Kazakhstan: long-term economic growth and the role of the oil sector.

Kalyuzhnova, Yelena ; Patterson, Kerry

INTRODUCTION

Over the last 10 years mineral wealth has played a significant role in the economic development of resource-rich countries, which have benefited greatly from their hydrocarbon reserves. Kazakhstan is an exemplar in this respect, and has successfully overcome a transition period from a central planned to a market-driven economy; overall, the Kazakhstani government has provided institutional and policy support to effect this transition and has delivered more in terms of economic prosperity than other Central Asian countries and, possibly, Russia.

This study examines the relationship between long-term economic growth and oil exports for Kazakhstan and, in so doing, it addresses two related questions for an 'oil-rich' country: (i) does 'oil matter' in the long run; (ii) is oil a resource curse, a blessing or neither? We first provide a brief review of the key arguments in the 'curse versus blessing' debate. Thereafter, we develop a narrative background to recent economic developments in Kazakhstan, which includes an overview of relevant institutional and policy initiatives. This narrative tends to support the view that, overall, these developments have been positive, a view that is supported by an examination of the governance indicators that have been developed by The Worldwide Governance Indicators (WGI) Project (2015), see also Kaufmann et al. (1999). The narrative analysis is then complemented by an econometric model that directly addresses the key questions in terms of cointegration and the estimated steady state between growth in real gross domestic product (GDP) and oil exports; the empirical model then provides the basis of some relevant long-term projection scenarios.

The article is organised as follows. The section 'Natural resources and economic growth' briefly reviews the literature on the subject of natural resources and economic growth. In the section 'Kazakhstan: Recent economic development' we examine and evaluate economic development before the 2008 economic crisis and beyond, to understand the evolution of the economy of Kazakhstan. The section 'Institutional and policy factors' discusses the roles of institutions and policies in providing a framework for the successful extraction of oil resources, the encouragement of foreign investment and the general improvement of governance indicators for Kazakhstan. The empirical analysis is reported in the section 'Long-term economic growth' and provides the basis for the scenario projections. The section 'Conclusions' concludes the article.

NATURAL RESOURCES AND ECONOMIC GROWTH

The role of resource abundance in economic growth is the subject of an ongoing debate, see, for example, Gelb (1988), Karl (1997a, b), Wood (1999), Auty (2001), Frankel (2010) and Mohaddes and Pesaran (2013), with mixed evidence about the relationship between resource abundance and its effect on economic growth. We briefly consider, in turn, the arguments that resource abundance is, respectively, bad and good, for economic growth.

Natural resources are a resource curse

Recent history provides examples where the role of the raw materials sector has been blamed for the underdevelopment or low growth rates of some economies (Bravo-Ortega and de Gregorio, 2005). (1) The worst example of such performance is Nigeria (Bevan et at, 1999), where the increase in oil revenue did not make a positive impact on income per capita. The experience of Nigeria is not an exception. During the last several decades other resource-rich countries, including Qatar, Venezuela, Iraq and Kuwait, have had similar experiences of negative growth.

As early as the 1950s it was noticed that there was a correlation between slow growth and a decrease in the relative prices of exports of the natural resources industries. The Singer--Prebisch (1959 [1962]) hypothesis on the declining terms of trade, states that the prices of export-orientated commodities decrease over time relative to the prices of imported manufacturing goods. This leads to a situation where the available revenue will be less and less able to buy the manufacturing output, hence preventing investment and economic development. The 1970s and 1980s were characterised as an era of adoption of import-substitution policies as a pathway to increase the competitiveness of manufacturing, although this strategy was often inefficient and caused further problems in the manufacturing sector.

Concerns have also been raised regarding 'Dutch disease', caused by an appreciation of the real exchange rate driven by a boom in natural resources, the negative consequence of which is the shrinking of manufacturing exports and production, see Gylfason et al. (1999), Sachs and Warner (2001). The validity of the Dutch disease hypothesis has, inter-alia, been tested by examining the relationship between real oil prices and real exchange rates; for example, Jahan-Parvar and Mohammadi (2011) used a sample of 14 oil-exporting countries, with a co-integration approach, to demonstrate the existence of a stable relationship between real exchange rates and real oil prices in all countries, suggesting strong support for the Dutch disease hypothesis.

The transmission mechanisms directly linking natural resource exploitation and economic growth suggested by Gylfason (2001) were: the Dutch disease, rent-seeking, government mismanagement and low levels of human capital. There have been attempts to identify empirically the potential channels of transmission for the resource curse (also known as the paradox of plenty) by regressing institutional quality, human capital and so on, on natural resource dependence, and then calculating the indirect effects of resource dependence on growth from the coefficients of these intermediate variables on growth, see Lay and Mahmoud (2004), Papyrakis and Gerlagh (2004) and, for a critique of this approach, see Van der Ploeg (2011).

The price volatility of raw materials has also been blamed for the increasing difficulties in the fiscal management of export revenues and, ultimately, limiting growth. Further arguments have been developed that foreign companies investing into the raw materials sector of the resource-rich countries only benefit themselves and their countries, leaving behind the host countries, see Singer (1950) and Humphreys et al (2007).

In the assessing the likelihood of the resource curse, Isham et al. (2003) suggest a distinction between point and diffused resources, where the former relate to natural resources extracted from a narrow geographic or economic base, such as oil, minerals, coffee and cocoa, and the latter to resources, such as agricultural produce, which are geographically or economically diffused, with the latter contributing more robustly in sustain economic growth.

Using cross-section regressions, Sachs and Warner (1995, 1999, 2001) found a negative relationship between economic growth and natural resources. They supported this relationship with measures of resource abundance, such as the share of mining production in GDP, land per capita, and the share of natural resource exports in GDP and found that an increment of 1 standard deviation in the contribution of natural resources exports to GDP implied a lower rate of growth of the order of 1 % point each year.

Natural resources are a blessing

In contrast to the view that natural resources are a curse, there are several reports of development where natural resources are considered to have been the driver of economic growth, Wright (1990), Sarraf and Jiwanji (2001). Gylfason (2001) reports that of 65 resource-rich developing countries, only Botswana, Indonesia, Malaysia and Thailand managed to achieve long-term investment exceeding 25% of gross domestic product (GDP), and an average GDP growth above 4% p.a. The success of these three resource-rich Asian countries could be explained by the economic policies of their governments, specifically by economic diversification and industrialisation, see Van der Ploeg (2011).

Another example of economic diversification policy can be found in the United Arab Emirates (UAE), see Fasano (2002), where oil and gas wealth have been used to improve standards of living, especially in the provision of free health care, education and the modernisation of infrastructure. The UAE government has been aware of the depletion of its natural resources and, therefore, the creation of its petrochemical and fertilizer industries has become a priority. In contrast, Dubai took the route for diversification into tourism, finance, light manufacturing and telecommunications.

