NATO Economic Colloquium 1997

Regional Cooperation Among Central Asian Economies Michael Kaser Institute for German Studies, University of Birmingham A region of five states It was a sign of Central Asian political cohesion that in 1993 the governments of the five states declared themselves to constitute 'Central Asia' (in Russian usage, Tsentral'naya Aziya) rather than the four states of literally 'Middle Asia' (Srednyaya Aziya) plus Kazakhstan. The previous year, when a United Nations mission was present in Tashkent at the moment of their admission to UN membership (March 1992), the acronym CARK had been coined (Central Asian Republics and Kazakhstan) which mirrored the Soviet classification of the two as separate 'large economic regions'. There had, however, been sound historical reasons for 'Middle Asia' to be taken as an economic entity, because (ignoring uninhabitable mountains and deserts) three of the states - Tajikistan, Turkmenistan and Uzbekistan - with southern Kyrgyzstan were based on settled agriculture, chiefly in the basins of the two great rivers renowned in antiquity, the Amu Darya (the classical Oxus) and the Syr Darya (the classical Jaxartes), and the Turkmen oases. However, desert again apart, much of Kazakhstan and northern Kyrgyzstan depended economically on nomadic steppe and semi-arid pasturing and upland transhumance. The only natural barrier to migration and invasion was the eastern mountains (with the Chinese Empire beyond them), and the entire zone was hence open to cultural, political and military impact from the south (Persians, Indian and Arab), the west (Romans) and the north (Mongols and Russians). But in four millennia of an exchange economy, the region has been divided by customs frontiers only in the past five years: on that vast historical scale the issue of economic re-integration has been posed in a period of a mere 0.1% of the time. A short way of putting the historical reason is that the region has an imperial rather than a balkanized past: it was part, and at times the metropole, of great empires: to name but a few, the Persian Emperors, Alexander of Macedon and Ghengis Khan ruling from outside, the Khorezmshahs and Timur from within. Tsarist Russia took Kazakhstan in the first half of the last century and 'Middle Asia' in the second half, leaving only Bukhara and Khiva in semi-independence. When the successor Soviet state effected the 'national delimitation of Central Asia' in 1924, it drew political, not economic, frontiers. Under the Five-year Plans, the allocation and distribution system was uniform throughout the 'socialist sixth of the world'; of the few economic powers allowed purely Republican ministries, none were discriminatory against other Union-Republics. The rouble was the single currency but the exchange of goods and services among Union-Republics did not depend on money: transactions were on official quota or informal negotiation, each with no concern for inner-Soviet frontiers. The basis of those relationships was swept away by the events of late 1991 and early 1992 - the collapse of the USSR and general price decontrol. In that last year of Soviet existence, gross budget transfers from the Union exceeded domestic tax revenue in three republics (Kyrgyzstan, Tajikistan and Uzbekistan), and were substantial in Turkmenistan and Kazakhstan (8% and 5% of GDP respectively). Without such inflow and no mechanism for public borrowing, budget deficits could only be monetised; the gap was of course widened where state expenditure proportionately grew (Kyrgyzstan, Tajikistan and Uzbekistan). The coincidental liberalisation of most wholesale and retail prices (after seven decades of largely arbitrary fixed pricing) further fuelled inflation, even to hyperinflation. Widely different rates of inflation in 1993 (they were still bunched together in 1992) and national quantitative controls applied in response to the many supply/demand imbalances that arose quickly disarticulated the previously-unified channels of production and consumption. The residuary body of the former Soviet Union, the Commonwealth of Independent States (CIS), sought to restore trade and payments relations within the new market environments by rational use of the existing common currency, the rouble, and in October 1992 three of the Central Asian states (Kazakhstan, Kyrgyzstan and Uzbekistan) joined with Armenia, Belarus and Russia to plan for an Interstate Bank of the CIS. It was argued both within the countries and by international advisers that the advantages of a single currency should not be readily discarded - an exemplar in the Maastricht Treaty envisaging Economic and Monetary Union for the EU had been signed in December 1991 - but within the space of a few months (May to November 1993) the CIS Bank proposal was a dead letter and the five states had separate currencies (1).

