The former Soviet Union Republic Kazakhstan is the ninth biggest country in the world, located in Central Asia. Kazakhstan is widely known for its massive amount of oil reserves. It possesses 30 billion barrels of proven reserves, which makes it the eleventh country in the world listed according to its proven reserves. It has 40 % more proven reserve than the number twelve on the list, the United States of America (EIA, 2008).
The oil of Kazakhstan is being exploited in fifteen oil fields. Of these, the Tengiz field, which is discovered in 1979, is the biggest. This field is with its size of 2500 square kilometre not only the biggest field in Kazakhstan, but also the sixth largest field in the world. It has an estimated 6-9 billion barrels of economically exploitable oil, and 24 billion barrels of oil available in total.
In the 80s the available Soviet technology, which was inferior to western technology, was unable to (economically) exploit most of the available reserves of Kazakhstan. As of 1991, after the independence from the Soviet Union, western based firms were granted the opportunity to drill for oil. For example, in 1993 the joint-venture Tengizchevroil, a joint venture between Chevron (with a 50 % share in the consortium); ExxonMobil (25 %); KazMunayGas (20 %); and LukArco (5 %), started producing in the Tengiz field. Due to the problems of the transition from communism to capitalism, the production of oil stayed at a relatively low level until 1998, but ever since the production has increased rapidly. In 2008, the production was almost 3.5 times as high as in 1994, when the production reached its lowest value (The Agency of Statistics of the Republic of Kazakhstan, 2010). In 2008, Kazakhstan was ranked number 19 in the world according to the production of crude oil (EIA, 2008). Until 2030, when Kazakhstan is expected to reach its oil peak, the production of oil is only likely to increase steadily.
In 1992 the oil production of Kazakhstan was just a bit higher than its oil consumption. With such levels Kazakhstan was able to be self-sufficient, but it was not able to export its production. Since 1994, the oil production grew steadily. In 2005 the production more than doubled compared to the production level in the 90s. Meanwhile, the consumption declined from 1992 until 1999 and stayed stable ever since. In 2008 the oil consumption was half as high as it was in the beginning of the 90s. With a decline in oil consumption and a huge leap in production Kazakhstan has turned into one of the biggest oil exporting countries (EIA, 2008). The net oil exports of Kazakhstan accounted for about 1.2 million barrels per day in 2008. That same year, the net inflows of the foreign direct investment (FDI) of Kazakhstan, mainly in the oil rich oblasts Atyrau, and Akatau, accounted for 11% ($14,647,520,460) of the GDP (The World Bank, 2010).
In 2003 the natural resources accounted for 86.1% of the total export of goods and services of Kazakhstan (The World Bank, 2005). Oil is accountable for 59% of the total export of Kazakhstan. Ferrous metals are second with a total of 19%. Together they account for 78% of the export of Kazakhstan (CIA, The World Factbook). It must be noted that mining itself, just like oil and gas, is a sector that can be held responsible for the presence of the Dutch Disease. For example, the Spanish industry suffered from huge inflows of gold in the 16th century. In other words, 78% of the exports of Kazakhstan come from sectors which promote the Dutch Disease. Kazakhstan should not only watch the potential harm of the Dutch Disease originating from the oil sector, but it should be aware that the mining sector itself might be an origin of the Dutch Disease. However, in this paper I will only focus on the Dutch Disease originating from the oil sector. But, first of all, what do we mean with the Dutch Disease?
Explanation of the Dutch Disease
"The term Dutch Disease refers to the adverse effects on Dutch manufacturing of the natural gas discoveries of the nineteen sixties, essentially through the subsequent appreciation of the Dutch real exchange rate." [1] (Corden; 1984). The term 'Dutch Disease' was coined in The Economist [2] in 1977, referring to a loss in competiveness in the non-gas sectors due to gas revenues in the 1960s in The Netherlands. According to Sachs and Warner (1995): "resource-poor economies often vastly outperform resource-rich countries in economic growth" [3] . Resource-rich countries must carefully manage its resources in order to stimulate economic growth. Famous examples of the Dutch Disease are: huge gold inflows from America into Spain in the 16th century, the discoveries of gold in Australia during the 1850's, and the discovery of natural gas in the Netherlands in 1959.