An analysis of a number of individual case studies suggests that the consistently successful countries among the resource-rich economies are the developed countries, such as Norway, the UK, Canada, Sweden and Finland, see Innis (1956), Andersen (1993), Kemp and Stephen (2005), Al-Kasim et al. (2006), Larsen (2006) and Blomstrom and Kokko (2007). An explanation of such success can be based on a low level of corruption and well-developed institutions that help to form and implement efficient governmental policies towards the oil and gas sectors, and make coherent managerial decisions regarding economic development, see Acemoglu et al (2005). The industrialisation in these countries was based on the initial export of raw materials, which later created linkages to other sectors of the economy, especially the manufacturing sector.

Other research, see Maloney (2002), Stijns (2005), Lederman and Maloney (2007), using different measures, such as net resource export per worker or reserves per capita, has found a positive effect on economic growth 'regardless of econometric technique ... several plausible indicators of the incidence of natural resource exports seem to have a positive rather than a negative effect on subsequent economic growth. Put bluntly, there is no resource curse', Lederman and Maloney (2007, p. 3). Continuing the thought further, Mohaddes and Pesaran (2013) have argued that it is the volatility in oil revenues and the government's inappropriate economic and political responses to these volatilities, which are the curse and not resource abundance in itself.

Further studies have tended to confirm these results, bringing good news for the resource-rich countries, namely that their natural resources can have a positive effect on economic growth, depending on the optimal use of them, see Cavalcanti et al. (2011) and Kurtz and Brooks (2011). In September 2013, an International Monetary Fund (IMF) conference on 'Harnessing Natural Resource Wealth for Inclusive Growth and Economic Development', highlighted that the key challenges faced by all countries in ensuring that resource wealth contributes in a sustained and inclusive fashion to growth and higher living standards for all, are macro-fiscal issues and ways to promote economic diversification (Singh, 2013). The role of strong institutional settings, with powerful enforcement mechanisms, is an essential element in sustaining long-term economic growth. In their work, Mehlum et al. (2006) argued that natural resources are negatively associated with economic growth in the case of weak institutions.

The research evidence also suggests that avoiding the resource curse requires sound and forward-looking policy in addition to the existence of strong institutions. This gives a primary role to the governments of resource-rich countries, which typically play a dominant role in the economy through their control of natural resources and associated income. Decisions regarding taxation expenditures, including public investment, fiscal deficits, savings mechanisms such as the role of sovereign wealth funds, and the governance framework, will have important consequences--not just for today but also for future generations--with the potential to ensure successful development or alternatively to destabilise the economy, see Hannesson (2005).

Gylfason et al. (1999) suggested that the natural resources sector needs less human capital than other productive sectors, so that diversification is a key area for economic policy; and the fact that the resource is exhaustible pushes the governments to think ahead. Singh (2013), Director of the Asia and Pacific Department, IMF noted: 'for inclusive growth in addition to wise use of the resources it is imperative that backward and forward linkages are developed between the natural resource sector and the wider economy. Achieving this objective involves financial sector deepening, building infrastructure, enhancing human capital, and promoting the agricultural sector'.

KAZAKHSTAN: RECENT ECONOMIC DEVELOPMENT

A number of papers have analysed economic development in the Caspian Sea Region, and Russia since its independence, with the common theme of highlighting the dependence of these economies on mineral wealth (2). The last (20) years have brought significant changes to their economic development, with the hydrocarbons sector in particular giving these economies a new shape, for example in the strategic importance of Russia, Kazakhstan and Azerbaijan in world energy markets.

The development of the oil and gas sector in Russia and the Caspian region economies has been particularly noticeable in the case of Kazakhstan from 2001 through to 2014, especially the increasing importance of oil and gas sector revenues in total exports. Starting from a share comparable to Russia in 2001, with 50% for Russia and 53% for Kazakhstan, this share had increased to 77% by 2014, compared with 56% for Russia. Kazakhstan is clearly now a 'resource-based economy' in the sense in which this term has been used by Ahrend (2006), with the share of oil and gas revenues in total exports regularly exceeding 40% and generally of the order of 70%, see Table 1.

Kazakhstan has attracted the largest flows of FDI in the Caspian Region, see Figure 1. In part this has resulted from the government's attempt to adopt a policy of continuous improvement of the laws on investment and taxes. Overall, since 1993, US$171.2 billion of FDI has been raised in Kazakhstan, with an FDI growth rate of 27% on average for the last 20 years, which compares with an FDI growth rate in the UAE of 26%, in Brazil of 15% and of 14% in Turkey. The geographical composition of the FDI in Kazakhstan is represented by 120 countries, the main contributors being the Netherlands, China, Switzerland, the United States and France.

[FIGURE 1 OMITTED]

This impressive picture is mainly because of investment into Kazakhstan's natural resources, although the composition of the FDI in other resource-rich Caspian economies is similar (Kalyuzhnova, 2008). Overall, during 1993-2012, 34.8% of FDI was invested in geological surveys and exploration, 30.2% in the production of raw materials, with the largest share in hydrocarbons, and 10.2% in final manufacturing production.

In order to ensure that the non-natural resource sectors of the economy were attracting sufficient FDI, the Accelerated Industrial and Innovative Development Programme for 2010-2014 (AAIIDP) was adopted by the Kazakhstani government, with the first results reported in 2013. According to Assett Issekeshev, the Minister of Industry and New Technologies, during the implementation of AAIIDP, the share of FDI into crude oil declined by 18%, while at the same time the share of FDI into manufacturing production, which includes such sectors as machine building, chemicals, petrochemical industries, and food processing, increased by 8%. Moreover, for the first time FDI was attracted into high-tech industries, such as pharmaceuticals, computer manufacturing, electronic and optical products. Overall, at the present time, the government is working together with foreign investors from 80 countries on more than 400 initiatives; 81 projects (with a value of $8.8 billion) of these initiatives have already been completed. As of 31 December 2014, FDIs in Kazakhstan reached 132.6 $ bn, primarily in the oil and gas sector. Kazakhstan is widely considered to have the best investment climate in the region.

The 1990s were very difficult years for the economic development of Kazakhstan; however, from 1999 onwards, following a substantial currency devaluation, an increase in proven oil reserves and an upturn in oil prices, Kazakhstan experienced stronger economic growth starting in 2000. Until 2007 economic growth was strong, but it slowed somewhat as a result of stress in the financial sector and the impact of this on the construction industry. 2007 was the eighth consecutive year of real GDP growth in excess of 7% p.a. Sharp increases in oil prices in the early 2000s, and subsequently the growth in non-oil revenues, allowed for a further substantial expansion in budgetary expenditures and in parallel to this a sizable increase in the overall fiscal surplus (Kalyuzhnova, 2008).

From July 2007, Kazakhstan's banking sector was affected by the more generally widespread financial crisis: the banks had borrowed heavily on the international capital markets and the speculative bubble in the residential property market burst. Falling oil prices and the deepening economic crisis also had a significant impact on the wider Kazakhstani economy. The 2007-2009 crisis was the first stress test for the newly established (since 2000) National Fund of the Republic of Kazakhstan (NFRK). By 2008, the NFRK had become one of the world's largest sovereign wealth funds, alongside those of the oil-producing countries of the Middle East and Russia. NFRK has worked for Kazakhstan overall, transfers to the fund did help mitigate the pressure on Kazakhstan's economy during the upswing, and left it better prepared for the 2007-2009 shock, see Kalyuzhnova (2011).

INSTITUTIONAL AND POLICY FACTORS

In this section, we make a distinction between the institutional and policy frameworks as they have affected the oil industry and the quality of governance in Kazakhstan. To give an example of the distinction, the rule of law is an 'institution' that provides the contractual basis for economic participation between companies and countries; whereas a particular law, for example, the 'Law on Investment' (detailed below), is the result of a policy to treat domestic and foreign investors equally. We consider each of these factors in turn and in section 4.3 consider the good governance indicators for Kazakhstan produced by the World Governance Indicators project, see WGI (2015).

Institutional factors

A strong institutional setting has been shown to be a necessary condition for the success of a national economy, with poor institutions having a negative effect on a country's growth. There is cross-country evidence that has demonstrated a significant negative impact of natural resources on income per capita after controlling for institutional quality, trade openness and geography. In cases where countries have poor institutions and a low degree of trade openness, the negative implications are quite drastic, see Arezki and van der Ploeg (2010); therefore, the strategy of trade openness and improving institutional quality could turn natural resources into the country's blessing.

Over the last 20 years, during which Kazakhstan has been formed as a state, the economy has been shaped and developed and the nation has become more interested in the terms and conditions of the oil and gas contracts concluded at the beginning of the 1990s. The verdict was that a number of them had unfavourable terms with regard to Kazakhstan. Why did this happen? This can be explained by a number of reasons, starting from the economic crisis of the early 1990s, the inexperience of the Kazakhstani negotiators, as well as the high risk associated with early investments into fields with uncertain reserve bases (Suleimenov and Osipov, 2010).

At the present time Kazakhstan is facing a challenge that will determine the success of long-term economic growth: namely the creation of an institutional framework, which is required by economic maturity and the growth of markets, to encourage external investing partners by providing a framework enabling transactions to take place in a secure and orderly manner. All players will know that the decisions they take and the contracts they make will be protected by law and enforced if necessary. For all the participants in the Kazakhstani market it is important to have an institutional framework that is rational, and provides a guarantee of economic stability and certainty. This can be achieved by good governance and sound economic policymaking.

Policy

The route through which the government operates its energy policy is the Ministry of Oil and Gas, which effectively manages the country's hydrocarbons industry. Overall, energy policy is based on sound economic and environmental practices, as well as good legal practices and technologies, which is how the government positioned itself in its relations with foreign and domestic investors. For example, during the investment conference of the 47th annual meeting of the Asian Development Bank in Astana (May, 2014), the Kazakhstan government highlighted that they would amend the corresponding laws to create a more favourable investment climate; the government also removed the barriers faced by foreign investors in the development of their activity in Kazakhstan and prepared a comprehensive package of incentives for foreign investors.

Although, overall, Kazakhstan's tax laws are among the most comprehensive in the CIS; in January 2009, Kazakhstan adopted a new Tax Code that lowered corporate-income and value-added taxes, replaced royalty payments with a mineral-extraction tax, and introduced excess-profits and rent taxes on the export of crude oil and natural gas. 'Due to the new Tax Code, new contracts cannot include any provisions on stability of contracts (stabilisation clauses). In addition, the Code only expressly preserves tax stability in Production Sharing Agreements and contracts approved by the RK President, meaning that other contracts made before the Code may not be stabilised for tax purposes', Suleimenov and Osipov (2010, p. 5).

On 8 December 2004, amendments to the Petroleum Law and Subsurface Law became effective, and, finally, the Production Sharing Agreement Law (applicable to the Caspian and Aral Seas). These changes have had a substantial impact on the petroleum industry: they reflect the government's policies with respect to increased participation of the National Oil Company (in production); greater attention to the use and development of local content and 'high technologies', a change in government (tax) take; and increased regulation and oversight.

During the 2000s, the Kazakhstani government addressed the problem of attracting domestic investors by creating favourable conditions for them. In 2003, it adopted the Law on Investment, which established a single investment regime for domestic and foreign investors and providers, which guaranteed the stability of existing contracts, with the qualification that new contracts will be subject to amendments in domestic legislation, certain provisions of international treaties, and domestic laws dealing with 'national and ecological security, health, and ethic'. This law contains incentives and preferences for government-determined priority sectors, providing customs duty exemptions and in-kind grants, but the weakness of this law is that it excludes all the norms regarding foreign investors. The Law's narrow definition of investment disputes, its lack of clear provisions for access to international arbitration, and certain aspects of investment contract stability guarantees, are also points of concern of foreign investors.

Several major acts of legislation have had an impact on oil and gas foreign investments in Kazakhstan: the Decree of the President of the Republic 'On Subsoil and Subsoil Use', in force from 1996, and the Decree of the President of the Republic 'On Petroleum', in force from 1995. One of the main principles of the Law on Subsoil Use is the creation of favourable conditions for foreign investment. In summary, the tax, custom and procurement regimes have also played an important role in investment in the oil and gas sectors, specifically: (i) the 2003 Customs Code and the Customs Code of the Customs Union, which came into force in July 2010; (ii) the Tax Code; (in) the Law on Currency Regulation and Currency control; and (iv) the Law on Government Procurement. These laws provide for non-expropriation, currency convertibility, guarantees of legal stability, transparent government procurement and incentives for priority sectors.

The Kazakhstan government is continuing its policy initiative for the next 5 years with the 'Concept for Industrial and Innovative Development, 2015--2020' programme, which is aimed at continuing to attract foreign and domestic investments and further developing an industrial strategy aimed at diversifying Kazakhstan's economy away from its overdependence on raw materials. In order to facilitate the work of foreign investors, especially in targeted non- extractive industries, the government has announced a set of industries that feature tax waivers and simplified procedures for acquiring visas and work permits.

Governance and economic indicators

The relationship between governance and economic growth has been the subject of some recent research, with the WGI, now produced annually by Kaufmann et al. (1999), particularly influential in this respect. The underlying components of the WGI indices (such as surveys) are organised in clusters into six indicators designed to cover the following aspects of institutions and policies: voice and accountability, political stability, government effectiveness (GE), regulatory quality (RQ), the rule of law (RoL) and control of corruption (CC). The governance indices are constructed to have the range (-2.5, +2.5), with improvements indicated by a numerical increase in the index.

The indicators of particular concern here are the two pairs comprising GE and RQ, capturing policy-based issues, and RoL and CC, which have a greater institutional focus. Of the first pair, Kaufmann et al. (1999) note: The main focus of this (GE) index is on the inputs required for the government to be able to produce and implement good policies and deliver public goods.' The second index (RQ) relates to 'perceptions of the burdens imposed by excessive regulation in areas such as foreign trade and business development'. The GE index relates to the ability to provide public goods, such as infrastructure, whereas the RQ index focusses on the policies. Institutional quality is assessed by the RoL and CC indices. The four relevant indicators are shown in Figure 2.

Considering the policy indicators, GE and RQ, after a dip in 2002, there was a general improvement from 2002 to 2009, followed by slight declines more recently; note that these two indicators tend to move quite closely together. The institutional indicators, RoL and CC, also show a general improvement, but from a slightly lower base than for GE and RQ. Overall, the most pronounced increase has been in the GE and RoL indicators. Generally, the indicators provide support for the narrative view expressed above that there have been general, if not uniform, improvements in the institutional and policy aspects of good governance, which have provided a framework conducive to growth and attractive to overseas investors.

This view is supported by reference to indicators produced by a number of other agencies; for example, the Index of Economic Freedom, IEF (2015), is an index based on 10 underlying components related to the following four sections: the Rule of Law (RoL); Limited Government (LG); Regulatory Efficiency (RE); and Open Markets (OM), the lower limit for each index 0 and the upper limit is 100, an increase in the numerical score is intended to indicate greater 'freedom'. The IEF adds to the WGI in the sense of explicitly scoring for freedoms in trade, investment and finance in the Open Markets area, which should encourage and enable overseas investment. The overall IEF and the indices for the four sections are shown in Figure 3, where the aggregated section indices average the respective components. All of the section indices point to a steadily improving environment for economic development and, as a result, a steady increase in the overall IEF. In a comparison of the Caspian Region economies, Kazakhstan ranks above Azerbaijan and Turkmenistan, which have IEF overall scores in 2014 of 61.3 and 41.4, respectively, compared with 63.7 for Kazakhstan. Interestingly, Kazakhstan, starting from a much lower base in the 1990s, now ranks close to, and above, some Western countries, for example in 2014, Italy had an overall IEF of 60.9, France 63.5, compared with 63.7 for Kazakhstan.

[FIGURE 2 OMITTED]

More specific indices are available from the Global Competitiveness Dataset, World Economic Forum (2015), although the available time span for Kazakhstan data is shorter than for the WGI and IEF. Of these their value added, compared with the WGI and IEF indices, is in the indices for technological progress, which are particularly relevant and comprise three separate indices relating to: availability of latest technologies, firm-level technology absorption and FDI and technology transfer. The technology indices are indicative of the developments that have enabled Kazakhstan in the production and export of its oil and gas resources, combined with diversification into other industries, each of which has been facilitated by FDI and the effective transfer of technology.

[FIGURE 3 OMITTED]

These data are graphed in Figures 4, 5 and 6, one for each index, a numerical increase in the index is an improvement and the upper limit is 7.0. By way of a numeraire, we also include the index values for the Russian Federation, which separated from Kazakhstan at the same time, but is generally second in all of these developments of its economy. The indices support the narrative background, although not without qualification, that since the late 1990s, Kazakhstan has developed its natural resources and other industries in ways that are likely to avoid the problem of the resource curse, especially with its ability to benefit from the transfer of technology in the natural resource and other industries. In the following section, we examine this argument in more detailed from an econometric point of view.

LONG-TERM ECONOMIC GROWTH

The narrative background of recent economic developments in Kazakhstan suggests that after transitional adjustments in the 1990s, following separation from the USSR, growth in FDI, improvements in governance and the design of oil sector policy, there was a sustained growth in oil exports and GDP. Against this background, this section considers a modelling and estimation framework, designed to address the key questions set out in the first two sections: is the abundance of oil a benefit to longer term economic development: is it a curse, a blessing--or neither? What are the implications of the growth in oil exports for longer-term economic growth?

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

The modelling framework we adopt is because of Esfahani et al. (2014), hereafter EMP (2014), a key point in their approach, with which we agree, concerns the role of political systems and institutions, they note: 'While the countries under consideration have very different political systems and institutional economic arrangements, our theory suggests that these differences can only affect the estimates of the parameters rather than the nature of the long-run relationship that holds between outputs (domestic and foreign) and oil export revenues'., (ibid., p. 100). What matters is the oil reserve-to-production ratio. Moreover they argue that cross-section analysis captures short-term deviations from equilibrium and can, therefore be misleading.

The modelling framework

The modelling framework we adopt is because of Esfahani et al. (2014), hereafter EMP (2014), whose theory and empirical results suggested a key distinction between countries with reserves, and hence future oil income, over the short to medium term (approximately less than 10 years) and over the longer term, broadly greater than 10 years. In economies, such as Kazakhstan, in which oil income is growing at least as fast as the natural rate of growth, [g.sup.n] = g + n, where g and n are, respectively, the rates of growth of technical progress and of the domestic population, EMP derive the following long-run relationship:

[y.sup.ss.sub.t] = [[mu].sup.*] + [[psi].sub.1][y.sup.*.sub.t] + [[psi].sub.2] [x.sup.0.sub.t] + [[gamma].sub.y]t + [[epsilon].sub.y,t] (1)

Where [y.sub.t] is the log of GDP in constant prices; [y.sup.*.sub.t] is the are the log of overseas output, in practice a weighted average of the real GDPs of key trading partners; [x.sup.0.sub.t] is the log of real oil income adjusted for exchange rate movements and the domestic price level, that is [x.sup.0.sub.t] [equivalent to] log ([RER.sub.t]([P.sup.o.sub.t] [X.sup.bo.sub.t])), where [RER.sub.t] [equivalent to] [E.sub.t]/[P.sub.t] is the real exchange rate, [E.sub.t] is the nominal exchange rate expressed as the number of domestic currency units per US$ (3), [P.sup.o.sub.t] is the unit price in US$ per barrel of oil, [X.sup.bo.sub.t] is domestic oil exports in 000s of barrels and [P.sub.t] is a domestic price index, here the CPI which is also used for the deflation of nominal GDP and t is a time trend. For future reference we define [rer.sub.t] [equivalent to] log ([RER.sub.t]) and [xo.sub.t] [equivalent to] log ([P.sup.o.sub.t] [X.sup.bo.sub.t]).

The long-run relationship in Equation 1 relates real domestic GDP (output), [y.sub.t], to 'overseas' output [y.sup.*.sub.t] with an elasticity of [[psi].sub.1], to real oil income with an elasticity of [[psi].sub.2] and a time trend, with coefficient [[gamma].sub.y] the latter term is absent in the case of common deterministic trends among the variables, which is referred to a co-trending. The model enables three important distinctions to be made, which are relevant to the key statements raised in the Introduction concerning the importance of oil to output, specifically: (i) oil does not matter, it is neither a resource curse nor blessing; (ii) oil is a 'blessing'; and (iii) oil is a 'curse'. These distinctions give rise to three hypotheses, as follows:

[H.sub.0] [[psi].sub.2] = 0 (oil does not matter);

[H.sub.A;1] [][psi].sub.2] > 0 (oil is a blessing);

[H.sub.A,2] [[psi].sub.2] <0 (oil is a curse).

Empirical results

The theory-consistent relationship in Equation 1 is a steady-state, relationship, which needs to be embedded in a dynamic estimation framework that enables transient movements to be captured. An appropriate estimation framework for the sample period in this case, is the autoregressive distributed lag (ARDL) method because of Pesaran and Shin (1999), see also Pesaran et al. (2001). (4)

The ARDL relevant to Equation 1 is as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)

The ARDL is referred to by the lag orders on the component variables on the right-hand side of Equation 2, ARDL(p, [q.sub.1], [q.sub.2], [q.sub.3]). Next, define [phi] (1) [equivalent to] [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. The long-run coefficient for [y.sup.*.sub.t] is given by [[psi].sub.1] = [[beta].sub.1](1)/[phi](1); Equation 2 allows separate transient adjustments from the real exchange rate, RER, and the US$ value of oil income, [xo.sub.t], while maintaining the steady-state restriction that [[psi].sub.2] = [[beta].sub.2](1)/[phi](1) = [[beta].sub.3](1)/[phi](1). The steady-state level of [y.sub.t], [y.sup.ss.sub.t], is then as given by Equation 1, where [[mu].sup.*] = [mu]/[phi] (1) and [[gamma].sub.y] = [beta]/[phi](1). (Lag lengths are selected using AIC.)

The ARDL model of Equation 2 can be reparameterised into the following error correction model (ECM), which captures both the transient dynamics and the steady-state solution:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

Where[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], which is interpreted as the deviation from the (estimated) steady state. The interpretation of the ECM is well known, with the key coefficient being that on the lagged [ecm.sub.t], [gamma], which is expected to be negative and governs the speed of adjustment when out of equilibrium.

While the overall sample period available with quarterly data was 1995ql to 2013q4, it is clear from narrative of the economic development of Kazakhstan that consideration would have to be given to obtaining a relatively homogenous period for estimation. There were four factors supporting the view of a structural break in 1999, which was best accommodated by treating the sample 1999q4 to 2013q4 as a relatively homogenous period. First, the substantial currency devaluation of the Kazakhstan tenge (Kt) in 1999, which started the year at 85Kt:US$ and ended at 140Kt:US$, second, as indicated above, a number of structural changes in the financial infrastructure of Kazakhstan; the developments in institutions and policies associated with the post-1999 period; and the general lack of output growth in the 1990s, compared with the 'take-off in 1999.

The proxy for 'overseas' output, y* was based on Kazakhstan's main trading partners, Russia, the EU and China, which together account for approximately 80% of trade (imports and exports), with weights of 0.4, 0.4 and 0.2, based on the mid-point of the sample. The estimates of the steady-state coefficients, together with diagnostic statistics for the corresponding ARDL equation, are reported in Table 2. The ARDL is well-specified in terms of the diagnostic statistics and the estimated steady-state coefficients are consistent with the theoretical model. (A prior regression included a trend, as suggested by Equation 1, but the coefficient was not significant, thus the hypothesis of cotrending was not rejected.) The bounds test for cointegration (see Pesaran et ai, 2001), in both the F and W versions, strongly support cointegration, with sample values well in excess of the 95 % upper bound. Forming the corresponding ECM, the coefficient on the estimate of the lagged disequilibrium, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] with a t statistic of -4.36 and, hence, a p-value of 0, indicating a relatively fast recovery from disequilibrium.

Conjectural projection

In this section, we consider a long-term conjecture or 'scenario' to assess the growth potential in real GDP based on the estimated version of Equation 1, which conditions the growth of output on the growth of 'world' output, [y.sub.t], and oil exports, [x.sup.o.sub.t]. It is relevant to consider projections using the steady state equation as dynamic adjustments are transient. The empirical results (see section Table 2) support the following general specification of steady state:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)

Where [g.sup.ss.sub.y] is the steady-state growth in [y.sup.ss.sub.y] (home-country GDP) and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and [g.sup.o.sub.ss] are the steady-state growth rates in [y.sup.*.sub.t] ('overseas' GDP) and [x.sup.o.sub.t] (real value of oil exports), respectively.

The real value of oil exports can be decomposed into its component parts as follows: [x.sup.o.sub.t] [equivalent to] [p.sup.o.sub.t] + [xo.sub.t] + [rer.sub.t], where, [p.sup.o.sub.t] [equivalent to] ln([P.sup.o.sub.t]), [x.sup.o.sub.t] [equivalent to] 1n ([X.sup.o.sub.t]) and [rer.sub.t] [equivalent to] 1n ([E.sub.t]/[P.sub.t]); thus, the growth in the real value of exports, [g.sup.o.sub.t], can be (identically) decomposed into the growth rates of the price of oil, [g.sup.po.sub.t], the number of barrels exported, [g.sup.po.sub.t] and the real exchange rate, [g.sup.rer.sub.t], so that: [g.sup.o.sub.t] [equivalent to] [g.sup.po.sub.t] + [g.sup.po.sub.t] + [g.sup.rer.sub.t]. A scenario projection will need to consider each of these components and the growth in overseas output, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII].

Over the sample period the price of oil in nominal terms has increased on average by just under 3 % p.q,; however, the recent decline in oil prices has led to a substantial revision to the potential course of oil prices over the coming decade. The Department of Energy and Climate Change (DECC, 2014), has recently summarised a number of sources, for example the International Energy Agency (IEA) and the US Energy Information Administration (EIA), and provided projections of the price of a barrel of oil. The 'central' projection for 2014 is based on the Brent futures curve, where the projections for 2015 to 2020 are an average of eleven expert projections and thereafter they are based on the IEA and EIA projections of constant growth.

The DECC projections, which are in terms of 2014 prices, US$2014, are converted to nominal prices using the IEA default rate of 2.3% p.a. in the numeraire price index and are shown in levels and logs in Figure 7. The oil price is assumed not to return to its 2014 level (in 2014 prices) until 2022 and thereafter increase to 135US$2014 in 2025, which corresponds to approximately $220 in nominal prices. The projection shows that the oil price declines through to 2017 and then grows at an assumed constant rate of 4.6% p.a. in nominal terms thereafter.

[FIGURE 7 OMITTED]

Although the three variables [rer.sup.t], [p.sup.o.sub.t] and [x.sup.o.sub.t] do not cointegrate (which affects the modelling of levels of the variables), we can exploit the relationship between these variables in differences, where [g.sup.po.sub.t] = [DELTA][p.sup.o.sub.t], [g.sup.bo.sub.t] [equivalent to] [DELTA][xo.sub.t] and [g.sup.rer.sub.t] [DELTA][rer.sub.t]. A scenario projection will need to consider each of these components to model the required growth rates. Empirically, we find that there is recursive causal structure, in which causality in differences runs from [g.sup.po.sub.t] to [g.sup.bo.sub.t], and then from [g.sup.po.sub.t] and [g.sup.bo.sub.t] to [g.sup.rer.sub.t]; thus, a projection of [g.sup.bo.sub.t] can be conditioned on the projected path of [g.sup.po.sub.t] and then the projection of [g.sup.rer.sub.t] can be based on the projected paths of [g.sup.po.sub.t] and [g.sup.bo.sub.t]. To illustrate, given the projected path of [g.sup.po.sub.t], the projected value of [g.sup.bo.sub.t] converges to 0.0205, that is 2.05% p.q, which is an estimate of the steady-state value [g.sup.bo.sub.ss] of [g.sup.bo.sub.t]; then, given the projected paths of [g.sup.po.sub.t] and [g.sup.bo.sub.t], the estimated [g.sup.rer.sub.t] converges to [g.sup.rer.sub.t] = - 0.0158, that -1.58% p.q. The DECC projections of the oil price imply [g.sup.po.sub.ss] = 0.0113, then [g.sup.o.sub.ss] [equivalent to] [g.sup.po.sub.ss] + [g.sup.bo.sub.ss] + [g.sup.rer.sub.ss] = 0.0113 + 0.0205 - 0.0158 = 0.016, that is 1.60% p.q (6.56% p.a.). This is less than the sample average of 3.16% p.q, but reflects the changed view on the consensus course of the oil price.

A scenario projection also requires a projection of overseas output, here comprising the weighted GDPs of Russia, the EU and China. Recent projections of World GDP by region prepared by the EIA (2014) suggest annual growth rates over the period 2015-2025 of 3.0% p.a., 2.45% p.a. and 6.0% p.a., respectively, for these countries. Applying the weights 0.4, 0.4 and 0.2, gives a projected weighted growth rate of 3.4% p.a., which is less than the average over the sample period of 5.4% p.a. and reflects the slow recovery following the 2008 financial crash and a slowing in the growth rate of China's GDP. At a quarterly rate [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] = 0.0083.

[FIGURE 8 OMITTED]

The steady-state growth projection [g.sup.y.sub.ss] can now be constructed from its components and the estimated counterpart of Equation 1; these depend on the projected growth rates and the coefficients, thus: [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], see Table 2. The projected steady-state growth rate after 2018 is [g.sup.y.sub.ss] = 0.53 x 0.0083 + 0.62 x 0.0160 = 0.0143, that is 1.43% p.q giving an annualised rate of 5.85% p.a. Although this is high by the standards of industrialised countries, the rate is much lower than has been the case for Kazakhstan over the sample, with an average annual growth rate of just under 13% p.a.

In order to project [g.sup.y.sub.t], and hence yt, over the period 2015-2025, we need to accommodate the dip in the projected oil price from 2015 until 2018, where [g.sup.po.sub.t] < [g.sup.po.sub.ss] = 0.0113. This is simple enough to do by simulating [g.sup.bo.sub.t] and [g.sup.rer.sub.t] using the estimated recursive model conditioned on the DECC projection of [g.sup.po.sub.t] (as shown in Figure 7). The resulting projected values of [y.sub.t], taking 2015q1 as the starting point, are [y.sub.p,t] = [y.sub.p,t-1] + [g.sup.y.sub.p,t], where the p subscript indicates a projection, and [y.sub.p,t-1] = [y.sub.t-1] for t - 1 = 2014q4. The projected values are shown in Figure 8; taking into account the slower growth in the early part of the projection period because of the dip in the oil price, the projected average growth rate, 2015-2025, is [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is 5.09% p.a. which, as anticipated, is slightly less than in the steady-state projection.

CONCLUSIONS

A major concern for nations with substantial oil reserves is whether their exploitation will prove to be a longer-term benefit. This concern has given rise to the question of whether such reserves are a blessing, a curse or neither. There is empirical evidence across a range of countries that all are indeed possible and have indeed occurred. The experience of recent countries to join the oil-rich 'club' is particularly relevant to an evaluation of this question as they have had the chance to learn from the experiences of earlier members. In this context, Kazakhstan is a leading player. Having separated from the USSR in 1991, it underwent a transition to a market-based economy, which involved not only the transition to market pricing decisions, but the development of institutions and policies similar to those in the western industrialised economies. These developments were concerned with building up the institutional and physical infrastructure necessary to compete in world markets, both in exporting and in attracting FDI, so as to develop existing resources and diversify away from an over dependence on natural resources.

The narrative of this article supports the general view that despite some autocratic elements of government, Kazakhstan has, generally, been successful in this aim, although there is still some way to go to match the European countries; nevertheless, it has been very successful amongst the newly transitioning economies in attracting FDI and, to a degree, in diversifying its economy. These developments are at least suggestive that Kazakhstan has avoided the danger that oil is a curse. To analyse its performance in this respect, we developed the modelling framework because of Esfahani et al (2014), in which the blessing/curse/neither trichotomy can be tested. The empirical results strongly suggest that oil has been a blessing for Kazakhstan, with a significantly positive steady-state relationship between oil exports and GDP. We concur with EMP that to the extent that policy and institutional arrangements have an impact they do so on the parameter estimates; they summarised the situation as follows: '... what matters most is the extent to which oil revenues are likely to be sustainable over the medium to long term, and in this regard it is the oil reserve-to-production ratio that could be important as to whether oil income enters the long-run output equation as a major determinant.' (EMP, p. 10).

This framework was then further developed to consider projections of the growth of GDP conditioned on projections of the oil price. The recent central projection of oil price produced by the UK's Department of Energy and Climate Change, based on an average of 11 expert projections and IEA and EIA projections, was used to provide an illustrative 'scenario' projection of GDP. On the basis of the estimated steady-state relationships, and conditional on the DECC oil price projection, GDP was projected to grow at an average annual rate of just over 5% p.a. for the period 2015-2025, in contrast to its previously much higher growth rates. This is a 'scenario' rather than a forecast, it is intended to answer or provide guidance to questions starting with 'what if' and to encourage discussion of the implicit sensitivities that it uncovers; scenarios can also be helpful to policy discussions that may lead to changes in potentially endogenous variables, such as the (real) exchange rate, and to reinforce the need for accompanying infrastructure developments.

Acknowledgements

The authors are grateful to Maxim Romanov for his excellent research assistance, to Azat Aituar and Dina Azhgaliyeva for data assistance, and to Tom Fletcher of the Department of Energy and Climate Change (DECC). They are also grateful to a referee for his encouragement and constructive comments on previous versions of this article.

REFERENCES

Acemoglu, D, Johnson, S and Robinson, JA. 2005: Institutions as the fundamental cause of long-run growth. In: Aghion, P and Durlauf, S (eds). Handbook of Economic Growth. Chapter 6. Elsevier: North Holland, pp. 385-471.

Ahrend, R. 2006: How to sustain growth in a resource based economy? The main concepts and their application to the Russian case. OECD Economics Department Working Paper No. 478, Paris.

Al-Kasim, F, Skinner, R and Arnott, R. 2006: Managing petroleum resources: The 'Norwegian model' in a broad perspective. Oxford, UK: Oxford Institute for Energy Studies.

Andersen, SS. 1993: The struggle over North sea oil and gas: Government strategies in Denmark. Scandinavian University Press: UK; Oslo, Norway.

Arezki, R and Van der Ploeg, F. 2010: Do natural resources depress income per Capita? Review of Development Economics 15(3): 504-21.

Asea, P and Lahiri, A. 1999: The precious bane. Journal of Economic Dynamic and Control 23(5/6): 823-49.

Aslund, A. 2007: Russia's capitalist revolution. Petersen Institute: Washington DC.

Auty, RM. 2001: The political economy of resource-driven growth. European Economic Review 46(4/6): 839-46.

Bevan, D, Collier, P and Gunning, JW. 1999: The political economy of poverty, equity and growth: Nigerian and Indonesia. Oxford University Press: Oxford.

Blomstrom, M and Kokko, A. 2007: From natural resources to high-tech production: The evolution of industrial competitiveness in Sweden and Finland. In: Lederman, D and Maloney, WF (eds). Natural Resources: Neither Curse nor Destiny. Stanford University Press: Stanford, CA, pp. 71-99.

Bravo-Ortega, C and de Gregorio, J. 2005: The relative richness of the poor? Natural resources, human capital and economic growth. World Bank Working Paper, 3484, Washington, USA.

Cavalcanti, TV, de Mohaddes, K and Raissi, M. 2011: Growth, development and natural resources: New evidence using a heterogeneous panel analysis. Quarterly Review of Economics and Finance 51(4): 305-18.

Department of Energy and Climate Change. 2014: DECC Fossil Fuel Price Projections, URN 14D/351, www.gov.uk/DECC, accessed on 26 June 2015.

Esfahani, HS, Mohaddes, K and Pesaran, MH. 2014: An empirical growth model for major oil exporters. Journal of Applied Econometrics 29(1): 1-21.

Fasano, U. 2002: Review of the experience with oil stabilisation and savings funds in selected countries. IMF Working Paper WP/00/112, Washington, USA.

Frankel, J. 2010: The natural resource curse: A survey. National Bureau of Economic Research Working Paper No. 15836, March, Cambridge, Massachusetts.

Gelb, A. (ed) 1988: Oil windfalls: Blessing or curse? Oxford University Press: Oxford.

Gylfason, T. 2001: Natural resources, education, and economic development. European Economic Review 45(4/6): 847-59.

Gylfason, T, Tryggvi, TH and Zoega, G. 1999: A mixed blessing: Natural resources and economic growth. Macroeconomic Dynamics 3(2): 204-25.

Hannesson, R. 2005: Saving petroleum wealth: Tales of three jurisdictions. In: Glomsrod, S and Osmundsen, P (eds). Petroleum Regulation with Stable Steady States. Ashagate: Farnham, UK, pp. 112-130.

Hanson, P. 2002: Barriers to long-run growth in Russia. Economy and Society 31(1): 62-84.

Hanson, P. 2008: The August 2008 Conflict: Economic Consequences for Russia. Chatham House Briefing Note, September, London, UK.

Hanson, P. 2009: Russia to 2020. Finmeccanica Occasional Paper, November, Rome, Italy.

Humphreys, M, Sachs, J and Stiglitz, JE. 2007: Escaping the resource curse. Columbia University Press: New York.

IEF (Index of Economic Forum). 2015: http://www.heritage.org/index/explorelview = by-region-country-year.

Innis, H. 1956: Essays in Canadian economic history. University of Toronto: Toronto.

Isham, J, Woolcock, M, Pritchett, L and Busby, G. 2003: The varieties of resource experience: Natural resource export structures and the political economy of economic growth. World Bank Economic Review 19(2): 141-74.

Jahan-Parvar, MR and Mohammadi, H. 2011: Oil prices and real exchange rates in oil-exporting countries: A bounds testing approach. Journal of Developing Areas 45(1): 309-18.

Kalyuzhnova, Y. 1998: The Kazakhstani economy: Independence and transition. Palgrave Macmillan: Basingstoke.

Kalyuzhnova, Y. 2002: Economies and energy. In: Kalyuzhnova, Y, Jaffe, AM, Lynch, D and Sickles, R (eds). Energy in the Caspian Region: Present and Future. Chapter 3. Palgrave Macmillan: Basingstoke.

Kalyuzhnova, Y. 2008: Economics of the Caspian oil and gas wealth: Companies, governments, policies. Palgrave Macmillan: Basingstoke.

Kalyuzhnova, Y. 2011: The national fund of the Republic of Kazakhstan (NFRK): From accumulation to stress-test to global future. Energy Policy 39(10): 6650-7.

Karl, TL. 1997a: The perils of the petro-state: Reflections on the paradox of plenty. Journal of International Affairs 53(1): 31-48.

Karl, TL. 1997b: The paradox of plenty: Oil booms and petro-states. University of California Press: Berkeley.

Kaser, M. 1997: The economies of Kazakhstan and Uzbekistan. The Royal Institute of International Affairs: London.

Kaser, M. 2003: The economic and social impact of systemic transition in Central Asia and Azerbaijan. Perspectives on Global Development and Technology 2(3): 459-73.

Kaufmann, D, Kraay, A and Zoido-Lobaton, P. 1999: Aggregating governance indicators. World Bank Policy Research Department Working Paper No. 2195, Washington DC.

Kemp, AG and Stephen, L. 2005: Petroleum policy and prospects in the UK continental shelf (UKCS). In: Glomsrod, S and Osmundsen, P (eds). Petroleum Industry Regulation Within Stable States. Ashgate Publishing: Aldershot, pp. 131-156.

Kurtz, MJ and Brooks, SM. 2011: Conditioning the 'resource curse': Globalization, human capital, and growth in oil-rich nations. Comparative Political Studies 44(6): 747-70.

Lane, P and Torrnell, A. 1996: Power, growth and the voracity effect. Journal of Economic Growth 1 (June): 213-241.

Larsen, ER. 2006: Escaping the resource curse and the dutch disease? When and why Norway caught up and forged ahead of its neighbours. American Journal of Economics and Sociology 65(3): 605-40.

Lay, J and Mahmoud, TO. 2004: Bananas, oil and development: Examining the resource curse and its transmission channels by resource type. Working Paper No. 1218. Kiel Institute for World Economics, Kiel, Germany.

Lederman, D and Maloney, WF. 2007: Natural Resources: Neither Curse nor Destiny. Stanford University Press: Stanford, CA.

Maloney, W. 2002: Missed opportunities: Innovations and resource-based growth in Latin America. Economia 3(1): 111-67.

Mehlum, H, Moene, K and Torvik, R. 2006: Institutions and the resource curse. Economic Journal 116(508): 1-20.

Mohaddes, K and Pesaran, MH. 2013: One hundred years of oil income and the iranian economy: A curse or a blessing? Cambridge Working Papers in Economics 1302, University of Cambridge: Faculty of Economics, Cambridge, UK.

Papyrakis, E and Gerlagh, R. 2004: The resource curse hypothesis and its transmission channels. Journal of Comparative Economics 32(1): 181-93.