The common post-independence shocks In their early years each of the Central Asian states suffered triple shocks and their variant responses further widened the disaggregation of the economic region: The fiscal shock required measures towards the equilibration of public-sector revenue and expenditure in matters over which Union-Republican governments had exercised little control or discretionary power. Thus taxes appropriate to a market economy had to be formulated, the relevant tax base surveyed and registered, and expenditures had to be radically restructured; The management shock required new institutions and staff to run them for political independence and for a market economy, both changes being simultaneous and de novo. The dispersion of property rights from state and collective ownership proceeded at a different pace in each state, multiplied the number of economic agents which the state had to tax or regulate, and introduced foreign investors with whom few officials previously had had to deal; The recession shock of declining production and investment retarded macroeconomic stabilization and (partly due to differential government protection) varied from country to country. Calibrated against 1989, measured GDP in 1996 in Uzbekistan was as high as 84%, but as low as 37% in Tajikistan. The other three states were around half (2). For the first year and a half of independence, the five governments partly protected themselves at Russian expense.Their new Central Banks (the former Union-Republican offices of the USSR Gosbank) and the specialised state banks (corresponding branches of such as Promstroibank and Vneshekonombank) extended to domestic entities credits with scant concern for ability to service the loans; indeed, such supportive funding was little more than outright subsidization and an exacerbation of inflation. It was the IMF's conditions to Russia for a US$1.5bn Systemic Transformation Facility in 1993 which precipitated the break-up of the rouble zone. The Central Bank of Russia (CBR) was required to target its base rate and monetary emission to an appreciation of the real rate of exchange: change in the rouble-dollar rate was to be half that of domestic inflation. The CBR would have had to control the money supply of other CIS central banks still using roubles (3). Bilateral negotiations between the CBR and the Russian Government on the one hand and each of the Central Asian authorities on the other showed mutual policy incompatibility, except, as already noted, for Tajikstan, where an economically weak administration controlling only part of a territory in civil war depended on continued Russian support. As part of the establishment of their economic autonomy, the newly-independent governments formulated customs tariffs, introduced exchange controls and applied their own regulations to foreign investment. The common external tariff of the former USSR and the central planning of state-owned enterprises had directed trade predominantly among the Union-Republics, but after independence the burgeoning private sector sought markets wherever profit beckoned. Between 1991 and 1995 the share of mutual trade in the aggregate GDP of CIS members fell from 14.4% to 4.0% (4). Between 1991 and 1996 the percentage share of the CIS in trade turnover (imports plus exports) fell in Uzbekistan from 87.5 to 27.0, in Tajikistan from 85.3 to 50.6 and in Turkmenistan from 90.6 to 51.0; lesser declines were exhibited by Kazakhstan (87.0 to 61.4) and Kyrgyzstan (87.9 to 65.5) (5). Payments arrears also contributed to the fall in intra-CIS trade for two reasons: one was the increase of arrears generally as recession made a differential impact on buyers and on sellers; the other was the approximation to world prices for such transactions after long isolation from the relativities prevailing outside the USSR or selected partners (CMEA members and politically-favoured developing countries). Finally, the impact of foreign direct investment (FDI) differed from state to state, and three countries opened special economic zones. The EBRD calculates that between 1989 (when more liberal Soviet joint-venture legislation came into force) and 1996 Kazakhstan received US$2,761m and Turkmenistan US$444m - below only that into Azerbaijan (mainly in oil and gas finance) in per capita terms (US$56, 32 and 80 respectively); the other Central Asian states received less - Uzbekistan US$342m, Kyrgyzstan US$146m and Tajikistan US$55m (per capita a mere US$2, 7 and 2 respectively). There are three special economic zones in Kazakhstan, four in Kyrgyzstan and seven in Turkmenistan, with a principal objective to encourage foreign investors to establish manufacturing facilities therein, imports and exports being