Political Regime Types and Economic Development in Nigeria: Significance of Conflict and Corruption

Abstract

This article investigates the short- and long-run effects of political regimes (POLREGs) on economic development in Nigeria between 1984 and 2015. It looks at the effects of the conflict (CONFL) and corruption (COR) on economic development indicators and examines the interactive effect of POLREGs and COR as well as CONFL on economic development. COR and CONFL seem to be more prevalent in Nigeria during democracy relative to the periods under dictatorship. Using the Autoregressive Distributed Lag (ARDL) approach to cointegration, it derived a number of robust conclusions. Democracy in the long run yields higher economic development when it is devoid of CONFL and COR, while autocracy hinders economic development. In the short run however, more autocracy fosters economic development in Nigeria while democracy hinders it. COR portends grave threat to the development of Nigeria’s economy as it reduces development in the long run. Effect of CONFL on economic development in Nigeria is unclear. These findings highlight the need to establish effective anti-graft agencies to fight COR to the barest minimum in Nigeria. They also highlight the need to employ CONFL resolution mechanisms in resolving CONFL issues in the democratization process of the country.

JEL: D74, K420, P160, O1, Q550

Keywords

Conflict corruption democracy economic development Nigeria

Introduction

There has been no consensus in the literature as regards the connection between political regime (POLREG) type and economic freedom. Some observers, such as Friedman (1962), believe that the two freedoms are mutually reinforcing. However, some other observers view it as that democracy has either a negative effect on economic performance or no overall effect. Countries with dictatorships have been predicted to grow as rapidly as democracies, perhaps even faster. Although most of the rich countries in the world are democratic, the direction of causality is unclear. Gerring, Bond, Barndt and Moreno (2005) argued that one must keep in mind that many rich countries have become rich under dictatorship.

A high degree of corruption (COR) therefore deters investment, and democracy is being claimed to reduce COR, especially in relatively rich countries (Fjelde & Hegre, 2007) and when democracy is consolidated (Rock, 2009). Democracy is commonly believed to reduce COR (Kolstad & Wiig, 2011; Rock, 2003). However, the situation in Nigeria seems not to agree with this assertion as the level of COR got aggravated whenever the opportunity of a democracy avails the country. Also, on the other hand, the effect of democratization is argued to be weakened when accounting for the incidence of conflict (CONFL). Cervellati and Sunde (2012) claimed that the growth effect of democratization is heterogeneous and depends on the democratization scenario. Peaceful transitions to democracy have a significant positive effect on growth that is even larger than reported in the previous literature, whereas violent transitions have no or even negative growth effects. The contentions whether democracy or dictatorship spurs economic performance, and in turn development motivates this article to investigate the effect of POLREGs on economic development in Nigeria. To the best of the knowledge of this article, none of the studies in the ample literature have investigated the short- and long-run dynamics of the effect of democracy on economic development. This article however examines the short- and long-run effect of the level of democracy on economic development in Nigeria.

It also investigates the interactive effect of COR and CONFL with POLREGs on economic development. Apart from the introductory part, the second section reviews the relevant literature on the effect of POLREGs on economic development. The third section appraises the trend of COR, CONFL, socio-economic development and POLREGs in Nigeria within the study period, while the fourth section presents the methodology and data source. The fifth section contains the analysis and discussion while the last part concludes and suggests policies for policymakers.

The Arguments and Empirical Links between Political Regimes and Growth

The argument of whether democracy could affect growth by Przeworski and Limongi (1993) was in four parts. First, the argument highlights how regime types might matter for property rights.¹ The overall assessment of Przeworski and Limongi is thus that while everyone seems to agree that secured property rights foster growth, it is controversial whether or not democracies or dictatorships better secure these and they further conclude that the idea that democracy protects property rights is a recent invention. However, Knutsen (2011b), North, Wallis and Weingast (2009) and Timmons (2010) disagreed with Przeworski and Limongi’s claims with a counter argument that the median-voter-based model on redistribution of property captures only one aspect of the politics of redistribution.

Second, Przeworski and Limongi (1993, p. 54) highlight how POLREG types undermine investment. They claimed that the first modern statements that democracy undermines growth are those by Galenson (1959) and de Schweinitz (1959), who argued that democracy unleashes pressures for immediate consumption, which occurs at the cost of investment, hence of growth. A counter argument was however given by Knutsen (2011b) against the claims that democracy is inimical to economic development. He argued that contrary to claims of Huntington and Dominguez (1975) and Przeworski and Limongi (1993), most dictatorship do not generate very high savings and investment rates because dictators are self-interested, foreign direct investment is sensitive to protection of property rights and democracy likely strengthens property rights protection and democracy reduce COR which deters investment.

Third, Przeworski and Limongi (1993, p. 56) noted that scholars studying Latin America and East Asia have linked the economic performances of some dictatorships in these two regions to the autonomy of the dictatorial state. However, Olson (1982) claims that democracies are prone to capture from special interest groups. This may possibly lead to policies that are incoherent with the interests of the general public; economic growth may be sacrificed for the protection of specific business sectors or pivotal voting blocs whose interest is not aligned with economic growth. Knutsen (2011b) also refuted the claim and argued that if there is a lack of free and fair elections linking the regime to the broader electorate, no dictator could survive without backing from specific groups, be it the party, the landlord elite or the military.

Finally, Przeworski and Limongi (1993, p. 57) argued that the dictatorships are a source of inefficiency. State autonomies are harmful for economic performance and state is always ready to prey on the society (North, 1990), and only democratic institutions can constrain it to act in general interest. A dictator spends excessive amounts on a repressive apparatus instead of productive investments (Acemoglu & Robinson, 2006b). In view of this, if a dictator believes that modernization theory is correct, with economic growth and industrialization leading to a strong middle class and calls for democracy, the dictator will be better off not industrializing (Acemoglu & Robinson, 2006a). In democracies on the other hand, leaders who engage in predatory activities are more likely to be detected because of freedom of media, more likely to be stopped by other institutions like the legislature and courts, and more likely to be thrown out of office in the next election.

Several academics and policymakers seem to believe strongly in the ‘Lee thesis’ (Helliwell, 1994; Leblang, 1997; Przeworski & Limongi, 1993; Przeworski, Alvarez, Cheibub, & Limongi, 2000; Sen, 1999), credited to former Singaporean Prime Minister, Lee Kuan Yew. The Lee thesis postulates that particularly in developing countries, a strong dictatorship is necessary for promoting economic development. However, some early studies found a negative effect of democracy on economic growth (Helliwell, 1994; Przeworski & Limongi, 1993; Rachdi & Saidi, 2015). In the recent time, statistical studies relying on more proper estimation techniques and data have found either no significant effect (Helliwell, 1994; Przeworski, Alvarez, Cheibub, & Limongi, 2000; Remmer, 1990) or a positive significant effect (Baum & Lake, 2003; Bueno de Mesquita, Smith, Siverson, & Morrow, 2003; Doucouliagos & Ulubasouglou, 2008; Leblang, 1997).

Studies of Arat (1988), Knutsen (2011a), Goldsmith (1995) found a positive correlation between democracy and growth. Other studies like Lake and Baum (2001), Bueno de Mesquita, Smith, Siverson, & Morrow (2003), Acemoglu and Robinson (2006b) have also found positive effects democracy on socio-economic indicators. Several other studies have also found the effects of dictatorship on growth and economic development (Gandhi, 2008; Knutsen, 2010; Olson, 1993; The World Bank, 1993; Wade, 1990).

Evidence from Nigeria

The Nigerian Democratic Experience

Before colonial rule and the introduction of Western democracy, different parts of Nigeria have inherent in them their indigenous political systems. In the Yoruba political system, the Obaship (Kingship) guarantees good governance and the representation of people through established institutions. The Alaafin (King) of Oyo, who many often praised as having the powers of life and death, is in practice, not so absolute in exercising his powers. The Basorun, who is the head of the Oyomesi, the committee responsible for the selection of the Alaafin, is by Oyo constitution, empowered to order an Alaafin to abdicate the throne, when the Alaafin is considered to have violated the Empire’s constitution (Aderibigbe, 1965). This checks and balances inherent in African political system, particularly Nigeria, prevent the occurrence of absolutism and misuse of power by their leader (Omoiya, 2012). On the other hand, emir’s decisions in the emirate political system of Northern Nigeria are directly subject to the agreement of his Council (Hunwick, 1965). The Emirate Council consists of the Emir himself, the Waziri, the Khadi, the Chief Imam and other prominent chiefs that vary from place to place (Hunwick, 1965).

In 1900, the British government established colonial rule on the colony of Lagos, protectorates of the South and the North. In 1906, the British Colonial administration formally amalgamated the colony of Lagos and the protectorate of the South (Obaro, 1977), which later accounted for the 1914 amalgamation of the colonies and protectorates of the south and north, which was named Nigeria. The colonial government entrenched in its administration various tenets of democracy in the then British West African Countries. The introduction of elective principle brought about increased political activities to Lagos and, in turn, resulted to the emergence of political parties. Richard Administration’s constitutional provision in 1986 extended the electoral principles to the Northern region. However, the seed of discord was sowed particularly on the electoral process of the Nigerian democracy as this is evident in the gradual way the colonial policies were implemented. After independence, the Nigerian state was compartmentalized into three main regions: North, East and West. Each region was committed to themselves rather than to the Nigerian project as a whole. The fragile unity in diversity encouraged each of the three regions to concentrate more on regional developments and programmes that will respectably sustain them, in case of eventual dismemberment of the Nigeria State.

The Nigerian democratization started experiencing set back in 1963, with the disagreement that accompanied the 1963 election and population census, which had a negative impact on the growth of Nigeria’s democracy (Paden, 1986). The political tumult that accompanied various disagreements naturally opened up the Nigeria State to events that culminated into the 1966 coup, which truncated the first elected civilian administration in Nigeria (Post & Vicker, 1973). There was a sectional perspective into the 1966 coup, which claimed the lives of the Premiers of both Western and Northern regions and spared the lives of their counterparts in the Mid-Western and Eastern regions, which therefore motivated a counter coup in July 1966. The resultant sectarian crisis and civil unrest metamorphosed into a Civil War consequent upon the decision of the eastern region to secede (Niven, 1970).

