i+{\gamma}_i^{\ast }{\dot{x}}_i+{\mu}_i $$

(10.6)

where $\begin{array}{l}\dot{\mathrm{A}}= \log \left(\frac{A_i}{X_i^{\ast }}\right)= IC{T}_i;{\alpha}_i^{\ast }=\frac{1}{\alpha};{\dot{y}}_i=\mathrm{economic}\;\mathrm{growth};\hfill \\ {}{\beta}_i^{\ast }=\frac{\beta_i}{\alpha_i};{\dot{s}}_i= \log \left(\frac{S_i}{X_i^{\ast }}\right)=\mathrm{economic}\;\mathrm{setting}\ \left(\mathrm{governance}\right);\hfill \\ {}{\gamma}_i^{\ast }=\frac{\gamma_i}{\alpha_i};{\dot{x}}_i=\frac{X_i}{X_i^{\ast }}=\mathrm{primary}\kern0.75em and\kern0.75em \mathrm{other}\kern0.75em \mathrm{drivers};\kern0.5em \hfill \\ {} and\kern0.5em {\mu}_i=\mathrm{the}\kern0.5em \mathrm{random}\kern0.5em \mathrm{classical}\kern0.5em \mathrm{error}\kern0.5em \mathrm{term}.\hfill \end{array}$

This all shows that in the main document, the ICT equation is the equivalent of 10.(6) Eq. (10.4) above, and the development equation is Eq. (10.1) 10.3.

1.2 Supplement 2: A Note on Country Classification by Income Level

The World Bank classifies countries as developing if they are low income ($0–1,045 per capita) and lower middle income ($1,046–4,125 per capita). Countries with upper middle incomes ($4,126–12, 735 per capita) and high incomes ($12,736 or higher) are classified as being developed. The classification is arbitrary. No particular line of reasoning is given for why the cutoff point between “developed” and “developing” is set at $12,735. There is no reason to believe that a country just below the cutoff line cannot be more “developed” than a country just above it. For instance, Equatorial Guinea has a higher average income level than both China and South Africa, but its industrial base and technological structure are miles far behind. This is one of the reasons we modified the World Bank and grouped African countries into two groups: low-income group consisting of 28 countries, and middle-income group made up of 21 countries. This reclassification is consistent with our understanding of both ICT and development in these countries.

1.3 Supplement 3: ICT Variable Definitions and Data Sources

Variables	Signs	Definitions	Sources
Mobile phone	Mobile	Mobile phone subscriptions (per 100 people)	WDI
Internet	Internet	Internet subscriptions (per 100 people)	WDI
Telephone	Telephone	Telephone subscriptions (per 100 people)	WDI
Political stability	PolS	“Political stability/no violence (estimate): measured as the perceptions of the likelihood that the government will be destabilized or overthrown by unconstitutional and violent means, including domestic violence and terrorism”	WGI
Voice and accountability	VA	“Voice and accountability (estimate): measures the extent to which a country’s citizens are able to participate in selecting their government and to enjoy freedom of expression, freedom of association, and a free media”	WGI
Government effectiveness	GE	“Government effectiveness (estimate): measures the quality of public services, the quality and degree of independence from political pressures of the civil service, the quality of policy formulation and implementation, and the credibility of governments’ commitments to such policies”	WGI
Regulation quality	RQ	“Regulation quality (estimate): measured as the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development”	WGI
Corruption control	CC	“Control of corruption (estimate): captures perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as ‘capture’ of the state by elites and private interests”	WGI
Rule of law	RL	“Rule of law (estimate): captures perceptions of the extent to which agents have confidence in and abide by the rules of society and in particular the quality of contract enforcement, property rights, the police, the courts, as well as the likelihood of crime and violence”	WGI
GDP growth	GDPg	GDP growth rate	WDI
Trade openness	Trade	Import plus exports of goods and services (% of GDP)	WDI
Population growth	Population	Total population growth (annual %)	WDI
Education	PSE	Primary school enrolment (% of gross)	WDI

1. WGI World Governance Indicators, WDI World Development Indicators, GDP gross domestic product

1.4 Supplement 4: ICT Summary Statistics

	Mean	SD	Min	Max	Obs
Mobile phone penetration	23.379	28.004	0.000	147.202	572
Internet penetration	4.152	6.450	0.005	43.605	566
Telephone penetration	3.039	5.810	0.005	32.455	565
Political stability	−0.543	0.956	−3.323	1.192	578
Voice and accountability	−0.646	0.737	−2.233	0.990	578
Government effectiveness	−0.771	0.620	−2.450	0.934	577
Regulation quality	−0.715	0.644	−2.665	0.983	578
Corruption control	−0.642	0.591	−1.924	1.249	579
Rule of law	−0.741	0.662	−2.668	1.056	578
GDP growth	4.714	6.322	−47.552	63.379	608
Trade openness	78.177	36.138	20.964	209.874	597
Population growth	2.361	0.948	−1.081	6.576	588
Education	97.446	25.895	32.199	181.700	470