duty-free. Disparities in attracting FDI and the facility of tax-concessional zones would divert or create trade in the medium, rather than the short, term, but they were further breaks with the uniformity that had marked external economic relations in Soviet Central Asia. Rebuilding economic relations within the CIS Remedial measures to seek to minimise the decline in intra-CIS trade were early on members' agenda. At the general CIS level, an Economic Court was authorized in July 1992 but it only started work in mid-1994 and not all members subscribe (6). In October 1992 an agreement was signed in Bishkek by eight CIS members to establish a payments union to prolong the use of the rouble as a common currency, but when this was overtaken by the separation of currencies, a series of CIS treaties were negotiated to attempt economic reintegration. First came that on the Establishment of Economic Union (September 1993), then two together, on the Establishment of a Payments Union and the Establishment of an Interstate Economic Committee (October 1994) and finally on an Interstate Monetary Committee. Together, these treaties were intended to assure a common customs regime, a multilateral payments mechanism and the facilitation of transnational enterprises. The Interstate Monetary Committee first met in February 1996 and set up a secretariat, but although presented with a draft for a mechanism for mutual payments by the Russian delegation at its second meeting (October 1996), no CIS-wide convertibility has been - nor is soon likely to be - achieved. The reasons are all too obvious - the considerable settlement arrears that have accumulated between CIS members, the divergent movements of exchange rates (in real terms, some have appreciated, others depreciated), exchange rate determination (some are unitary, some dual, some controlled, others floating) and finally their degree of liberalisation (ranging from free capital and current accounts to arbitrary control). Recognising the immediate problems, a CIS leaders' meeting in Moscow in April 1997 resolved on a longer-term 'Concept of Economic Integration of CIS States'. The resolution recognises that each member should independently determine the extent of its participation in each of three stages - a free trade area, a customs union and a payments union. No 'single market' (on the EU model) assuring free movement across member frontiers of labour and capital is apparently envisaged (as took place within the confines of Soviet central planning). Between 1989 and the beginning of 1996 some 4.2 million people moved within, between and from the five Central Asian republics, driven by political, economic and ethnic factors, and following the general abolition of residence permits (7). Such migration has been slackening, but a fully-free labour market within the CIS would require non-national residents to have the same rights as nationals - as is the case within the EU; Kazak and Uzbek legislation does not permit dual nationality. Capital movements are still subject to control in Central Asian states, save in Kyrgyzstan, but the agreement by Kazakhstan and Uzbekistan to accept IMF Article VIII on current account convertibility (not yet implemented in Uzbekistan) is a step towards it. In fact, the limitation on transnational capital flow arises more from national constraints on ownership of land and of privatised state enterprises. Cooperation across CIS frontiers At a humbler level, a large number of functional bodies have been set up under CIS auspices to foster and enhance mutual cooperation - on banking, railways, agriculture, engineering, coal-mining, oil and gas. Relations more profound than trade contracts are being re-established through 'financial-industrial groups', which are particularly numerous between Kazakhstan and Russia (in ferrous and non-ferrous mining and metallurgy, heavy engineering and telecommunications), with some between Uzbekistan and Russia. The more grandiose infrastructure projects are all bilateral or trilateral with participation by the private-sector corporations involved. The Central Asian states especially seek to diversify their export markets away from the CIS, but need to cross CIS territory to do so. Kazakhstan has described itself as 'a country at the heart of East-West rail lines' (8). These comprised the newly-exploited railways between Turkmenistan and Uzbekistan to the Iranian network and from Russia across Kazakstan into China; a Kazakhstan-Turkmenistan-Iran link is also under construction, such that, with four major trunk-lines already into Russia and thence to Europe or to the Northern Pacific, Central Asian freight has a range of options to ocean ports. One minor issue that has caused conflict has been the charges levied by the Tajik authorities for trains crossing Tajik territory between parts of Uzbekistan in the Fergana Valley. The relevant governments have substantial equity in all the oil and gas pipelines connecting CIS states. Those originating in Central Asia include one already settled - the Caspian Pipeline Consortium (CPC) for a line, to be operated by Russia's Transneft, from the north-west Kazakh oilfields to the Russian Black Sea port of Novorossisk - and one still to be finalized - the Kazakhstan Pipeline Consortium to take other Kazakh oil under the Caspian, collect Azerbaijan crude, and exit at the Georgian port of Supsa on the Black Sea. Kazakh oil is to transit Turkmenistan to northern Iran, the United States government conceding on its sanctions against Iran because the transaction is only a swap for southern Iranian oil on a US company account. The gas pipelines built in the Soviet period to take Uzbek gas into Russia will be supplemented as new deposits are exploited, but the only major agreement (of May 1997) involving other CIS transit is for Turkmen gas to be routed through Iran to supply Turkey and thence Georgia; the agreement overlaps an earlier Iran-Turkey contract for Iranian gas and may supersede plans for a pipeline under the Caspian to transit Azerbaijan and Armenia into Turkey. Russia already supplies Turkey via Ukraine, Romania and Bulgaria and if Turkey becomes over-supplied, Turkmen gas could be sold on to European users. Intergovernmental agreements on water have been more controversial and in significant cases rational usage and appropriate charges have yet to be negotiated. The principal cotton-growing state, Uzbekistan, uses 76% of the total freshwater resources available within itself and from neighbouring countries (compared with Kazakhstan's 30% and Kyrgyzstan's 24%) (9). Irrigation methods for cotton plantations have grossly wasted water and drained into waterways a noxious mix of chemicals. Intergovernmental negotiations particularly concern the shrinkage of the Aral Sea, the timing of water release and charges for water to rationalise consumption (10). An agreement between Kazakhstan, Uzbekistan and Kyrgyzstan of May 1996 provided for scheduling the flow of the Syr-Darya, and one between Turkmenistan and Uzbekistan of January 1996 regulates the lower Amu-Darya; if the peace agreement holds in Tajikistan, an agreement may become possible over the upper Amu-Darya. Conflict arises not only on the amount of water released but its timing: Kyrgyzstan and Tajikistan want to retain water for autumn and winter hydroelectricity generation, whereas the downstream states need it in spring and summer for irrigation. Environmental problems The water allocation problem has generated Central Asia's worst environmental damage - the drying out of the Aral Sea, which now holds only one-quarter of its 1960 volume as a direct result of the diminution of the annual input of water from the Amu-Darya and the Syr-Darya from 60 cu. km to 5 cu. km. Water pollution from farm chemicals, soil salination from inefficient canalization and airborne salt from the dessicated bed of the Aral Sea are further problems, now the subject of an agreement between the five Presidents and the World Bank in March 1997: although agreement on the Tajik contribution is awaited, the other four countries will devote 0.3% of their fiscal revenue to an International Fund for Saving the Aral Sea, to which the World Bank will over four years contribute US$380m. At least two other environmental problems have international dimensions. One is the emission of fluor by the South Tajik Aluminium Works at Tursunzade causing fluorosis downwind of the plant in Uzbekistan. Contamination problems arise from the chemical warfare installations and test grounds near Aralsk in Kazakstan and could affect neighbouring areas of Uzbekistan. Radiation from the Semipalatinsk nuclear-weapon test 'polygon' (disused since 1963) has required the movement since Kazakh independence of nearly 200,000 people elsewhere in the republic and 116,000 to other countries; the neighbouring Altai region of Russia has also been affected by the residual radiation (11). The World Bank and the EBRD have made environmental protection part of their lending programmes and the latter is advising each government on a National Environmental Action Plan. Two integration unions The inadequacy to date of CIS-wide collaboration has led governments to seek smaller groupings for more intensive economic integration, of which the most inclusive, but domestically-disputed, is the Customs Union of Belarus with Russia. In Central Asia, a trilateral agreement was signed in 1994 on a 'united economic zone' of Kazakstan, Kyrgyzstan and Uzbekistan, and some progress has been achieved in harmonizing rates of value-added tax and on double taxation (12). Kazakhstan and Kyrgyzstan are members of a quadrilateral customs union embracing also Belarus and Russia, established in March 1996; a threat by Kazakhstan in early 1997 to leave, due to problems of tariffs in the transition period before projected tariff harmonisation, appears to have been withdrawn. Kazakhstan raised some tariffs to the Russian level which protected Russian production, but of course resulted in dearer imports: domestic interests made Kazakhstan cut them sharply, and hence (as is also the case with Belarus) allowed Russian buyers to import more cheaply than through their own country's entry-points. Central Asian complementarities Prospects for integration among the five central Asian states depend on their economic complementarities. They all need to generate more value-added to their rich natural-resource exploitation and to enlarge other sectors - especially services - to employ their expanding working age population (15-64 years of age): by 2010 it will in four of the republics be between two-fifths and two-thirds larger than it is now (13). Demographic movement is less dynamic in Kazakhstan, where a large Slavic population has long had a lower fertility. Because a high proportion of the active labour force is occupied in agriculture (one-third or more in four states, other than Kazakhstan) and the post-independence recession has brought redundancies in many industrial plants geared to a Soviet-wide and non-cost-conscious clientele, underemployment may add seven million people to the 11 million projected from demographic growth. Some industries are already established which could be redeveloped in a policy of diversification, but they had their origin in a centrally planned economy which assured them of outlets by and large irrespective of their production or transport costs. To nurture some of them into a competitive relationship with imports may require a period of protection when WTO membership is negotiated and as the two customs unions with some of the states are finalized. Utilisation of crude oil and natural gas in refining and petrochemicals is already being undertaken. Other examples include processing Uzbek wool into high- quality textiles and to make fabrics, carpets and blankets for export; and a Turkmen plant to weave high-quality denim. Financial and tourist services are expanding. Foreign investment is increasing domestic value-added, especially among consumer goods which use local materials, and may be a sound mode of assuring that the trilateral and quadrilateral customs unions evoked for political reasons do not bypass the signals of competition and profitability. Footnotes The Kazakh tenge, the Kyrgyz som, the Turkmen manat and the Uzbek sum were national units; Tajikistan maintained use of the Russian rouble until 1995. These are the estimates shown in EBRD, Transition Report Update 1997, Table 1. Those of the UN ECE, Economic Survey of Europe in 1996-1997, Table B.1 are lower, except for Kyrgyzstan (slightly higher) and Turkmenistan (much higher, because it reflects the Turkmenstan Statistical Office's revaluation of oil and gas to world prices by 73% in 1992-3, a change which strictly should not influence an indicator of real economic activity). Ukraine had already disengaged and Kyrgyzstan was in the process of doing so. See paper by Yuri Shishkov (Panel II). See Table 2 of paper by Yuri Shishkov (Panel II). This section draws on Mark Webber, CIS Integration Trends: Russia and the Former Soviet South, London: Royal Institute of International Affairs, 1997. UNHCR, CIS Conference, Geneva, HCR, 1996, p. 5. A few residence permit controls have been retained (e.g. for Tashkent) to inhibit urban unemployment and shanty dwelling. Financial Times, 11 July 1996 (Supplement on Kazakhstan, p. vi). World Bank, World Development Report 1997, Table 10 (no data for Tajikistan or Turkmenistan). On rationalizing water consumption by user charges, see Michael Kaser, The Economies of Kazakhstan and Uzbekistan, London, Royal Institute of International Affairs, 1997, pp. 61-2, and EBRD, Transition Report 1996, pp. 42-3. UNHCR, CIS Conference, p.15. From a detailed survey of its subsequent operation, Heribert Dieter, 'Regional Integration in Central Asia: Current Economic Position and Prospects', Central Asian Survey, vol.15, no.3/4, 1996, pp.369-386, has high expectations of the 'trilateral union' and points out that its Interstate council has funding from the European Union. UN ECE, Economic Survey of Europe in 1996-1997, Table 4.3.2: the 1995-2010 increments are projected at 64% for Tajikistan, 53% for Uzbekistan, 52% for Turkmenistan and 42% for Kyrgyzstan, but only 21% for Kazakhstan.