A number of coup took place in the 1970s after the end of Civil War and the beginning of another democracy in 1979, which led to a change in government from Gowon’s to Muritala’s administration and then to Obasanjo’s administration. An election was conducted in 1979 and brought in Shagari as the Second Republic President in October 1979. The element of segmentation along regional and ethnic divide that characterized the polity since independence was also visible in the second republic.

After Shagari’s first tenure, another election conducted in 1983 was marred with electoral malpractices and created another opportunity for the military to launch another coup that brought in Buhari (Akinbobola, 2000). However, it is pertinent to note that from 1999, Nigeria has been enjoying the longest period of democracy since independence. Figure 1 shows the trend of POLREGs’ characteristics in Nigeria from 1984 to 2015. The purple line indicates the trend of the extent of democracy, the red-dotted line represents autocracy, while the blue dashed-line POLREG indicates POLREG trend. The measures are composite indices derived from the coded values of authority characteristic component variables² according to the formulas, originally designed by Gurr.

Figure 1 shows that there was strong dictatorship between 1984 and 1999; DEMOC line was at its minimum (0) while AUTOC was close to its maximum score, and POLREG moves around its minimum value implying that Nigeria experienced a strong dictatorship in this period. Between 1984 and 1998, the mean score of the democracy score was 0 showing that there were little or no characteristics of democracy in place during that period. The transition from dictatorship to a civil rule took place in 1999, which marked the beginning of an upwards trend in POLREGs and democracy. The transition also marked the beginning of the fourth republic which is the longest period of civil rule in Nigeria after other democracies were short-lived. The DEMOC and POLREG lines rose further in 2015, showing more democracy, the first time in the history of the country’s that power is being transited from one political party to another. The government of Jonathan of People’s Democratic Party lost the 2015 election to Buhari of All Progressive Congress. Despite the positive remarks on 2015 election in Nigeria by Freedom House (2015), Nigeria was categorized as partly free using ratings from political and civil rights enjoyed by the citizens.³

Figure 1.

Source: The author.

Democracy and Socio-economic Developments in Nigeria

Historically, the dearth of democratic experience has created enormous challenges to institutionalizing democracy and national integration for national development in Nigeria (Egbefo, 2015, p. 60). After the Nigeria’s transition to democracy in 1999, the expectations of the majority of Nigerians was that democracy would engender efficient, accountable, transparent and participatory governance. It was thought that democracy would promote sustainable socio-economic development. However, contrary to the belief of many Nigerians, the 17 years of democracy has not significantly improved the socio-economic conditions of Nigerians.

COR became a major bane on development in the Forth Republic. Rather than popular expression of power by the people, there was obvious disconnect between the government and the ruling elite on the one hand and the masses on the other. This development fosters rampant COR and economic sclerosis because there is no investment in infrastructure as the country’s leader’s cream off its wealth (Burleigh, 2013, p. 1). COR in the public sector degenerated to outright looting of the nation’s treasury and wealth by unscrupulous politicians and public servants at the different level of the country’s governance (Unumen & Emordi, 2012). It was reported that 136 million barrels of crude oil worth $11billion (£7.79 billion) were illegally siphoned off in first 2 years from 2009 to 2011 (Burleigh, 2013, p. 1).

A nation with abundance of potentials, both in human and natural resources, is rated among the 60 poorest nations in the world with a purchasing power parity (PPP) per capita of $5929 (Gregson, 2017). Between 2004 and 2010, the economy grew strongly at an average annual growth rate of 6.6 per cent making it the fifth fastest growing economy in the world. By 2010, the country’s growth rate stood at 7.8 per cent and by 2014 it had become the largest economy in Africa (Unumen & Oghi, 2016, p. 39). In 2015, the growth rate dropped to 2.7 per cent while it dropped further to −1.7 per cent by 2016 (IMF, 2017, p. 7). However, in its own report, the Nigeria’s National Bureau of Statistics (NBS) reported that for the full year 2016, GDP contracted by −1.51 per cent, indicating real GDP of N67,984.20 billion for the year, the worst in more than 30 years (Obasi & Taiwo-Obalonye, 2017).

Unumen (2014) and Unumen and Oghi (2016) stated that by all indices of development, the Nigeria remains as an underdeveloped country. The country’s relative poverty rate increased from 54.5 per cent in 2004 to 69 per cent by 2010. The percentage of Nigerians living in abject poverty increased from 54.7 per cent in 2004 to 61.2 per cent in 2010 (National Bureau of Statistics [NBS], 2012) and per capita poverty rate registers at 35.2 and 33.1 per cent of the population in 2009/2010 and 2012/2013, respectively (The World Bank, 2014, p. 17). Life expectancy (LEXP) at birth rose from 46.6 in 2000 to 53.1 by the end of 2015 (UNDP, 2016, p. 2), which is still very low compared with what we have in developed countries. Mean years of schooling increased by 0.8 years, from 5.2 in 2005 to 6.0 in 2015 and expected years of schooling also increased 2.0 years, from 8.0 in 2000 to 10.0 in 2015.

Figure 2 shows the trend of the POLREGs, COR, gross national income (GNI) and GDP per capita growths in Nigeria between 1984 and 2015. The GNI per capita growth is the orange-line that has almost identical fluctuating trend with the GDP per capita growth in black line. The POLREGs line is the red-line while COR⁴ is the blue-line representing COR measure sourced from the political risk ratings of the International Country Risk Guide (i.e., ICRG), from the PRS group report. In Figure 2, the GNI and GDP per capita growths exhibited high level of fluctuations unlike POLREGs and COR trends. The movement of POLREGs either towards full democracy or full autocracy does not reflect in the movement of GNI or GDP per capita growths. However, one thing that is noticed in this period of study is that GNI and GDP per capita growths had the highest percentage growth of 29.5 per cent and 30.4 per cent consecutively in 2004 during democracy and the lowest percentage growth of −13.1 per cent and −15.8 per cent consecutively in 1987 during Babangida’s administration. Also, after the 2015 elections and the transition of power from one political party to another, the GNI and GDP per capita growths became negative implying a negative growth. It also shows that COR has a relatively higher risk point total between 1984 and 1998 (i.e., relatively lower risk of COR), and a relatively lower risk points total between 1999 and 2015 (i.e., relatively higher risk of COR). Though, the COR risk point total of Nigeria is generally low over the years, Figure 2 indicates that the period of dictatorship experienced lesser risk of COR compared with the period of democracy in Nigeria.

Figure 2.

Source: The author.

Political Regimes and Conflict: The Nigerian Experience

Since Nigeria gained its independence from British colonialism, and advanced to a post-colonial order which was replete with socio-economic and political dilemmas, one major problem post-colonial Nigeria faced was the obstinate task of governing a multifaceted nation, comprised of 36 regional states divided along ethno-religious lines, up to 300 ethnic groups and a plethora of linguistic dialects, in addition to three distinct religious groupings. Nigeria was confronted with the efficient administration and governance of a broad-based society with a multiplicity of interests, political ideologies, values, traditions and cultural inclinations. The peak of an atmosphere of mutual mistrust from different regions of Nigeria came with the advent of the Biafra secessionist battle of 1967, which has resurfaced again more recently.

After about 30 years of dictatorship, Nigeria got back into conventional democracy. While this development was seen by some people as an avenue to explore dividends and goodies of democracy, others saw it as an opportunity to express grievances (Adetoye & Omilusi, 2015), the outcome of which is the occurrence and re-occurrence of ethno-religious, resource-base, socio-economic and political CONFLs. The diversity which has been threatening the unity of the country since the pre- and post-independence periods and militated against the deepening of her nascent democracy also persisted in the present fourth republic. This has manifested in form of call for Sovereign National Conference in some parts of the country, agitations for secession by some regions of the country in the case of Biafra, violent fight against Western philosophy and education, resource control as well as persistent wave of political, Herdsmen–Farmer CONFL, inter-ethnic and sectional violence among others. Nigeria’s fourth republic has been adjudged very chaotic; Elaigwu (2005) identified 17 major violent CONFLs in Nigeria from May 1985 to 1 May 1999, and from 31 May 1999 to June 2005 he identified at least 121 cases of CONFLs in Nigeria. Adebanwi (2004) in the similar view claimed that it appeared the dawn of democracy provided the atmosphere to ventilate bottled-up frustrations, grievances and fears generously and often times recklessly. Another study also heaped the blame of Nigeria’s CONFLs on COR and the abrupt termination of the late Abacha, one of the ruthless dictators Nigeria has ever had. His demise triggered spontaneous culture of revivalism and agitation among different social groupings (Osita, 2007, p. 21).

Figure 3 shows the trend of POLREGs⁵ characteristics, and the trend of CONFL⁶ I made use of ACTOTAL the total summed magnitudes of all (societal and interstate conflicts) MEPV.⁷ CONFL represents ACTOTAL in Figure 3. Figure 3 shows that CONFL was high before the 1980s, but later dropped to ‘0’ between 1994 and 1996 during dictatorship. Within the period of the study, the CONFL value got its lowest value of ‘0’ between 1994 and 1996, implying no episodes of CONFL, and its highest magnitude score of ‘5’ between 2009 and 2010, implying a high episode of CONFL. It should also be noted that since the transition to civil rule in 1999, CONFL episodes have though been fluctuating but remain high Nigeria. One can adduce the rising trend of CONFL during democracy in Nigeria to the high rate of political violence bewildering her democratic process. Nigeria’s democracy in the view of this article can also be described as what Vreeland (2008) referred to as anocracy. The increasing CONFL trend during democracy may be one main reason why it has not delivered economic growth in Nigeria.

Figure 3.

Source: The author.

Methodology

The study applies the ADRL bounds testing approach developed by Pesaran, Shin and Smith (2001) to investigate the primary objectives of the article. For the purpose of achieving the objectives of this article, the study is anchored on the basic theoretical underpinning of Cobb–Douglas production function as adapted and developed by Fosu (2001) which states that:

q = a_{1} + a_{2} p + b_{1} l + b_{2} p l + c_{1} k + c_{2} p k + u

(1)

Where q is output growth, p measures Political instability, l and k are the respective growth rates of labour and capital, and u is the appended stochastic perturbation term.