In this paper, I will investigate whether Kazakhstan suffers from the paradox suggested by Sachs and Warner. The first part of the paper introduces the Core model, which gives a theoretical foundation of the Dutch Disease. The second part explains why the Dutch Disease is harmful for the economy. The third part puts forward empirical evidence which shows whether Kazakhstan suffers from the Dutch Disease.
The Core model
The Core model of Corden (1984) explains two effects that are accountable for the Dutch Disease: the spending effect, and the resource movement effect. The latter can be divided into the direct de- industrialization effect, and the indirect de- industrialization effect. This model takes two tradable sectors and a non-tradable sector into account. Namely: a booming sector (i.e. the oil sector), the lagging sector (i.e. the manufacturing sector), and the non-tradable sector (i.e. the services sector). The first two sectors encounter given world prices. The non-tradable sector sets a country specific price.
Labour is mobile between the three sectors, and moves in order to equalize the wage between the three sectors. Capital, however, is immobile since it is assumed to be sector specific. A boom in the booming sector will raise the aggregate incomes of the initially employed.
According to Corden, a boom can happen in three ways: via technical improvements in the booming sector (1), via a discovery of new resources (2), or the booming sector exports all its output, it has no sales domestically, and price of its product on the world market rises relative to the price of imports (3). In two ways a resource boom will affect the economy via the "spending effect" and via the "resource movement effect". The first effect refers to the supply side, while the second effect refers to the demand side of an economy.
Spending effect (demand side)
A boom leads to extra income of the oil sector. The income of employees, profits of the oil sector, and/or government tax revenues will go up. Therefore, the demand for all goods in the economy will increase. The price of tradable goods is determined by the world market and will stay stable, but the price of non-tradable, relative to tradable goods, will increase. Therefore, the real exchange rate appreciates further. Labour will shift from the tradable sector into the non-tradable sector. This results into a contraction of the non-tradable sector.
Resource movement effect (supply side)
A boom will improve the marginal product of labour in the booming sector. This leads to an increase in the demand of labour in this sector. Labour will move out of the lagging sector and the non-tradable sector, and into the booming sector. Since the use of the production factors of the lagging sector declines, this effect results into a contraction of the lagging sector. Corden divides the resource movement effect into two parts: direct de-industrialisation and indirect de-industrialisation.
Direct de-industrialisation
Labour, if it is mobile, moves out of the lagging sector and into the booming sector. In this case, the non-tradable sector is not affected. An appreciation of the currency is not required. Therefore, the real exchange rate stays stable.
Indirect de-industrialisation
Labour moves out of the non-tradable sector and into the booming sector. This creates an excess demand for the non-tradable sector in addition to the spending effect, and results in an additional real appreciation. This will move labour about of the lagging sector and into the non-tradable sector, strengthening the de-industrialisation which results from the spending effect. The combination of the two, which make labour moving from the lagging sector to the non-tradable sector, is called the indirect de-industrialisation.
The resource movement effect only takes place if labour is mobile, and will move between the different sectors. However, the oil sector uses a relatively low amount of labour. Furthermore, the labour mobility in Kazakhstan is assumed to be low. Therefore, Kazakhstan will only marginally be affected by the resource movement effect.
The spending effect appears regardless of whether labour is either mobile or not. This effect will come into force, due to a rise in the aggregate demand. The aggregate demand has risen since a higher oil price generates higher wages and/or profits.
Table 1: Overview of the resource movement effect and spending effect:
Output
Employment
Wage
Price
Spending Effect
Booming sector
-
-
+
World price
Lagging sector
-
-
+
World price
Non-tradable sector
+
+
+
+
Resource Movement Effect
Booming sector
+
+
+
World price
Lagging sector
-
-
+
World price
Non-tradable sector
-
-
+
+
Combined
Booming sector
Undetermined
Undetermined
+
World price
Lagging sector
-
-
+
World price
Non-tradable sector
Undetermined
Undetermined
+
+
By combining the two effects, the Dutch Disease hypothesis assumes the following results to happen due to a boom: an appreciation of the real exchange rate (1), a likely though ambiguously increase of the output of the non-tradable goods sector (2), an unambiguous decline in the output and employment of the lagging sector (3), a drop in the exports of the lagging sector (4), and, if labour is mobile, an increase in the overall wage level (5). It seems likely that (4) a drop in the exports of the lagging sector happens due to presence of (3) a decline in the production of the lagging sector, but this not necessarily have to be the case. As Stijns (2003) points out, the production of the lagging sector could increase while the export decreases. This could happen if, due to a resource boom, the demand for the lagging sector domestically grows more than the exports shrink. A drop in the exports of the lagging sector (4) is necessary for an unambiguous decline in the output and employment of the lagging sector (3) to happen, but not sufficient.
