The history of the financial sector in Nigeria is characterised by set-backs and reforms which may lead to nonlinear effect on growth. Previous studies predicated on linear models yielded conflicting results on the effect of finance on economic growth. Thus, there is need for further evidenced-based research and to achieve this; the paper employed the recently developed non-linear autoregressive distributed lag (NARDL) methodology. The main objective of the study is to evaluate the existence of asymmetric long-run macroeconomic relationship between financial sector development and economic growth in Nigeria between 1990 and 2015. The findings show that a long run nonlinear or asymmetric cointegration relationship exist between economic growth and two financial proxy indicators (total banking asset to GDP and currency in circulation to GDP) while the outcome for the other indicators were inconclusive.
The authors are staff of the Central Bank of Nigeria, Abuja. The views expressed in the paper are those of the authors and do not reflect those of the Bank or its policy.
An efficient financial system has tremendous role to play in the process of capital accumulation and technological innovation, which in turn promotes economic growth. Some of the well-acknowledged functions of the financial system that are growth-enhancing include: facilitating the mobilization of savings as well as investment in the economy; providing efficient payment mechanism for transactions, especially through the use of money which lubricates the wheel of economic progress; providing a veritable system of protecting economic agents from risks; and by overcoming incentive-related problems that undermine financial contracts such as moral hazards, adverse selection and information asymmetry (Merton and Bodie 1995).
The race to build an efficient financial system started in pre-colonial Nigeria during the twilight years of the nineteenth century. However, the severe banking crisis that occurred between 1940 and 1960 left the nascent financial system prostrate with the closure of several banking institutions (Moh and Eboreime 2011) The post-independence experience with financial sector development in Nigeria was characterised by weak institutions that operated under the ambit of direct control policies which negatively affected financial intermediation.
A turning point in the history of financial sector development in Nigeria came with the introduction of the structural adjustment programme (SAP) in 1986 which resulted in financial sector deregulation and liberalization. The number of commercial and merchant banks leaped from less than 40 before SAP to 120 in 1992. But according to Alade (2016), “the fragility of the financial system showed up when the number of banks declined to 89 as a result of the liquidation of over 30 terminally distressed banks.” Aside the liberal licensing of banks by the Central Bank of Nigeria (CBN), the imperative for a financial system that is supportive to growth brought about further reforms such as: universal banking in 2000; banking system consolidation in 2004; several other policies of the CBN aimed at regulatory over-hauling between 2005 and 2008; quantitative easing during the period of global financial crisis in 2008/2009 and the ‘Alpha Project Initiative’ that resulted in a new banking model (Alade 2016).
The preceding overview shows that setbacks continued to plague the orderly progress of the nation’s financial system even after the introduction of SAP. Given this vacillating scenario, it is germane to ask, what is the long run impact of financial development on economic growth in Nigeria? To answer this question, the prevailing finance-growth literature has focused on multivariate linear econometric models which may fail to account for the observed non-linear dynamics between growth and financial development. Therefore, “existing findings could become grossly inadequate should symmetry restrictions be rejected, which provide evidence of misspecification error, and thus, associated policy prescriptions there from would have misled policymakers overtime”.
The fluctuating experiences or rather, the successive acceleration and deceleration in the trend of financial sector performance in Nigeria calls for innovative and evidenced-based research initiatives to determine the effect of financial development on economic growth and to draw relevant policy conclusions. In this regard, several Nigerian studies produced conflicting outcomes with either finance promoting or impeding growth (Ndebbio 2004 and Nnanna 2004, cited in Madichie et al. 2014). Furthermore, divergent opinions on this issue have existed amongst economists since the time of Schumpeter (1912), Robinson (1952), Patrick (1966), Goldsmith (1969), McKinnon (1973) and Shaw (1973), and this makes the finance-growth controversy a subject of further empirical investigation.
One of the major methodological shortcomings of previous studies lies in the application of linear-based analytical approaches to possibly non-linear economic relationships. This present study differs markedly from previous ones by employing the recently developed non-linear autoregressive distributed lag (NARDL) methodology of Shin, Yu and Greenwood-Nimmo (2014) to investigate the asymmetric relationship between financial development and economic growth in Nigeria. The specific objectives of the study are to:
to evaluate the asymmetric long-run macroeconomic relationship between financial sector development and economic growth in Nigeria between 1990 and 2015;
to assess the asymmetric short-run dynamics in the finance-growth nexus;
To ascertain the veracity of symmetric effects based on Nigerian data.