Thus, this study adapts Fosu’s model and estimates the following models:

\begin{matrix} q_{1} = a_{1} + b_{1} P O L R E G_{t} + b_{2} C O N F L_{t} + b_{3} C O R_{t} + b_{4} G D P G_{t} + \\ b_{5} G F C F G_{t} + b_{6} G E_{t} + b_{8} I N T_{t} + b_{9} E X C H_{t} + u_{1} \end{matrix}

(2)

\begin{matrix} q_{n} = a_{1} + b_{1} P O L R E G_{t} + b_{2} C O N F L_{t} + b_{3} C O R_{t} + b_{4} G D P G_{t} + \\ b_{5} G F C F G_{t} + b_{6} G E_{t} + b_{8} I N T_{t} + b_{9} E X C H_{t} + u_{1} \end{matrix}

(3)

Where q₁……q_n are the regressands⁸ for each of the model; POLREG_t represents POLREG measuring the type of political system operating in the country; CONFL_t is CONFL measuring the level of absence of peace; COR_t is COR measuring the presence of COR in the country; GDPG_t is the growth rate of gross domestic product; EXCH_t is the exchange rate (EXCH); GE_t is government spending; GFCFG_t is gross fixed capital formation growth measuring investment level; INT_t is real interest rate and u is the appended stochastic perturbation term. a₁, b₁ … … … … . . b₆, are the parameters to be estimated.

While for the ‘interactive effects’, the following models were estimated:

\begin{matrix} q_{1} = a_{1} + b_{1} P O L R E G_{t} * C O R_{t} + b_{2} P O L R E G_{t} * C O N F L_{t} + b_{3} P O L R E G_{t} + \\ b_{4} G D P G_{t} + b_{5} G F C F G_{t} + b_{6} G E_{t} + b_{8} I N T_{t} + b_{9} E X C H_{t} + u_{1} \end{matrix}

(4)

\begin{matrix} q_{n} = a_{1} + b_{1} P O L R E G_{t} * C O R_{t} + b_{2} P O L R E G_{t} * C O N F L_{t} + b_{3} P O L R E G_{t} + \\ b_{4} G D P G_{t} + b_{5} G F C F G_{t} + b_{6} G E_{t} + b_{8} I N T_{t} + b_{9} E X C H_{t} + u_{1} \end{matrix}

(5)

Where q₁ … …q_n are the regressands as it is in Equation 11; POLREG_t * COR_t is the integration of POLREGs and COR; POLREG_t * CONFL_t is the interaction between POLREGs and CONFL; POLREG_t * CONFL_t, POLREG_t, GDPG_t, GFCFG_t, GE_t, INT_t, EXCH_t and u₁ are the same as we have in Equation 11.

The article estimates eight models using the same set of regressors. It uses economic development index (i.e., ECNDEV_t) as the first regressand and other single economic development indicators (i.e., AVAD_t, CPER_t, EGPC_t, GDPPC_t (per capita GDP), GNIPC_t (Gross National Income per Capita), LEXP_t, SCHENROL_t) were used as regressands in other models. The same process was repeated for the interactive effects of both POLREGs and COR on one hand, and POLREGs and CONFL on the other hand on economic development. Therefore, q₁ … … … q_n indicate the list of regressands as used in the study while u₁ … … … u_n represent the list of error terms. In order to conserve space, the Autoregressive Distributed Lag (ARDL) models for Equations 10 and 17 are not presented.

Data Measurement and Source

Due to the inaccessibility of data, this article uses time series data ranging from 1984 to 2015. Data on POLREG were sourced from Polity IV Project of Marshall and Jaggers, Center for Systemic Peace,⁹ while the data on CONFL were sourced from MEPV magnitude scores, in the CSP database.¹⁰ The data on COR¹¹ were sourced from political risk ratings of the ICRG, of the PRS group report. Data of LEXP, Agricultural Value-added (AVAD), real per capita GDP (RGDP), GNI per capita, Consumption Per Capita (CPER), Electric power consumption (kWh per capita) (EGPC), secondary school enrolment (SCHENROL) are sourced from World Development Indicators (henceforth WDI) of the World Bank. Economic development index was computed from data of LEXP, AVAD, RGDP, CPER, EDU, using the Principal Component Analysis (PCA). Other data include GDP growth (GDPG), EXCH, Government Expenditure (GE), Interest rate (INT) and Gross Fixed Capital Formation Growth (GFCFG), which were also sourced from the WDI and are used as instrument and regressors for the models.

Many economic development indicators have been introduced in the literature. There is serious contention about which of these indicators best fit to measure economic development. Hence, there is need to construct a comprehensive measure of economic development. However, this study uses six different components to represent different aspects of development. AVAD per worker captures the level of rural development in Nigeria while household final consumption expenditure per capita and GDP per capita capture resources need for a decent standard of iving or poverty (used by Chirino, Valdivielso, & Melian, 2006). Electric power consumption per capita captures social or infrastructural development; The LEXP at birth captures longevity or capability to leading a long and healthy life (Chirino, Valdivielso, & Melian, 2006). And finally, SCHENROL captures the level of iteracy. Before undertaking PCA, I checked the factorability of variables with the Barlett’s test for sphericity and Kaiser–Meyer–Oklin (KMO) coefficient. The Barlett’s test converts the calculated determinants of the matrix to a c² statistic, which is tested for significance. The null hypothesis of the test is that variables are collinear. The KMO test, on the other hand, entails the comparison of the size of the variables’ correlation coefficients with the size of the partial correlation coefficients. In the KMO test, a minimum value of 60.0 is necessary for an acceptable PCA. The results on Table 1 present Barlett’s and KMO tests as well as the PCA. The results show that the six variables may be assembled into another set of factor using the PCA. Therefore, the values of the first PCA are to calculate the weights for the ECNDEV.

Table 1.

Construction of Economic Development Index

Test for factorability
Determinant of the matrix of correlation							0.0023
Barlett’s test for sphericity							365.718
							(0.000)***
Kaiser-Meyer-Oklin measure							0.777
Principal Components/Correlation						Number of Obs = 32
			Number of comp. = 6
			Trace = 6
		Rotation: (unrotated = principal)			Rho = 1.0000
		Component	Eigenvalue	Difference	Proportion	Cumulative
		Comp1	5.48021	5.19945	0.9134	0.9134
		Comp2	0.280759	0.142891	0.0468	0.9602
		Comp3	0.137868	0.0766372	0.0230	0.9831
		Comp4	0.0612304	0.0263701	0.0102	0.9933
		Comp5	0.034860	0.029788	0.0058	0.9992
		Comp6	0.005072		0.0008	1.0000
Principal Components (eigen vectors)
Variable	Comp1	Comp2	Comp3	Comp4	Comp5	Comp6	Unexplained
AVADPW	0.4204	−0.0598	0.0854	−0.5158	0.5845	−0.4523	0
CPER	0.3745	0.8996	0.1472	0.1388	−0.0006	0.0979	0
EGPC	0.4003	−0.3575	0.7468	0.3812	−0.0276	0.0902	0
GDPPC	0.4210	−0.0785	−0.1390	−0.2072	−0.7757	−0.3907	0
LEXP	0.4216	−0.1687	−0.2127	−0.3701	−0.0110	0.7819	0
CHENROL	0.4095	−0.1573	−0.5905	0.6241	0.2361	−0.1167	0

Source: Author’s computation with SPSS and STATA.

Data Analysis and Empirical Results

Effects of Political Regimes on Economic Development in Nigeria

To validate the applicability of the ARDL bounds testing method, I employed the unit root tests to evaluate the order of integration of the variables. Both the Augmented Dickey–Fuller (ADF) and Phillip–Perron (PP) tests are employed. In Table 2, all variables are of order I(0) and I(1), and/or a combination of both, and none is integrated of higher order. Therefore based on these results, it is acceptable to apply the ARDL technique.

Table 2.

Unit Root Test Result

Variables	Augmented Dickey–Fuller			Philip–Peron
Variables	Levels	1st Diff	Remarks	Levels	1st Diff	Remarks
AVADPW	1.855547	−4.435753***	I(1)	1.779715	−4.508832***	I(0)
CONFL	−1.873426	−4.861046***	I(1)	−2.0372	−4.970959***	I(1)
COR	−2.075726	−3.810971***	I(1)	−1.4094	−3.816396***	I(1)
CPER	−0.632861	−9.162554***	I(1)	−1.3996	−9.098266***	I(0)
ECNDEV	0.948957	−4.224064***	I(1)	0.7951	−4.240506***	I(1)
EGPC	−1.281544	−6.766623***	I(1)	−1.2513	−6.880446***	I(1)
EXCH	0.092548	−5.050588***	I(1)	0.104743	−5.048326***	I(1)
FDI	−3.547315**		I(0)	−3.531109**		I(0)
GDPg	−4.582585***		I(0)	−4.589829***		I(0)
GDPPC	0.948953	−4.224067***	I(1)	0.795047	−4.24051***	I(1)
GDPPCg	−4.588589***		I(0)	−4.595895***		I(0)
GE	−3.173381**		I(0)	−3.171054**		I(0)
GFCF	−1.686344	−2.732537*	I(1)	−1.749478	−5.580418***	I(1)
GFCFg	−2.690807*		I(0)	−5.519003***		I(0)
GNIPC	1.008125	−2.918069*	I(1)	0.979097	−5.336634***	I(1)
GNIPCg	−5.797174***		I(0)	−5.790433***		I(0)
INT	−5.497889***		I(0)	−5.497789***		I(0)
LEXP	−0.684513	−1.821222*	I(1)	2.107317	−1.082292*	I(1)
POLREG	−0.866274	−3.400099**	I(1)	−0.568395	−3.282385**	I(1)
POLREG* CONFL	−0.704549	−5.300389***	I(1)	−0.101768	−5.902282***	I(1)
POLREG* COR	−0.415487	−4.013274***	I(1)	−0.540695	−4.048815***	I(1)
POPg	−4.767169***		I(0)	−1.259856	−1.992171*	I(1)
CHENROL	−0.035657	−0.986311*	I(1)	−0.169627	−3.981691***	I(1)

Source: Author’s computation with Eviews 9.

Notes: ***, **, * indicate the level of significance at 1 per cent, 5 per cent and 10 per cent, respectively. Unit root was conducted with intercept and no trend.