Empirical research has been performed in order to prove whether the prediction of the Dutch Disease theory is a correct one. Chef and Rogoff (2002) investigate whether the real exchange appreciates (1). Spatafore and Warner (1995) examine whether the output of the non-tradable sector goes up (2). Stijns (2003) tests whether it is the case that the export of the lagging sector declines (4). Chen and Rogoff (2002) show that the real exchange did appreciate (1) in Australia, Canada, and New Zealand. These are all countries in which minerals attribute to a significant share of the total export. It must be noted that they investigated only three mineral-rich countries. More research should be done in order to provide information about other resource abundant countries. Spatafora and Warner (1995) control whether it is the case that the output of the non-tradable sector increases (2). They did so by looking at 18 oil-exporting countries in the period of 1973 - 1989. Indeed, for the main categories of the non-tradable goods sector a (significant) positive effect on the output is present. Stijns (2003) checks whether the export of the lagging sector declines (4). According to his results: "A one percent increase in world energy price is estimated to decrease a net energy exporter's real manufacturing exports by almost half a percent. Similarly, after instrumentation, a one percent increase in an energy exporting country's net energy exports is estimated decrease the country's real manufacturing exports by 8 percent."
To summarize, an increase of the production of the booming sector leads to increase in the demand for non-tradable goods. This in turn leads to a decrease in the allocation of labour and capital of the lagging sector. Both the spending effect and the resource movement effect imply that a high endowment of natural resources will move labour into the non-tradable goods sector. Labour which otherwise would have been employed in the lagging sector. So, if a country experiences a resource boom, the size of the lagging sector will decline, and the size of the non-tradable goods sector will increase. The downsizing of the lagging sector is referred to as the Dutch Disease.
The Dutch Disease does not necessarily need to be a real 'disease'. By itself, it is not an explanation why a country would suffer from a natural resource curse. In other words, a contraction in the lagging sector does not need to have a negative impact on the economy. A country is just specializing into the sector in which it has a comparative advantage. If oil prices are high, it is only optimal for resources to move out of the lagging sector and into the booming and non-tradable sector. Furthermore, if the price of oil would stay high forever, the optimal thing for a country to do would be by eliminating the lagging sector as a whole and specialize in the production of oil. Why would de-industrialization have to lead to lower economic growth?
Increased volatility
De-industrialization makes the economy more vulnerable for volatility. Hausman and Rigobon (2003) show that volatility has a negative relationship on economic growth. An economy with a small lagging sector faces difficulties to adapt to shocks via the mobility of labour. Let us assume an extreme situation with high oil prices, and in which the booming sector uses no labour. The lagging sector will disappear. Now, the non-tradable sector will be the only employer. Since all labour is employed in one sector the economy is highly vulnerable for volatility. According to this theory the growth will only be temporary lower, since the lagging sector will recover once the oil price is on a low level. However, Krugman (1987) shows that if industries move abroad for a sufficiently long period, they will stay abroad; even when the favourable conditions abroad have ended.
Production externalities
The main reason why de-industrialization leads to lower economic growth is due to the fact that the lagging sector is characterized by positive externalities. In comparison to the other sectors, it tends to be more competitive and innovative, and it is characterized by technological spillovers. Three reasons can be given. First of all, it is relatively easy for a firm to enter the lagging sector. This is the case, since the lagging sector is exempted from having to pay large rents. This improves competition, and, therefore, efficiency. Second of all, the lagging sector is more open to innovation than the booming sector. The production of manufacturing products has more room for innovation. Third of all, the lagging sector comes along with vertical as well as horizontal spillovers (e.g. learning by doing). I will explain these reasons in more detail in the following part.