Pesaran, MH and Shin, Y. 1999: An autoregressive distributed lag modelling approach to cointegration analysis. In: Storm, S. (ed). Econometrics and Economic Theory in the 20th Century: The Ragnar Frisch Centennial Symposium. Chapter 11. Cambridge University Press: Cambridge.

Pesaran, MH, Shin, Y and Smith, R. 2001: Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics 16(3): 289-326.

Pomfret, R. 1995: The economies of central Asia. Princeton University Press: Princeton, NJ.

Pomfret, R. 2003: Economic performance in Central Asia since 1991: Macro and micro evidence. Comparative Economic Studies 43(4): 442-465.

Pomfret, R. 2006: The central Asian economies since independence. Princeton University Press: Princeton, NJ.

Prebisch, R. 1959 [1962]: The economic development of Latin America and its principal problems. Reprinted in. Economic Bulletin for Latin America 7(1): 1-22.

Rodriguez, F and Sachs, JD. 1999: Why do resource-abundant economies grow more slowly? Journal of Economic Growth 4(3): 277-303.

Rosser, A. 2006: The Political Economy of the Resource Curse: A Literature Survey. IDS Working Paper 268. Brighton, UK: Institute of Development Studies, www.ids.ac.uk/publication/ thepolitical-economy-of-the-resource-curse-a-literature-survey, accessed 9 December 2013.

Rutland, P. 2008: Putin's economic record: Is the oil boom sustainable? Europe-Asia Studies 60(6): 1051-72.

Sachs, JD and Warner, AM. 1995 [1997]: Natural resource abundance and economic growth. In: Meier, G and Rauch, J (eds). Leading Issues in Economic Development. Oxford University Press: Oxford.

Sachs, JD and Warner, AM. 1997: Fundamental Sources of Long Run Growth. American Economic Review 87(2): 184-188.

Sachs, JD and Warner, AM. 1999: The big push, natural resource booms and growth. Journal of Development Economics 59(1): 43-76.

Sachs, JD and Warner, AM. 2001: The curse of national resources. European Economic Review 45(4/6): 827-38.

Sarraf, M and Jiwanji, M. 2001: Beating the resource curse: The case of Botswana. Environment Department, World Bank: Washington DC.

Singer, H. 1950: The distribution of trade between investing and borrowing countries. American Economic Review 40(2): 473-85.

Singh, A. 2013: Keynote Speech. Conference on Harnessing Natural Resource Wealth for Inclusive Growth and Economic Development. International Monetary Fund, Dili, 18 September, Washington, https://www.imf.org/extemal/np/speeches/2013/091813.htm.

Stijns, J-PC. 2005: Natural resource abundance and economic growth revisited. Resources Policy 30(2): 107-30.

Suleimenov, MK and Osipov, YB. 2010: Review of the Legislative Base for Investments in Oil and Gas Sector of the Republic of Kazakhstan--Executive Summary (English). Project document. IIED/British Embassy, Kazakhstan.

Van der Ploeg, F. 2011: Natural resources: Curse or blessing? Journal of Economic Literature 49(6): 366-420.

WGI. 2015: http://www.worldbank.org/wbi/governance/pubs/aggindicators.htm.

Wood, A. 1999: Natural resources, human resources and export composition: A cross- country perspective. In: Mayer, J, Chamber, B and Farooq, A (eds). Development Policies in Natural Resource Economies. Edward Elgar: Cheltenham, UK.

Wright, G. 1990: The origins of American industrial success, 1879-1940. American Economic Review 80(4): 651-68.

World Economic Forum. 2015: www3.weforum.org/docs/GCR2014 ... /GCI_Dataset_2006-07-201415.xls, accessed on 21 June 2015.

YELENA KALYUZHNOVA [1] & KERRY PATTERSON [2]

[1] The Centre for Euro-Asian Studies, Henley Business School, University of Reading, Whiteknights, PO Box 218, Reading, RG6 6AA, UK.

E-mail: y.kaluyzhnova@reading.ac.uk

[2] Department of Economics, University of Reading, Reading, RG6 6AA, UK.

E-mail: k.d.patterson@reading.ac.uk

(1) Further references include Sachs and Warner (1995 [1997], 1997), Lane and Torrnell (1996), Asea and Lahiri (1999), Gylfason et al. (1999), Rodriguez and Sachs (1999), Sachs and Warner (1999, 2001) and Rosser (2006).

(2) For further references see Pomfret (1995,2003,2006), Kaser (1997,2003), Kalyuzhnova (1998, 2002, 2008), Hanson (2002, 2008, 2009), Aslund (2007) and Rutland (2008).

(3) Here we follow EMP in referring to Et/P, as the real exchange rate; however, the latter term is usually defined as [E.sub.t][P.sup.*.sub.t]/[P.sub.t], where [P.sup.*.sub.t] is the overseas price level, so that the numerator is in effect the cost of a 'unit' of overseas output in domestic currency units relative to the domestic cost in domestic currency units.

(4) The ARDL approach is parsimonious with respect to the lag structure compared with VARbased approaches, such as Johansen's maximum-likelihood method, which are more costly in terms of degrees of freedom.

Table 1: Caspian economies: Share of oil and gas revenues
in total exports

Country         2001 (%)    2002 (%)    2003 (%)    2004 (%)

Azerbaijan         91          88          86          82
Kazakhstan         53          56          59          63
Russia             50          51          53          53
Turkmenistan       89          90          89          88

Country         2005 (%)    2006 (%)    2007 (%)    2008 (%)

Azerbaijan         76          84          81          97
Kazakhstan         68          67          64          67
Russia             60          61          60          64
Turkmenistan       90          86          91          82

                2009 (%)    2010 (%)    2011 (%)    2012 (%)

Azerbaijan         93          94          95          93
Kazakhstan         68          71          70          69
Russia             60          63          64          67
Turkmenistan       70          68          90          93

                2013 (%)    2014 (%)

Azerbaijan         93          92
Kazakhstan         75          77
Russia             68          56
Turkmenistan       90          91

Source: http://www.trademap.org/BUateral_TS.aspx

Table 2: Estimates of the long-run coefficients
and diagnostic tests

Variable                [y.sup.*.sub.t]     [x.sup.o.sub.t]
                         [[??].sub.1]         [[??].sub.1]
Estimate                     0.53                 0.62
't' [p]                   2.99 [0.00]         5.82 [0.00]
Cointegration tests       F-statistic         W-statistics
                          8.16 {4.60}         32.62{18.4}
Diagnostic tests      Serial correlation       Normality
Test statistic [p]       2.62 [0.624]         4.96 [0.09]

Variable

Estimate
't' [p]
Cointegration tests

Diagnostic tests      Heteroscedasticity   [??]
Test statistic [p]       1.95 [0.163]      0.044

Notes: [p] is the p-value for the null hypothesis that the population
regression coefficient is 0 against the one-sided positive alternative
for [[psi].sub.1] and [[psi].sub.2]; {x} are the respective 95% upper
bound critical values for the test of cointegration; [??] is the
estimated eguation standard error for the ARDL model.