Next, I test the presence of long-run relationships among the variables used. Table 3A reports the results of the ADRL bounds cointegration tests. The Wald tests (F tests) for joint null hypothesis that the coefficients of the lagged variables in level form are zero (no cointegration between the variables), and the results of the calculated F-statistics and the values for both upper and lower bounds are presented. The critical value bounds of the F-statistic with k = 8 at 5 and 10 per cent levels of significance are presented in Table 3A. All calculated F-statistic values for each model are greater than critical values at upper bound I(1), thus implies that there exist long-run cointegration among the variables used in each models (i.e., models 1–8). The results of the F-statistic for the models used in checking the ‘interactive effect’ are presented in Table 3B.

Table 3.

Testing for Long-run Cointegration; F-statistic

Model 1: (Dependent variable: ECNDEV)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		149.1226
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55
Model 2: (Dependent variable: AVAD)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		3.46738
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	2.79	1.66
5%	3.11	1.91
Model 3: (Dependent variable: CPER)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		3.968063
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55
Model 4: (Dependent variable: EGPC)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		3.738762
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55
Model 5: (Dependent variable: GDPPC)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		3.699965
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55
Model 6: (Dependent variable: GNIPC)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		12.81282
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55
Model 7: (Dependent variable: LEXP)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		10,848.78
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55
Model 8: (Dependent variable: SCHENROL)		F-Statistic
F(POLREG, COR, CONFL, GDPG, GFCFG, GE, INT, EXCH,)		37.89263
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.06	1.95
5%	3.39	2.22

Source: Author’s computation with Eviews 9.

Table 3B.

Testing for Long-run Cointegration; F-statistic

Model 1: (Dependent variable: ECNDEV)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		229.6012
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 31
10%	3.06	1.95
5%	3.39	2.22
Model 2: (Dependent variable: AVAD)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		4.015585
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 31
10%	3.06	1.95
5%	3.39	2.22
Model 3: (Dependent variable: CPER)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		11.30096
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55
Model 4: (Dependent variable: EGPC)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		3.442853
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 31
10%	3.06	1.95
5%	3.39	2.22
Model 5: (Dependent variable: GDPPC)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		406.2466
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 31
10%	3.06	1.95
5%	3.39	2.22
Model 6: (Dependent variable: GNIPC)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		3.490507
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 31
10%	3.06	1.95
5%	3.39	2.22
Model 7: (Dependent variable: LEXP)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		32.20468
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 31
10%	3.06	1.95
5%	3.39	2.22
Model 8: (Dependent variable: SCHENROL)		F-Statistic
F(POLREGCOR, POLREGCONFL, POLREG, GDPG, GFCFG, GE, INT, EXCH)		3.861362
Critical Values	Upper Bound I(1)	Lower Bound I(0)
K = 8; n = 30
10%	3.34	2.26
5%	3.68	2.55

Source: Author’s computation with Eviews 9.

Having found the existence of long-run relationship, I obtained the long-run dynamics of Equations 4 to 11. Table 4A reports the empirical findings of the estimated long-run coefficients for the eight different models of the economic development equations of the effect of POLREGs on economic development Nigeria. All models include the same set of regressors. Table 4B presents the short-run dynamics of these models, and it is important to note that all equations as well pass all the diagnostic tests of Breusch–Godfrey test of serial correlation, functional form test, the autoregressive conditional heteroskedasticity (ARCH) test and normality (JB(N)). All the long-run coefficients of POLREGs are statistically significant for all the models in Table 4A with the exception of model 7 (with LEXP as its regressands). The unified polity scale values used ranges from +10 (strongly democratic to −10 (strongly autocratic)¹² implies that an increase in POLREG tends towards democracy while a decrease tends towards autocracy. The result in Table 4A shows that POLREG has positive effect on ECNDEV, AVAD, GDPPC, GNIPC and SCHENROL, which implies that an increase to democracy increases these economic development indicators, and in turn economic development, while a decrease to autocracy reduces these economic development indicators in the long run. For instance, 1 unit/percentage increase in POLREG will lead to 13.2, 46.1 and 13.9 unit/percentage increases in ECNDEV, AVAD and GDPPC, respectively, and vice versa. On the other hand, POLREG has a negative on CPER and EGPC, indicating that a move towards democracy reduces these indicators and a move towards autocracy increases CPER and EGPC in the long run. That is, a percentage increase in POLREG will reduce CPER and EGPC by 8.1 and 5.9 per cent, respectively, and vice versa. However, in the short run, POLREGs have a negative effect on all economic development indicators except GDP per capita and GNI per capita.¹³ Thus, more democracy decreases economic development in the short run but increases GDP and GNI per capita. The positive effect of POLREGs on GDP and GNI per capita in the short run may be attributed to the method the two indicators are being calculated, which does not necessarily reflect the standard of living of people in a particular country. The GDP and GNI per capita have been criticized by economic scholars as not being a true reflection of the standard of living of people in an economy. If per capita income is the measurement, the population problem may be concealed, as the population has already been divided out. As Kuznets (1995) warns, the choice of per capita, per unit or any similar measure to gauge the rate of economic development carries with it the danger of neglecting the denominator of the ratio.

Also in the result reported in Table 4A, CONFL has statistically significant effects on all the economic development indicators except AVAD and GNIPC in the long run. The study uses the MEPV2015 ACTTOTAL¹⁴ to measure CONFL Nigeria. CONFL has a positive effect on ECNDEV, GDPPC and SCHENROL, while it has negative effects on CPER, EGPC and LEXP. That is, a percentage increase in CONFL will increase ECNDEV, GDPPC and SCHENROL by 21.4, 6.4 and 8.5 per cent, respectively, and reduce CPER, EGPC and LEXP by 3.3, 4.6 and 21.4 per cent, respectively, and vice versa. The positive effect of CONFL on ECNDEV and GDP per capita implies that CONFL increases development in Nigeria, contrary to theoretical believes and empirical findings of Collier and Hoeffler (2004), and Polachek and Sevastianova (2010). However, Chauvet (2003) argues that while violent instability attracts foreign aid, social instability discourages it. Therefore, to the extent that aid stimulates economic growth, CONFL might affect aid allocation decisions and therefore growth rates in aid recipient countries. Also the impact of CONFL on development depends on the typology and coverage of the CONFL (Collier & Hoeffler, 2004). In the short run, CONFL indicates a negative effect on ECNDEV but still maintains a positive effect on GDPPC,¹⁵ which may also be attributed to the deflating effect of CONFL on population, a denominator for deriving GDPPC.

Result in Table 4A also shows a statistically significant effect of COR on all the economic development indicators except LEXP. In Table 4A, the COR data sourced from ICRG¹⁶ show that COR has positive and statistically significant effect on all the economic development indicators except CPER. For instance, a percentage increase in COR (which implies lower potential risk of COR) increases ECNDEV, EGPC, GDPPC and SCHENROL by 188.1, 49.7, 89.6 and 49.8 per cent, respectively, and vice versa. In other words, the higher the number of points of COR indicating a lower potential risk, the higher the economic development in the long run (as in the work of Mauro, 1995; Ugur & Dasgupta, 2011). As such, a lower potential risk of COR in Nigeria generally boosts economic development and other economic development indicators as GDP and GNI per capita, school enrolment, power consumption per capita and AVAD per worker in the long run. However, it displays a mix result in the short run as COR has both negative and positive effects on economic development indicators.¹⁷ A negative effect of COR implies that a higher number of points of COR indicating a lower potential risk reduce economic development by 34.3 per cent in the short run and vice versa. Some works in the literature that tried to find whether there is a positive effect of COR¹⁸ found that aside the negative effect of COR on growth, which is the general belief, there can also be positive effects of COR on growth. Leff (1964), Huntington (1968), Lui (1996), Mo (2001), Teles (2007) have at one point or the other found a positive effect of COR on economic growth.

The results of the effect of other regressors used in the models are presented in Tables 4A and 4B. The GDPG and INT show a statistically significant effect on ECNDEV, while GFCFG and EXCH are not statistically significant in both short- and long-run dynamics. In view of the results in Table 4A, more democracy yield higher economic development in the long run particularly in per capita GNI and GDP as well as AVAD per worker, while autocracy yields lower economic development. However, in the short run, more autocracy boosts up economic development, while democracy reduces economic development in Nigeria. A reduced level of COR fosters economic development in Nigeria in the long run, while its effect in the short run is not clear, as it displays varying effect on economic development indicators. CONFL shows varying effects on economic development indicators due to the fact that some of these indicators are per capita variables, which have population as a component. CONFL affects the two components of these per capita variables, thus making the effect of CONFL on these economic development indicators depends on the magnitude of effect it has on each of the component used in computing it.

Table 4A.

Estimated Long-run Coefficients from the ARDL Models

Variable	Model(1): ECNDEV ARDL (1,2,2,2, 2,2,1,0,0)	Model(2): AVAD ARDL (2,2,2,0, 0,2,1,2,2)	Model(3): (CPER) ARDL (2,0,2, 1,2,2,2,0,2)	Model(4): (EGPC) ARDL (1,0,1,0,2, 2,2,2,0)	Model(5): (GDPPC) ARDL (2,1,2,2, 2,2,2,1,0)	Model(6): (GNIPC) ARDL (2,2,2,2, 2,2,2,1,2)	Model(7): (LEXP) ARDL (2,2,2,0, 2,2,2,1,2)	Model(8): (SCHENROL) ARDL (2,2,2,2, 2,1,2,2,2)
POLREG	0.132***	0.461**	−0.081***	−0.0595***	0.139***	0.175**	−0.0800	0.023*
	(0.039)	(0.017)	(0.002)	(0.000)	(0.004)	(0.068)	(0.166)	(0.081)
CONFL	0.214***	−0.1952	−0.033**	−0.046***	0.064**	0.0622	−0.214**	0.085***
	(0.001)	(0.143)	(0.036)	(0.000)	(0.026)	(0.137)	(0.018)	(0.001)
COR	1.881***	5.056***	−0.413***	0.497***	0.896***	1.045***	−0.5666	0.498***
	(0.001)	(0.000)	(0.006)	(0.000)	(0.000)	(0.003)	(0.259)	(0.001)
GDPG	−0.284***	0.115***	−0.4127	0.0026	0.111***	0.0019	0.0052	0.0053
	(0.002)	(0.005)	(0.235)	(0.347)	(0.000)	(0.760)	(0.351)	(0.163)
GFCFG	−0.0002	−0.054***	0.005***	0.004***	−0.003**	−0.0044	0.015**	−0.006***
	(0.964)	(0.002)	(0.003)	(0.002)	(0.014)	(0.139)	(0.016)	(0.002)
GE	0.0178	0.0182	−0.0059	−0.012***	0.014**	0.031*	−0.058**	0.016***
	(0.107)	(0.614)	(0.467)	(0.003)	(0.010)	(0.053)	(0.026)	(0.003)
INT	0.013***	0.0188	0.005**	−0.0012	0.0013	0.006**	0.028***	0.007***
	(0.002)	(0.195)	(0.033)	(0.411)	(0.265)	(0.027)	(0.002)	(0.000)
EXCH	0.0006	−0.0035	0.004**	0.017***	−0.0077	−0.0104	0.039**	0.003***
	(0.855)	(0.709)	(0.019)	(0.000)	− (0.138)	(0.265)	(0.010)	(0.004)

Source: Author’s computation with Eviews 9.