Backward and forward linkages effects
Hirschman (1958) introduced the concept of backward and forward linkages effects of industries. The first effect relates to derived demand. The second effect relates to the utilisation of output. The total linkage effect can be interpreted as the probability that the establishment of industry i will lead to the establishment of n additional industries j.
The backward linkages effect
The newly established industry i requires the input product y. This product is being produced by firms belonging to industry j. Therefore, the increase of the production of a downstream manufacturer present in industry i gives rise to the production of a upstream manufacturer present in industry j.
The forward linkages effect
The newly established industry i produces the output product z. This product is being used as the input product by firms belonging to a, possible new, industry k. Therefore, the increase of the production of a manufacturer present in industry i gives rise to the production of a downstream manufacturer present in, a possible new, industry z.
The beneficial linkage effects of an export sector to the economy as a whole are considered to be small. In the Core model, production in the lagging sector, as opposed to the booming sector, will mean a division of labour in a more complex way, and therefore, a higher standard of living. Therefore, a boom diminishes the backward and forward linkages effects.
Learning-by-doing
Matsuyama (1992) presents a model containing of two sectors: the agricultural sector, and the manufacturing sector. The productivity of the manufacturing sector rises over time, due to learning by doing. An increase in agricultural productivity will move labour out of the manufacturing sector. This leads to a decrease in the productivity of manufacturing and, therefore, in lower economic growth. By applying the same logic for the Core model, we will end up with a diminishing economic growth. The spending effect and the resource moving effect will shift labour out of the lagging sector, and into the non-tradable sector. The learning-by-doing effect of the non-tradable sector is smaller compared to the lagging sector. This way the shift of labour will lower total economic growth.
The Dutch Disease in Kazakhstan
I will investigate whether the Dutch Disease is present by performing a Granger causality test, and by comparing the Real Exchange Rate - with and without oil exports - of Kazakhstan with other countries. By making a comparison between the economic growth and economic development I will check whether Kazakhstan is likely to suffer from the natural resource curse. This would mean that the economy of Kazakhstan suffers from producing minerals like oil.
Granger causality test
This analysis of the presence of the Dutch Disease in the economy of Kazakhstan is performed based on the Granger causality test. This methodology records whether the preceding values of X (in our case oil) can explain the development of the succeeding values of Y (another sector). The Granger causality test is the strongest approach I use. It checks whether the one variable influences the other. Using this method I can test whether the production of oil influences other sectors of the economy. With a 10% confidence interval I will check of the production of which, if any, sectors are harmed by the production of oil. With this method I prove whether we can state that the Dutch Disease is present in Kazakhstan, and whether the presence has a big impact on the economy or not.
Real Exchange Rate
A feature of the Dutch Disease is the appreciation of the real exchange rate. An appreciation of the national currency is necessity for the Dutch Disease to be present. Therefore, I shall make a comparison between the real exchange rate including oil trade and the real exchange rate excluding oil trade. An appreciation in the exchange due to oil trade will negatively influence the other sectors. These sectors might face a decline of export, since their products become more expensive for foreign countries. The appreciation of the real exchange rate is a characteristic of the Dutch Disease. This comparison does not prove the existence of the Dutch Disease, but it shows whether the symptoms are present.
Economic Growth vs. Economic Development
An increase in the production of oil generates a higher GDP, increasing the economic growth. However, if only a small portion of the society gains benefits from the oil production then the overall economic development stays stable or might even diminish. This problem is known as the natural resource curse. We can measure economic growth by looking at the GDP per capita. Economic development can be measured by looking at the Human Development Index (HDI). The HDI is an index published by the United Nations Development Programme. The HDI compares the life expectancy at birth, adult literacy rate, education, and GDP per capita. A country with a much lower level of HDI in comparison with the level of GDP per capita is likely to suffer from the natural resource course.
Results of the research
Does Kazakhstan suffer from the Dutch Disease?
Granger causality test
The Granger causality test checks whether the one variable influences the other. Using this method we can test whether the production of oil sector influences other sectors of the economy. The first differences of the variables have been used in order for the removal of the non-stationary nature of the time series data. The study is based on data of the years 1991 until 2008. Kudebayeva (2003) based a similar research on the main export commodities of Kazakhstan for the years 1990 until 2002. She did so by checking for the period 1990 - 2002, and of the period 1996 - 2002. In order to compare my results with her Granger causality test results, I will take the period 1991 - 2008 into account, as well as the period 1996 - 2008.
The calculations have been performed by using both a lag of 1, and a lag of 2. A lag of 1 means that in the calculation we are used the current period (t) and the previous period (t-1). A lag of two 2 means we are using the current data (t), and the previous periods (t-1) and (t-2). For this test we use the null hypothesis: The production of raw oil does not Granger cause the production of sector "X". A confidence level of 10 % has been used in order to reject the hypothesis.
Table 2: Granger causality test between the oil sector and other sectors for the period 1990 - 2002
Sample: 1990 - 2002
Null Hypothesis
Lags
Probability
DOIL does not Granger Cause DGrain
1
0.0957
DOIL does not Granger Cause DGrain
2
0.0697
Table 3: Granger causality test between the oil sector and other sectors for the period 1996 - 2002
Sample: 1996 - 2002
Null Hypothesis
Lags
Probability
DOIL does not Granger Cause DGrain
1
0.0346
DOIL does not Granger Cause DRcopp
1
0.0676
DOIL does not Granger Cause DRometal
1
0.1013
DOIL does not Granger Cause DRometal
2
0.0900
Research based on data of the years 1990 - 2002
This research is based on the main export commodities of Kazakhstan. These main export commodities are raw oil, refined copper, rolled ferrous metals, and grain. The results of the research are shown in table 2, and table 3.
For the period 1990 - 2002:
The hypothesis stating that the production of oil does not influence the production of grain (DGrain) can be rejected. By both a lag of 1 (9.57 %), and a lag of 2 (6.97 %) we obtain a percentage below the 10 %.
For the period 1996 - 2002:
The hypothesis stating that the production of oil does not influence the production of grain (DGrain), refined copper (DRcopp), rolled ferrous metals (DRometal) can be rejected. If we use a lag of 1, we obtain for both grain (3.46 %), and refined copper (6.76%) a percentage below the 10 % confidence level. Furthermore, if we use a lag of 2 we obtain for rolled ferrous metals (9.00 %) a percentage below the 10 %.
By looking at both periods we reject the null hypothesis thesis for the following three sectors: grain, refined copper, and rolled ferrous metals. Therefore, we conclude that it is impossible to claim the Dutch Disease is not present in Kazakhstan.
Table 4: Granger causality test between the oil sector and other sectors for the period 1991 - 2008
Sample: 1991 - 2008
Null Hypothesis
Lags
Probability
DOIL does not Granger Cause DGrain
1
0.0090
DOIL does not Granger Cause DGrain
2
0.0151
DOIL does not Granger Cause DRcopp
1
0.6477
DOIL does not Granger Cause DRcopp
2
0.7973
DOIL does not Granger Cause DCereal
1
0.0044
DOIL does not Granger Cause DCereal
2
0.0539
DOIL does not Granger Cause DEnergy
1
0.0012
DOIL does not Granger Cause DEnergy
2
0.0158
DOIL does not Granger Cause DMining
1
0.2104
DOIL does not Granger Cause DMining
2
0.0971
DOIL does not Granger Cause DCoal
1
0.0147
DOIL does not Granger Cause DCoal
2
0.1403
DOIL does not Granger Cause DOcopper
1
0.0893
DOIL does not Granger Cause DOcopper
2
0.1549
Source: The test results are based on data published by The Agency of Statistics of the Republic of Kazakhstan (www.stat.kz)
Table 5: Granger causality test between the oil sector and other sectors for the period 1996 - 2008
Sample: 1996 - 2008
Null Hypothesis
Lags
Probability
DOIL does not Granger Cause DGrain
1
0.2946
DOIL does not Granger Cause DGrain
2
0.3428
DOIL does not Granger Cause DRcopp
1
0.0345
DOIL does not Granger Cause DRcopp
2
0.2192
DOIL does not Granger Cause DCereal
1
0.0918
DOIL does not Granger Cause DCereal
2
0.0539
DOIL does not Granger Cause DEnergy
1
0.0633
DOIL does not Granger Cause DEnergy
2
0.0158
DOIL does not Granger Cause DMining
1
0.1811
DOIL does not Granger Cause DMining
2
0.0971
DOIL does not Granger Cause DElectricity
1
0.4323
DOIL does not Granger Cause DElectricity
2
0.0833
DOIL does not Granger Cause DOchromite
1
0.0769
DOIL does not Granger Cause DOchromite
2
0.2236
Source: The test results are based on data published by The Agency of Statistics of the Republic of Kazakhstan (www.stat.kz)
Research based on data 1991 - 2008
I performed a similar research with more recent data in order to check whether the same conclusion is still valid for Kazakhstan. I expanded the investigation by using not only the data of the production of the main export commodities of Kazakhstan, but by using data of any large production sector of Kazakhstan. Furthermore, I included the total production of the whole mining industry (DMining) to see whether this industry as a whole is affected by the production of raw oil. With the exemption of the production of grain, and refined copper I did not include any product which is not proven to be significantly affected by the oil production.The results of the research are shown in table 4, and table 5.