1. SD standard deviation, Min minimum, Max maximum, Obs observations, Adj adjusted

1.5 Supplement 5: ICT Correlation Matrix (Uniform Sample Size: 407)

Governance variables						Control variables				Dependent variables
PolS	VA	GE	RQ	CC	RL	GDPg	Trade	Popg	PSE	Mobile	Internet	Telephone
1.000	0.636	0.605	0.538	0.614	0.767	−0.084	0.253	−0.271	0.255	0.298	0.312	0.470	PolS
	1.000	0.740	0.727	0.612	0.787	0.018	0.014	−0.250	0.248	0.274	0.325	0.459	VA
		1.000	0.845	0.979	0.874	0.030	0.021	−0.335	0.212	0.293	0.320	0.504	GE
			1.000	0.649	0.772	−0.025	−0.002	−0.247	0.217	0.264	0.176	0.286	RQ
				1.000	0.817	−0.090	−0.014	−0.309	0.118	0.273	0.342	0.565	CC
					1.000	−0.044	0.109	−0.286	0.219	0.274	0.332	0.530	RL
						1.000	0.029	0.157	0.083	−0.043	−0.002	−0.052	GDPg
							1.000	−0.380	0.167	0.259	0.158	0.228	Trade
								1.000	−0.172	−0.331	−0.414	−0.581	Popg
									1.000	0.288	0.224	0.181	PSE
										1.000	0.690	0.479	Mobile
											1.000	0.695	Internet
												1.000	Telephone

1. PolS Political stability, VA voice and accountability, GE government effectiveness, RQ regulation quality, CC corruption control, RL rule of law, GDPg GDP per capita growth rate, Popg population growth, PSE primary school enrolment, Mobile mobile phone penetration, Internet Internet penetration, Telephone telephone penetration

1.6 Supplement 6: IHDI Variable Definitions and Data Sources

Variables	Signs	Definitions	Sources
Inclusive development	IHDI	Inequality-adjusted human development index	UNDP
Mobile phone	Mobile	Mobile phone subscriptions (per 100 people)	WDI
Internet	Internet	Internet subscriptions (per 100 people)	WDI
Telephone	Telephone	Telephone subscriptions (per 100 people)	WDI
Foreign aid	Aid	Total official development assistance (% of GDP)	WDI
Private credit	Credit	Private credit by deposit banks and other financial institutions (% of GDP)	WDI
Remittance	Remit	Remittance inflows (% of GDP)	WDI
Foreign investment	FDI	Foreign direct investment net inflows (% of GDP)	WDI

1. UNDP United Nations Development Program, WDI World Development Indicators, GDP gross domestic product

1.7 Supplement 7: IHDI Summary Statistics

	Mean	SD	Min	Max	Obs
Inequality-adjusted human development	0.721	3.505	0.129	0.768	485
Mobile phone penetration	23.379	28.004	0.000	147.202	572
Internet penetration	4.152	6.450	0.005	43.605	566
Telephone penetration	3.039	5.810	0.005	32.455	565
Foreign aid	11.687	14.193	−0.253	181.187	606
Private domestic credit	18.551	22.472	0.550	149.78	507
Remittances	3.977	8.031	0.000	64.100	434
Net foreign direct investment inflows	5.332	8.737	−6.043	91.007	603

1. SD standard deviation, Min minimum, Max maximum, Obs observations, Adj adjusted

1.8 Supplement 8: IHDI Correlation Matrix (Uniform Sample Size: 324)

Foreign aid	Credit	Remittances	FDI	Mobile	Internet	Telephone	IHDI
1.000	−0.173	−0.037	0.411	−0.165	−0.196	−0.223	−0.382	Foreign aid
	1.000	−0.084	−0.065	0.514	0.511	0.614	0.529	Credit
		1.000	0.115	−0.050	−0.035	−0.062	−0.027	Remittances
			1.000	0.111	0.072	−0.029	−0.001	FDI
				1.000	0.749	0.504	0.626	Mobile
					1.000	0.669	0.649	Internet
						1.000	0.747	Telephone
							1.000	IHDI

1. Credit Private domestic credit, FDI foreign direct investment, Mobile mobile phone penetration, Internet Internet penetration, Telephone telephone penetration, IHDI inequality-adjusted human development index