Notes: ***, **, * indicate statistically significant at 1%, 5% and 10% levels, respectively. The test for serial correlation is the LM test for autocorrelation, the test for functional form is Ramsey’s RESET test, the test for normality is the test proposed by Bera and Jarque (1981), the test for heteroskedasticity is the LM test. Lag length is based on SBC.

Table 4B.

Estimated Short-run Coefficients from the ARDL Models

Variable	Model(1): ECNDEV ARDL (1,2,2, 2,2,2,1,0,0)	Model(2): AVAD ARDL (2,2,2,0, 0,2,1,2,2)	Model(3): (CPER) ARDL (2,0,2,1, 2,2,2,0,2)	Model(4): (EGPC) ARDL (1,0,1,0, 2,2,2,2,0)	Model(5): (GDPPC) ARDL (2,1,2, 2,2,2,2,1,0)	Model(6): (GNIPC) ARDL (2,2,2, 2,2,2,2,1,2)	Model(7): (LEXP) ARDL (2,2,2,0, 2,2,2,1,2)	Model(7): (SCHENROL) ARDL (2,2,2, 2,2,1,2,2,2)
d(POLREG)	−0.019***	−0.0083	−0.102***	−0.061***	0.004***	0.043*	0.002**	−0.024**
	(0.004)	(0.697)	(0.001)	(0.000)	(0.005)	(0.069)	(0.028)	(0.048)
d(COR)	−0.343***	−0.1935	−0.740**	−0.586***	0.055***	0.799*	0.033**	−0.0262
	(0.001)	(0.400)	(0.013)	(0.001)	(0.006)	(0.065)	(0.049)	(0.864)
d(CONFL)	0.0026	0.0104	0.0189	−0.047***	−0.00003	−0.043*	−0.004**	0.050***
	(0.516)	(0.558)	(0.369)	(0.001)	(0.969)	(0.052)	(0.027)	(0.008)
d(GDPG)	0.031***	−0.008**	0.000001	0.0008	0.009***	0.010**	0.0006***	0.004***
	(0.000)	(0.042)	(0.999)	(0.618)	(0.000)	(0.012)	(0.006)	(0.004)
d(GCFCG)	0.0005	0.004***	0.002**	0.002***	−0.00004	−0.002*	0.0003***	−0.001**
	(0.105)	(0.007)	(0.021)	(0.008)	(0.462)	(0.063)	(0.002)	(0.011)
d(INT)	−0.001***	−0.0011	−0.0013	−0.0007	−0.00005	0.0015	0.0003*	−0.00005
	(0.001)	(0.295)	(0.395)	(0.340)	(0.348)	(0.421)	(0.009)	(0.949)
d(EXCH)	−0.00008	0.0006	0.005**	0.008***	−0.0004**	−0.002	0.0002**	−0.004***
	(0.856)	(0.718)	(0.036)	(0.000)	(0.033)	(0.315)	(0.027)	(0.002)
d(GE)	−0.0022	−0.0029	−0.017	−0.012***	0.001***	0.0005	−0.002***	0.008**
	(0.156)	(0.628)	(0.017)	(0.003)	(0.005)	(0.882)	(0.002)	(0.048)
CointEq(-1)	−0.125***	−0.159***	−1.252***	−1.029***	−0.097***	−0.824**	−0.030***	−1.059***
	(0.004)	(0.003)	(0.000)	(0.000)	(0.005)	(0.047)	(0.003)	(0.001)
Adj R-squared	0.9958	0.3693	0.6817	0.519	0.9955	0.9264	0.9999	0.9592
DW-statistics	2.6848	2.7353	1.9118	2.2375	2.5469	2.9687	2.1347	2.4264
LM (χ²) Version
erial Correlation	χ²(2) = 24.929 [.000]	χ²(2) = 8.385 [.015]	χ²(2) = 26.358 [.000]	χ²(2) = 14.903 [.001]	χ²(2) = 16.025 [.000]	χ²(1) = 29.959 [.000]	χ²(1) = 24.877 [.000]	χ²(2) = 29.338 [.000]
Functional Form	χ²(7) = 0.471 [.652]	χ²(8) = 1.042 [.328]	χ²(6) = 1.003 [.356]	χ²(9) = 0.707 [.497]	χ²(5) = 0.681 [.526]	χ²(2) = 0.123 [.913]	χ²(1) = 1.467 [.381]	χ²(3) = 0.715 [.526]
Normality	χ²(1) = 0.151 [.927]	χ²(1) = 0.402 [.818]	χ²(1) = 0.404 [.817]	χ²(1) = 0.252 [.888]	χ²(1) = 0.984 [.611]	χ²(1) = 8.411 [.015]	χ²(1) = 0.426 [.808]	χ²(1) = 0.059 [.971]
Heteroskedasticity	χ²(21) = 19.034 [.583]	χ²(21) = 20.532 [.488]	χ²(22) = 23.473 [.376]	χ²(19) = 22.765 [.248]	χ²(23) = 21.260 [.565]	χ²(26) = 25.873 [.470]	χ²(27) = 27.790 [.421]	χ²(25) = 29.461[.245]
F-Statistics
erial Correlation	F(2,6) = 14.747 [.005]	F(2,7) = 1.358 [.318]	F(2,5) = 18.095[.005]	F(2,8) = 3.949 [.064]	F(2,4) = 2.293 [.217]	F(1,5) = 365.674[.037]	F(1,1) = 4.856 [.271]	F(2,2) = 44.323 [.022]
Functional Form	F(1,7) = 0.222 [.652]	F(1,2) = 1.086 [.328]	F(1,6) = 1.006 [.356]	F(1,9) = 0.500 [.497]	F(1,5) = 0.464 [.526]	F(1,2) = 0.015 [.913]	F(1,1) = 2.152 [.381]	F(1,3) = 0.511 [.526]
Normality	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable
Heteroskedasticity	F(21,8) = 0.661[.788]	F(21,8) = 0.826[.659]	F(22,7) = 1.144[.458]	F(19,10) = 1.656[.208]	F(23,6) = 0.565[.801]	F(26,3) = 0.723[.729]	F(27,2) = 0.932[.644]	F(25,4) = 8.742[.024]

Source: Author’s computation with Eviews 9.

Notes: *, **, *** indicate statistically significant at 1%, 5% and 10% levels, respectively. The test for serial correlation is the LM test for autocorrelation, the test for functional form is Ramsey’s RESET test, the test for normality is the test proposed by Bera and Jarque (1981), the test for heteroskedasticity is the LM test. Lag length is based on SBC.

Interactive Effects of Political Regimes with Conflict and Corruption on Economic Development

Scholars do believe that if democratic tenets are entrenched, there is more transparency and accountability which in turn reduces the COR. Figure 2 indicates that there was higher risk of COR during democracy relative to the dictatorship in Nigeria. Since the Nigeria assumed democracy in 1999, despite the various reforms of government and the establishment of anti-COR agencies, the country has experienced high profile COR cases by public office holders.

Over and above COR as the main bane of democracy to achieving development in most developing countries, CONFL has also been identified as a major challenge to democratization in Africa. CONFL is undoubtedly a major cause of underdevelopment, especially in developing countries. In this subsection, I estimated the interactive effects of POLREGs and COR as well as the interactive effect of POLREGs and CONFL, on ECNDEV and other economic development indicators. Also, other variables were added as regressors to improve the robustness of the result of the estimation. I conducted the same pre-estimation tests and follow the same procedure (as in Effects of Political Regimes on Economic Development in Nigeria section). The results in Table 3B show that there exist long-run relationships among the variables used in each models (i.e., models 1–8). Next, I obtained the long-run dynamics of Equation 10 to 17. In the results as presented in Table 5A, the interactive variable of POLREGs and COR (POLREG_t * COR_t) shows a statistically significant positive effect on all the economic development indicators, including the ECNDEV (Table 5A). However, the effect is not statistically significant on household CPER.

In other words, an increase in the value of the variable interacting POLREGs and COR (POLREG*COR) denotes a move towards more democracy and lower level of COR, while a decrease denotes a move towards autocracy and higher risk level of COR. The result of the long-run dynamics shows that an increase in POLREG*COR increases ECNDEV and AVAD by 1.365 units and 13.9 per cent in the long run and are statistically significant at 5 per cent and 1 per cent levels, respectively. With this result however, one can say that the long-run democracy increases economic development in Nigeria essentially when it is devoid of COR. Though on the contrary, an autocratic regime with or without COR drains development in Nigeria, particularly in the long run. There is positive effect of POLREG*COR on the indicators of economic development and ECNDEV in the short-run dynamics except on CPER (with a negative effect), which thus was statistically insignificant in the long-run dynamics (Table 5B in the Appendix). That is to say, even in the short run an increase in the value of the variable interacting POLREGs and COR (POLREG*COR) tends to an increase in economic development and other economic development indicators.

