The long‐run impact of idiosyncratic and common shocks on industry output in Ghana

Published date01 March 2015
DOIhttp://doi.org/10.1111/opec.12039
Date01 March 2015
The long-run impact of idiosyncratic
and common shocks on industry output
in Ghana
Philip Kofi Adom,* Kwaku Amakye,** Charles Barnor*** and
George Quartey****
*Assistant lecturer, Department of Banking and Finance, University of Professional Studies, P.O.Box LG
149, Legon-Accra, Accra, Ghana. Email: adomonline@yahoo.co.uk; adomonline12@gmail.com;
Philip.adom@upsamail.edu.gh
**Assistant lecturer, Department of Banking and Finance, University of Professional Studies, Accra,
Ghana. Email: amakye2002gh@yahoo.com; Kwaku.amakye@upsamail.edu.gh
***Head, Department of Banking and Finance, University of Professional Studies, Accra, Ghana. Email:
charlesbarnor@yahoo.com; charles.barnor@upsamail.edu.gh
****Dean, Faculty of Accounting and Finance, University of Professional Studies, Accra, Ghana. Email:
geoquartey@yahoo.com; george.quartey@upsamail.edu.gh
Abstract
This study explores the long-run impact of idiosyncratic and common shocks on industry output
in Ghana while controlling for the effects of investment. In order to deal with the second-order
bias problem, this study employed canonical cointegration and fully modified ordinary least-
squares (OLS) regressions, which are more robust to second-order bias problems. Different
models are, therefore, specified and estimated. Fully modified OLS and canonical cointegration
are extended in successive steps in order to verify if the inclusion of idiosyncratic and common
shocks improves the statistical properties of the model. Secondly, a backward approach, in which
idiosyncratic and common shocks are excluded successively, is also adopted. Preliminary findings
showed signs of long-run equilibrium. The a priori expectation and the statistical importance of
investment are established in both fully modified OLS and canonical cointegration models. This
result is robust using both the Bartlett and Parzen kernels. However, while the elasticity value for
investment is invariant to the model and kernel type used for fully modified OLS, the opposite
result is found for canonical cointegration. Importantly, the absolute value of the investment elas-
ticity is kept within the limits of 0 and 1. The impacts of idiosyncratic and common shocks are
negative and statistically significant in the long run for both fully-modified OLS and canonical
cointegration. This result is robust to the Bartlett and Parzen kernels. Result based on the fully
modified OLS also showed that the sizes of the elasticity values for both idiosyncratic and
common shocks are sensitive to the model type and kernel type used. Despite the differences in
the elasticity values, result for both models are qualitatively similar.
17
© 2015 Organization of the Petroleum Exporting Countries. Published by John Wiley & Sons Ltd, 9600 Garsington
Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
1. Introduction
The growth of Ghana has been cyclical in nature, and this has been reflected in various
subsectors of the economy. The industrial sector continues to show few prospects of
gaining the first place in terms of contribution to the nation’s growth and development.
The major causes of this have included both domestic and international developments.
Domestically, exchange rate volatility and frequent power outages continue to pose
serious downside risk to the industrial sector. On average, import of goods and services
consumed in Ghana accounts for about 70 per cent of business input costs. In 1992–
1993 and 1993–2006, the nominal exchange rate depreciated by 52 per cent and 22
per cent, respectively. The rate of depreciation persisted but declined during the period
2006–2009 (10.11 per cent). Recently, in 2012, the economy of Ghana was hit by
another excessive exchange rate shock. The cedi started out at about GH¢1.58 to the
dollar in the beginning of January 2012 but went up to GH¢1.66 by the close of January
2012. By the end of the first quarter, the cedi had crossed to GH¢1.7 per dollar, repre-
senting a depreciation rate of about 8.3 per cent. By the end of June 2012, the cedi hit
GH¢1.89 per dollar. Between January and August 2012, the cedi had depreciated by
18.3 per cent.
