THE DUTCH DISEASE IN KUWAIT: THEORY AND EMPIRICAL STUDY

1.1       BACKGROUND

The thesis that abundant resources can translate to a curse rather than a blessing has long been held in economics from as early as the 16^th century mercantilist era especially with the experience of Spain which suffered economically destabilizing consequences as a result of an influx of gold from the newly discovered lands of the Americas (Benjamin, Devarajan and Weiner, 1989). In more recent times, economists have observed that most of the fastest growing economies of the post-World War II era were largely resource poor while resource abundant countries suffered virtual economic collapse. As observed by Auty (2001), “between 1960 and 1990, the generality of observed countries with scarce natural resources grew at least two to three times faster than countries with abundant resources”. Examples are often cited of the so called East Asian tigers – countries with very little natural resources, which serve as the best examples of countries in this category having averaged growth rates of 4 to 5% per annum between 1970 and 2000. By contrast, GDP per capita in resource abundant countries like Venezuela, Zambia and Sierra Leone, shrank by between 1.3 and 1.9% per year despite having large reserves of oil, copper and diamonds, over same period (See Table 1.1). This narrative often leads to the presence of a so called “Resource Curse” among resource abundant countries.

 

Table 1.1 Relationship between Resource Abundance and Economic Growth Rates

	Average annual growth rate 1970-2000 (%)	Number of countries
Resource exporters	0.955	56
Oil, Mineral Exporters	0.758	40
Agricultural raw materials, Food Exporters	1.448	16
Non-resource exporters (Manufactures)	3.738	8

Source: Gylfason (2001)

The resource curse theory goes contrary to another widely agreed strand of economic theory which maintains that resources contribute to growth rather than hinder it. Three growth channels through which resources cause economic growth are often cited for this theory. Firstly, revenue earned from resources either through direct sales or a marked appreciation in global prices can be reinvested in other sectors of the economy to promote growth. In the spirit of the so called “big push” models (Misra and Puri, 2010), increases in aggregate demand following a surge in revenue as a result of resource exports can act as an economic stimulus that helps countries push themselves out of poverty traps and move towards economic growth and development. Secondly, revenues from trading natural resource may help relieve development constraints that arise as a result of growth demands exceeding the capacity of domestic economies as captured by the “gap models” Humphrey (1970). In the two-gap model for example, growth is constrained by insufficient savings to finance domestic investment thereby creating the “savings gap” while the misalignment of export revenues and import expenditures may create financing gaps that constrain domestic economic growth. The three-gap model goes on to introduce a fiscal gap that arises as a result of insufficient fiscal earnings. Earnings from natural resource sales can bridge these gaps by providing an additional source of savings, foreign exchange, and revenue to ease fiscal constraints.

Natural resources may also serve as sources of useful inputs and raw materials for domestic industries that stimulate local industrial production. Economic history is replete with accounts of nations that turned the corner as a result of the direct influence of resource endowments while a number of countries (e.g., Botswana with its rich diamond fields, Chile with copper and Norway with vast petroleum resources) continue to enjoy sustained economic growth as a result of their natural resource endowments. Finally, natural resources provide an input for domestic industry that other countries may lack, and throughout history, resources have contributed to successful industrialization and development as seen in Western Europe, the U.S. and Australia.

While the resource curse thesis is therefore not uncontroversial, what is clear from our introductory  review is that while natural resources can be a welcome source of economic growth and development, it could also be a ‘poisoned chalice’, directly responsible for economic backwardness and retrogression. This confounding paradox has left economists searching for answers with regards the underlying factors responsible for the varying experiences of countries and the mechanisms through which resource endowments could either encourage or impede economic growth. Four key factors (channels) have been pointed to by economist as being responsible for the fortunes of resource abundant countries. These are; decline and volatility in terms of trade (Prebisch, 1950 and Singer, 1950), the Dutch disease (Corden and Neary, 1982), Political and economic institutions (North, 1990; Rodrik, Subramanian, and Trebbi, 2002) as well as Education, Health and other social variables (Gylfason, 2001;  Gylfason, Herbertsson and Zoega, 1999). While terms of trade volatility is a largely exogenous variable with little domestic control, the other variables (Dutch disease effect, political and economic institutions and socio-political conditions) are mostly endogenous and could be appropriately investigated in an individual country’s context. In this light, this study sets out to investigate the presence or otherwise of the resource curse phenomena in the Gulf state of Kuwait.

While the evidence of a resource curse is obvious in a number of countries, it is not so obvious in others and requires some level of investigation to be determined. The Gulf state of Kuwait has over the years tried to shield itself from the deleterious effects of oil exports by establishing one of the richest sovereign wealth funds under the Kuwait Investment Agency (KIA) with funds reaching over $500 billion. However, the country continues to be overly reliant on the commodity as Oil accounts for about 60% of her GDP and 95% of her export earnings (OPEC, 2016) thereby making her a candidate for the undesirable resource curse effect. This study will examine if Kuwait has been a victim of its own blessing by determining if it’s oil wealth has subjected the country to the theorized negative consequences ie; the Dutch disease, deterioration in her political and economic institutions as well as negative social consequences, all subsumed as an enquiry into the presence or otherwise of the Resource Curse in Kuwait.

This research paper will consist of four sections with the first being a general introduction to the research work, the second will be a comprehensive literature review comprising an examination of the core issues surrounding the resource curse in general and the Dutch disease effect in particular as well as a sectoral analysis of relevant sectors of the Kuwaiti economy. The third section consists of our methodology and regression analysis while our final section will cover our summary and conclusions.

 

 

LITERATURE REVIEW

2.1       Kuwait Economy and Sectoral Growth

Kuwait, a country with a total land area of just under 18,000 km² lies at the head of the Persian Gulf and is bordered in the North (North West) by Iraq, South (South East) by Saudi Arabia while overlooking the Persian Gulf to the East. With a population put at 4.4million (Trading Economics, 2016) comprising mostly of foreign immigrants (70%), the country’s economy has evolved from the days of being a modest economy sustained by subsistence fishing and rather large-scale pearl gathering, to become one of the top oil producers in the world. With proven reserves in excess of 98 billion barrels (about 10% of global proven reserves), oil today accounts for over 50% of the country’s GDP, 80% of her foreign exchange earnings and 95% of her exports (World Bank, 2016) buttressing the critical role of the resource to the country.

While Kuwait’s oil sector has flourished over the years, other sectors have not fared as great. For a country not known for its agricultural exports given its lack of arable land and fresh water, its economic progress from the early days of large scale pearl gathering which had sustained the economy has been marked. Paradoxically, while the oil sector has grown in leaps and bounds, other sectors have remained relatively comatose. Agriculture for instance now accounts for just 1% of the country’s economy and 8% of GDP translating to an annual import of 96% of its food (Nations Encyclopedia, 2016). What is clear from a casual observation of employment figures between sectors is that increased revenues from oil sales has drawn workers away from historical pursuits  with less than 10,000 people employed in the agric sector in the early 1990’s (Country studies, 2016). While the government has made efforts to boost agriculture including the adoption of the hydroponic system, widespread pollution of arable land caused by retreating Iraqi forces during the costly Gulf war in the early 1990’s further hampered these efforts. The result is limited agricultural capacity while fishing continues to provide a minor but important source of economic value.

Manufacturing Sector

On the manufacturing front, for a country whose economy was sustained by pearl gathering before the discovery of oil, appreciable progress has been made in the industrial sector even though the country still lags behind its Gulf country peers. According to the IMF, Kuwait’s manufacturing industry contributed 5.33% to the country’s GDP in 2010 with activities in the sector dominated by the overall health of the economy; slowing in times of slow economic growth and expanding in periods of economic expansion. The manufacturing sector is dominated by building manufacturing companies which constitute 47% of the total sector with construction and chemical sector following at 16% (Capital standards, 2016).

Source: Capital Standards (2016)

Overall, while the Kuwaiti economy has expanded over the years, its journey like that of most nations has not been that of a steady increase as major setbacks have been experienced along the way. Aside the usual vagaries of international oil prices and its concomitant negative impact on price dependent countries like Kuwait, the country suffered a devastating invasion from neighbours Iraq in 1991 that saw losses amounting to $25 billion and took the better part of three years to recover from. The country’s economic growth as captured by its GDP, has fared considerably well over the years averaging 3.65% from 1965 until 2015, reaching an all time high of 33.99% in 1993 and a record low of -20.62 percent in 1980 (Trading Economics, 2016).

Fig 1.2: Kuwait: Distribution of Employment by Economic Sector for Selected Years

Source: World Bank, 2016

 

Health Sector:

Kuwait is home to one of the more advanced and modern healthcare systems in the Middle East even though considerable challenges remain. The Oxford Business Group (OBG) in its 2015 Kuwait country report noted that the country’s health sector continued to be shaped by the demographic composition of its population arguing that while birth rates have been relatively high, fertility rate has continued to fall, dropping from an estimated 4.8 children per woman in 1983 to 2.6 per woman in 2013. The country’s mortality rate has also declined considerably owing to better healthcare and overall progress which has resulted in a demographic shift that is seeing an increase in the ageing population – a growing pattern across the GCC (Gulf Cooperation Countries). Kuwait follows the general population demographic pattern of the GCC, where the proportion of those aged 65 and over is due to increase from 2% in 2015 to 20% by 2050 according to figures from Alpen Capital. Oxford Business Group’s country health report on Kuwait for 2015 noted that while the State’s medical system has expanded hugely over the past fifty years, many Kuwaiti’s continue to seek specialized care outside the country especially for less common conditions (OBG, 2016).

While healthcare costs have risen considerably, the bulk of these costs continue to be borne by the government with private healthcare accounting for only about a fifth of total healthcare expenditures in the country. When compared to its GCC neighbours, the Kuwaiti government spends considerably more on healthcare with its expenditure as a proportion of GDP rising steadily over the years from 2.2% in 2006 to reach 3% of GDP in 2015 translating to an average per capita expenditure of $1,507 in 2013 ($2,320 in 2015) which is considerably higher than the GCC average of $1,279 (World Bank, 2016). Kuwait’s prosperous economy is also altering the nature of health care in the country with State sponsored medical services expanding considerably over the past fifty years. According to a forecast by Alpen Capital, the country’s healthcare market is set to rise in value from $39.4bn to $69.4bn between 2013 and 2018. In light of this anticipated expansion and growing need for specialized care given a marked rise in non communicable diseases and a growing ageing population, there’s a desperate need for health personnel with Kuwait counting 2.7 physicians per 1000 heads of population in 2012, according to the latest figures available from the World Bank. In 2012, Kuwait had a total of 47,380 medical staff, of which 21,853 were Kuwaiti, or roughly, just 46% of the total.

Education Sector

Education delivery is seen as an obligation of government in Kuwait and over the years, the government has tried to match this expectation. The government as at 2015 accounted for about 62% of schools and religious educational institutions in the country with private institutions accounting for about 37.8% (Global, 2017). Schooling is mandatory for all children aged 6 – 14 in Kuwait with free primary, secondary and tertiary education available for nationals in public schools. The country has over the years, raised its Cumulative Annual Growth Rate (CAGR) contribution to the education sector to 14.8% between 2010 and 2014 and by 2015, had apportioned the equivalent of 15% of its total budget to education (World Bank, 2016).

While the increase in funding underlines the intent of the country’s government, the country still lags considerably behind when compared to its GCC (Gulf Cooperation Council) peers which average 17.5% (Global, 2017). While it could be argued that Kuwait does not have to spend as much as its GCC contemporaries given that its education expenditure works out to an average of USD14,300 per student (due to its lower population) as against the GCC regional average of USD 11,000 per student, observed challenges go beyond monetary expenditures. According to World Bank data, Kuwait’s youth literacy rate (population between 15 – 24 years) was 99.5% in 2015, indicating a decline from a peak of 99.7% in 2005 with the proportion of literate female youths put at 99.4% in 2015 also indicating a decline from a 2005 peak of 99.8%. The country’s adult literacy rate which represents those aged 15 years and above, stands at 95.2% for females and 96.8% for male.

A key indicator of the state of education in Kuwait is the World Economic Forum’s (WEF) Global Competitiveness Report for 2016 – 2017. The report among other metrics, ranked countries in terms of the quality of their education system and while the likes of Qatar, The UAE and Bahrain were ranked considerably high among the GCC countries, Kuwait and Oman were lowly ranked as indicated in the table below.

 

Table 1.3        Education Assessment in the GCC Region

Parameters/Countries	Bahrain	Kuwait	Qatar	Oman	KSA	UAE
Quality of primary education	36	103	6	78	64	12
Quality of the education system(higher education and training)	25	86	5	80	48	10
Quality of math and science education	29	105	5	89	65	10
Quality of management schools	34	92	5	119	55	16
Internet access in schools	39	91	11	84	65	4
Availability of specialized training services	35	113	18	112	76	22
Extent of staff training	26	80	8	49	57	16

Source: WEF Competitiveness Report 2016–17

Tertiary Education

Tertiary education enrollment in Kuwait remains considerably low with enrollment figures put at just 27%. The country lags well behind the likes of regional powerhouse Saudi Arabia which has a 61% enrollment rate and relatively poor Bahrain with a 36.8% rate. Global (2017) notes that the Kuwait university is the only tertiary institution running at full capacity in the country and in the secondary education sector, there’s still considerable room for growth with current enrollment rates put at 93.6% compared to the GCC average of between 101 – 110% recorded in Oman, Qatar and Saudi Arabia.

Institutional Issues

Kuwait has the oldest elected parliament of all Arab Gulf countries; its National Assembly was

established in 1963. The parliament according to the CIA World Factbook, is hardly a mere “rubber-stamp” body, and openly discusses vital issues – a rarity in the region. In particular, its parliament continues to press for more democracy and transparency as regards the State’s financial matters, a greater role for itself in defining oil policies, and more parliamentary oversight (BTI, 2016). While the legal framework governing Kuwaiti elections are largely in line with international standards, vote inequality remains a major flaw with the lack of political party participation in the country’s political process posing considerable challenges. The limited powers of Kuwait’s parliament is also inconsistent with international best practice and statutes. BTI (2016)  notes that limitations exist to voting rights for  soldiers  and  naturalized citizens who are barred from taking part in the electoral process for at least 20 years after naturalization.

Judiciary and Property Rights

Although Kuwaiti law provides for an independent judiciary, over the years, the Judicial arm has continued to experience sustained pressure from the executive (BTI 2016). GCR (2016 ) reports that bribes and irregular payments have been allegedly exchanged to influence court rulings with the country’s opposition publishing damaging records of illicit financial transfers between politicians and  judicial figures (Alakhbar, 2014). The country’s dispute resolution and court processes also have a reputation for being excessively long and time-consuming, with enforcement of court rulings problematic. Kuwaiti law adequately defines and protects property rights, but problems persist with the protection and enforcement of these rights and laws (BTI 2016,). While citizens enjoy full access to lands and properties, non–Gulf Cooperation Council citizens’ are restricted from owning land in Kuwait. Despite obvious setbacks, deliberate efforts continue to strengthen property laws and the judicial process with regular and obligatory anti corruption and transparency trainings (USDS, 2015).

Taxation and Social Security

Kuwait’s relatively small population and high oil earnings means the country can afford an elaborate welfare package for its citizens with little need to impose any financial obligations on them. The Kuwait government disregards taxation as a source of revenue, with little to no tax burden imposed on citizens and limited rates for foreign companies (BTI 2016). Currently, residents pay no income taxes while sales taxes for goods and services are non-existent. Foreign investors under new laws can enjoy up to 10 years of tax exemptions under Kuwait’s new Investment Law (ICS 2015).

In terms of social security, citizens enjoy extensive state help covering medical and maternity cover, child care, pensions, unemployment benefit and in some instances housing and disability benefits (KSSS, 2016).

 

Corruption

The rapid rise in oil prices and the accompanying oil boom is believed to have fueled corruption in Kuwait with the Kuwaiti parliament often fingered as a source of electoral and institutional corruption (Global Security, 2013). Government contracts and contract processes are notorious for being dotted with corrupt activities with most contracts containing some level of corruption Global security (2013) observes. While Kuwait’s judges are not generally perceived to be corrupt, corruption is widespread in the bureaucratic processes of the judiciary while governments lengthy procurement process often results in accusations of bribery and other corrupt activities (Global security, 2013). As an indicator of the state of corruption in the country, Transparency International in its 2011 Corruption Perceptions Index (CPI) ranked Kuwait 54 out of 180 countries (75^th position in 2016) with the country placed second to last among the GCC countries (Transparency International, 2016). Kuwait’s persistently high Corruption Perception Index Score ranging between 4.6 (4.1 in 2016) and 5.0 (out of 10) indicates the country has a worrisome corruption level.

 

On the whole, Kuwaitis are generally proud of their level of political participation ina aregion not known for political freedom. The country’s political structure stands out among its peers and a large portion of its oil wealth continues to be used to fund a comprehensive welfare state with extensive education, health facilities as well as infrastructural projects. Though severe challenges remain in terms of implementing institutional reforms and the long term sustainability of the broader economy especially in the face of declining oil prices, there’s no doubt appreciable progress has been made in terms of human capacity development and economic advancement from an obscure pearling sustained economy to one of the most developed countries in the region.

2.3       THE DUTCH DISEASE LITERATURE

At its narrowest interpretation, observes Ahmed, Bucifal, and George (2008), “Resource Curse” is used interchangeably with the phrase “Dutch Disease”, but Davis (1995) observes that the Dutch disease and the Resource Curse thesis are two distinct issues, though most times thought to be synonymous. Davis (1995) describes the Dutch disease as a morbid term that denotes the existence of both booming and lagging sectors in an economy due to temporary or sustained increases in export income. The Dutch Disease primarily refers to a condition in which a booming single export sector acts to increase the relative price of non-tradable goods and services, in effect, impacting the tradable goods sector negatively (Stijns, 2003).

The term ‘Dutch Disease’ was made popular by the unpleasant experience of the Netherland economy in the 60’s and early 70s. The name arose from the effects arising from the discovery of North Sea gas on the Dutch manufacturing sector which led to the shrinking of the country’s tradables sector with increased production from her newly discovered Groningen oil fields. With natural gas, the country’s main export during the period, experiencing increases in prices leading to large inflows of foreign exchange into the country, the result was a rise in domestic money supply and concomitant inflation that led to higher prices for Dutch imports. For a relatively open economy like the Dutch had, high import prices translated to higher costs of production which ultimately led to stagnation in aggregate supply and large scale unemployment (Smirnova and Kulkani, 2013)

Mineral exporting countries in the following decades have appeared to generate the ideal environment for the disease, with a single prominent mineral sector in the face of other relatively declining sectors. In the core theory as observed by Stijns (2003), the mining sector booms while both the manufacturing and agricultural sectors shrink. Essentially, the Dutch disease results in short-term deindustrialization of the economy with long term implications.

 

2.3.1    Dutch Disease Models

The classic Dutch Disease models as observed by Amuzegar (2001), embraces the neo-classical assumptions of wage-price flexibility, internal mobility of labour, initial general equilibrium, free entry and exit, etc. We examine two prominent Dutch Disease models in this section.

 

The first of our two models was proposed by Sachs and Warner (1995) and comprised three sectors i.e., two Tradable sectors comprised of both a natural resource and non-natural resource (manufacturing) sectors and a Non-traded sector. In their model, while labour and capital are consumed in the traded as well as the manufacturing sectors, they are not used in the natural resource sector. With a booming natural resource sector, demand for non-traded goods increases leading to lesser and lesser allocations to the manufacturing sector. The model holds that with a booming natural resource sector, production activities increasingly shift from the manufacturing sector to the natural resource sector with labour and capital that ordinarily would be drawn to the manufacturing sector concentrated on the non-traded sectors. The result is a shrinking of the country’s manufacturing sector while the non-traded sector expands (Sachs and Warner, 1995).

 

Although Meade and Rusell (1957) were credited with the first paper on the resource boom paradox, what became known as the core model of the Dutch Disease theory was developed by Corden and Neary (1982). Their model (our second model) like that by Sachs and Warner (1995), assumed an economy divided into three sectors; the booming export sector, the lagging export sector both of which are the traded goods sectors; and the non-traded goods sector that may be Services, existing in a small and open economy which produces three goods. According to Stijns (2003), while two of the goods are traded at exogenously determined prices, the third, a non-traded good, has its price determined by the forces of demand and supply. One of the two traded goods enjoys a boom while the other does not. Of the two goods, the non-traded good is produced by the service sector (though it could be extended to other sectors e.g. building and construction, agriculture etc.). While capital is sector specific in the main model, labor is assumed to be mobile with a boom in resource affecting the economy in two ways: i.e.

Resource movement effect:  This occurs firstly, on the supply side, when an exogenous increase in the value of output from the booming sector increases marginal product of labor in the sector causing a shift of labor to that sector from other sectors. What follows is a concomitant contraction in the tradable sector as a result of a reduction in the consumption of production factors as well as an increase in the price of non-traded goods due to a resultant excess demand. The rise in price of non-tradables translates to an appreciation of the real exchange rate. On the demand side, the boom, leads to increased income at home and hence increased demand for all goods.

Corden and Neary (1982) further explains this effect as arising due to the perfect mobility of labour and capital feature of the economy which allows for the movement of these factors of production from the manufacturing sector to the booming and service sectors with relative ease. According to Benkhodja (2011: 2), this effect occurs because “an increase in oil prices generates a rise in wages and/or profits thereby triggering a rise in aggregate demand in the economy.” Since a considerable portion of this demand would accrue to the service sector, a rise in the price of non-tradable goods would follow causing the real exchange rate to appreciate with the result being some degree of definite deindustrialization concludes Benkhodia (2011).

 

Spending Effect: The spending effect arises as a result of the price of tradables being exogenously determined. As a result, extra spending raises the relative price of non-tradables causing further appreciation of the real exchange rate and a shift in labour from the tradable to the non-tradable sector and a resultant contraction in the tradables sector.

 

Since each sector contains one mobile factor (labour) as well as a non-mobile factor (capital), the resource movement effect and the spending effect both combine to cause a movement of labour away from the manufacturing sector, leading to declines in manufacturing output. On the other hand, output in the booming sector increases as it absorbs factors of production from other sectors.  Corden and Neary (1982) identified two forms of deindustrialization as a result of the movement of labour between the tradable and non-tradables sector. They referred to the movement of labour from the manufacturing sector to the booming sector as “direct deindustrialization” while according to the researchers, a combination of the flow of labour out of the non-tradable sector and increased demand for goods from same sector as a result of the ‘spending effect’, would lead to further movement of labor from the manufacturing sector to the non-tradable sector or what they referred to as “indirect deindustrialization”.

Stijns (2003) highlights 4 major impacts of a combination of the spending and resource movement effects:

1. An unambiguous appreciation of the real exchange rate
2. A theoretically ambiguous increase in non-traded output

• An unambiguous fall in production in the manufacturing sector and

1. A fall in manufacturing exports.

These 4 highlighted impacts listed by Stijns (2003) constitute the key channels through which the Dutch Disease affects economies in the classic Dutch disease model. A combination of these effects in countries where they manifest ensures that instead of reaping the rewards of a booming mineral sector, affected countries end up suffering severe economic and socioeconomic consequences as a result of a natural resource boom. Considering that the tradable and non-tradable sectors represent the manufacturing and service sectors respectively, Altamirano (1999) asserts that the classic Dutch disease model indicates a definite de-industrialization effect after a natural resource boom. This ‘shrinking’ of the manufacturing sector is what triggers the ‘Disease’.

Mahmudov (2002) however draws attention to the argument by Sachs and Warner that the shrinking of the manufacturing sector does not in itself constitute a harmful effect if neoclassical competitive conditions prevail in the economy. Mahmudov (2002) identified two conditions where the manufacturing sector possesses special features under which the Dutch Disease could constitute a real disease that causes a chronic condition of retarded economic growth. These are:

1. The existence of backward and forward linkages especially if such linkages constitute positive production externalities.
2. Where the manufacturing sector revolves around ‘learning by doing’ (an economic concept through which productivity is achieved through practice, continual self-perfection and minor innovations). Yang and Borland (1991) showed the concept (learning by doing) plays an important role in the evolution of countries to greater levels of specialisation in production thereby acting as an important engine for long-run growth. According to Mahmudov (2002: 15) “if most economic growth is caused by learning by doing which is mainly the case in the non-oil tradable sector, a decline in that sector would lead to lower productivity and therefore to lower national income in the future”.

In a nutshell, Corden and Neary showed that the traditional tradable sector (manufacturing or agriculture) is in effect, ‘crowded out’ by the other two sectors as a result of an appreciation of the domestic currency’s real effective exchange rate (REER) which renders the traditional exports less competitive and attractive to importers.

 

The effects of the Dutch Disease identified by various researchers include;

1. A recession in the traditional sectors (agricultural, manufacturing etc.) or what Stijns (2003) refers to as “de-industrialisation”
2. Rise in domestic inflation
3. Increase in Wage rate
4. A transfer of resources from the booming sector at the collapse of the boom
5. The overvaluation of the local currency that acts to impede the export of non-primary products, total exports and in effect, leads to unfavourable terms of trade.
6. Appreciation of the Real Exchange Rate
7. Heavy dependence on natural resource and primary goods exports which triggers the Dutch Disease in times of export booms and leads to the overvaluation of the domestic currency.

While economists largely accept the existence of the concept of the Dutch Disease, resource rich countries have proven to be more diverse than what the theory actually predicts with sometimes contradicting results. According to Mahmudov (2002), while countries like Nigeria, Congo, Venezuela and Sierra Leone fit into the Dutch Disease classification, others like Australia and Chile contradicts the Dutch Disease hypotheses while according to Altamirano (1991), countries like Peru and Malaysia show no evidence at all to either accept or reject the Dutch disease hypotheses leading Amuzegar (2001) to conclude that the Dutch Disease hypotheses does not always offer a conclusive explanation for the poor economic performance of resource-rich countries.

 

In the next section, we examine some of the evidence gathered from empirical studies by various researchers.

 

2.3.2    Empirical Review of the Dutch Disease Hypotheses

In conducting an empirical review of the Dutch Disease hypotheses, we follow Olusi and Olagunju (2005) in breaking down the review to cover first, developed countries before examining studies conducted on developing countries. We then shift our focus to studies on the Middle East and North Africa (MENA) region before concluding with a review of studies carried out on our subject country – Kuwait.

 

Developed Countries

Considering that the Dutch disease arose from the observed impact of abundant natural gas resources in the Netherlands, a flurry of researches were carried out on the country with varying results. Ellman’s (1981) research on the effect of the exploitation of large deposits of Natural gas in the Netherlands showed it led to a near collapse of the country’s textile sector while her manufacturing and construction industries declined significantly. Subsequent studies by Barker (1981) and Kremers (1985) also discovered the country’s textile and manufacturing industries had recorded major declines but both researchers were hesitant to attribute the declines to the effect of the Dutch disease as other countries in the region had also suffered similar declines without having booming or large-scale discoveries of natural resources. A similar study carried out by Hutchinson (1990) on The Netherlands, Norway and The United Kingdom indicated the presence of the ‘crowding out’ effect common in the Dutch disease hypotheses. According to the researcher, increased exploitation of natural resources had led to sharp increases in the percentage of value added contributed by the natural resource sector since the 1970’s. Manufacturing sector contributions on the other hand, had declined significantly. Another study by Bjørnland (1997) which investigated the effects of oil and gas booms on the manufacturing sectors of both the Norway and the UK economies however only found ‘weak evidence’ of declines in the UK manufacturing sector while empirical results showed that the Norwegian manufacturing sectors had actually benefitted from the energy booms that characterized the study period. The study by Ross (1986) on the UK concluded that the effects of the Dutch Disease were apparent in the country as between 1977 and 1980 (the years following commercial exploitation of petroleum), the country’s Real Effective Exchange Rate appreciated sharply, while manufacturing output not only grew sluggishly but declined.

Developing Countries

A number of studies have also been carried out on less developed countries. Beginning with Russia which today occupies the position of the second highest exporter of petroleum, Oomes and Kalcheva (2007) tested for the presence of four selected symptoms of the Dutch Disease in the Russian economy; real exchange rate appreciation, slower manufacturing growth, faster service sector growth; and higher overall wages. Their results showed the economy tested positive for each of the symptoms but the researchers were hesitant in attributing the symptoms to the Dutch disease as they could have also been triggered by other factors as well. For example they pointed out, ‘an increase in the relative size of the service sector may be a natural “transition” phenomenon, given that the manufacturing sector had received significant state support during Soviet times, while the service sector remained artificially undeveloped. Moreover, they continued, “deindustrialization has been a natural phenomenon even in the United States and other advanced industrial countries that are not necessarily resource-rich, simply because, as households become richer, demand naturally tends to shift away from goods toward services” (Oomes and Kalcheva, 2007: 22). Therefore, their study was inconclusive. Smirnova and Kulkarni (2013) however successfully diagnosed the Russian economy as being deeply affected by the Dutch disease.

Studies carried out on Indonesia by Warr (1985) were ambiguous with regards the effect of the Dutch Disease. Although his study showed an energy boom had a distinctive effect on domestic prices, it was unclear if the boom had affected the structure of the country’s economy. Further study by Roemer (1994) was clearer with the researcher stating that the “Indonesian government had avoided the worst effects of the DD through careful exchange rate management.” Furthermore and in the African context, a study by Söderling (1999) showed that Cameroon’s poor performance in manufacturing exports from the 1980s onwards could be attributed to its overvalued REER which arose as a result of a combination of poor economic management and exogenous shocks caused primarily by increased natural resource exports (Dutch disease). The researcher called for more emphasis on developing the country’s manufacturing industry as well as improved REER management to help improve the country’s manufacturing sector, and boost economic growth. Benjamin, Devarajan and Weiner (1989) also researched on Cameroon and while identifying traces of the Dutch Disease, also made some interesting findings. According to the researchers, while exports (agricultural) had contracted, as a result of REER appreciation, most of the manufacturing sector was insulated from foreign competition as a result of the unavailability of perfect substitutes. The country’s industrial sector therefore expanded considerably. Also, considering that a large share of the country’s foreign exchange earnings was lodged abroad, most of the effects of the Dutch disease were avoided the researchers observed.

Roemer’s (1988) study on Nigeria, Mexico and Venezuela observed that the countries respective exchange rates had appreciated considerably following increased natural resource exports arising from a boom period, which led to contractions in the industrial/manufacturing sector in at least one country. In the case of Nigeria and Mexico however, manufacturing output had at least grown at the same rate as the service sector or even surpassed it due to deliberate industrialization objectives of the respective governments. Olusi and Olaguju’s (2005) study on Nigeria substituted the conventional tradable sector (manufacturing) with the more appropriate agricultural sector arguing that agriculture was the traditional mainstay of the Nigerian economy. Their research successfully identified the presence of the Dutch disease in the Nigerian economy as the agricultural sector that had hitherto acted as the base of the economy had shrunk to near insignificance with the increased exploitation of petroleum.

Middle East and North Africa (MENA)

Turning our attention to the region of our study country; the Middle East, a number of studies have been carried out on countries in the region. Baky-Haskuee (2011) investigated the effect of oil income on the Iranian economy adopting a co-integration approach. His study showed that oil windfalls have had a telling effect on the country’s economy by virtually changing the structure of the economy as well as relative prices such that the share of agriculture which had hitherto been a driving force in the economy and industry have decreased while the shares of services and construction have increased. The Dutch disease effect in Iran according to the researcher is quite apparent. Esfahania, Mohaddesb, and Pesaranc (2013) investigated the impact of oil exports on the Iranian economy and their results showed that output in the real sectors of the economy is largely influenced by oil exports and foreign output. The study impulse responses also showed that “the Iranian economy adjusts quickly to shocks in foreign output and oil exports,” (Esfahania, Mohaddesb, and Pesaranc, 2013: 221) a condition that was partly attributed to the relatively underdeveloped nature of the country’s financial markets by the researchers.

In a mega study, Apergis, El-Montasser, Sekyere, Ajmi, and Gupta (2014) investigated the effect of petroleum exports on agricultural value added in the MENA region. Using annual data from 1970 to 2011, their analysis showed a negative relationship between oil rents and agricultural value added in the long run. Apergis et al (2014) also observed “a rather slow rate of short run adjustment of agriculture value added back to equilibrium after a boom in oil rents was also noticed.” This result indicated that a boom in the oil sector was associated with observed contractions in the agricultural sectors of sampled countries in the long run. Apergis et al (2014) attributed this to the classic Dutch disease symptom of resource movement effect from other economic sectors to the booming oil sector.

Kuwait Empirical Studies

There’s a relative dearth of empirical studies primarily focused on the economy of Kuwait with regards the effect of the Dutch disease (or the absence of one). While a number of regional studies (Apergis et al. 2014, etc.) successfully identified limited effects of the Dutch Disease in the country, there’s still some doubt as to the degree the country may have suffered from the Dutch disease or averted it. One of the earlier studies on the country was carried out by Dr. Al Sabah in 1988 where he wrote on the ‘Dutch Disease in an oil exporting country; Kuwait.’ According to his findings, there was an unambiguous decline of the tradeable goods sector and a relative increase in the non-tradeable sector after the 1973-74 oil price shock. He also concluded that the real exchange rate was significantly instrumental in reallocating resources between sectors as well as influencing consumption patterns. Furthermore, he observed that “in periods of real depreciation, the tradeable sector somehow increased its relative share in non-oil GDP, while the non-oil trade deficit exhibited a downward trend” (Al-Sabah, 1988: 141). During periods of real appreciation on the other hand, the tradeable sector lost a significant amount of its share to the non-tradeables sector. These led the researcher to conclude that there appeared to be “a considerable correlation between the relative demise of the tradeable sector and the oil boom,” which is totally consistent with the classic Dutch disease predictions. Al Sabah however noted that the large-scale disequilibrium shocks during the period of study (like the collapse of the Bretton Woods System and destabilizing government expenditures), made it difficult to determine the direction of causality (Al-Sabah, 1988).

Another interesting study was conducted on Kuwait by Looney (1991) where he primarily investigated the impact of the Dutch disease on the Kuwaiti industrial sector. His study showed that while Kuwait had suffered inflationary pressures following the oil boom of the 70’s, her inflation rate was relatively low when compared with its oil producing contemporaries. The country’s real exchange rate was found to have appreciated by 31% and while the country went on to devalue her real exchange rate by 17%, her large foreign assets allowed the country control its domestic inflation and enabled inflationary financing to be avoided. In a nutshell, Looney’s (1991) analysis showed through the movements in real exchange rates and relative prices that the Dutch Disease was active in Kuwait. As a result, the service sector consisting mainly of non-tradables experienced positive Dutch disease effects while output in most traded sectors declined. Also, primary activities like agriculture, fishing and mining exhibited mixed results with agriculture and mining suffering weak Dutch disease effects while fishing actually experienced positive effects due to the country’s fishing self-sufficiency. The country’s manufacturing sector as anticipated, experienced negative impacts as a result of the appreciated exchange rate, and relative price increases, all of which emphasized the susceptibility of the country to Dutch Disease effects (Looney, 1991).

Another study by Dousari, Falah and Aswar (2012) aimed to assess the presence of the Dutch disease by examining the impact of increased oil exports on the value of the Kuwaiti domestic currency, the Dinar. Their study turned out to be ambiguous with no definite answer to the question of the presence of the Dutch Disease in Kuwait. According to the researchers, while there was an initial decline in the non-oil tradeable sector,  the sector rebounded considerably in the last decades of the study leading the researchers to conclude that the Dutch disease effect was absent in the sector. Moreover, the researchers investigation of Kuwait’s real exchange rate indicated, debatably by their own admission, that rather than appreciate, the country’s real exchange rate had depreciated in the period considered and this absence of a core feature of the Dutch disease led the researchers to conclude that the Dutch disease effect was absent in Kuwait even though there were traces of the Disease in a few sectors of the economy (Dousari, Falah and Aswar, 2012).

 

 

2.3.3    Agricultural, Industrial and Service Sector Performance, Exchange Rate Management, and the Dutch Disease in Kuwait

 

Fig 2.1 GDP contribution per Sector

Source: Statista 2018

 

Agricultural Sector

While in the classic theories of the Dutch disease the traditional tradable sector is often assumed to be the industrial/manufacturing sector (Corden and Neary, 1982: Sachs and Warner, 1995), in most developing countries however, the agricultural sector as observed by Olusi and Olagunju (2005) had hitherto been the productive base and should therefore represent the tradables sector in these countries. The case of Kuwait however somewhat defeats the argument for the adoption of agriculture as the tradables sector. As observed by a Kuwait business environment publication, the Kuwaiti agricultural sector is in a parlous state with the sector historically contributing only about 0.4% to the country’s GDP. For Kuwait, a combination of a small portion of arable land, scarce fresh water, deficient soil, low manpower etc., have ensured agricultural productivity in Kuwait has over the years remained largely insignificant (KBLH, 2012). While the Arabian Gulf is known to be rich in fish supplies which had hitherto been considerable sources of exports, widespread environmental catastrophe following the Gulf war of 1991 effectively destroyed an already struggling industry while Pearling which was also a mainstay of the economy before the discovery of oil had gradually waned out of significance.

Manufacturing Sector

Turning to the manufacturing sector in Kuwait, Stevens (1986) observed that countries like Kuwait may not experience classic Dutch disease problems associated with a booming sector owing to its low productive industrial activity before the discovery of oil. This low pre-oil industrial activity means there was very little in terms of industrial production for the shift in focus and activities to the oil sector to damage. Given the low initial level of industrialization observes Looney (1991,) an appreciating currency as a result of a booming oil sector and associated effects “could provide a net subsidy rather than a cost to indigenous manufacturers, by reducing the cost of imported capital and intermediate goods” (Looney, 1991: 22). These conditions make it difficult to uniquely assess the Dutch disease effect on both the development of the agricultural and manufacturing sectors in Kuwait.

Service Sector

The Kuwaiti service sector which constitutes the non-tradable sector in this context, has consistently been the second most productive sector following the discovery of oil in the country. However, the sector highlights some of the deepest problems with the Kuwaiti economy. According to a 2015 report issued by the Manpower Public Authority in Kuwait, the total number of migrant workers in Kuwait by mid-September stood a little above 1.5 million employees and workers, divided between the private and the public sectors, with the private sector’s share reaching up to 96%. A breakdown of employment in the various sectors in the country showed that the foreign workforce is distributed unevenly among the various economic sectors with transportation and warehousing at 5.4%, manufacturing 9.7%, agriculture and fishing 5.1%, power, gas and water 0.3%, finance, insurance, real estate and business services 42.2%, building and construction 13.3% other activities 1.2%, social and personal services 15.9%, and mining and quarries 0.9%.

The percent distribution indicates that the service and commodity distribution sectors combined make up a troubling 70% of the foreign workforce. Other sectors such as agriculture and fishing — which do not account for a significant share in the country’s GDP — employ 80,000 workers. Such distribution of foreign labor highlights the gap in the economy, the prevalence of the service sector and the tertiary ones and more worrying, the rentier nature of the economy. Little or no studies have yet examined the implication of having such a high migrant population dominating a key economic sector in the country but without a doubt, the scale of potential migrant remittances would have acted as a drain on the economy. While the government has over the years committed huge amounts including tax waivers as well as other incentives to improve and encourage the country’s industrial and manufacturing sectors, it is no surprise therefore that the service sector has enjoyed far less incentives and direct investments perhaps due to the dominance of foreign and migrant workers.

Exchange Rate Management

On exchange rate management, there is very little unanimity with regards the effect of exchange rate regimes on Dutch disease effects on a country.  While a set of economists advocate a fixed exchange regime to tackle Dutch Disease effects, others maintain that a flexible exchange regime is more effective when the effects of the Disease are anticipated. For example, Lartey (2008) while testing the role of monetary policies in tackling Dutch Disease effects found that a fixed nominal exchange regime stimulates Dutch Disease effects while Al-mulali and Che sab (2012) argued that a fixed exchange regime is a far more effective policy to tackle Dutch Disease effects.

While there are compelling arguments and evidence for both sides of the argument, Ebrahim-Zadeh (2003) however notes that the weakening of the competitiveness of the traditional tradable sector (Dutch Disease effect) occurs irrespective of the exchange rate regime adopted by a country. This according to the researchers is because either way, (fixed or floating) the Real Effective Exchange Rate appreciates. Ebrahim-Zadeh further buttressed his argument with evidence from the case of the Nigerian economy.

While an investigation of the impact of exchange rate regimes on the Dutch disease effect in Kuwait is beyond the scope of this study, it is worth noting that during the period under study, the country operated variants of the fixed exchange rate regime. Having adopted a regional peg to the dollar in line with regional demands for economic integration in 2003, the country jettisoned the Dollar peg in May 2007 (Marzovilla and Mele, 2010) and returned to its previous Basket peg system as a result of inflationary pressures arising from the continued depreciation of the Dollar. Our study would therefore subsume any impact of varying exchange rate regimes as it covers periods of varying exchange rate management systems.

 

DATA SOURCES AND DESCRIPTION, EMPIRICAL MODEL AND ESTIMATION TECHNIQUE

3.1       Description and Source of Data

Data used for the analyses of this study were primarily obtained from the World Bank World Development Indicator Dataset as obtained from the World Bank’s website except data on the Real Effective Exchange Rate which was obtained from Bruegel , a European think tank that specialises in Economics, as well as data on the Total Factor Productivity (TFP) obtained from the Federal Reserve Bank of St. Louis. The data which span from 1985 through 2015 comprises the following variables:

Real Effective Exchange Rate: The World Bank defines the Real Effective Exchange Rate as ‘the nominal effective exchange rate (a measure of the value of a currency against a weighted average of several foreign currencies) divided by a price deflator or index of costs’.

Total Factor Productivity: To measure Industrial/Manufacturing output, data on Total Factor productivity is obtained. The TFP is the portion of output not explained by the amount of inputs used in production. As such, its level is determined by how efficiently and intensely the inputs are utilized in production. Several studies have served to provide both empirical and theoretical evidence that manufacturing exports benefit total factor productivity. Studies by Edwards (1997), Biggs, Shah and Srivastava (1995), as well as that by Tybout (1992) are replete with empirical and theoretical evidence. Following Soderling (1999), the variable is used herein to assess manufacturing output and productivity.

Services output: Includes value added in wholesale and retail trade (including hotels and restaurants), transport, and government, financial, professional, and personal services such as education, health care, and real estate services. Also included are imputed bank service charges, import duties, and any statistical discrepancies noted by national compilers as well as discrepancies arising from rescaling.

Oil Revenue: Total oil rents are defined by the World Bank as the sum of oil rents, natural gas rents, coal rents (hard and soft), mineral rents, and forest rents, measures as a percentage of GDP (Primarily oil in the case of Kuwait).

3.2       Model Specification

In order to determine the existence or otherwise of the Dutch Disease in Kuwait, a model is developed to assess the impact of oil export on;

1. i) The Economy (The tradable and the non-tradable sectors as measured by Total Factor Productivity, Service sector output and Exchange Rate)

Dutch disease literature affirms that the effects of the Dutch disease are primarily observed in economic variables (Real Effective Exchange Rate, Competitiveness variables as well as in the manufacturing and service sectors). Drawing from the theoretical framework and extant literature reviewed in line with the objective(s) of the study, the empirical model is hereby specified in its functional form as

Y = f(R, Z)                                                            (1)

Where Y is output, R and Z are input with R denoting revenue from oil and Z, a vector of other inputs that produce output Y. The output Y can be considered from three different perspectives namely: economic, institutional and social. The vector of inputs (Z) may vary for a particular perspective while oil revenue (R) remains an input in perspective considered.

Specifically from a Dutch disease point of view, equation 1 can be modified as:

EXR = E (OIL, TFP, SER)                                      (2)

Where

EXR = Exchange rate measured by real effective exchange rate

OIL = Oil revenue proxied by total oil revenue percent of GDP

TFP = Total factor productivity

SER = Activities in the service sector measured by service sector output percent of GDP

3.2.1    Ȧ priori Expectations

Ȧ Priori expectation for the above model is that for model (2) and in line with classical Dutch Disease theory, increased oil rents would over time lead to a decline in the industrial sector as measured by the TFP while both service sector output (SER) and Real Effective Exchange rate (EXR) would increase, indicating the presence of the Dutch Disease.

3.2.2    Estimation Techniques and Procedure:

The estimation procedure for this study will involve the use of the following techniques; Unit root test, cointegration test, Vector Autoregression (VAR) or Vector Error Correction methodology (VECM) which will depend on the outcome of our cointegration test.

Stationarity test of time series is necessary to ascertain their characteristics since such variables are usually trended. Stationarity time series do not yield spurious results in regression analysis. Specifically, the Augmented Dickey-Fuller (ADF) test is employed to ascertain the stationarity of the variables. The specification can be expressed as one of the following;

Equation (5) is specified without intercept () or trend (). Equation (6) has on the intercept term while equation (7) contains both intercept and trend.  Represents the variables and  is the residual term.

Johnson and Juselius (1992) test of cointegration or long-run relationship among variables is adopted for this study. It employs both the maximum Eigen statistic and Trace statistic used to test the rank (r) of matrix of parameters of the system of equations and they can be expressed as;

Where  is the estimated values of the characteristic roots (eigen values).

Equations (8) and (9) are the maximum eigen and trace tests respectively.

The VAR model in a specified linear function of P and P lags of other variables in the system. Let   denote a column vector of four variables of interest in equation (2) through (4) above, than  process can be specified as

+  + …, + + K +              … (10)

Where

Vector (n x 1), Y is a column vector

(n x 1) of all endogenous variables

K is (n x 1) vector of constants

p is the number of lags and

the parameters to be estimated

If a cointegration exists among variables in a system of equations then the VAR(p)  process can be expressed as a VECM(p) model.

Where  and π are parameters matrices and Y is a vector of first differences of Y.  The short run relationship is captured by the element.

While the long run efforts are captured by

After estimating a VAR or VECM model, impulse response function (IRF) and forecast error variance decomposition (FEVD) associated with the model can be analysed.

The IRF simulates over time, the effect of a one-time shock in one equation on itself and on other equations in the entire equation system. It is used to detect interactions among variables. On the other hand, FEVD shows the contributions of innovations in variables to the error variations over time in a given variable. Thus it measures the magnitudes of shocks in variables on the error variables on the error variation of a particular variable.

3.3       Empirical Results

Table 3.1: Unit Root Test of Variables.

VARIABLES	LEVELS		REMARK	FIRST DIFFERENCES		REMARK
	ADF STAT	5% CRITITCAL VALUE		ADF STAT	5% CRITITCAL VALUE
EXR	-2.37538	-3.57424	NS	-4.24177	-3.57424	S
SER	-0.84857	-1.95247	NS	-7.18414	-1.95291	S
TFP	-2.02389	-3.57424	NS	-4.73608	-3.57424	S
COR	-2.40526	-3.57424	NS	-5.14223	-3.57424	S
BQ	-0.73816	-1.95641	NS	-7.32677	-1.95291	S
LO	-1.26961	-1.95641	NS	-3.34412	-1.95291	S
HEXP	-2.70146	-2.96397	NS	-5.94222	-2.97185	S
GESS	-0.05744	-1.95247	NS	-2.29822	-1.95502	S
LE	-2.72555	-3.58062	NS	-5.40037	-3.57424	S
OIL	-2.34915	-2.96397	NS	-5.95555	-2.97185	S

Note: NS = Non-stationary; S = stationary.

Source: Author’s

 

 

The Table 1 above presents the results of the unit root test to examine the time series characteristics. It shows that all the ADF statistical values were less than their corresponding 5 percent critical values in absolute terms at levels. This implies that the series are not stationary at levels. However, after differencing all the series once, all the ADF statistical values became greater than their counterpart 5 percent critical values in absolute terms. It indicates that all the time series considered are difference stationary or integrated of order one, I(1). It is therefore necessary to test if a long run or cointegrating relationship exists among variables in the specified models.

 

Table 3.2: Cointegration Test: Model 1.

Hypothesized	Eigenvalue	Trace Statistic	Prob.	Max-Eigen Statistic	Prob.
No. of CE(s)
None	0.538702	34.75250	0.4611	22.43761	0.1988
At most 1	0.238686	12.31489	0.9204	7.908570	0.9092
At most 2	0.119839	4.406316	0.8681	3.701859	0.8894
At most 3	0.023999	0.704457	0.4013	0.704457	0.4013

Note: CE = Cointegrating Equation.

Source: Author’s

 

Table 2 provides both the Trace and Maximum Eigen statistical tests of cointegration proposed by Johansen (1988). All the probability values of the Trace and Maximum Eigen statistics were not significant even at the 10 percent level. It connotes that a hypothesis of no cointegration among the variables in Model 1 cannot be rejected.

 

 

Table 3.3: Lag Length Criteria.

	Model 1
Optimal Lag Length	1

Source: Author’s

 

Before estimating either the VAR or VECM techniques, it is vital to ascertain the optimum length of lags to be included in the analyses. Table 5 showcases the optimum lag length suitable for the model. The values are arrived at considering the various selection criteria used which includes: Akaike information criterion (AIC), Schwarz information criterion SIC), Hannan-Quinn information criterion (HQ), Final prediction error (FPE) and sequential modified LR test statistic (LR) each at the 5 percent significance level. Hence, the VAR (1) and VAR (1) processes are applied for Models 1 and 2 respectively while VECM (1) process is applied for Model 3 in this study.

Table 3.4: Impulse Response Results: Model 1.

Periods	EXR	TFP	SER
1	0.000000	0.000000	0.000000
2	1.540219	-0.01297	-0.12938
3	2.283267	-0.01892	-0.2259
4	2.452389	-0.02017	-0.25828
5	2.246356	-0.01839	-0.24569
6	1.834866	-0.01496	-0.20687
7	1.348651	-0.01095	-0.15676
8	0.877858	-0.00709	-0.10603
9	0.476176	-0.0038	-0.06132
10	0.167995	-0.0013	-0.02599

Source: Author’s

 

Figure 3.1: Graph of Impulse Responses of EXR, TFP and SER to Oil Shock. 

Table 6 presents the responses of EXR, TFP and SER to a Cholesky one standard deviation shock in OIL. The second column shows that a shock in OIL caused the REER to rise or depreciate over the entire periods from the positive values estimated. The REER rose sharply and stayed positive throughout the period under consideration. REER rose to its peak at the fourth period but declined gradually in the long run as depicted in the first panel of Figure 1. The implication is that a rise in oil revenues led to a rise in exchange rate in the country during the period under review.

On the other hand, increases in OIL income led to a drop in TFP throughout the periods as observed from the negative values on the third column of Table 6 (indicated by the blue line in the TFP Impulse Response chart). In Figure 1 the second panel also depicts that a shock in OIL made TFP to decline, reaching its trough in the fourth period. It therefore implies that despite a positive shock in oil revenue in the economy, the broader implication was a decline in manufacturing productivity throughout the period both in the short and long run.

On the third column of Table 6 SER values were all negative. This indicates that a positive innovation in oil revenues reduced activities in the service sector though not significantly in the entire periods. The third panel of Figure 1 (indicated by the blue line in the graph) highlights this downward trend in the sector.

 

Table 3.5: Variance Decomposition of EXR, TFP and SER: Model 1.

Periods	EXR	TFP	SER
1	0.000000	0.000000	0.000000
2	6.653278	0.672601	0.036754
3	14.64300	1.614395	0.137666
4	20.91040	2.465334	0.258997
5	25.02547	3.072558	0.361539
6	27.34721	3.423300	0.429537
7	28.40148	3.577175	0.465259
8	28.69170	3.612876	0.478886
9	28.61884	3.596124	0.480981
10	28.44483	3.567827	0.478988

Source: Author’s

Table 7 provides the Forecast Error Variance Decomposition (FEVD) of EXR, TFP and SER in the Model due to variations in OIL. It shows the proportion of the forecast error variance for each variable as it progressed from the second to the tenth periods in other words, it helps us determine how much of the forecast error variance of each of the variables can be explained by exogenous shocks to the other variables. The result shows that in the second period, OIL accounted for about 6.65, 0.67 and 0.04 percent of the variations in forecast error of EXR, TFP and SER respectively. In the tenth period, variations in OIL accounted for forecast error variation of 28.4, 3.57 and 0.48 percent in EXR, TFP and SER respectively. It indicates that the forecast error variation of EXR is highest while that of the service sector activities is least due to variations in oil revenue. In other words, among our variables, oil exports had the greatest effect on the Exchange Rate (REER) and had the least or a largely insignificant effect on the service sector in Kuwait. These findings corroborate the results of the impulse response discussed above.

 

SUMMARY OF FINDINGS, DISCUSSION AND CONCLUSION

Our model’s result indicates that the direct impact of oil exports in Kuwait has been an appreciation in the Exchange Rate as indicated by increases in the Real Effective Exchange Rate (REER), a decline in the manufacturing sector as indicated by declines in Total Factor Productivity (TFP) as well as a marginal but insignificant decline in the service sector. In other words, while exchange rate had appreciated over the period as a result of oil exports, both the tradable and non-tradable sectors had suffered declines even though the decline in the non-traded sector was statistically insignificant.

Usually, and in line with the Dutch Disease theory, the channel of effect is that increased oil earnings leads to an increase in the Exchange Rate which in turn automatically triggers a decline in the traded sector as increases in relative prices mean traded goods are more expensive while the non-traded service sector would expand as a result of an overall rise in domestic activities. Our result while recognizing both an increase in exchange rates (REER) and concomitant decline in the traded sector (manufacturing productivity as assessed by the TFP), also indicated a muted effect in the non-traded service sector.  Our Variance Decomposition Model shown in Table 3.5 also highlights the muted effect of Oil exports on the service sector with insignificant effects in the entire periods considered.

A number of reasons could be adduced for the non-effect of oil exports on the Service sector in the face of a contraction in the traded sector. As pointed out by Looney (1991: 32), “Government expenditures in Kuwait, particularly direct investment in productive capacity, have obviously played a large and direct role in the country’s industrial expansion”. Furthermore, considering the parlous state of the Kuwaiti manufacturing sector before the advent of oil exploration, there was very little going on in the sector and increased oil earnings inevitably translated to increased investment in the sector and concomitant growth. Looney (1991) listed 4 identifiable industrial segments that may have benefitted from increased oil earnings:

1. Industries with abundant local materials
2. Industries with natural trade protection where factors such as shipping and import costs makes domestic production attractive to manufacturers
3. Strategic industries considered critical to national development and security
4. Industries with comparative cost advantage.

The listed manufacturing segments were beneficiaries of the Kuwaiti government’s deliberate investment and expansion policies and would most likely have benefitted from increased oil earnings leaving the sector susceptible to the vagaries of Oil exports and exchange rate fluctuations. The service sector on the other hand, has been chiefly dominated by foreign workers (up to 70%) and may not have enjoyed significant direct investments as a result of oil exports. While the sector has doubtlessly expanded over the years, the effect of oil exports on the sector have been marginal hence our indifferent result.

While we admit that our explanation of the service sector behavior in the face of an increasing exchange rate and declining manufacturing sector is simplistic at best, our VAR results makes it abundantly clear that the service sector has remained relatively unshaken as a result of oil exports and the service sector behavior should at the minimum, elicit further enquiry.

4.1       CONCLUSION

This study investigated the presence of the Dutch Disease in Kuwait by examining if the classic symptoms of the Dutch disease were present in the Kuwaiti economy. The main symptoms we tested to detect the Dutch Disease included:

• Manufacturing sector decline
• Increase in Service sector productivity and
• Appreciation of the Real Effective Exchange Rate (REER)

Our model result however failed to conclusively detect the presence of the Dutch disease. While the Kuwait economy has exhibited obvious traits of the Dutch Disease with considerable increases in the exchange rate as well as declines in the manufacturing sector as direct consequences of oil exports, in line with the Classic Dutch Disease theory, the country’s Service sector had stayed relatively unchanged and even declined insignificantly when it should have expanded in order to fulfill the entire classic Dutch Disease conditions.

Our study however highlighted the considerable effect of the Oil exports on the economy with oil exports directly leading to increases in the exchange rate and declines in the manufacturing sector. Although a number of reasons could be adduced for the contradiction in the behavior of the Service sector, the insignificant effect of oil exports on the sector leaves us to conclude that there is no conclusive evidence of the existence of the Dutch Disease in Kuwait.

 

 

 

REFERENCES

Ahmed, N., Bucifal, S. and T. George (2008) ‘An Analysis of the Relevance of the Resource Curse to Australia, and Policy Implications’. Australian National University Publication.

Alakhbar (2014) ‘KuwaitGate. Alleged coup, billions of dollors abroad, and struggles for succession’, 18 June.

Al-mulali U. and Che Sab C. (2012) ‘Oil Prices and the Real Exchange Rate in Oil-Exporting Countries.’ OPEC Energy Review, Vol. 36, Issue 4, pp. 375-382.

Al-Sabah M. (1988)    “The ‘Dutch disease’ in an oil-exporting country: Kuwait.” OPEC Energy Review, Vol. 12, Issue 2, pp. 129 – 144.

Altamirano N. (1999) ‘Sustainable Development and Exhaustible Resources: The Dutch Disease is not a Disease at all”. IR/RS Research Report, No. 99 – 08, August.

Amuzegar J. (2001) Managing the oil Wealth; OPEC’s Windfalls and Pitfalls. New York, St. Martin Press.

Apergis N, El-Montasser G, Sekyere E, Ajmi A, and Gupta R (2014) Dutch disease effect of oil rents on agriculture value added in Middle East and North African (MENA) countries. Energy Economics Vol. 45, pp. 485–490

Auty, R M., (1993). Sustaining Development in Mineral Economies: The Resource Curse Thesis. Routledge, London.

Auty, R.M. (ed.) (2001) Resource Abundance and Economic Development. Oxford: Oxford University Press

Baky-Haskuee M. (2011) ‘A Co-integration Approach to the Effect of Oil Income on Resource Allocation in the Iranian Economy.’ International Journal of Economics and Finance. Vol. 3, No. 2; pp 194 – 199

Benkhodja M. (2011) ‘Monetary Policy and the Dutch Disease in a Small Open Oil Exporting Economy’. Groupe D’analyse et de Théorie Économique. Lyon ‐ St Étienne. W P No. 1134, December.

Bjørnland H. (1997) The Economic Effects of North Sea Oil on the Manufacturing Sector. Scottish Journal of Political Economy, Vol 45, No. 5, pp. 553-585.

BTI (2016) Bertelsmann Foundation: Transformation Index – Kuwait 2016.

Capital Standards, (2016) Capital standards industry research report. March 2016.

Corden, W.M, and Neary, J.P. (1982), “Booming Sector and De-industrialisation in a Small Open Economy”, Economic Journal, Vol. 92, No. 368, pp. 825-848

Davis G. (1995) ‘Learning to Love the Dutch Disease: Evidence from the Mineral Economies.’ World Development, Vol. 23, No. 10, pp. 1765-1779.

Dousari A, Falah F and Aswar A (2013) Dutch Disease Effect on Small Open Economy. Kuwait University

Esfahania S, Mohaddesb K., Pesaranc H (2013) ‘Oil exports and the Iranian economy.’ The Quarterly Review of Economics and Finance. Vol. 53 No. 213, pp 221– 237

GCR (2015) Global Construction Review: ‘Kuwaiti Construction Review’, 30 September 2015.

Greif, A. (1994). Self-Enforcing Political Systems and Economic Growth: Later Medieval Genoa. Princeton, Princeton University Press.

Gylfason, T. (2001), “Natural Resources, Education, and Economic Development”, European Economic Review, Vol. 45, No. 4-6, pp. 847-859

Gylfason, T., and Zoega G. (2006). Natural Resources and Economic Growth: The Role of Investment. The World Economy, Vol. No. 29, pp. 1091–115.

Gylfason, T., Herbertsson, T.T., Zoega, G. (1999), “A Mixed Blessing: Natural Resources and Economic Growth”, Macroeconomic Dynamics, Vol. 3, No. 2, pp. 204-225

Humphrey D. (1970), ‘The determinants and structure of import substitution’, Western Economic Journal, Vol. 8 (Sept.), pp. 246-58

Humphreys M., Sachs J., and Stiglitz J. (EDs.) (2007) Escaping the Resource Curse. Columbia University Press. New York.

Hutchinson M. (1990) Manufacturing Sector Resiliency to Energy Booms: Empirical Evidence from Norway, Netherlands and the United Kingdom.  BIS Working Paper No. 13, September.

Kaufman, D., Kraay, A. and Mastruzzi, M. (2010), “Governance Matters III: Governance Indicators for 1996-2002”, World Bank Policy Research. Working Paper No 57, Washington DC: World Bank. KBLH (2012) Kuwait Business Law Handbook: Strategic Information and Laws. International Business Publications, USA.

KSSS (2016) Kuwait Social Security System. Just Landed. Accessed July 18th 2017. Available @ https://www.justlanded.com/english/Kuwait/Kuwait-Guide/Jobs/Social-Security

Lartey E. (2008) ‘Capital inflows, Dutch Disease Effects and Monetary Policy in a Small Open Economy.’ Review of International Economics, Vol. 16, No. 5, pp. 971 – 989.

Looney R. (1991) ‘Diversification in a small oil exporting economy: The impact of the Dutch disease on Kuwait’s industrialization’. Resources Policy, Vol. 17, Issue 1, Pp. 31-41

Mahmudov .S. (2002) Practice and Application of oil funds: Azerbaijan and Kazakhstan as Case Studies in Addressing the Dutch Disease. Duquesne University Economic Issues Publication.

Marzovilla O. and Mele M. (2010) From Dollar Peg to Basket Peg: the Experience of Kuwait in view of the GCC Monetary Unification. MPRA paper No. 22484.

Meade, J., and E. Russel (1957) Wage Rates, The Cost of Living and the Balance of Payments. Economic Record No. 33, pp. 23–28. In J. Meade and S. Howson (eds.) Collected Papers of James Meade. Vol. 3, Winchester, Mass., London and Sydney: Unwin Hyman.

Misra S and Puri V. (2010) “Economics of Development and Planning – Theory and Practice. (12^th Ed.) Himalaya Publishing House.

Olusi J. and A. Olagunju (2005) ‘The Primary Sectors of the Economy and the Dutch Disease in Nigeria’. The Pakistan Development Review, Vol. 44, No. 2, pp. 159–175

Oomes N., and K. Kalcheva (2007) Diagnosing Dutch Disease: Does Russia Have the Symptoms? IMF Working Paper No. WP/07/102. Middle East and Central Asia Department, April.

OPEC (2017) Kuwait Facts and Figures. Organisation of Petroleum Exporting Countries. Accessed 9^th August 2017.

Prebisch, R. (1950) The Economic Development of Latin America and its Principal Problems. Lake Success, New York: United Nations

Rodrik, D., Subramanian, A., and Trebbis, F. (2002) “Institutions Rule: The Primacy of Institutions over Geography and Integration in Economic Development,” National Bureau of Economic Research (NBER) Working Paper 9305, Cambridge, Massachusetts.

Roemer, M. (1994) ‘Dutch Disease and Economic Growth: The Legacy of Indonesia, Harvard Institute for International Development.’ (June), pp. 1–14. Development Discussion Paper No. 489.

Ross M (2001). Does Oil Hinder Democracy? World Politics 53 (April): 325–61.

Ross, A. (1986) The United Kingdom’s Experience with North Sea Oil and Gas. IDS Bulletin (October), pp 14–21

Sachs, J. (2001). “Tropical Underdevelopment”. NBER Working Paper No 8119.

Sachs, J. D. (2007). How to handle the macroeconomics of oil wealth. Chapter 7 in Escaping the Resource Curse, edited by M. Humphreys, J.Sachs and J.Stiglitz (Columbia University Press: NY), 173-193.

Singer, H.W. (1950) “The Distribution of Trade between Investing and Borrowing Countries”, American Economic Review, Vol. 40, No. 2, pp. 473-485

Smirnova O., and Kulkani K., (2013) ‘A Test Of Dutch Disease Phenomenon In Case Of Russia: 1998-2005.’ Graduate School of International Studies, University of Denver, Publication.

Söderling L. (1999) Structural Policies for International Competitiveness in Manufacturing: the Case of Cameroon. OECD Development Centre, Technical Paper No. 146.

Statista (2016) Kuwait Distribution of employment by economic sector. Accessed 13^th July 2017.

Stevens P. (1986), ‘The impact of oil on the role of the state in economic development: a case study of the Arab world’. Arab Affairs, Vol. 1, pp. 87-9l.

Trading Economics (2017) Kuwait Population Facts and Figures

Transparency International (2016) Global Corruption Index 2016. Accessed 17 Septemeber 2017. Available @ https://www.transparency.org/country/KWT

USDS (2015) US Department of State: Human Rights Practices Report – Kuwait 2015

Warr, P. (1985) Indonesia’s Other Dutch Disease. In P. Neary and S. V. Winbergen (eds.) Natural Resources and the Macroeconomy. Oxford: Basil Blackwell Ltd.

Yang X. and Borland J. (1991) ‘A Microeconomic Mechanism for Economic Growth’. Journal of Political Economy, Vol. 99, issue 3, pp. 460-82.

APPENDICES

 

Unit Root Test

 

Null Hypothesis: EXR has a unit root
Exogenous: Constant, Linear Trend
Lag Length: 1 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-2.375376	0.3836
Test critical values:	1% level		-4.309824
	5% level		-3.574244
	10% level		-3.221728
*MacKinnon (1996) one-sided p-values.

 

Null Hypothesis: D(EXR) has a unit root
Exogenous: Constant, Linear Trend
Lag Length: 0 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-4.241765	0.0117
Test critical values:	1% level		-4.309824
	5% level		-3.574244
	10% level		-3.221728
*MacKinnon (1996) one-sided p-values.

 

Null Hypothesis: TFP has a unit root
Exogenous: Constant, Linear Trend
Lag Length: 1 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-2.023893	0.5644
Test critical values:	1% level		-4.309824
	5% level		-3.574244
	10% level		-3.221728
*MacKinnon (1996) one-sided p-values.

 

Null Hypothesis: D(TFP) has a unit root
Exogenous: Constant, Linear Trend
Lag Length: 0 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-4.736080	0.0037
Test critical values:	1% level		-4.309824
	5% level		-3.574244
	10% level		-3.221728
*MacKinnon (1996) one-sided p-values.

 

Null Hypothesis: SER has a unit root
Exogenous: None
Lag Length: 0 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-0.848567	0.3400
Test critical values:	1% level		-2.644302
	5% level		-1.952473
	10% level		-1.610211
*MacKinnon (1996) one-sided p-values.

 

Null Hypothesis: D(SER) has a unit root
Exogenous: None
Lag Length: 0 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-7.184136	0.0000
Test critical values:	1% level		-2.647120
	5% level		-1.952910
	10% level		-1.610011
*MacKinnon (1996) one-sided p-values.

 

Null Hypothesis: OIL has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-2.349154	0.1641
Test critical values:	1% level		-3.670170
	5% level		-2.963972
	10% level		-2.621007
*MacKinnon (1996) one-sided p-values.

 

Null Hypothesis: D(OIL) has a unit root
Exogenous: Constant
Lag Length: 1 (Automatic – based on SIC, maxlag=7)
			t-Statistic	Prob.*
Augmented Dickey-Fuller test statistic			-5.955553	0.0000
Test critical values:	1% level		-3.689194
	5% level		-2.971853
	10% level		-2.625121
*MacKinnon (1996) one-sided p-values.

 

 

Cointegration Test Model 1

 

 

Date: 01/30/18   Time: 14:47
Sample (adjusted): 1987 2015
Included observations: 29 after adjustments
Trend assumption: Linear deterministic trend
Series: EXR TFP SER OIL
Lags interval (in first differences): 1 to 1
Unrestricted Cointegration Rank Test (Trace)
Hypothesized		Trace	0.05
No. of CE(s)	Eigenvalue	Statistic	Critical Value	Prob.**
None	0.538702	34.75250	47.85613	0.4611
At most 1	0.238686	12.31489	29.79707	0.9204
At most 2	0.119839	4.406316	15.49471	0.8681
At most 3	0.023999	0.704457	3.841466	0.4013
Trace test indicates no cointegration at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Unrestricted Cointegration Rank Test (Maximum Eigenvalue)
Hypothesized		Max-Eigen	0.05
No. of CE(s)	Eigenvalue	Statistic	Critical Value	Prob.**
None	0.538702	22.43761	27.58434	0.1988
At most 1	0.238686	7.908570	21.13162	0.9092
At most 2	0.119839	3.701859	14.26460	0.8894
At most 3	0.023999	0.704457	3.841466	0.4013
Max-eigenvalue test indicates no cointegration at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Unrestricted Cointegrating Coefficients (normalized by b’*S11*b=I):
EXR	TFP	SER	OIL
0.137103	3.724208	0.303118	0.018311
0.076946	3.568687	-0.075579	-0.140695
0.002455	6.927151	0.180594	0.054362
0.137277	0.430663	-0.060880	-0.013852
Unrestricted Adjustment Coefficients (alpha):
D(EXR)	-1.078197	-1.155380	0.047332	-0.325381
D(TFP)	0.076381	0.011100	-0.019553	-0.005281
D(SER)	-4.282556	0.013832	-0.718742	0.305890
D(OIL)	3.905407	1.682718	1.100335	-0.741021
1 Cointegrating Equation(s):		Log likelihood	-216.8685
Normalized cointegrating coefficients (standard error in parentheses)
EXR	TFP	SER	OIL
1.000000	27.16368	2.210887	0.133555
	(10.6949)	(0.47962)	(0.20367)
Adjustment coefficients (standard error in parentheses)
D(EXR)	-0.147824
	(0.09489)
D(TFP)	0.010472
	(0.00284)
D(SER)	-0.587149
	(0.13980)
D(OIL)	0.535441
	(0.21754)
2 Cointegrating Equation(s):		Log likelihood	-212.9142
Normalized cointegrating coefficients (standard error in parentheses)
EXR	TFP	SER	OIL
1.000000	0.000000	6.724874	2.907207
		(2.09749)	(1.09695)
0.000000	1.000000	-0.166177	-0.102109
		(0.07203)	(0.03767)
Adjustment coefficients (standard error in parentheses)
D(EXR)	-0.236726	-8.138621
	(0.10201)	(3.34685)
D(TFP)	0.011326	0.324070
	(0.00324)	(0.10615)
D(SER)	-0.586085	-15.89977
	(0.16032)	(5.25965)
D(OIL)	0.664920	20.54964
	(0.24329)	(7.98180)
3 Cointegrating Equation(s):		Log likelihood	-211.0633
Normalized cointegrating coefficients (standard error in parentheses)
EXR	TFP	SER	OIL
1.000000	0.000000	0.000000	-0.950888
			(0.34152)
0.000000	1.000000	0.000000	-0.006772
			(0.00752)
0.000000	0.000000	1.000000	0.573705
			(0.13558)
Adjustment coefficients (standard error in parentheses)
D(EXR)	-0.236610	-7.810742	-0.230950
	(0.10201)	(5.60331)	(0.23411)
D(TFP)	0.011278	0.188624	0.018782
	(0.00317)	(0.17422)	(0.00728)
D(SER)	-0.587850	-20.87860	-1.428966
	(0.15860)	(8.71112)	(0.36396)
D(OIL)	0.667622	28.17183	1.255335
	(0.24063)	(13.2170)	(0.55221)

Lag Length Selection Criteria Model 1

 

 

VAR Lag Order Selection Criteria
Endogenous variables: EXR TFP SER OIL
Exogenous variables: C
Date: 01/30/18   Time: 14:52
Sample: 1985 2015
Included observations: 29
Lag	LogL	LR	FPE	AIC	SC	HQ
0	-283.3230	NA	4740.907	19.81538	20.00398	19.87445
1	-225.9505	94.96148*	277.1366*	16.96210*	17.90507*	17.25743*
2	-210.7110	21.01994	313.4696	17.01455	18.71189	17.54614
* indicates lag order selected by the criterion
LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion

VAR Model 1

 

 

Vector Autoregression Estimates
Date: 01/30/18   Time: 14:55
Sample (adjusted): 1986 2015
Included observations: 30 after adjustments
Standard errors in ( ) & t-statistics in [ ]
	EXR	TFP	SER	OIL
EXR(-1)	0.591486	0.004670	-0.263641	0.005384
	(0.11809)	(0.00332)	(0.17480)	(0.27098)
	[ 5.00866]	[ 1.40665]	[-1.50827]	[ 0.01987]
TFP(-1)	-9.087206	1.110192	-20.58403	19.74163
	(5.81672)	(0.16353)	(8.60976)	(13.3472)
	[-1.56225]	[ 6.78910]	[-2.39078]	[ 1.47908]
SER(-1)	-0.169371	0.010296	-0.129190	0.665555
	(0.21279)	(0.00598)	(0.31497)	(0.48828)
	[-0.79594]	[ 1.72105]	[-0.41016]	[ 1.36305]
OIL(-1)	0.283087	-0.002384	-0.023779	0.800193
	(0.09810)	(0.00276)	(0.14521)	(0.22511)
	[ 2.88557]	[-0.86436]	[-0.16376]	[ 3.55463]
C	44.77079	-0.740198	75.80774	-29.08932
	(18.3873)	(0.51692)	(27.2164)	(42.1921)
	[ 2.43487]	[-1.43193]	[ 2.78537]	[-0.68945]
R-squared	0.774322	0.698325	0.304638	0.512226
Adj. R-squared	0.738214	0.650057	0.193380	0.434182
Sum sq. resids	462.6534	0.365654	1013.633	2436.022
S.E. equation	4.301876	0.120939	6.367520	9.871214
F-statistic	21.44433	14.46766	2.738120	6.563316
Log likelihood	-83.60487	23.54083	-95.36963	-108.5220
Akaike AIC	5.906991	-1.236055	6.691309	7.568134
Schwarz SC	6.140524	-1.002522	6.924842	7.801667
Mean dependent	102.5515	1.066841	22.99910	38.32605
S.D. dependent	8.407837	0.204440	7.089828	13.12298
Determinant resid covariance (dof adj.)		185.0670
Determinant resid covariance		89.24911
Log likelihood		-237.6441
Akaike information criterion		17.17627
Schwarz criterion		18.11040

 

Impulse Response Function  Model 1

 

 

Response of EXR:
Period	EXR	TFP	SER	OIL
1	4.301876	0.000000	0.000000	0.000000
	(0.55537)	(0.00000)	(0.00000)	(0.00000)
2	2.608770	1.140568	-2.582778	1.540219
	(0.83277)	(0.76008)	(0.93655)	(0.56960)
3	1.576959	1.447108	-2.520616	2.283267
	(0.99581)	(0.97981)	(1.05931)	(0.86346)
4	0.781385	1.546815	-2.053672	2.452389
	(1.09731)	(1.13269)	(1.16383)	(1.07231)
5	0.215770	1.503836	-1.454336	2.246356
	(1.12718)	(1.22803)	(1.21184)	(1.18679)
6	-0.140578	1.364262	-0.858931	1.834866
	(1.08962)	(1.27110)	(1.20900)	(1.21475)
7	-0.325854	1.170499	-0.347952	1.348651
	(0.99801)	(1.26580)	(1.15913)	(1.17506)
8	-0.384765	0.957500	0.040721	0.877858
	(0.87215)	(1.21820)	(1.06957)	(1.09094)
9	-0.360474	0.750909	0.300068	0.476176
	(0.73403)	(1.13702)	(0.95208)	(0.98365)
10	-0.289800	0.567118	0.442861	0.167995
	(0.60338)	(1.03250)	(0.82174)	(0.86887)
Response of TFP:
Period	EXR	TFP	SER	OIL
1	-0.022063	0.118909	0.000000	0.000000
	(0.02190)	(0.01535)	(0.00000)	(0.00000)
2	0.000614	0.077321	0.065103	-0.012970
	(0.02307)	(0.02116)	(0.02488)	(0.01510)
3	0.006047	0.060542	0.059302	-0.018918
	(0.02464)	(0.02404)	(0.02491)	(0.02201)
4	0.009578	0.048249	0.049335	-0.020174
	(0.02578)	(0.02684)	(0.02611)	(0.02601)
5	0.011667	0.038966	0.039215	-0.018393
	(0.02559)	(0.02878)	(0.02700)	(0.02748)
6	0.012436	0.032001	0.029976	-0.014964
	(0.02412)	(0.02973)	(0.02697)	(0.02722)
7	0.012143	0.026761	0.022140	-0.010951
	(0.02175)	(0.02977)	(0.02589)	(0.02582)
8	0.011103	0.022755	0.015898	-0.007087
	(0.01890)	(0.02907)	(0.02395)	(0.02371)
9	0.009626	0.019605	0.011208	-0.003804
	(0.01597)	(0.02779)	(0.02144)	(0.02117)
10	0.007977	0.017031	0.007885	-0.001295
	(0.01321)	(0.02611)	(0.01869)	(0.01840)
Response of SER:
Period	EXR	TFP	SER	OIL
1	0.436501	-4.057404	4.887970	0.000000
	(1.16118)	(1.03479)	(0.63103)	(0.00000)
2	-0.731158	-2.052303	-0.484063	-0.129379
	(0.82045)	(0.79775)	(1.22855)	(0.79024)
3	-0.598871	-1.721846	-0.556032	-0.225897
	(0.71485)	(0.70516)	(0.73355)	(0.73702)
4	-0.446233	-1.485028	-0.474389	-0.258283
	(0.67882)	(0.72501)	(0.71739)	(0.72641)
5	-0.325771	-1.274281	-0.423573	-0.245690
	(0.63998)	(0.74657)	(0.70112)	(0.71959)
6	-0.236685	-1.085425	-0.393085	-0.206869
	(0.59156)	(0.75707)	(0.67893)	(0.69541)
7	-0.174084	-0.917677	-0.370539	-0.156764
	(0.53451)	(0.75305)	(0.64290)	(0.65350)
8	-0.132212	-0.770521	-0.348458	-0.106029
	(0.47287)	(0.73526)	(0.59453)	(0.59869)
9	-0.105455	-0.643201	-0.323318	-0.061324
	(0.41090)	(0.70599)	(0.53840)	(0.53600)
10	-0.088905	-0.534562	-0.294303	-0.025993
	(0.35198)	(0.66800)	(0.47930)	(0.46973)
Response of OIL:
Period	EXR	TFP	SER	OIL
1	-0.220040	5.418534	-6.199161	5.440804
	(1.80200)	(1.66044)	(1.27563)	(0.70240)
2	-0.297961	3.982906	-1.707312	4.353694
	(1.58052)	(1.52774)	(2.07292)	(1.34760)
3	-0.698888	3.353750	-0.417008	3.149926
	(1.61607)	(1.61030)	(1.67455)	(1.66097)
4	-0.829960	2.740661	0.453392	2.009033
	(1.59346)	(1.70616)	(1.65884)	(1.73194)
5	-0.767832	2.165531	1.009961	1.050646
	(1.47284)	(1.73127)	(1.62960)	(1.66818)
6	-0.599755	1.662091	1.292600	0.326184
	(1.29386)	(1.68904)	(1.53908)	(1.54985)
7	-0.392695	1.246677	1.359867	-0.162201
	(1.10320)	(1.59720)	(1.39688)	(1.41962)
8	-0.192129	0.921416	1.276743	-0.443060
	(0.93447)	(1.47457)	(1.23013)	(1.28977)
9	-0.024616	0.678870	1.103789	-0.560287
	(0.80054)	(1.33699)	(1.06607)	(1.15863)
10	0.098216	0.506213	0.890947	-0.561685
	(0.69586)	(1.19633)	(0.92294)	(1.02398)
Cholesky Ordering: EXR TFP SER OIL
Standard Errors: Analytic

 

 

 

 

 

Variance Decomposition Model 1

 

 

Variance Decomposition of EXR:
Period	S.E.	EXR	TFP	SER	OIL
1	4.301876	100.0000	0.000000	0.000000	0.000000
2	5.971242	70.98948	3.648489	18.70875	6.653278
3	7.197462	53.66170	6.553650	25.14165	14.64300
4	8.064642	43.68059	8.898820	26.51019	20.91040
5	8.631786	38.19165	10.80314	25.97973	25.02547
6	8.971800	35.37628	12.31208	24.96443	27.34721
7	9.160206	34.06256	13.44361	24.09235	28.40148
8	9.259942	33.50541	14.22479	23.57810	28.69170
9	9.314350	33.26490	14.70902	23.40723	28.61884
10	9.348105	33.12120	14.97103	23.46293	28.44483
Variance Decomposition of TFP:
Period	S.E.	EXR	TFP	SER	OIL
1	0.120939	3.328202	96.67180	0.000000	0.000000
2	0.158151	1.947747	80.43373	16.94592	0.672601
3	0.180522	1.607112	72.98096	23.79754	1.614395
4	0.194548	1.626111	68.98799	26.92056	2.465334
5	0.203420	1.816290	66.77108	28.34008	3.072558
6	0.208999	2.074665	65.59782	28.90422	3.423300
7	0.212496	2.333499	65.04289	29.04643	3.577175
8	0.214705	2.553131	64.83415	28.99985	3.612876
9	0.216138	2.717769	64.80042	28.88568	3.596124
10	0.217102	2.828688	64.84179	28.76170	3.567827
Variance Decomposition of SER:
Period	S.E.	EXR	TFP	SER	OIL
1	6.367520	0.469926	40.60279	58.92729	0.000000
2	6.748548	1.592181	45.39560	52.97547	0.036754
3	7.016160	2.201602	48.02134	49.63940	0.137666
4	7.205740	2.470781	49.77502	47.49520	0.258997
5	7.341143	2.577400	50.96884	46.09222	0.361539
6	7.438001	2.611969	51.77959	45.17890	0.429537
7	7.507208	2.617805	52.32356	44.59338	0.465259
8	7.556588	2.614316	52.68168	44.22512	0.478886
9	7.591781	2.609429	52.91218	43.99741	0.480981
10	7.616830	2.605918	53.05729	43.85780	0.478988
Variance Decomposition of OIL:
Period	S.E.	EXR	TFP	SER	OIL
1	9.871214	0.049689	30.13162	39.43889	30.37981
2	11.63025	0.101431	33.43422	30.56607	35.89828
3	12.53374	0.398258	35.94753	26.42889	37.22532
4	13.02062	0.775335	37.73988	24.61060	36.87419
5	13.30186	1.076098	38.81124	24.15739	35.95527
6	13.48477	1.244919	39.28474	24.42534	35.04500
7	13.61701	1.304023	39.36361	24.95054	34.38182
8	13.71624	1.304844	39.24739	25.45727	33.99049
9	13.78873	1.291480	39.07823	25.83112	33.79917
10	13.83850	1.287243	38.93142	26.06013	33.72120
Cholesky Ordering: EXR TFP SER OIL

 

 

Year	Oil Rents									Services									TFP									REER
1985	33.91665524									24.36060183									0.86197644472									120.71
1986	18.36378828								27.44721095									0.90899527073									105.44
1987	24.22850715								22.83189915									0.95774716139									97.56
1988	24.58097179								25.9101227									0.85217386484									92.04
1989	38.52758819							22.47143888									1.07076239590									90.16
1990	36.23508677							25.17126696									0.78812581301									96.63
1991	7.128014719							55.24925487									0.45656371117									102.40
1992	24.59661903						30.63724166									0.82735311985									93.71
1993	33.60377909						24.14744953									1.08684492110									95.63
1994	30.95793203						23.80443311									1.16190886500									95.52
1995	31.62526422					24.79637959									1.15312707420									89.69
1996	35.59825328					20.84294496									1.16007304190									92.56
1997	32.64121012					22.6976977									1.15814661980									97.55
1998	22.0580477				28.1863684									1.14498865600									103.18
1999	29.43492727				22.35226171									1.09652066230									102.92
2000	44.94719333				17.88268381									1.12827968600									107.23
2001	38.30580102			20.11521045									1.11810529230									112.81
2002	32.33429439			19.63065835									1.12660980220									112.11
2003	37.11036948			20.38401602									1.27794420720									103.15
2004	43.76629173		18.95358794									1.35558295250									97.34
2005	54.22018069		16.69520621									1.41770076750									99.18
2006	53.16843145		18.79115605									1.39634609220									99.96
2007	49.33048497	20.50971979									1.34722578530									100.00
2008	54.16617077	18.81766585									1.24299407010									104.68
2009	36.80162558	23.80434594									1.08206069470									105.09
2010	48.02838859	21.48201837									0.98356479406									106.16
2011	60.2357162	18.89933089									1.00000000000									107.32
2012	59.81890007	17.19636204									1.00261974330									110.49
2013	56.08821637	15.60822778									0.94745546579									113.20
2014	53.39691894	18.48038509									0.87770503759									116.60
2015	38.48250644	26.17652963																		126.22

For more academic help please check a wide range of services our Economics Writing Help team offers:
– Economics Assignment Writing Services
– Economics Essay Writing Services
– Economics Dissertation Writing Services
– Buy An Economics Research Paper

Students working on case studies or might need academic help, might find our custom Case Studies Writing Services helpful.

Also look at some of our business services
– Business Essay Writing Service
– Business Dissertation Writing Services
– Business Report Writing
– Business Assignment Help
– Business Planning Writing Service
– Business Assignment Writing Service

In addition, for Accounting Writing Help please make sure check out services offered by our qualified accounting team:

– Accounting Coursework Help
– Accounting Dissertation Writing Services
– Accounting Essay Writing Services