The result in Table 5A is quite interesting as the interactive variable of POLREGs and CONFL (POLREG*CONFL) has a statistically significant effects only on electric power consumption per capita (EGPC), LEXP and SCHENROL. From this result, one can infer that POLREG*CONFL has no statistically significant income output and consumption-related variable, which make bulk of the indicators that formed the ECNDEV. The result in Table 5A shows that POLREG*CONFL has a negative effect on EGPC and LEXP implying that an increase in POLREG*CONFL reduces EGPC and LEXP by 0.009 per cent and 0.119 per cent, respectively. In other words, a move towards full democracy with a higher incidence of CONFL tends to reduce electric power consumption per capita and LEXP. This corroborates the findings of Plümper and Neumayer (2006), ACAPS (2012) who argued that armed CONFLs have important indirect negative consequences on agriculture, infrastructure, public health provision and social order. Recall also that in Table 4A, POLREG has a statistically insignificant negative effect on LEXP, while CONFL has a statistically significant negative effect on it. This suggests that POLREG has no effect on LEXP and the negative effect of the interaction of the two variables (POLREG*CONFL) is basically caused by the effect of CONFL on LEXP.

Also some studies have shown a significant positive effect of democracy on per capita household electricity consumption (Acemoglu & Robinson, 2006b; Ahlborg, Boräng, Jagers & Söderholm, 2015; Collier, 1999; Schmitter & Karl, 1991). On the other hand, electricity producing infrastructures are considered to be of generally recognized military importance and are targeted for destruction on grounds of ‘military necessity’ (Gellman, 1991).

Result in Table 5A also shows that the variable interacting POLREGs and CONFL (POLREG*CONFL) has a significant positive effect on SCHENROL in the long run. This corroborates the results in Table 4A, where POLREG and CONFL have statistically significant positive effects on SCHENROL, suggesting that a move to more democracy increases SCHENROL, likewise more incidence CONFL. The result in Table 5A suggests that a move towards more democracy as well as higher incidences of CONFL tends to increase SCHENROL in Nigeria in the long run. Contrary to the result in Table 5A, several studies found a negative effect of armed CONFL on education (UNESCO, 2011; Bennett, 2009; Stewart, Huang, & Wang, 2001). The direct impact of armed CONFL on education ranges from the fact that educational facilities are damaged and destroyed, schools occupied by armed forces, school buildings become shelter for IDPs, lack of qualified personnel caused by displacement of teachers and other staff, curriculum changed to support the violent societal dynamics, recruitment of children into armed groups and other indirect impacts (UNESCO, 2011).

Table 5A.

Estimated Long-run Coefficients from the ARDL Models

Variable	Model(1): ECNDEV ARDL (1,1,0, 0,1,1,1,1,0)	Model(2): AVAD ARDL (1,0,1, 0,0,0,0,1,0)	Model(3): (CPER) ARDL (2,2,2, 2,1,2,2,2,2)	Model(4): (EGPC) ARDL (1,0,0, 0,0,1,0,0,0)	Model(5): (GDPPC) ARDL (1,1,0, 1,1,1,1,1,1)	Model(6): (GNIPC) ARDL (1,1,0, 1,0,0,0,0,0,)	Model(7): (LEXP) ARDL (1,0,0, 1,0,1,1,0,0)	Model(8): (SCHENROL) ARDL (1,1,2,2, 2,2,1,2,2)
POLREG*COR	1.365**	0.139***	0.0462	0.087***	0.237***	0.218***	1.368***	7.368***
	(0.024)	(0.004)	(0.303)	(0.005)	(0.000)	(0.000)	(0.000)	(0.004)
POLREG*CONFL	−0.0474	0.0119	−0.0057	−0.009*	0.009	−0.0001	−0.119**	0.654**
	(0.200)	(0.141)	(0.351)	(0.055)	(0.215)	(0.986)	(0.019)	(0.027)
POLREG	−2.522**	−0.278***	−0.343*	−0.209***	−0.406***	−0.425***	−2.366***	−15.406***
	(0.015)	(0.001)	(0.051)	(0.000)	(0.000)	(0.000)	(0.000)	(0.001)
GDPG	1.2189	0.0034	−0.040*	0.004	0.196**	0.025**	0.061*	−0.448**
	(0.117)	(0.432)	(0.058)	(0.214)	(0.023)	(0.042)	(0.077)	(0.036)
GFCFG	0.0064	0.0026	0.015**	0.0002	−0.0006	−0.0010	0.0138	−0.168*
	(0.584)	(0.115)	(0.029)	(0.892)	(0.679)	(0.766)	(0.299)	(0.077)
GE	0.1222	0.0073	0.059**	−0.0061	−0.0123	−0.0028*	0.251**	−0.656**
	(0.194)	(0.474)	(0.014)	(0.373)	(0.171)	(0.766)	(0.027)	(0.017)
INT	0.0188	0.0037	0.0039	−0.0013	−0.0029	0.0006	−0.0022	0.322***
	(0.203)	(0.195)	(0.114)	(0.281)	(0.293)	(0.732)	(0.843)	(0.005)
EXCH	0.045*	0.009***	0.042**	0.009***	0.002	0.007***	0.089***	0.349***
	(0.011)	(0.000)	(0.032)	(0.000)	− (0.466)	(0.009)	(0.000)	(0.006)

Source: Author’s computation with Eviews 9.

Though the article argued that the time series data in the study do not show a strong negative correlation between CONFL and the provision of education, it is still possible that there is an unobserved relationship.¹⁹ Thus, it will be instructive to note however that SCHENROL data from WDI used in this study is incomplete data, which was interpolated with the aid of EVIEWS. The incompleteness of the data used presents a reason why the effects of interactive variable between POLREGs and CONFL (POLREG*CONFL) on school enrolment may be contrary to a priori expectations.

In the short run, POLREG*CONFL maintains a significant negative effect on LEXP and electric power consumption, while same significant positive effect is maintained on SCHENROL (Table 5B). Also in the short run, POLREG*CONFL has a significant positive effect on per capita consumption and overall effect on each per capita variable (GDP or GNI) depends on whether the denominator, which in this case population, is most deflated by the effect of CONFL. For instance, if population is most deflated by the effect of CONFL rather than GDP or consumption, the effect of CONFL and/or the interaction of POLREGs and CONFL will increase per capita GDP and consumption. This may be the case in Nigeria as the bulk of the GDP is derived from high-tech production (i.e., oil production, ICT and other services) with little need of labour force, and also the bulk of her consumption is from importation. Other regressors used (except EXCH and GDPG) in the eight models in Table 5A as reported did not show impressive statistically significant effect on the ECNDEV and other economic development indicators.

Table 5B.

Estimated Short-run Coefficients from the ARDL Models

Variable	Model(1): ECNDEV ARDL (1,1,0,0, 1,1,1,1,0)	Model(2): AVAD ARDL (1,0,1,0, 0,0,0,1,0)	Model(3): (CPER) ARDL (2,2,2,2, 1,2,2,2,2)	Model(4): (EGPC) ARDL (1,0,0,0, 0,1,0,0,0)	Model(5): (GDPPC) ARDL (1,1,0,1, 1,1,1,1,1)	Model(6): (GNIPC) ARDL (1,1,0,1, 0,0,0,0,0)	Model(7): (LEXP) ARDL (1,0,0,1, 0,1,1,0,0)	Model(7): (SCHENROL) ARDL (1,1,2,2, 2,2,1,2,2)
d(POLREG*COR)	0.026***	0.064*	−0.419**	0.072**	0.0028	0.043**	0.105***	6.413*
	(0.004)	(0.072)	(0.029)	(0.029)	(0.269)	(0.042)	(0.011)	(0.084)
d(POLREG*CONFL)	−0.0011	−0.0012	0.029**	−0.008**	0.0004*	−0.00004	−0.009***	0.508**
	(0.189)	(0.703)	(0.038)	(0.048)	(0.089)	(0.986)	(0.006)	(0.047)
d(POLREG)	−0.056***	−0.128*	0.846**	−0.175***	−0.0045	−0.089**	−0.211***	−14.657*
	(0.002)	(0.056)	(0.029)	(0.008)	(0.396)	(0.031)	(0.001)	(0.062)
d(GDPG)	0.028***	0.0016	0.022**	0.0033	0.009***	0.008***	0.005**	−0.0549
	(0.000)	(0.448)	(0.013)	(0.202)	(0.000)	(0.000)	(0.026)	(0.644)
d(GCFCG)	−0.0001	0.001*	−0.0023	−0.0011	0.0001	−0.0003	0.0001	−0.109*
	(0.474)	(0.069)	(0.177)	(0.178)	(0.178)	(0.488)	(0.852)	(0.081)
d(GE)	−0.0004	0.0034	−0.065**	−0.0051	0.00036	−0.0009	0.0079	−0.3078
	(0.771)	(0.462)	(0.023)	(0.388)	(0.293)	(0.765)	(0.145)	(0.375)
d(INT)	−0.0001	0.0002	−0.0648	−0.0011	−0.0001	0.0002	−0.0002	0.0716
	(0.663)	(0.798)	(0.152)	(0.275)	(0.356)	(0.732)	(0.840)	(0.283)
d(EXCH)	0.001**	‘0.004**	−0.044**	0.007***	−0.00015	0.0024*	0.007***	0.0799
	(0.049)	(0.025)	(0.018)	(0.000)	(0.354)	(0.064)	(0.002)	(0.556)
CointEq(-1)	−0.022**	−0.459***	−1.226**	−0.836***	−0.046**	−0.330**	−0.077***	−1.343**
	(0.011)	(0.003)	(0.035)	(0.000)	(0.022)	(0.015)	(0.002)	(0.012)
Adj R-squared	0.9917	0.9909	0.9987	0.8981	0.9998	0.9847	0.9994	0.9444
DW-statistics	2.1026	2.1368	2.3173	2.2878	2.8251	2.2675	1.3307	2.1606
LM (χ²) Version
erial Correlation	χ²(2) = 14.196[.001]	χ²(2) = 8.792 [.012]	χ²(2) = 29.129[.000]	χ²(2) = 4.024 [.134]	χ²(2) = 6.050 [.049]	χ²(2) = 3.687 [.158]	χ²(2) = 7.208 [.027]	χ²(2) = 24.152 [.000]
Functional Form	χ²(15) = 0.773[.451]	χ²(18) = 0.233[.818]	χ²(2) = 0.294 [.797]	χ²(19) = 0.326[.748]	χ²(13) =1.656 [.122]	χ²(18) = 0.459[.652]	χ²(17) = 7.224[.000]	χ²(4) = 3.316 [.029]
Normality	χ²(1) = 1.113 [.573]	χ²(1) = 032..589[.000]	χ²(1) = 0.387 [.824]	χ²(1) = 3.261 [.195]	χ²(1) = 0.351 [.839]	χ²(1) = 0.899 [.638]	χ²(1) = 3.522 [.172]	χ²(1) = 0.139 [.932]
Heteroskedasticity	χ²(14) = 10.622[.715]	χ²(11) = 10.714[.468]	χ²(26) = 27.401[.388]	χ²(10) = 13.262[.209]	χ²(16) = 22.952[.115]	χ²(11) = 12.661[.316]	χ²(12) = 13.390[.341]	χ²(24) = 26.231 [.342]
F-Statistics
erial Correlation	F(2,14) = 5.914[.013]	F(2,17) = 3.365[.059]	F(2,5) = 16.735[.170]	F(2,18) = 1.342[.286]	F(2,4) = 1.455 [.272]	F(2,17) = 1.148[.037]	F(2,16) = 2.423[.120]	F(2,3) =6.194 [.086]
Functional Form	F(1,15) = 0.598[.451]	F(1,18) = 0.055[.818]	F(1,2) =0.086 [.797]	F(1,19) =0.106 [.748]	F(1,13) = 2.742[.122]	F(1,18) =0.211 [.652]	F(1,17) = 52.182[.000]	F(1,4) = 10.992 [.029]
Normality	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable	Not applicable
Heteroskedasticity	F(14,16) = 0.596[.832]	F(11,17) = 0.9123[.548]	F(26,3) =1.216 [.506]	F(10,20) = 1.495[.213]	F(16,14) = 2.496[.046]	F(11,19) = 1.193[.354]	F(12,8) =1.141 [.389]	F(24,5) =1.449 [.364]