For the period 1991 - 2008:
It can still be concluded that it is impossible to claim that the production of raw oil does not influence the production of grain. By both a lag of 1 (0.9 %), and a lag of 2 (1.51%) we obtain a percentage far below the confidence level of 10 %. However, the same can no longer be concluded for refined copper. By both a lag of 1 (64.77 %), and a lag of 2 (79.73 %) we obtain a percentage far above the confidence level.
Now, we will expand the research to the production of cereal and vegetable flour (DCereal), thermal energy (DEnergy), mining (DMining), coal (DCoal), copper ore (DOcopper) . If we use a lag of 1, for the following sectors we obtain a percentage below the confidence level: cereal and vegetable flour (0.44 %), thermal energy (0.12 %), coal (1.47 %), and copper ore (8.93 %). By using a lag of 2, we conclude that the production of cereal and vegetable flour (5.39 %), thermal energy (1.58 %), and mining (9.71 %) are influenced by the production of raw oil.
For the period 1996 - 2008:
In contrast to what we would expect, the production of grain is no longer influenced by the production of raw oil (29.46 % by using a lag of 1, 34.28 % by using a lag of 2). This result is counterintuitive, since grain is affected in all other periods. Furthermore, the production of refined copper is affected (3.45 % by using a lag of 1) a sector which showed no sign of being affected for the period 1991 - 2008. Other sectors which are affected by using a lag of 1 are: cereal and vegetable flour (9.18 %), thermal energy (6.33 %), chromite ore (7.69 %). The following sectors are influenced by using a lag of 2: cereal and vegetable flour (5.39 %), thermal energy (1.58 %), mining (9.71 %), and electric power (8.33 %).
By looking at both periods we reject the null hypothesis thesis for the following nine sectors: grain, refined copper, cereal and vegetable flour, thermal energy, mining, coal, chromite ore electric power, and copper ore. Therefore, we conclude that it is impossible to claim the Dutch Disease is not present in Kazakhstan.
Overall, the two performed Granger causality tests prove that it is impossible to claim that the Dutch Disease is not present in Kazakhstan. The first test does so by the following three sectors: grain, refined copper, and rolled ferrous metals. The second test does the same with the following nine sectors: grain, refined copper, cereal and vegetable flour, thermal energy, mining, coal, chromite ore electric power, and copper ore. Both tests conclude that is impossible to claim for grain as well as refined copper that they are not suffering from the Dutch Disease. The results of both tests point strongly towards the presence of the Dutch Disease in the main export sectors of Kazakhstan.
Real Exchange Rate
Both graph 1, and graph 2 show a comparison of Kazakhstan and other countries, comparing the real exchange rate including oil trade and the real exchange rate excluding oil trade. Graph 3 shows a combination of the two. An appreciation in the exchange due to oil trade will negatively influence the other sectors. These sectors might face a decline of export, since their products become more expensive for foreign countries. An appreciation of the real exchange rate due to the influence of oil trade is a characteristic of the Dutch Disease.