This study extends the frontiers of literature and is unique in several respects. One, the study not only applies the recent NARDL approach, but it is also the first study we know that applied this methodology to the finance-growth nexus. Two, we applied the recently rebased GDP using the 2010 constant prices for entire scope of the study from 1990 to 2015.
Following the introduction, section two focuses on trend analysis while section three dwells on the review of literature. Section four provides an exposition on the methodology and the fifth section relates to empirical analysis and discussion of findings. Finally, section six concludes the paper.
2. Trend Analysis of Financial Sector Development in Nigeria
Since the introduction of SAP in 1986, Nigeria began to implement financial sector reforms as part of broader market oriented reforms. The objective of the reforms was to build a more efficient, robust and deeper financial sector. Although, the financial sector seems to have improved since the commencement of reforms, yet, the depth of financial development in Nigeria is still questionable.
Figure 1. Relationship between TBAGDP, PSCGDP and RGDPG
Source: The Authors Figure 1 indicates some form of co-movement between total banking assets to GDP (TBAGDP) and real GDP growth (RGDPG) but some divergences could be observed. For instance, the peaks in TBAGDP in 1991, 2001 and 2009 correspond well to that of RGDPG. Similarly, at several points in time when TBAGDP fell, we note that RGDPG fell as well. Thus, TBAGDP seems to be a driver of economic growth.
The trend noticed in private sector credit to GDP (PSCGDP) represents a significant level shift and it shows a steady rise following the banking consolidation exercise in 2004, which resulted in an upswing in economic activities, while the RGDPG reveals that the global economic crisis of 2008/2009 triggered slower growth in the Nigerian economy. Furthermore, the recent plunge in crude oil prices starting from July 2014 affected economic activities to such an extent that the economy began to show signs of negative growth in 2015.
Figure 2. Relationship between CICGDP, M2GDP, MKCGDP and RGDPG
Source: The Authors Figure 2 shows that at different points in time from 1990 to 2003, real GDP growth (RGDPG) seems to be a key driver of market capitalization (MKCGDP). However, between 2003 and 2007, the MKCGDP ratio rose significantly and reached its highest peak while RGDPG fell on the average. The substantial rise in market capitalization was due mainly to the bank consolidation exercise at that time. During the period covered by the study, the currency in circulation to GDP ratio (CICGDP) and broad money supply to GDP ratio (M2GDP) co-moved with RGDPG and diverged at different times. In summary, the simple trend analysis captured above portrays a nonlinear relationship between real GDP growth and other financial variables, thereby lending credence to the application of a nonlinear technique to data analysis.
Investigations into the nature of relationship between financial system development and economic growth has spanned over a century. There are several competing and divergent paradigms on the finance–growth nexus but two of these are dominant, namely the supply-leading and demand-following paradigms.
Marwa and Zhanje (2015) and Levine (1997) undertook an extensive literature survey on the relation between financial development and economic growth. The reviewed works include Bagehot (1873), who showed that capital spillovers from the financial system increased the rate of industrial revolution in Britain as resources were pooled together and allocated to entrepreneurs with the most profitable opportunities. In a ground-breaking work, Schumpeter (1912) contends that a well-functioning financial system will drive technological innovations (growth) through efficient distribution of resources from unproductive to industrious sectors. This view is often considered as the first strategy in analyzing the finance-led growth hypothesis. However, Robinson (1952) avers that causality proceeds from growth to finance as summarized by the frequently quoted statement in this regard that “finance follows where enterprise leads”. Robinson (1952) stressed that a growing real sector will turn into a great response for the services of the financial sector, and this implies that, a developed financial sector is a direct result of the demands of the expanding real sector of the economy. This view asserts a unidirectional causation from economic growth to financial sector development.
Patrick (1966) emphasized the “stage of development hypothesis” in which finance plays the initial catalytic role in propelling the expansion of the real sector and afterwards, growth takes over as the main driver of financial development. Thus, Patrick (1966) avers that a robust financial sector can induce real investment in the early stages of economic progress but this becomes less important as development gets well underway and the demand-following response takes over and becomes dominant. Accordingly, this view gives cause to believe that a bi-directional relationship exists between growth and finance.