Source: Author’s computation with Eviews 9.

Notes: *, **, *** indicate statistically significance at 1%, 5% and 10% levels, respectively. The test for serial correlation is the LM test for autocorrelation, the test for functional form is Ramsey’s RESET test, the test for normality is the test proposed by Bera and Jarque (1981), the test for heteroskedasticity is the LM test. Lag length is based on SBC.

In view of the result of these estimations, democracy will only improve economic development in the long run if there is a reduced level of COR in Nigeria. Also, democracy will boost economic development in the long run if there are little or no incidences of CONFL in Nigeria. Likewise in the short run, democracy tends to boost economic development in Nigeria if there is a reduced level of COR. The effect of POLREGs with CONFL has not been clearly distinguished, especially on the per capita variables used. This is due to the fact that the two dicators used in computing the per capita variable can both be affected by CONFL; thus, the actual effect depends on the magnitude to which CONFL affects each of the indicators. With the presence of COR and CONFL, POLREG has a negative effect on ECNDEV and other economic development indicators used short run. This in other words infers that more democracy in Nigeria with the presence of CONFL and COR reduces economic development in the short run.

Conclusion

This study explores the relationship between POLREGs and economic development in Nigeria. The situation in Nigeria seems to disagree with some of these scholars’ assertion as the level of COR got aggravated since the inception of democracy in 1999. The country experienced high profile COR cases among public office holders it assumed democracy.

This article investigates the short- and long-run dynamics of the effect of POLREGs and economic development. It also examines the interactive effects of COR and POLREGs as well as CONFL and POLREGs on economic development. With the aid of graphical representations, the study found that there has been fluctuations in the political system in Nigeria and concludes that political system in Nigeria has experienced instability during the study period. The study also found, as against the a priori knowledge, that the level/volume of COR increased considerably after the transition to a democratic system. Thus, democratic inclinations tend to induce COR level in Nigeria. The article found increased level of CONFL during democratic periods as against dictatorship. The incidence of CONFL seems higher since the advent of a democratic dispensation in 1999, as against the a priori expectations. However, many authors (Adetoye & Omilusi, 2015; Elaigwu, 2005) gave reasons why the incidence of CONFL got higher in a democracy.

There exists long-run relationships among the variables used in all the models estimated in this study. In the long run, more democracy yields higher economic development in Nigeria particularly in per capita GNI and GDP, as well as AVAD per worker when the issue of CONFL and COR are being addressed, while autocracy yields lower economic development. In the short run however, more autocracy fosters economic development in Nigeria while democracy hinders it. COR portends grave threat to the development of Nigeria’s economy as higher risk of COR reduces economic development indicators in the long run. Nevertheless, a higher risk of COR leads to economic development in the short run in Nigeria.

The effect of CONFL on economic development in Nigeria is unclear; thus, t is determined by the magnitude of its impact on the components of the per capita variables used. Yet it reduces per capita consumption, LEXP and electric power consumption per capita, while it increases ECNDEV and per capita GDP. A move towards more democratization in Nigeria fosters economic development in the long run and short run if there is a reduced level of COR and vice versa. The effect of more democracy with the presence of CONFL on economic development seems ambiguous, especially on the per capita variables. Thus, democracy with the presence of CONFL reduces electric power consumption per capita and LEXP, while it increases SCHENROL in Nigeria in the long run. Conclusively, with the rise in COR and CONFL levels in Nigeria democratic experience, democracy tends to hinder development in both short run and long run. For the purpose of policymaking, the findings in this study highlight the need to establish effective anti-graft agencies in order to fight COR to the barest minimum in Nigeria. They highlight the need to the entrenched cardinal tenets of democracy (i.e., rule of law, transparency and accountability, equity and equal representations, etc.) to reduce the incidence of CONFLs and to allow Nigeria partake in the dividends of democracy. They also highlight the need to employ CONFL resolution mechanisms as well as enhance various institutions and research think-tanks (i.e., Nigeria’s institute for peace and CONFL resolution) in resolving CONFL issues in the democratization process of the country. There are other factors that may hinder or foster democracy and dictatorship in achieving economic development which this article will suggest for further investigation.

Footnotes

Notes

References

ACAPS (2012). Disaster summary sheet armed conflict (ACAPS Report, pp. 1–17). ACAPS. Retrieved 2 July 2017, from https://www.acaps.org/sites/acaps/files/resources/files/disaster_summary_sheet-armed_conflict_november_2012.pdf

Acemoglu

& Robinson

J. A.

(2006a). Economic backwardness in political perspective. American Political Science Review, 100(1), 115–131.

Acemoglu

& Robinson

J. A.

(2006b). Economic origins of dictatorship and democracy. New York, NY: Cambridge University Press.

Adebanwi

(2004). The City, Hegemony and Ethno-Spatial Politics: The Press and The Struggle for Lagos in Colonial Nigeria. Nationalism and Ethnic Politics, 9(4), 25–51.

Aderibigbe

A. B.

(1965). Peoples of Southern Nigeria. In Ajayi

A J. F.

& Espie

(Eds.), A thousand years of West African history (pp. 18–20). Ibadan: Ibadan University Press.

Adetoye

& Omilusi

M. O.

(2015). Ethno-religious conflicts and democracy in Nigeria. Global Journal of Arts Humanities and Social Sciences, 3(1), 51–58.

Ahlborg

Boräng

Jagers

S. C.

& Söderholm

(2015). Provision of electricity to African households: The importance of democracy and institutional quality. Energy Policy, 87(1), 125–135.

Akinbobola

(2000). Trends of democratization in Africa: An analysis of the challenges of political institutionalization. The Constitution: A Journal of Constitutional Development, 3(2), 2–17.

Arat

Z. F.

(1988). Democracy and economic development: Modernization theory revisited. Comparative Politics, 21(1), 21–36.

10.

Baum

M. A.

& Lake

D. A.

(2003). The political economy of growth: Democracy and human capital. American Journal of Political Science, 47(2), 333–347.

11.

Bennett

(2009). Ten years on, the Machel Study cites continued abuse against children in conflict, At a glance: United States of America. New York: UNICEF. Retrieved 10 September 2017, from https://www.unicef.org/infobycountry/usa_50027.html

12.

Bera

A. K.

& Jarque

C. M.

(1981). Efficient tests for normality, homoscedasticity and serial independence of regression residuals: Monte Carlo Evidence. Economics Letters, 7(4), 313–318.

13.

Bueno de Mesquita

Smith

Siverson

R. M.

& Morrow

J. D.

(2003). The logic of political survival. Cambridge, MA: MIT Press.

14.

Burleigh

(2013, August). A country so corrupt it would be better to burn our arid money. Daily Mail Online. Retrieved 19 November 2015, from www.dailymail.co.uk/debate/article-2387359/nigeria-corrupt-better-burn-air)-money.html

15.

Cervellati

& Sunde

(2012). Civil conflict, democratization, and growth: Violent democratization as critical juncture. The Scandinavian Journal of Economics, 116(2), 482–505.

16.

Chauvet

(2003). Socio-political instability and the allocation of international aid by donors. European Journal of Political Economy, 19(1), 33–59.

17.

Chirino

J. B.

Valdivielso

M. W.

& Melian

J. M. B.

(2006). Analysis of the effectiveness of official development assistance. Atlantic Review of Economics, 5(12), 1–21.

18.

Collier

(1999). On the economic consequences of civil war. Oxford Economic Papers, 51(1), 168–183.

19.

Collier

& Hoeffler

(2004). Greed and grievance in civil war. Oxford Economic Papers, 56(4), 563–595.

20.

de Schweinitz

Jr . (1959). Industrialization, labor controls and democracy. Economic Development and Cultural Change, 7(4), 385–404.

21.

Doucouliagos

& Ulubasoglu

M. A.

(2008). Democracy and economic growth: A meta-analysis. American Journal of Political Science, 52(1), 61–83.

22.

Egbefo

O. D.

(2015). Fifteen years of democracy, 1999–2014: Reflections on Nigeria’s quest for national integration. An International Multidisciplinary Journal, Ethiopia, 9(2), 59–77.

23.

Elaigwu

J. J.

(2005). Crises and conflicts management in Nigeria since 1980. In Yakubu

A. M.

Adegboye

R. T.

Uba

C. N.

& Dogo

(Eds.), Crises and conflict management in Nigeria since 1980 (pp. 16–17). Kaduna: Nigeria Defence Academy.

24.

Fjelde

& Hegre

(2007, February 28–March 3). Democracy depraved. Corruption and institutional change 1985–2000. Paper Presented to the 48th Annual Convention of the International Studies Association, Chicago.

25.