Graph 1: A comparison of the real exchange rate of Kazakhstan with the real exchange rates of the CIS countries
Source: The graph is based on data published by The Agency of Statistics of the Republic of Kazakhstan (www.stat.kz)
Commonwealth of Independent States (CIS)
Comparing the real exchange rate with the exchange rates of the CIS countries (Belarus, Kyrgyz Republic, Russian Federation, and Ukraine) we see that the exchange rate including oil trade lies above the exchange rate including oil trade. Until 2003 both graphs are moving together. As of the second half of 2003 the real exchange rate including is slightly higher than the real exchange rate excluding oil. This suggests that Kazakhstan is likely to suffer from the Dutch Disease.
Graph 2: A comparison of the real exchange rate of Kazakhstan with the real exchange rates of non-CIS countries
Source: The graph is based on data published by The Agency of Statistics of the Republic of Kazakhstan (www.stat.kz)
Other countries (Non-CIS countries)
However, if we compare Kazakhstan with non-CIS countries (UK, Germany, Italy, Netherlands, Finland, France, Hungary, Latvia, Lithuania, Poland, Czech Republic, Switzerland, Estonia, Iran, China, Korea, Turkey, Japan, USA, and offshore territories (British Bermuda Islands, British Virgin Islands) we get a different picture. The exchange rate including oil trade lies below the exchange rate including oil trade. Until the beginning of 2002 both graphs are moving together. As of the second half of 2002 the real exchange rate excluding oil is clearly much higher than the real exchange rate excluding oil. This suggests the absence of the Dutch Disease in Kazakhstan.
Graph 3: A comparison of the real exchange rate of Kazakhstan with the real exchange rates of both the non-CIS countries and CIS countries
Source: The graph is based on data published by The Agency of Statistics of the Republic of Kazakhstan (www.stat.kz)
Total
If we combine both data we get a graph with in which the exchange rate including oil trade lays above the exchange rate including oil trade. Until 2005 both graphs are moving together. As of the second half of 2005 the real exchange rate including oil is at a higher level than the real exchange rate excluding oil. Therefore, if we consider the total picture Kazakhstan is likely to suffer from the Dutch Disease.
Overall, Kazakhstan faces an appreciation of the real exchange rate due to the trade of oil, a feature of the Dutch Disease. The comparison of the real exchange rate with the CIS countries does show an appreciation of the national currency due to oil trade. However, the comparison with the non-CIS countries does contradict this conclusion. If we combine the two, and look at the complete picture for Kazakhstan, than this feature of the Dutch Disease is clearly present. However, since the results are contradicting each other, our conclusion is not a strong one.
Human Development Index
The Human Development Index (HDI) tests the overall quality of life. [4] It does so by looking at the life expectancy at birth, the adult literacy rate (% aged 15 and above), the combined gross enrolment ratio for primary, secondary and tertiary education (%), and the GDP per capita (PPP US$). By comparing the economic growth (GDP per capita) with the economic development (HDI) we can check whether a high GDP per capita results in a in a better overall quality of life.
Countries with a low rank of the HDI, and a high rank of GDP per capita are likely to suffer from the natural resource curse. The Human Development Reports ranks countries according to their HDI and their GDP per capita. In order to check for the presence of the natural resource curse, we will subtract the HDI rank from the GDP per capita rank.
The following results arise by checking the results for OPEC countries (except for Iraq, due to a lack of data): Algeria (-22), Angola (-33), Ecuador (21), Iran (-23), Kuwait (-8), Libya (4), Nigeria (4), Qatar (-12), Saudi Arabia (-19), the United Arab Emirates (-12), and Venezuela (14). Algeria, Angola, Iran, Kuwait, Qatar, Saudi Arabia, and the United Arab Emirates all have a number far below zero, suggesting that the Dutch Disease is present in these countries.
Ecuador and Venezuela face a high positive number. Both countries are under a socialist rule. These socialist regimes try to improve the situation for the poor, thereby increasing the overall HDI. Furthermore, due to high taxes of a socialist regime the rank of the GDP is relatively low. For example, the communistic country Cuba, which has an artificially low GDP, has a positive GDP rank minus HDI rank number of 43. A low GDP and a high HDI are the explanation for the high positive number of Ecuador and Venezuela. Therefore, this method is inconclusive whether the natural resource curse is present in either Ecuador or Venezuela.