McKinnon (1973) and Shaw (1973), emphasized the importance of financial liberalization in the enhancement of domestic savings and hence, investment. Consequently, the works of McKinnon (1973) and Shaw (1973) constitute the foundations for the financial sector liberalization in many developing countries, including Nigeria. The financial liberalization argument is based on the premise that financial repression reduces the real rate of economic growth and the real size of the financial sector compared to the size of non-financial sector. Therefore, while financial liberalization promotes financial sector development, financial repression through governmental controls, planned restrictions and price alterations hinder it.
The theoretical analysis of Eggoh and Villeu (2014) seems to provide some insight on possible asymmetric relationship between finance and growth. They focused on the role of financial intermediation in a simple endogenous growth model.
The analysis reveal that the finance-growth nexus can result in conflicting impacts as various levels of financial development can generate multiple endogenous growth paths due to reciprocal externality between financial sector and the real economy.
The Finance-Growth Nexus
To illustrate the possible effects of financial development on economic growth, we follow the exposition of Pagano (1993) which is predicated on the simple ‘AK’ model where aggregate output, Yt is dependent on aggregate capital stock, Kt:
where A is the total factor productivity. The gross investment, It is given by:
It = Kt+I - (1- δ) Kt. ………………………………..….…(2)
If the economy’s output is invested, the depreciation at end of t period is given by δ. The equilibrium condition for a two-sector economy without government and an external sector is given as gross saving equal gross investment:
In line with Pagano’s framework, the proportion of gross saving channeled to investment is φ while another proportion of 1- φ is assumed to be lost in the process of financial intermediation;
The economic growth rate is given by the ratio of change in output (see AK model in Todaro and Smith, 2009)
Given the proportional relation between output and capital, Pagano (1993) posits that growth rate at time t + 1 is given as:
gt+1 = Yt+1/Yt-1= Kt+1/Kt-1.
With appropriate substitutions in the above equations, the steady-state growth rate is given as (time subscripts dropped):
g=AI/Y - δ= Aφs - δ………………………………………(6)
s is the saving rate. Equation (6) provides the intuition on the channels through which financial sector development (FSD) impact on growth: FSD can increase the value of φ; it can improve the value of A through innovation; and FSD can positively influence the saving rate, s.
The outcome of an extensive review of empirical studies by Levine (1997) indicates that cross country comparisons, specific country studies, industry-level and firm-level investigations overwhelmingly suggest that financial system development is vitally associated with economic growth. For instance, countries with larger banks and more active stock markets grow faster in succeeding decades, while, industries and firms that rely heavily on borrowed funds “grow disproportionately faster in countries with well-developed banks and securities markets than in countries with poorly developed financial systems”. In the same vein, country-level studies show that countries with better financial sector have better growth performance.
Liang and Reichert (2006) studied the causal relationship between economic growth and financial sector development by using a panel of twenty advanced countries and 70 emerging/developing countries with data spanning from 1960 to 2000. The methodology employed includes Granger causality and the ‘Odedokun (1996) multifactor aggregate production function models which controlled for some other notable variables which influence real GDP in order to eliminate the problem of omitted variable bias. Inter alia, the Granger causality test involving data set of the emerging and developing countries provides strong evidence to the effect that causality runs from economic growth to financial sector development – a demand following relationship. However, the results of the panel least squares and fixed effect regressions based on the multivariate model reveals that the finance proxy, M3, has a positive and highly significant impact on growth, thus supporting the supply–leading paradigm.
Shen and Lee (2006) re-studied how financial development affects real GDP growth per capita in a sample of 48 countries. The result of the linear regression model show that stock market developments (SMD) produce positive impact on economic growth while banking sector development (BSD) has a negative impact or no impact on growth. However, further investigations indicate that when conditional variables such as financial liberalization, high income level (economy) and good shareholder protection were built into the model, the unfavourable effect of BSD on growth was lessened. On the contrary, when the conditional variable combination include middle income level, banking and currency crises, good creditor protection, higher corruption as well as specific regional grouping (Latin America, sub-Sahara Africa and East Asia), the negative effect of BSD on growth exacerbated but if only the regional grouping and SMD are integrated into the model, growth is promoted. Furthermore, Shen and Lee (2006) used a quadratic functional form to provide evidence that the finance-growth nexus may actually be non-linear.