Fosu

A. K.

(2001). Political instability and economic growth in developing economies: Some specification empirics. Economics Letters, 70(2), 289–294.

26.

Freedom House (2015). Nigeria’s presidential election hopeful sign for Nigerian democracy. Retrieved 23 September 2016, from https://freedomhouse.org/article/nigerias-presidential-election-hopeful-sign-nigerian-democracy

27.

Friedman

(1962). Capitalism and freedom (pp. 1–166). Chicago: The University of Chicago Press Ltd.

28.

Gandhi

(2008). Dictatorial Institutions and their Impact on Economic Growth. European Journal of Sociology, 49(1), 3–30.

29.

Galenson

(Ed.). (1959). Introduction. Labor and economic development (pp. 1–18). New York, NY: Wiley.

30.

Gellman

(1991). Allied air war struck broadly in Iraq; officials acknowledge strategy went beyond purely military targets. The Washington Post. New York, NY: Global Policy Forum. Retrieved 12 October 2017, from https://www.globalpolicy.org/component/content/article/169/36375.html

31.

Gerring

Bond

Barndt

W. T.

& Moreno

(2005). Democracy and economic growth: A historical perspective. World Politics, 57(3), 323–364.

32.

Goldsmith

A. A.

(1995). Democracy, property rights and economic growth. The Journal of Development Studies, 32(2), 157–174.

33.

Gregson

(2017). The poorest countries in the world Global Finance. Retrieved 17 October 2017, from https://www.gfmag.com/global-data/economic-data/the-poorest-countriesin-the-world?page=12

34.

Helliwell

J. F.

(1994). Empirical linkages between democracy and economic growth. British Journal of Political Science, 24(2), 225–248.

35.

Huntington

S. P.

(1968). Political order in changing societies. New Haven, CT: Yale University Press.

36.

Huntington

S. P.

& Dominguez

J. I.

(1975). Political development. In Greenstein

F. I.

& Polsby

N. W.

(Eds.), Macro political theory (pp. 1–114). Reading, MA: Addison-Wesley.

37.

Hunwick

J. O.

(1965). The nineteenth century Jihads. In Ade Ajayi

J. F

& Espie

(Eds.), A thousand years of West African history (pp. 267–288). Ibadan: Ibadan University Press and Nelson.

38.

IMF (2017). A shifting global economic landscape (pp. 1–7). World Economic Outlook Update, an Update of the Key WEO Projections. Washington, DC: IMF.

39.

Knutsen

C. H.

(2010). Investigating the lee-thesis: How bad is democracy for Asian economies. European Political Science Review, 2(3), 451–473.

40.

Knutsen

C. H.

(2011a). Democracy, dictatorship and protection of property right. Journal of Development Studies, 47(1), 164–182.

41.

Knutsen

C. H.

(2011b). The economic effects of democracy and dictatorship (PhD thesis). Norway: Department of Political Science, University of Oslo.

42.

Kolstad

& Wiig

(2011). Does democracy reduce corruption? (CMI Working Paper). London: The Institute and Faculty of Actuaries (IFoA).

43.

Kuznets

(1995). Economic growth and income inequality. American Economic Review, 45(1), 1–28.

44.

Lake

D. A.

& Baum

M. A.

(2001). The invisible hand of democracy: Political control and the provision of public services. Comparative Political Studies, 34(6), 587–621.

45.

Leblang

D. A.

(1997). Political democracy and economic growth: Pooled cross-sectional and times-series evidence. British Journal of Political Science, 27(3), 453–466.

46.

Leff

(1964). Economic development through bureaucratic corruption. The American Behavioral Scientist, 8(3), 8–14.

47.

Lui

F. T.

(1996). Three Aspects of Corruption. Contemporary Economic Policy, 14(3), 26–29.

48.

Marshall

M. G.

(2016, May 25). Codebook: Major Episodes of Political Violence (MEPV) and Conflict Regions, 1946–2015. Centre for Systemic Peace, pp. 1–17. Retrieved 1 April 2017, from http://www.systemicpeace.org/inscr/MEPVcodebook2015.pdf

49.

Marshall

M. G.

& Jaggers

(2007). Polity IV project: Political regime characteristics and transitions, 1800–2006 (Dataset Users’ Manual, pp. 1–83). Fairfax, VA: George Mason University, Center for Systemic Peace. Retrieved 27 March 2017, from https://home.bi.no/a0110709/PolityIV_manual.pdf

50.

Mauro

(1995). Corruption and growth. The Quarterly Journal of Economics, 110(3), 681–712.

51.

Pak-Hung

(2001). Corruption and Economic Growth. Journal of Comparative Economics, 29(1), 66–79.

52.

National Bureau of Statistics (NBS) (2012). Nigeria Poverty Profile 2010. Abuja: NBS.

53.

Niven

(1970). The War of Nigerian Unity 1967–70 (pp. 7–175). Ibadan: Evan Brothers Ltd.

54.

North

D. C.

(1990). Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press.

55.

North

D. C.

Wallis

J. J.

& Weingast

B. R.

(2009). Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge: Cambridge University Press.

56.

Obaro

(1977). The fall of Nigeria. Ibadan: Heineman Educational Books Ltd.

57.

Obasi

& Taiwo-Obalonye

(2017, March 1). Nigeria’s GDP growth rate contracts by −1.3% in Q4, 2016—NBS, The Sun Newspaper. Retrieved 2 November 2017, from http://sunnewsonline.com/nigerias-gdp-growth-rate-contracts-by-1-3-in-q4-2016-nbs/

58.

Olson

(1982). The rise and decline of nations. Economic growth, stagflation and social rigidities. New Haven, CT: Yale University Press.

59.

Olson

(1993). Dictatorship, democracy, and development. American Political Science Review, 87(3), 567–576.

60.

Omoiya

S. Y.

(2012). An historical appraisal of Nigerian democratic experience. Research on Humanities and Social Sciences, 2(9), 9–15.

61.

Osita

O. O.

(2007). Conflict management in Nigeria issues and challenges. London: Adonis and Abbey Publishers Ltd.

62.

Paden

J. N.

(1986). Ahmadu Bello, Sardauna of Sokoto: Values and Leadership in Nigeria (pp. 1–799). Zaria, Nigeria: Hudahuda.

63.

Pesaran

M. H.

Shin

& Smith

R. J.

(2001). Bounds testing approaches to the analysis of level relationships. Journal of Applied Economics, 16(3), 289–326.

64.

Plümper

& Neumayer

(2006). The unequal burden of war: The effect of armed conflict on the gender gap in life expectancy. International Organization, 60(3), 723–754.

65.

Post

K. W. J.

& Vicker

(1973). Structure and conflict in Nigeria 1963–65. London: Heinemann.

66.

Polachek

S. W.

& Sevastianova

(2010). Does conflict disrupt growth? Evidence of the relationship between political instability and national economic performance (IZA Discussion Paper No. 4762). Bonn: Institute for Labour Economics.

67.

Przeworski

& Limongi

(1993). Political regimes and economic growth. Journal of Economic Perspectives, 7(3), 51–69.

68.

Przeworski

Alvarez

M. E.

Cheibub

J. A.

& Limongi

(2000). Democracy and development. Political institutions and well-being in the world, 1950–1990. Cambridge: Cambridge University Press.

69.

Rachdi

& Saidi

(2015). Democracy and economic growth: Evidence in MENA countries. Procedia—Social and Behavioral Sciences, 191(2), 616–621.

70.

Remmer

(1990). Democracy and economic crisis: The Latin American experience. World Politics, 42(3), 315–335.

71.

Rock

M. T.

(2003). The politics of development policy-making in new order Indonesia (Working Paper No. 632). Ann Arbor, MI: William Davidson Institute.

72.

Rock

(2009). Corruption and democracy. The Journal of Development Studies, 45(1), 55–75.

73.

Schmitter

P. C.

& Karl

T. L.

(1991). What democracy is…and is not. Journal of Democracy, 2(3), 75–88.

74.

Sen

(1999). Development as freedom. Oxford: Oxford University Press.

75.

Stewart

Huang

& Wang

(2001). Internal wars in developing countries: An empirical overview of economic and social consequences. Oxford: Oxford University Press.

76.

Teles

V. K.

(2007). Institutional quality and endogenous economic growth. Journal of Economic Studies, 34(1), 29–41.

77.

The World Bank (1993). The East Asian miracle: Economic growth and public policy. Oxford: Oxford University Press.

78.

The World Bank (2014). Nigerian economic report (No. 2). Washington, DC: The World Bank Group.

79.

Timmons

J. F.

(2010, October). Does Democracy Reduce Economic Inequality? British Journal of Political Science, 40(4), 741–757.

80.

Ugur

& Dasgupta

(2011). Corruption and economic growth: A meta-analysis of the evidence on low-income countries and beyond (MPRA Paper No. 31226). Munich: Munich University Library.

81.

UNDP (2016). Human development for everyone: Briefing note for countries on the 2016 Human development report for Nigeria (Human Development Report 2016). New York: UNDP.

82.

UNESCO (2011). The hidden crisis: Armed conflict and education. Retrieved 15 July 2017 from http://unesdoc.unesco.org/images/0019/001907/190743e.pdf

83.

Unumen

J. O.

(2014, March 15). Emerging culture of negative values in contemporary Nigeria: Implications for national development. Paper Presented at the 2nd Faculty of Arts International Conference on the Humanities and the Dynamics of African Culture in the 21st Century, Ambrose Alli University, Ekpoma, Edo State, Nigeria.

84.

Unumen

J. O.

& Emordi

E. C.

(2012, October 28–31). Increasing spate of financial corruption in the public sector: Implications on democracy and governance in Nigeria’s fourth republic. Paper Presented at the 57th Annual Congress of the Historical Society of Nigeria (HSN), IBB University, Lapai, Niger State.

85.

Unumen

J. O.

& Oghi

F. E.

(2016). Democracy, development and insurgency: The Nigerian experience in the fourth republic. An International Multi-disciplinary Journal, Ethiopia, 10(2), 31–46.

86.

Vreeland

J. R.

(2008). The effect of political regime on civil war. Journal of Conflict Resolution, 52(3), 401–425.

87.

Wade

(1990). Governing the market: Economic theory and the role of government in East Asian industrialization. Princeton, NJ: Princeton University Press.