The positive numbers of Nigeria and Libya are more difficult to explain. Their moderate positive numbers suggest that the rank of both their GDP and HDI are, more or less, in line with each other. This implies that these countries do not suffer from the natural resource curse. Further research should be done in order to explain the results of both Nigeria and Libya.
Kazakhstan is highly influenced by the Russian Federation. Therefore, we shall take a look at the results of the Russian Federation. If Russia has a negative number and is likely to suffer from the natural resource curse, then Kazakhstan is likely to suffer from the natural resource curse too. Indeed, Russia has a negative number of 9. This number implies the presence of the Dutch Disease in Russia.
In order for the test to be as accurately as possible we shall perform an expanded version of the described test. For Kazakhstan we shall perform the test for each year, as of 1995, in which the Human Development Report has been published.
Graph 4: The GDP per capita rank minus the HDI rank for Kazakhstan
Source: The graph is based on data from the Human Development Reports as published by the United Nations Development Programme
Graph 5: The GDP per capita rank, and the HDI rank for Kazakhstan
Source: The graph is based on data from the Human Development Reports as published by the United Nations Development Programme
From 1995 until 2000 Kazakhstan had a high positive difference between the GDP per capita rank and the HDI rank. In these years, the lowest number was reached in 1996 (6), and the highest in 1995 (15). From 1995 until 1997 Kazakhstan suffered from a steep decline in both the GDP per capita and HDI. This is due to the difficulties the country faced of transition from a planned economy to a market-oriented economy. In the years 1998 and 1999 the level of the HDI recovered. The same happened for the level of GDP level from 1998 until 2001. From 1995 until 2001, due to the high variations of the levels of GDP and HDI, the data in is unreliable.
As of 2001 both the GDP per capita and HDI stabilize (making the data more reliable). Therefore, for this test we shall take the results as of 2001 into account. With the exception of 2004 (a number of 4) and 2007/2008 (a number of 1), Kazakhstan had a negative GDP per capita minus HDI rank. The lowest number (-5) was reached in both 2003 and 2006. However, all numbers stay relatively close to the zero-line.
Overall, this test shows no clear sign of the presence of the natural resource in Kazakhstan. In five of the seven years Kazakhstan had a negative number, but only in 2003 and 2006 the number was relatively far below zero. Therefore, the results are not strong enough to show a clear signal of the presence of the natural resource curse.
Conclusion
A Granger causality test was performed in order to test for the presence of the Dutch Disease in Kazakhstan. This test checks whether it is possible to claim that the Disease is present in Kazakhstan. It proved that it is impossible to claim that the Dutch Disease is not present in Kazakhstan. Therefore, the results point strongly towards the presence of the Dutch Disease in the main export sectors of Kazakhstan. A feature of the Dutch Disease is an appreciation of the national currency due to the export of oil. Therefore, we wanted to know whether the tenge, the national currency of Kazakhstan, has appreciated due to the trade of oil. For this reason, we compared the real exchange rate including and excluding the oil trade. Overall, Kazakhstan faced an appreciation of the real exchange rate, but not all data clarifies this. The results do show an appreciation of the national currency, but the results are not particularly strong. However, we can conclude that the currency did appreciate due to the export of oil. By looking at both tests, it is safe to conclude that Kazakhstan does suffer from the Dutch Disease.
We went further by putting the economic growth and the economic development of Kazakhstan side to side. A positive difference between the two points implies the presence of the natural resource curse. This test showed no clear sign of the presence of the natural resource. Kazakhstan is likely to suffer from the Dutch Disease, but it goes too far to conclude that Kazakhstan suffers from having oil. Kazakhstan benefits from having oil, but it should put in more effort to diminish the potential harm of the Dutch Disease.
Further research
The Granger causality tests showed whether the production of oil influences the production of other sectors. Similar tests should be performed to check whether the production of other minerals (e.g. natural gas) can also be found to be accountable for the presence of the Dutch Disease in Kazakhstan.