Also, Ghana has an unusually long history of power crises. The most significant and
prolonged ones occurred in 1984, 1998, 2002, and 2006–2007. The unreliable nature of
power supply in the country continues to threaten the sustainability and growth of indus-
tries in Ghana. For instance, the 2007 powercrisis led to the collapse of some industries,
with attendant effects on employment. The economic impact of such outages has been
noted in the literature to be enormous, especially in developing countries, where the situa-
tion is very rampant. The World Bank research group on Africa has shown that there is
enormous economic loss due to power outages. According to the World Bank, power
outages cost African countries about 2 per cent of GDP.This implies that power outages
cost the Ghanaian economy about US$ 320 million in 2008 (see Adom and Bekoe, 2012;
Adom et al., 2012). Also, Power Systems Energy Consulting (PSEC) and Ghana Grid
Company Ltd (GRIDCo) estimate the valueof lost load (VOLL) for the industrial sector in
2009 to be US$ 598.4 million (Adom and Bekoe, 2012).Aside from these domestic devel-
opments, the industrial sector is also susceptible to developments on the worldoil market.
As oil drives all economic activities,either directly or indirectly, price shocks due to either
demand or supply shocks also pose serious downside risk to industries. Despite her recent
oil discovery,Ghana is still dependent on imported oil, and therefore increases in world oil
prices will contract the output growth of Ghanaian industries, hence the decline in eco-
nomic growth. Looking ahead,oil prices are less likely to fall below $100 per barrel due to
promising recovery in the United States, increased demand from China and India, and
unabated geopolitical risks.
Philip Kofi Adom et al.18
OPEC Energy Review March 2015 © 2015 Organization of the Petroleum Exporting Countries
How do the simultaneous occurrences of these shocks affect industry output in
Ghana? Previous studies have focused the discussion on the impact of exchange rate and
oil price shocks on economic growth and other macroeconomic variables (see, inter alia,
Routava, 2004; Aghion et al., 2009; Gomez-Loscos et al., 2011; Peersman and Robays,
2012). These studies ignore the role of power crises.This makes such studies limited in
scope in terms of appreciating the interactive role of these shocks. There is evidence of
the simultaneous occurrence of these shocks in history. For instance, between 1986 and
1991 and between 2006 and 2009 Ghana experienced a power crisis and exchange rate
and oil price shocks concurrently. The concurrent occurrence of these shocks creates a
situation of possible reinforcement among these shocks. Therefore, in a regression that
excludes any of these shocks, the impact of the included shocks on economic growth
may be overstated.
The purpose of this study is to determine the long-run impacts of common shocks (i.e.
crude oil price shocks and power crises) and idiosyncratic shocks (i.e. exchange rate vola-
tility) on industrial output in Ghana. The major concern with this approach is that there
could be an endogeneity problem, which stems from the fact that these shocks may
reinforce each other. Forinstance, world oil price spikes could fuel domestic power crises,
as the generation of power in the country is oil-driven.Also, world oil price spikes could
spike exchange rate shocks. The consequence of this is that, by applying models that are
not robust to endogeneity problems, the actual impact of these shocks cannot be identified.
This can be handled by applying instruments or using models that are known to be robust
to endogeneity and serial correlation problems. The difficulty with the former is that
finding appropriate instruments is a hard task. In this study, the authors apply the fully
modified ordinary least-squares (OLS) method and canonical cointegration regression
(CCR), which are more robust to endogeneity and serial correlation problems. These
methods are known to be asymptotically equivalent.
Another issue of concern is that existing studies on cointegration are based on statisti-
cal tests of no cointegration. However, as argued in Shin (1994), the limiting distribution
of test statistics for no cointegration depends only on a function of Brownian motion and
contains spurious regression distribution. In contrast, the limiting distribution of test sta-
tistics for cointegration involves a combination of a Brownian bridge and a function of
Brownian motion and depends on the compound normal distribution (Phillips, 1991). In
this study, the authors compare the result from statistical tests of no cointegration (Engle
and Granger, 1987; Phillips and Ouliaris, 1990; Johansen, 1991) with statistical tests of
cointegration (Phillips and Hansen, 1990). This makes it possible for the authors to cross-
check for possible ambiguity in the cointegration test.
The rest of the article is structured as follows. Section 2 reviews relevant literature;
Section 3 provides a setting and context analysis on the economyof Ghana and develop-
ments in the world economy; Section 4 describes the data and method used; Section 5
The long-run impact of idiosyncratic and common shocks on industry output in Ghana 19
OPEC Energy Review March 2015© 2015 Organization of the Petroleum Exporting Countries

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT