Topic: impact of oil price on the current account of GCC
Pages: 11, Double spaced
Order type: Dissertation
Language: English (U.K.)
The discussiоn should reflect the literature review and the findings. and interрretation to a good standard.
discuss how endogeneity can be solved by future research by reрlacing savings (S) with (the lag of savings). The lag is called a “рredetermined” variable. It means it is determined in the past. Therefore, it is resolves the endogeneity problem.
the recommendation should reflect the results such as how total factor productivity was insignificant by increasing it it will positively affect the current account. I am assuming this is chapter 5 discussion and chapter 6 conclusion and recommendation
Equation (4) which sums up the regression with all the variables presents the following illustration of the model:
The coefficient indicates that as savings (as % of GDP) increases by 10%, the current account (as% of GDP) increase by 3% on average (across the six countries).
The oil price shock measured by a 10% increase in oil prices increase the current account (as % of the GDP) by half of 1% percent on average.
As by 1% increase in TFP shock, which is measured by the difference between GCC and U.S shocks, increases the current account (as % to GDP) by a quarter of a percent on average.
It is also observed that when the fiscal policy shock increase by 1% the current account deficit grows by 1% on average.
The Cooperation Council for the Arab states of the Gulf originally known as Gulf Cooperation Council (GCC) is an intergovernmental economic and political organization, established in May 1981, and is currently made up of six states namely: Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates (Reiche, 2010). The following four of the GCC member states are also member states of the Organization of Petroleum Exporting Countries (OPEC) .
Saudi Arabia is the world’s largest oil producers after Russia (Bloomberg, 2017) accounting for an average production of 10.46 million barrels per day, with the highest U.S dollar value worth of crude oil at $U.S 136.2 billion, which accounts for 20.1% of the total crude oil exports (Worlds top exports, 2017).
The UAE and Kuwait are among the top 10 in the world in oil production and net oil exporters (World top exports, 2017). Qatar, Oman and Bahrain are considerably lower smaller oil producers. Although Qatar possesses relatively less oil, it commands the worlds’ third proved reserve of natural gas (Central Intelligence Agency, 2017).
Evidently, the GCC are key players in the global energy market, in the sense that they could influence global oil prices. They have become part of the international policy debate on global imbalances because oil price shocks affect the current accounts of oil producers differently from oil consumers. Oil is also the main source of revenue to the GCC, thus an increase (decrease) in oil prices creates surplus (deficit) in their current account, which the opposite effect on the current accounts of the oil consumers.
Indicators of oil revenue and oil exports among the GCC (International Monetary Fund, 2016)
The GCC states share a number of specific structural economic features, such as high dependency on exporting oil. Figure 1.2 conveys that approximately 70% of all GCC fiscal revenues rely on oil . Similarly on oil exports, all GCC countries (except United Arab Emirates), accounts for at least 60% of oil export goods.
In mid-year of 2014, the oil of price fell sharply from $145 to $45 and remained below $50 a barrel. The GCC economies have been adversely affected; the oil revenue, which account for the fiscal revenue fell sharply in 2015 by about 24% as shown in figure 1.3; its over-reliance on oil had exacerbates macroeconomic volatility, causing a fall in saving, external debts have been introduced (figure 1.4), resulting in a current account deficit, i.e., a balance of payment arose.
The GCC revenue structure as of 2012-2015 (International Monetary Fund, 2016)
While many studies such as of Kilian (2010) and Bodenstein et al. (2008) (Mohanty et al., 2011) Allegret et al.’s (2014) have undertaken on the wider subject matter of the correlation between oil price fluctuations and current account it was notable in the literature that most studies on this matter have been conducted on the economies of the Western world and have disregarded developing countries, in particular the GCC.
External debts of the GCC are arising ever since the drop of oil price in 2014 (Dijkhuizen, A. 2016)
The effect of oil price fluctuation is much more prevalent than ever among key players in the oil industry as well (especially politicians and economists) which have questioned the GCC reliance on oil. With GCC being a global player in the global oil market and being the most affected countries by oil-driven crisis; the world economy is jeopardize.
Therefore, the objective of this paper will measure the impact of oil price on the current account against the percentage of GDP of GCC countries. Which will fill a gap in the literature of studying the impact of oil price on current account precisely on the GCC countries. This paper also differs from most of the available studies of the impact of oil price on the current account of the GCC, reviewed below, mainly on methodological grounds by using a cross-country panel data regression, which in this approach will better over come the unobserved heterogeneity.
The empirical analysis is based on estimating an equation of the current account derived from theoretical model of Reuven Glick and Kenneth Rogoff (Glick and Rogoff, 1995) their model shows that the current account, whereby their model, the current account depends on investment, productivity shocks and fiscal policy shocks. In this study, we will extend their model by adding the effect of oil price shocks on the current accounts of the GCC.
The Data of this study have been collected for the GCC from the (International Monetary Fund, 2016), and (The Conference board, 2016). The program stata have been used for producing data results.
The results of this study have found the effect of oil price shocks on the current account to be positive and significant, however considering the issue of endogeneity arises due to the single-equation bias.
The remainder of this paper is organized as follow. Chapter 2 outlines a critical review of the literature pertaining to the subject matter under investigation. Furthermore, Chapter 3 covers the methodology that examined the methods that will be used to achieve the objective of this study followed by Chapter 4 presents the empirical results, then further Chapter 5 covers a discussion of the findings and Lastly, Chapter 6 comprises of the conclusions and recommendations of this study.
In this chapter, a critical review of the literature pertaining to the subject matter under investigation is undertaken. The review aims to locate the current study in the body of knowledge; and this is achieved by identifying the gaps in the knowledge and demonstrating how the current study fills those gaps. The chapter is organized thematically and covers aspects such as definitions and key theories, the theoretical and contextual framework, and related past studies relevant to the subject matter (impact of oil price fluctuations on the current accounts of GCC states).
In contextualizing the relationship between oil price fluctuations and current accounts, it is important to establish the manner in which oil price shocks affect the various dynamics of oil exporting countries. In essence, it is important to determine how variations (or shocks) in oil prices get transmitted to the current account balance of the country. In this respect, it is imperative that a distinction be made between changes in the price of oil that are occasioned by oil demand shocks and those occasioned by oil supply shocks. This is a very critical demarcation because, according to Kilian et al. (2009), the actual effect that oil price fluctuations have on the current account are influenced by, among other factors, the actual cause of the oil shocks.
Indeed both Kilian (2010) and Bodenstein et al. (2008) argue that there are five channels through which oil price shocks are transmitted to the current account. These are (1) the supply side channel; (2) the demand-side channel; (3) the monetary policy channel; (4) the trade channel; and (5) the valuation channel. They also argue that two critical factors influence smooth operation as well as the effectiveness of any given channel. These are the level of a country’s economic development and whether the country is an oil-exporting developing or developed economy or an oil-importing developing or industrialized economy.
This specific conceptual underpinning suits this study for various reasons. First and foremost, all GCC countries are classified as developing; and even though some of them have very prosperous economies, they are still considered to be developing countries (Nusair, 2016). Therefore, at least three of the five channels tend to be applicable to them. For most net oil exporting countries in the GCC, the trade channel, the valuation channel, and the monetary policy channel are among the key channels through which oil price shocks get transmitted to the current account dynamics.
It is indeed true that most of these countries usually resort to their monetary policies to seek to stabilize their economies especially in times of declining oil prices. In this regard, anticipated responses by these countries’ central banks (or any other monetary policy agencies) tend to amplify the recessionary pressures on the economy (resulting in recessions) (Cherif and Hasanov, 2013). With regard to the trade channel, these countries also tend to respond to oil price declines by mostly increasing either the quantity or prices of their other exports and/or reducing the amount of goods they import. Some countries will change how they value their currencies so as to increase the flow of foreign assets into the country’s economy (Mohanty et al., 2011). Therefore, this conceptual framework is best placed to explain the impact of oil price fluctuations on current accounts of GCC states.
Different theories could be used as the theoretical foundation of the current study. However, no other theoretical underpinning best suits the current study than the intertemporal approach to the current account that was put forth by Obstfeld and Rogoff (1995) which they rely on that approach. According to intertemporal approach, increases in oil revenue maybe invested or saved. According to the permanent income hypothesis, when open economies produce exhaustible natural resources (in the case of GCC), they ought to save most of their windfalls so as to be able make their consumption smooth, to preserve resource wealth, and make sure that there is intergenerational equity. Therefore, countries with oil windfalls will tend to run large current account surpluses in the immediate wake of their getting their oil income windfalls (Obstfeld and Rogoff, 1995).
The other channel for savings operates via the precautionary motive where a significant amount of additional savings can be made (Morsi, 2012). The best way to explain this channel is that oil exporting countries tend to regard increases in prices of oil as temporary. Therefore, they tend to hold the view that they need to save up as a precautionary measure so as to avoid running out of revenue when the prices either normalize or fall below the market value. Essentially, as Bems and Carvalyo (2011) contend, savings are used as way of dealing with uncertainty.
Within the framework of the Harberger–Laursen–Metzler effect, it may also be worth arguing that large oil income windfalls are likely to be saved especially if they are temporary and to be consumed when they are permanent (Allegret et al., 2014). This as a result of a marginal propensity to consume less than unity and thereby giving rise to an increase in current consumption that is less than the current level of income in the follow-up to a temporary increase in a country’s terms of trade. With regard to investments, it is also possible for oil revenue windfalls to be regarded as helping to relax constraints on borrowing and to expand financing sources for investment. In this case, deficits in the current account are induced (Allegret et al., 2014).
Definition of Key Terms
Some critical terms require defining in relation to this study in order to properly ground the study in knowledge. These include follows:
Oil price fluctuation: Oil price fluctuation as used in this study refers to the changes in prices of oil on the international market (Baumeister and Kilian, 2016). The changes are driven by different factors and include both long-term and short-term increases and decreases in the prices of oil (Baumeister and Kilian, 2016).
Current account: The current account is the summation of the country’s balance of trade, net cash transfers, and the net primary income (known as factor income) which have occurred over a given period of time (McCombie and Thirlwall, 2016). Put differently, it is the sum of the balance of trade (goods and services exports less imports), net income from abroad, and net current transfers. Together with the capital (or financial account), the current account forms a major component of a country’s balance of payments and is therefore a major indicator of the health of a country’s economy (Cheung, Furceri and Rusticelli, 2013).
OPEC: OPEC (Organization of the Petroleum Exporting Countries) is an intergovernmental organization that currently consists of 14 nations namely Algeria, Angola, Ecuador, Equatorial Guinea, Gabon, Iran, Iraq, Kuwait, Libya, Nigeria, Qatar, Saudi Arabia, United Arab Emirates, and Venezuela (Colgan, 2014). OPEC has its headquarters in Vienna, and its member states account for approximately 73% of the world’s total proven oil reserves and 44% of global oil production (Colgan, 2014). Therefore, OPEC has immense influence on global oil prices. This happens partly because OPEC can regulate global supply of oil and therefore easily influence the price level (assuming demand remains constant). This is in line with OPEC’s core mission which, according to Mattar (2004), is “to coordinate and unify the petroleum policies of its member countries and ensure the stabilization of oil markets, in order to secure an efficient, economic and regular supply of petroleum to consumers, a steady income to producers, and a fair return on capital for those investing in the petroleum industry.”
The GCC’s Responses to Oil Price Fluctuation
One of the issues that have attracted a lot of attention among scholars and practitioners alike is the most recent (2012-2014) oil price fluctuation and the different ways in which different GCC nations responded to it. According to Compagnie Française d’Assurance pour le Commerce Extérieur (COFACE) (COFACE, 2016) the GCC states, with 30% of the world’s proven oil reserves, were among the worst affected by the oil price decline that began gradually in 2012, accelerated sharply in the second half of 2014, and ended in 2016. COFACE argues further that in spite of this, different GCC countries were affected differently and therefore responded differently to the decline in oil prices with Saudi Arabia, Qatar, and the UAE being the least affected while Bahrain and Oman being the most affected. Their core argument with respect to this is that the effects were determined to a very large extent by the amount of financial buffers a country has in place (COFACE, 2016).
The arguments presented by COFACE (2016) are very important in that financial buffers play a critical role in determining a country’s exposure to oil price fluctuations. Generally, countries with more financial buffers tend to be less affected by such shocks compared those with fewer buffers (Momani, 2008). However, it would have been important if COFACE also clarified the relationship between financial buffers and economic diversification. After all, it is known that through greater economic diversification, major oil producing countries (such as those in the GCC) can better deal with declines in oil prices (Awartani and Maghyereh, 2013). It may also be important to add that even though COFACE does not directly address the issue of the impact of oil price fluctuations on the current account of GCC states, it still offers very important information pertaining to the relationship between oil price fluctuations and the economic performance of the different GCC states. The importance and value of such information to the present study is not in doubt.
On his part, Nusair (2016) contends that the GCC countries are net exporters of oil and as such reductions in oil prices will inevitably result in unfavourable terms of a negative trade shock. In the same vein, he argues further, the price inelastic nature of oil demand means that the aforementioned decline in prices of oil is often associated with worsening of both the balance of oil trade and the current account of both the GCC and individual GCC states all of which are net oil exporters. This argument is also correct because whenever oil prices decline, all countries that are net exporters of oil tend to be negatively affected both in their balance of trade and current account (Mohanty et al., 2011). On the contrary, a rise in oil prices improves the current account and balance of payments (Mohanty et al., 2011). However, it would have been important to add that different countries even within the GCC responded differently to the oil price declines. While some partially offset the direct effects on their current accounts (caused by falling oil prices) by cutting their imports, others did so by increasing exports in order to compensate for lost oil revenues (Reiche, 2010).
Relationship between Current Accounts and oil Prices: the Role of OPEC
To underscore the great importance that the relationship between oil prices and current accounts has, it may be worth noting that even as way back as the 1970-s and 1980s, economic scholars, policymakers, and other stakeholders had immense interest in understanding how the variations in oil prices affected the current accounts of different nations and groups of nations. The immense interest in the matter undoubtedly explains why Sachs, Cooper and Fischer (1981) investigated the factors that determined the direction and size of current account imbalances of different nations in the 1970s. According to them, the size and direction of current account imbalances in the 1970s could be largely explained in terms of increases in oil prices as well as current account surplus of OPEC member states. They argue further that since there was need for the large surpluses to be balanced in the aggregate demand through deficits of net oil importing countries, current account difficulties (deficits) in oil importing countries could be attributed to those countries’ imports of oil. Subsequently, a reduction in oil imports was largely considered to be a key policy objective (Sachs, Cooper and Fischer, 1981).
The arguments by Sachs, Cooper and Fischer (1981) are only correct to some limited extent (but not in their entirety). Indeed it is true that OPEC member states have had immense influence over the direction of oil prices on the international market (Colgan, 2014). Moreover, they have usually controlled demand of oil (and by extension the price of oil) by regulating output. Since it has always been to the advantage of OPEC member states to keep oil prices high, their most frequent policy has been to keep oil supply at levels that ensure that prices remain high (or increase) even if demand remains constant. Therefore, OPEC countries have tended to enjoy surpluses in their current accounts as a result (Momani, 2008).
It is also true – as Sachs, Cooper and Fischer (1981) argue – that the current accounts of oil importers tend to worsen in times when oil prices are high and when OPEC countries are enjoying surpluses in their current accounts (Cheung, Furceri and Rusticelli, 2013). However, it is not entirely true that all current account difficulties experienced by oil importers could be attributed to high oil prices. On the contrary, oil price increases only affect current accounts to some extent; but there are other factors that could explain the difficulties. In essence, current accounts of oil importers do not just decline because of excessive (and expensive) oil imports but also because of other factors such as general deprecation in the local currency.
The other argument that is fronted by Sachs, Cooper and Fischer (1981) and that may not be entirely true is that net oil importers can improve their current accounts solely through a policy shift such that they become less dependent on oil (by generally reducing oil imports). While reducing oil imports would undoubtedly result in more favourable balance of trade and therefore better current accounts for oil importing countries, this policy in itself – if not accompanied by others – may not necessarily address the problem of unfavourable balance of payments and a current account deficits. On the contrary, such a policy may need to be supported, or even preceded – by increasing savings (Colgan, 2014). The gist of this argument is that while rationalization of imports of oil would help importers to have better current accounts, any benefits realized from such a change in energy policy are likely to be offset by lack of savings or even a boom in investments (McCombie and Thirlwall, 2016). Therefore, the ultimate way of ensuring that oil importing countries have better and more favourable balances of trade and current accounts would inevitably have to include ensuring that they increase their savings or investments. As such, it may be worth arguing that cross-country differences in the level of investments and/or savings can be used to explain variations in current accounts just as much as fluctuations in oil prices.
The Oil Price-Current Account Nexus
Gnimassoun, Joets and Razafindrabe (2017) investigated the link between prices of oil and the current accounts for oil exporting countries with the major focus being on the time-varying nature of this link. Using the time-varying parameter vector autoregressive model with sign restriction, they found that even though an oil supply shock does not have a significant impact on the current account, an oil demand shock has both a positive and significant impact on the current account (which increases with time). The importance and relevance of Gnimassoun, Joets and Razafindrabe’s study lies in the fact that it underscores the very important role that is played by prices of oil in the forming of either deficits or surpluses (external imbalances) in the current accounts of nations. This is in turn very important because countries always endeavour to avoid global imbalances such as those that were witnessed between 2003 and 2008 (and that were attributed to the unprecedented increase on oil prices at the time.
The study by Gnimassoun, Joets and Razafindrabe (2017) also support findings by COFACE (2016) that to a very large extent, the link between the price of oil and the current account especially for oil exporting countries depends on the level of economic diversification of the countries concerned. It is this economic diversification that influences a country’s propensity to absorb oil shocks. Having commended the study, it is worth adding that its main delimitation is that it does not focus on a GCC country or even OPEC member state. Although Canada is classified among the largest oil exporters in the world, its oil exports only account for less than 20% of total exports. Therefore, the use of Canada as a case study, while clearly serving the interests of the researchers in terms of using an economically diversified country, may have the effect of rendering the study’s findings less generalizable and therefore less applicable to the GCC.
On their part, Berument, Ceylan and Dogan (2010) investigated how oil price shocks affect the output growth of some countries in the Middle East and North Africa (MENA) region. The focus of the study was on countries that are either net importers or net exporters of oil but which are too small to affect prices of oil. They found that increases in prices of oil had a positive and statistically significant effect on the outputs of nine countries namely Algeria, Iran, Iraq, Kuwait, Libya, Oman, Qatar, Syria, and the UAE. On the contrary, there seemed to be no positive and statistically significant effect of oil price shocks on the outputs of Bahrain, Djibouti, Egypt, Israel, Jordan, Morocco, and Tunisia.
It is to be appreciated that Berument, Ceylan and Dogan’s (2010) study does not directly address the impact of oil price fluctuations on the current account; and that this could be a limitation in some way. However, it is still very relevant because output growth – which the study deals with – is a very important determinant of a country’s current account. If anything, output growth is a measure of a country’s economic growth which is in turn indicated by (among other variables) the current account. More specifically, it is to be appreciated that oil is considered to be a commodity and as such a key determinant of a country’s balance of trade. This balance of trade is in turn one of the key determinants or a country’s current account (Baumeister and Kilian, 2016). Therefore, the importance of investigating and establishing the relationship between oil price shocks and economic growth needs no emphasis. The study is also relevant because it analyses the effects of oil shocks on the economic growth of a variety of countries with different economic structures and levels of diversification. This is commendable as it breaks with the tradition of most other studies that tend to focus on only one country or even on a set of countries with similar economic structures (or on oil producing countries in the developed world). Moreover, the focus on countries in the MENA region is commendable as most countries in this region are also major oil exporters.
Impact of Oil Price Fluctuations on Different Macroeconomic Aspects of OPEC Countries
The other study that seeks to establish the relationship between oil price fluctuations and economic growth is that by Aliyu (2009). The main aim of the study was to undertake an assessment of the impact of real exchange rate volatility and oil price shock on the real economic growth of Nigeria. It was found that there was unidirectional causality from oil prices to real GDP and bidirectional causality from real exchange rate to real GDP (and vice-versa). Results also indicated that oil price shock and appreciation in the level of exchange rate tend to have a positive impact on real economic growth (Aliyu, 2009).
The findings of Aliyu’s (2009) study are commendable especially because they are in agreement with findings from similar studies (such as Berument, Ceylan and Dogan, 2010; Gnimassoun, Joets and Razafindrabe, 2017, Sachs, Cooper and Fischer, 1981). Of greatest importance is the finding that when all other factors are held constant, increases in oil prices has a positive impact on economic growth in oil exporting countries but a negative impact in oil importing countries. This usually happens because when prices of oil increase, exporting countries earn more money from sales of oil while importing countries pay more for the same amount of oil imports (Baumeister and Kilian, 2016). The higher the prices of oil (which means more earnings from oil for oil exporting countries) the better the current accounts of these countries. On the contrary, the current accounts of oil importing countries decrease when oil prices increase (Baumeister and Kilian, 2016). It is also worth pointing out that the case study may be deemed inappropriate as Nigeria is not a member state of the GCC. However, Nigeria’s relevance as a case study is that it is one of the major oil exporting countries and also a member state of OPEC. Like all the other GCC states, Nigeria is a developing country and oil exports account for a very significant proportion of the country’s GDP. This essentially means that Nigeria’s economic structure is not very different from that of the GCC states.
Chuku et al. (2011) also use Nigeria as the case study in investigating oil price shocks and the dynamics of current account balances. The aim of their study was to ascertain the relationship between oil price shocks and current account balances. It was found that there were significant short-term impacts of oil price shocks on the current account of Nigeria with approximately 15.77% of variations in the current account being attributed to or caused by fluctuations in oil prices. However, they also found that there was no one-for-one relationship between oil price fluctuations and current account dynamics.
The relevance of Chuku et al.’s (2011) study lies more on its investigation of the impact of oil price fluctuations on the current account. This is important as this relationship is also at the core of the current study (albeit with a different case study). Chuku et al.’s study is also relevant because by using Nigeria as the case study, it is able to investigate the aforementioned relationship from a country that is both a net exporter and net importer of oil. After all, Nigeria is a leading exporter of crude oil (the 8th largest in the world) and a net importer of refined oil (it lacks refining infrastructure). The implication of this is that more information (on the impact of oil price shocks on current accounts) is likely to be learnt from such a case study than would have been possible by using a case study of a country that is just a net exporter or net importer of oil.
Other than that, the study by Chuku et al. (2011) deserves to be commended for specifying that even though oil price fluctuations have a significant impact on different dynamics of the current account, these impacts are only short-term. It follows, therefore, that whenever the impacts of oil price shocks on the current account are investigated, time has to be taken into consideration as time clearly mediates (or moderates) this relationship. This could be accounted for by the fact that in the long-term countries adopt measures necessary to improve their current account balances; and these measures inevitably counteract any effects that oil price shocks may have on the current account (Schubert, 2014). These findings are therefore in consistent with those by Schubert (2014) that following an increase in oil prices, the current accounts of small open economies usually exhibits the J-curve property whereby it first deteriorates before improving.
Interaction between Oil Prices and Economic Growth/Current Account
Just like Aliyu (2009), Ghalayini (2011) investigated the interaction between oil price and economic growth. However, Ghalayini uses as his case study all OPEC and G-7 countries as well as Russia, India and China unlike Aliyu who only focuses on Nigeria. The aim of the study was to investigate whether there existed any differences in oil price effects on economic growth between not just different countries (China, Russia, and India) but also groups of countries (OPEC and G-7). In essence, he sought to assess countries and groups of countries based on whether they are oil importing or oil exporting countries. Ghalayini hypothesized that in oil importing countries, increases in prices of oil had a negative impact on economic growth while for oil exporting countries increases in oil prices had a positive impact on economic growth (holding other factors constant). He found that for most countries, there was no proof of any interaction between changes in prices of oil and economic growth. The only exception was for the G-7 countries where there was proof of a unidirectional relation from prices of oil to economic growth (in terms of GDP).
The findings of Ghalayini’s (2011) study are important as they challenge conventional wisdom that in all circumstances oil price fluctuations affect economic growth on different ways. While it cannot be denied that oil is a major influencer of economic growth because of many sectors of the economy depend on oil and by extension oil prices (Cherif and Hasanov, 2013), changes in prices of oil may not necessarily affect economic growth both in exporting and importing countries. In exporting countries, increases in oil prices may not add to economic growth especially when the oil revenues are spend outside the country and not on economic development projects (Kilian, 2010). Similarly, it is possible for a decrease in oil prices not to enhance the economic growth of oil importers because the savings incurred due to lower oil prices are channelled to other forms of expenditure that may not have anything to do with economic development (Kilian, Rebucci and Spatafora, 2009). It follows, therefore, any relationship between oil price shocks and economic growth is country-specific and cannot be generalized across countries or groups of countries. This is no doubt a very important finding and one which scholars and policymakers need to pay a lot of attention to.
Allegret et al. (2014) seem to disagree with the arguments of Ghalayini (2011) and instead contend that it is almost inevitable for oil exporting countries to experience large improvements in their current accounts in the wake of sharp increases on oil prices. In a study whose aim was to investigate the relationship between oil price and the current account in 27 oil exporting countries, they argue that even though oil price variations positively affect current accounts the effect is not linear but is dependent on the degree of financial development for oil exporting countries. According to them, less developed countries usually experience a stronger impact on their current account as a result of variations in oil prices than more developed countries. They argue further that this impact tends to decline in tandem with financial depth of the country.
The relevance of Allegret et al.’s (2014) study lies mostly in its emphasis of the important and moderating role played by financial development in influencing the relationship between oil price and current account. While it cannot be denied that increases in oil prices means more revenue earnings for oil exporting countries, these oil windfalls’ actual effect on current accounts cannot be generalized a they depend on the country’s financial development (Morsy, 2012). Perhaps the GCC states best illustrate this. Those countries that are more financially developed tend to be less impacted by changes in oil prices compared to those countries that are less financially developed (Bems and de Carvalho Filho, 2011). This can in turn be explained not just in terms of the ability of the more financially developed countries to diversify their economies and therefore absorb oil price shocks better than less diversified economies but also in terms of achievement of stabilization and precautionary savings. Financial deepening helps to achieve precautionary savings through creation of a sovereign fund as well as through an increase in foreign reserves; and stabilization by doing away or minimizing borrowing constraints (McCombie and Thirlwall, 2016).
Overall, it may be justified to contend that Allegret et al.’s (2014) study breaks with traditional analyses of the oil price-current account nexus by focusing on the indirect role played by financial development. Such a non-traditional approach to investigating the oil price-current account relationship is very welcome as it adds new insights to the findings. The study is also worth commending for its use of a large sample of 27 oil exporting countries and over a long period of time (from 1980 to 2010). With such a large sample, it is easier to actually generalize the study’s findings to other case studies.
Conclusion and Gaps in the Body of Knowledge
From the study critical literature review undertaken in this chapter, it is evident that the subject of oil in general and oil price fluctuations in particular has attracted a lot of attention from scholars and key players in the oil industry as well (especially politicians and economists). It is also evident that the relationship between oil price fluctuations and different macroeconomic indicators – including the current account – has been a subject matter of immense scholarly practitioner interest. With oil arguably being one of the world’s most important commodities, its demand and supply – as well as the factors that affect its price – are understandably subjects of great importance. Moreover, the relationship between current accounts and oil prices of oil exporting countries (especially those that are member states of the OPEC and/or those found in major oil producing countries of the world) is a matter of great importance. This no doubt explains why it has been widely researched especially among scholars of economics.
In spite of the fact that many studies have been undertaken on the wider subject matter of the correlation between oil price fluctuations and current account, it is also noteworthy that most such studies have been to a very large extent focused on the economies of the Western world. This has left out most developing countries. Even more noteworthy is that the available empirical studies on the subject matter tend to focus on major oil exporting countries that are not necessarily member states of OPEC. Finally, the few studies that focus on the relationship between oil price fluctuations and current account hardly ever use GCC nations as their case study (with the notable exception of those GCC states that are also member countries of OPEC). Therefore, the current study fills a very important gap in knowledge by empirically investigating the impact of oil price fluctuations on the current accounts of GCC countries which provides a panel data evidence for the effect of oil prices on the current account.
In this chapter, a presentation of the research methodology pertaining to the subject matter under investigation is undertaken. The methodology aims to achieve the objective of this study, which is to measure the size of the impact of oil price fluctuation on the current accounts of the GCC. This is achieved by examining a secondary data with a quantitative technique.
The chapter is organized thematically and covers aspects such as research philosophy, Empirical model, research approach, data source, data transformation and data clean up.
3.2 Research philosophy
A realistic economic model incorporating all elements relevant to a typical country’s current account would be hopeless. In the current literature, intertemporal models are usually simulated due to the absence of closed-form solutions. Estimation, on the other hand, could measure on the context of GCC its responsiveness of the current account to oil price fluctuation.
The intertemporal approach on current account analysis extends to the absorption approach through the recognition that savings, investment, government spending results from forward looking that is calculated by accounting for macroeconomic determinants of relative current and future prices of oil, future productivity growth and government spending demand. As perceived earlier in this paper, The GCC countries external debts levels sore after the oil price shock significantly increased; that have led naturally to an intertemporally optimal current account deficit.
In order to pertain the intertemporal model to this papers empirical model, we have estimated rather than simulated an extension of (Glick and Rogoff, 1995 model. The model is similar to Sachs 1981, Obstfeld, 1986 and Frenkel and Razin, 1987, who have studied the intertemporal effect of government spending and productivity shocks. In their aggregated empirical model the current account is a function of the lagged investment level, a country-specific Total Factor Productivity shock (TFP), and a global TFP shock. In their findings, the country-specific TFP had a large effect on the current account than the global TFP. We extend this equation to estimate the effect of oil price shocks on the current accounts of the GCC. This study is consistent with (Chuku et al , 2011) which studied the short-term effect of oil price shocks on the current account. In their version of the model, it suggests that a country’s current account depends primarily on investments and productivity shocks. But since the GCC countries’ fiscal policy is heavily dependent on oil revenues, this empirical analysis will control for a fiscal policy shocks. Essentially, the current account fluctuation is a function of investment, TFP shock, fiscal shocks, and oil price shocks.
3.3 Empirical model
The approach by Glick and Rogoff can be extended as given in equation 3.3 for empirical analysis of the effect of fluctuations in oil prices on current accounts of GCC countries.
〖CA〗_it= 〖a_1 S〗_it+〖α_2 A ̇〗_it^C+α_3 A ̇_it^G+a_4 F_it+α_5 P_it^O+u_it (3.3)
ν_it= Normal random error term
i=1…6 (6 represents the GCC countries)
CA=Current account (% of nominal GDP)
S_it=Investment (% of nominal GDP) replaced by savings
A ̇_it^c= Country-specific productivity shock
(The dote presents growth rate and the superscript ‘C’ denotes as country-specific).
A ̇_it^G= Global productivity shock
(The dote presents growth rate and the superscript ‘G’ denotes as global-specific).
F_it= Fiscal policy
P_it^O= The real oil price
It is noted that among the GCC countries, Qatar was missing data on investment; therefore in this study, investment will be replaced by saving (% of nominal GDP). In the contextualizing of closed-economy models, savings are equal to investments. Although the GCC countries are not closed economies, it is noteworthy that the GCC countries have not claim for external debts before the 2014’s drop in oil prices. The GCC countries always maintained a current account surplus (i.e., savings > investments) the fiscal policy have financed investments through oil revenues. Therefore, replacing investments by savings is adequate.
The USA data have been chosen as a suitable proxy for global shocks due to the fact that the US is a major oil trade partner with the GCC.
3.4 Defining of shocks
On the context of the GCC, the current account and savings are determinant variables, on the other hand, the fiscal policy (government expenditure), oil price and total factor productivity are rational that are defined as shocks:
Fiscal policy shocks (F), real oil price shocks (P^O), country-specific shock (A ̇_it^c) and global TFP shock ((A) ̇_it^G).
These shocks are random shocks derived as follow:
〖log F〗_t=logF_(t-1)+Ω_t (3.4)
〖log P^O〗_t=log〖P^O〗_(t-1)+ϵ_t (3.5)
Where: Ω_t& ϵ_t= random normal (iid) shock
Equation (3.4) (3.5) illustrates that the variables of fiscal policy shock and oil price shocks are stationary by the treatment of differencing. Note that the global and country productivity shock data have been collected as growth rates.
3.5 Research approaches
The anticipation as stated in the introduction, the originality of this paper consists in the adoption of panel data econometrics analysis. The panel data namely as longitudinal data is a marriage of regressions and time series analysis, as with many regression sets allows this study to observe the group GCC countries overtime (Frees, E. W. 2004).
The data is classified as pooled time series and cross sectional. The empirical study will estimate the coefficients in the theoretical model (3.3) both fixed effect model and random effect model, we will choose the most suitable one on the basis of the Hausman test (Hausman, J. 1978), which checks whether the fixed effect or the random effect model estimators return similar results.
Further approach will be devoted to examine the normality of the residuals of the model, by the The skewness-kurtosis test (S-K test) of normality of the residual. In addition, the (Pesaran, M.H., 2004) will be tested in order to capture whether the residuals of the model are correlated ( cross-section dependence of residuals).
The beneficial of the panel data underlies in this study particularly is that the GCC countries level time series data are relatively short. The panel data pooled time series and cross-sectional allows the GCC grouped data to be observed overtime, which creates a larger T, hence larger observation.
Another benefit of using a panel data is that it solves the problem of omitted variable problems, allowing unobserved heterogeneity to be controlled for. For example taking a panel approach is desirable for estimation, otherwise, an inequality value in cross-section data would be associated with impact of the current account for different reasons rather than being caused by oil plunging alone. Such as unobserved country characteristics such as the size, political aspects of each country could explain inequality levels, nonetheless, panel approach will solve this issue.
3.6 Data sources
The panel data consists of N=6 (GCC countries) and T= 1995-2016.
The GCC data are taken from the IMF World Economic Outlook, April 2017 (International Monetary Fund, 2017). The data are annual time series from (1980-2016) and include: The current account (% of nominal GDP), total investment (% of nominal GDP), gross national saving (% of GDP), government expenditure, oil price measured in U.S. dollar (simple average of three spot prices; Dated Brent, West Texas Intermediate, and the Dubai Fateh, US$ per barrel (nominal)). Total factor productivity growth rate (TFP) is from the Conference Board (The conference Board, 2017).
Definition and Measurement of Variables
3.7 Data transformation
The nominal oil price is converted to real variable as follow:
Real oil price=nominal oil price/US deflator, where the year 1990 is the base line.
3.8 Data clean up
The times series analysis of the current accounts for all the GCC countries in (International Monetary Fund, 2017) run from 1980-2016. However, fiscal policy, TFP US, TFP GCC are only available from 1990. The fiscal policy is substantial variable to explain the current account of the GCCs; therefore data cannot be commenced from 1980 because it will bias the results by excluding fiscal policy. Therefore, the data will commence according to the alignment of the availability of all the data.
The study issue of endogeneity arises due to the single equation bias, this is due to the fact that the explanatory variables assume exogenous while in fact they are not. Such as savings is an endogenous decision; because people use a lot of information to decide whether how much to save and how much to consume, however, fiscal shocks are exogenous, because no one has any control over them and are unpredictable, The same is true for Total Factor Productivity and oil price shocks are considerably random, highly unpredictable and are exogenous to everyone.
Under this chapter, it presents a critical analysis of the data relating to the objective that is to testify the contextualizing impact of oil price fluctuation on the current account of the GCC that guided the empirical study. The results of the data were exploiting through the Stata soft ware version 14.
The sections under this chapter are organized thematically in order to obtain a substantial finding to fulfill the objective of the empirical study. Firstly, data plotting are illustrated of the variables under investigation, a descriptive statistics that presents a summary on the variables of interest, following a stationarity analysis test to identify a presence of unit root if any, lastly, the model specifications.
4.2 Data plot
An illustration of the variables features can be explored by plotting the data. Under this section three graphs are created of the GCC countries of the period 1990-2016 as shown below.
Figure (1.5) displays the GCC variables of current account, savings and real oil price of all the GCC countries. Firstly, it can be noted that the variables current account and savings have had a significantly large drop for Kuwait in 1990-1991. The observed outlier is attributed due to the Gulf War I 1990-1991. Moreover, Qatar appeared to deviate away from the mean in 1990-1998, the possibility of the divergent could be attributed to the period where oil exports had decreased. However, Qatar had recovered ever since by exporting gas with top ranking as illustrated in the introduction.
For the other GCC the variables appear to be stationary with no obvious trend.
Log (G1) log fiscal policy shock 1 lag, TFP (Country specific Total factor productivity)
Figure (1.6) displays the GCC variables of the fiscal policy shock and country specific total factor productivity. Similarly to figure (1.5) it can be seen that Kuwait shows a significantly large drop in the fiscal policy and a large increase in county’s total factor productivity during the period of the Gulf War I. the other GCC countries appears to be stationary with no obvious.
4.4 Pre-estimation Tests
Unit root test
The panel data was tested to identify any non-stationarity variables, the variables CA, S were investigated using a panel unit root test of Levin, Lu & Chu (Levin et al., 2002) and Breitung (Breitung, 2001).
Although the previous data plotting have not captured and obvious trend, the unit root test will test two specifications: with trend and without trend, in order to mediate any issue regarding the observed outlier of Kuwait that had a significant downward drop in the data during the Gulf War I.
The other variables however would not be tested for presence of a unit root. This due to the fact that he shocks are defined stationary by construction (refer to page 31 defining shocks). The fiscal policy shock and the oil price shock are measured as log-differenced variable, hence, they are differenced-stationary. TFP shocks are taken from the Conference Board as TFP growth rate, which are also stationary.
4.4 Panel Unit root test
The Levin, Lu & Chu and Breitung tests assume that there is a common unit root process across the panel, i.e., ρ_i is identical across the cross-section.
y_it=ρ_i y_it+X_it δ_i+ϵ_it (1)
Where: i= 6
X_it= Exogenous variables
ρ_i= Autoregressive coefficient
ϵ_it= Error term (assumed to be mutually independently idiosyncratic disturbance)
|ρ|< 1 weakly (trend) stationary
|ρ|= 1 unit root
According to table 4.0, under the Levin Lu and chu test, the p-values of the variables CA and S indicates their ρ is statistically significant at 5% level of significance on the circumstance that trend is not included. Therefore, we can reject the null hypothesis of the presence of a unit root. However, savings (s) appears to be insignificant at the specification of trend.
Table 4.1 displays the unit root test under the Breitung test for the variables CA and S. The P-values indicate that ρ is statistically significant at the 5% level of significance when trend have not been included. Therefore, we can reject the null hypothesis of the presence of a unit root under the specification of without trend. However, savings (s) appears to be insignificant at the specification of trend.
All and all, the variables appears to be stationary and concludes as observed in the plotted data that their wasn’t an obvious trend in the graphs.
4.5 The model regressions
In practice, panel data are estimated using pooled Ordinary least square (OLS) the estimators, fixed effect and random effect model. The FE and the RE are designed to remove omitted variable bias by measuring changes within the group. By measuring within group (across time) it controls for potential omitted variables unique to the group.
The key difference between FE and RE is that the FE assumes the individual specific effect is correlated to the independent variable. In contrast, the RE assumes the individual specific effect is uncorrelated to the independent variable.
The fixed effect model (FE)
〖CA〗_it= 〖a_1 S〗_it+〖α_2 A ̇〗_it^C+α_3 A ̇_it^G+a_4 F_it+α_5 P_it^O+u_it
The random effect model (RE)
〖CA〗_it= 〖a_1 S〗_it+〖α_2 A ̇〗_it^C+α_3 A ̇_it^G+a_4 F_it+α_5 P_it^O+ϵ_it+u_it
In order to select between the fixed effect and random effect that is most suitable for the empirical data, the Hausman test is conducted. The Hausman test is designed to detect the violation of the RE assumption that the explanatory variables orthogonal to the unit effects. If there wasn’t a detection of a correlation between the independent variables and the unit effect, then the estimates of β in the fixed effect model ((β_FE ) ̂) should be similar to the estimates of the random effect ((β_RE ) ̂).
The Hausman test:
((β_RE ) ̂-(β_FE ) ̂) [β_FE- β_RE]^(-1)((β_RE ) ̂-(β_FE ) ̂)
Under the null hypothesis of orthogonality, the test statistics (H) measures the difference between the two estimates, where the (H) is distributed chi-square with the degrees of freedom equal to the number of the regressors in the model which in this case (4).
According to figure 4.3 which presents a summary of the Hausman test, we failed to reject the null hypothesis that there is no difference between the random effect and fixed affect against the alternative that is fixed effect, with (p-value= 0.999), this conveys that both fixed and random models are consistent with the model and both of the coefficients are similar. The coefficients presented in figure 4.2 also provide additional support, with minimal difference. Under the ground of this test, this paper will be reporting the both the FE and the RE.
Note: rounded to 3 decimals places, statistically significant at 1%*** 5%** 10%*, Δ log_G1 is lagged once, (A ̇^C-A ̇^G) in order to reduce degrees of freedom, DP=real oil price*dummy variables for each GCC countries, standard error are robust to control for heteroskedasticity.
Table 4.8 presents the fixed effect model and table 4.9 presents the random effect model.
Each table comprises of four equations.
Regression (1) we began with a regression of current account on savings and TFX shocks, The variable TFX is the difference between the countries’ minus global total factor productivity shock (A ̇^C-A ̇^G), the reason of doing so instead of separately is having to save degrees-of-freedom by estimating fewer coefficient (one instead of two).
Regression (2) in the second regression we added fiscal policy shocks as an additional regressor.
Regression (3) The third column we added the oil price shock.
Regression (4) The last column we used a dummy variable (3.6) to capture the effect of oil price shocks on each country individually.
〖CA〗_it= 〖a_1 S〗_it+〖α_2 A ̇〗_it^C+α_3 A ̇_it^G+a_4 F_it+α_5 P_it^OB×dum1+P_it^OK×dum2+〖P_it^OO 〖×dum3+P〗_it^OQ×dum4〖+P〗_it^OS×dum5〖+P〗_it^OU×dum6+u〗_(it ) (3.6)
dum=real oil price×dummy variable for each GCC country (3.7)
presents the value of current account estimated from regression (4) and the residual line presents the difference between the actual and fitted line. It could be concluded that the model fits the data certainly well and for every country. The fitted data and the actual data seem to be highly correlated.
Figure 1.11 illustrates an approximation of normal distribution of the residual. The space cells represent the data, which appears to be plotted each cell on the cell frequency versus the center of the data. The residuals appear to follow a normal distribution by approximation.