# 15 Investment frictions in Africa: identifying the

structural and institutional determinants

Chuku Chuku _1, Kenneth Onye y2, and Hycent Ajah z3

1Department of Economics and Centre for Growth and Business Cycle

Research, University of Manchester, Manchester, U.K.

1,2,3Department of Economics, University of Uyo, Uyo, Nigeria.

March, 2015

Extended Abstract

As shown in a recent World Bank study, the cross-country variation in investment activity

and returns is widening and the variation is even more pronounced in Africa. Between 1980

and 2010, the rate of gross capital formation ranged between 1 and 90 percent of production

worldwide (Lim, 2014). This widening variation in investment activity is mostly due to the

di_erent kinds of frictions present in di_erent economies which prevents a normalization

of the returns from investment activities across countries. This eventually, inhibits the

potential for regional integration and investment competitiveness. In order to facilitate e_orts

towards regional integration in Africa, it is important to correctly identify the factors that are

responsible for the investment related frictions in African economies. Hence, in this study, we

endeavour to provide answers to questions such as, what factors are responsible for relative

investment activity in Africa?

Addressing this question for the African experience generally requires a slightly broader

approach than is used in the literature (see for examples Ndikumana, 2005; Love & Zicchino,

2006), this is particularly because of the greater diversity that exists in the region in terms

of political and institutional frameworks which is di_erent from the relative homogeneous

characteristics of developed economies in Europe and America. The proposition we make

is that in addition to the usual economic factors that determine investment frictions and





activity, there exist a wider set of factors including political, security, legal and institutional

dimensions that should be accounted for in understanding the dynamics of investment activity

and competitiveness in Africa.

The objective of this study is to empirically identify the broad set of factors and frictions

that explain the di_erences in investment activity and competitiveness in Africa in the last

three decades. The study is particularly di_erent from others in the literature because it

considers a broader set of structural and institutional determinants that are important to

characterize the African experience and does not lump developed and developing countries

together in a panel.

Empirical strategy and data

The empirical strategy adopted in the study is theoretically motivated from a standard

neoclassical growth formulation (see Lim, 2014, for a similar application), where production

is constant returns to scale and given by the Cobb-Douglas speci_cation

Yit = ezK_


it (1)

Where Yit is level of output in country i, ez is technology which is subject to a stochastic

AR(1) shock process thus; zt = _zt􀀀1 +_. While Kit and Lit are the capital and labour used in

production in country i and _ is the share of capital in output. Where capital stocks evolves

according to the following equation of motion

Ki;t+1 = (1 􀀀 _)Kit + Iit (2)

The optimal capital stock in country i at time t is given as the weighted ratio of real output

Yit and the cost of capital Rit hence


it =





Using the familiar result from neoclassical growth theory that in steady state with a balanced

growth path _, the growth rate of output, capital and consumption are equal, we can plug in

the optimal level of capital (3) into the steady state equation of motion for (2) to obtain an

expression for investment as

Iit =

_(_ + _)Yit




By taking the logarithm of both sides of (4), we obtain an estimable equation for investment

given as

ln Iit = ln _ + ln(_ + _it) + ln Yit 􀀀 _ lnRit (5)


Where ln _ is the constant term and ln(_ + _it) _ git is the depreciation-adjusted growth rate

in country i. To account for the additional structural and institutional variables which the

neoclassical growth theory abstracts from, we include additional economic and structural

variables in the vector Xit and institutional variables in the vector1 Zit, plus an error term _it

so that the complete econometric estimation equation becomes

iit = _ + _ii;t􀀀1 + _git + 'yit 􀀀 _rit + 0Xit +           0Zit + _it (6)

Where the lower-case letters indicate the logarithms of the variables and bold letters are

vectors. Further, an interest rate smoothing term ii;t􀀀1 is also introduced to account for

partial-adjustment behaviour in capital formation observed in the literature (see Eberly,

Rebelo, & Vincent, 2012)

The baseline regression equation in (6) is primarily estimated by system generalized

method of moments (GMM) with robustness tests conducted using pooled regression and

standard instrumental variable (I.V) techniques. The main advantage of using the system

GMM technique is to enable us exploit the e_ciency gains that arise from considering the

instrument set as a system especially given that the number of cross section identi_ers are

less that the time series (i.e N<T). This method also allows us to take care of potential

endogeniety problems.

Most of the data series that will be used in the study are retrieved from the World

Development Indicators (WDI) of the World Bank. We provide a brief description and

sources of the not too common data series to be included in the structural and institutional

vectors. We proxy institutional quality using indices of corruption and rule of law from

International Country Risk Guide (ICRG), while institutional structure is measured by

democratic accountability taken from ICRG. To capture political stability we use the Polity

IV project dataset from systemic risk. Business environment is measured using the index of

strength of investment protection from ICRG. The Human Development Index representing

the index of human capital per person based on years of schooling is obtained from (Barro &

Lee, 2013) while returns to education will be obtained from the Penn World Table (PWT).

Tax structure is captured by the corporate tax rate from WDI. Variables for economic controls

include; Gross _xed capital formation, output measured by real Gross domestic product

(GDP), Output growth (Growth in real (RDGP), Government consumption, Ination, Cost

of capital measured using Real interest rate (lending rate adjusted for ination) and Trade

openness (Imports plus exports divided by GDP) are taken from the WDI. Financial openness

Index of restrictions on capital account openness is obtained from Chinn and Ito (2008). The

dataset will cover several West Africa countries over the period 1990 to 2012.

1The complete set of variables to be included in these vectors would be decided depending on certain

criteria discussed latter.



Barro, R. J., & Lee, J. W. (2013). A new data set of educational attainment in the world,

1950{2010. Journal of Development Economics, 104 , 184{198.

Eberly, J., Rebelo, S., & Vincent, N. (2012). What explains the lagged-investment e_ect?

Journal of Monetary Economics, 59 (4), 370{380.

Lim, J. J. (2014). Institutional and structural determinants of investment worldwide. Journal

of Macroeconomics, 41 , 160{177.

Love, I., & Zicchino, L. (2006). Financial development and dynamic investment behavior:

Evidence from panel var. The Quarterly Review of Economics and Finance, 46 (2),


Ndikumana, L. (2005). Financial development, _nancial structure, and domestic investment:

International evidence. Journal of International Money and Finance, 24 (4), 651{673.

Recent Publications - View all