Huang and Lin (2009) studied the growth effects of financial intermediary development for a data set of 71 countries with distinct levels of income by employing the instrumental variable threshold regression approach. The finding shows a non-linear positive effect of finance on economic growth, with the impact on low income countries being more pronounced than high income economies. In terms of the channel of transmission, the result indicate that finance affect growth in high income countries through capital accumulation, whereas, for low income countries, financial development impact positively on growth through the mechanism of productivity growth and capital accumulation.
Odeniran and Udeaja (2010) explored the relationship between financial sector development and economic growth in Nigeria using time series data from 1960 to 2009. The financial sector proxies used include: broad money to GDP, bank deposit liability to GDP, net domestic credit to GDP and private sector credit to GDP. Evidence from the cointegration analysis suggest that a long run relationship exist between real output and the financial sector variables which is in harmony with the finance-led theories. Furthermore, the granger test reveals that causality runs from financial sector development to growth – supply leading. However, the evidence in respect of a bidirectional relationship stems from the fact that causality also runs from per capita output to net domestic credit and private sector credit.
Chea’s (2011) study distills some key factors that can accelerate or impede economic growth in sub-Saharan African (SSA) countries. The findings show that financial deepening could reduce the problems associated with income inequalities and poverty provided that financial liberalization (interest rates and lending) is accompanied by institutional reforms that, in particular, emphasizes stronger property rights and reduction in information asymmetries with respect to borrowers and lenders. Furthermore, SSA must tackle the challenges of low saving and impediments to expansion of domestic bond markets to promote growth.
Similarly, DFID (2004) observe that the development of the financial sector can make important contributions to long-run economic growth and poverty reduction, especially in developing countries through enhancement in savings, encouraging inflows of foreign capital and optimizing the allocation of capital between competing uses.
Adua, Marbuah and Mensah (2013) investigated the long-run growth effects of financial development in Ghana. The study employed the autoregressive distributed lag (ARDL) approach and principal component analysis. Findings based on cointegration test shows that out of the eight financial proxy variables, long run growth effects were established with respect to only three financial indicators, namely: credit to private sector to GDP ratio, credit to private sector as a ratio to total domestic credit and broad money supply to GDP ratio. However, results from the principal component analysis indicate that the growth effect of financial development is sensitive to the choice of proxy. For example, both the private sector credit and aggregate domestic credit variables were found to be conducive to growth while broad money supply to GDP is not growth-promoting. Thus, they concluded that whether financial development is good or bad for growth depends on the indicator used as proxy for financial development.
Kapingura (2013) examined the dynamic relationship between financial development and economic growth in South Africa. The results suggest that financial intermediaries and financial markets have different impacts on economic growth given their different roles in the economy. In particular, there is bi-directional causality between stock market and economic growth. Also, a unidirectional causality from the bond market to economic growth was established.
Kiprop, Kalio, Kibet and Kiprop (2015) studied the finance-growth nexus in Kenya based on the autoregressive distributed lag method. The findings validates the supply-leading hypothesis in Kenya as the effect of financial development on economic growth was positive and statistically significant, both in the short and long run.
Nkoro and Uko (2013) examined the finance-growth relationship in Nigeria. The study employed the Johansen cointegration/ECM methodology with annual dataset covering the period, 1980-2009. The findings show that financial variables such as market capitalisation to GDP, interest rate and broad money supply to GDP are stimulators of the Nigerian economy in the long run. The outcome of the error correction mechanism confirms the importance of the financial sector in the short run with economic growth adjusting speedily to financial sector development. However, the authors noted that there exists an inverse relationship between credit to GDP and Nigeria’s economic growth. This was attributed, inter alia, to the crowding out of private investment due to large government borrowing.
Balago (2014) studied financial sector development and economic growth in Nigeria. The result shows that development in the financial sector as measured by banking sector credit, total market capitalisation and foreign direct investment positively affect economic growth. This result is consistent with a number of earlier studies reviewed in the literature that found financial sector to positively affect real gross domestic product.
A major shortcoming of majority of studies is the implicit assumption that the relationship between finance and growth is linear. Hence, this present work takes a different investigative approach by adopting a non-linear methodology.