Chapter

11 Low Pressure, High Tension: The Energy-Water Nexus in the CIS-7 Countries

Author(s):
Sarosh Sattar, and Clinton Shiells
Published Date:
April 2004
Share
  • ShareShare
Show Summary Details
Author(s)
David Kennedy, Samuel Fankhauser and Martin Raiser1 

Overview

Energy and water have emerged as critical issues for the CIS-7 countries—Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan—and their neighbors for at least two reasons. The first is that energy and water constitute the region’s main natural resources, and the exploitation of both was and still is a key to these countries’ mode of production. The second is that the distribution of these resources is very unequal across countries. Azerbaijan, Kazakhstan, Turkmenistan, and Uzbekistan benefit from rich energy reserves, while Armenia, Georgia, the Kyrgyz Republic, and Tajikistan have substantial water resources. This unequal distribution gives rise to potential gains from trade but it is also the source of recurrent conflict between neighboring states in the region. Energy and water issues are closely linked given that the latter can be used, inter alia, for hydropower generation and/or irrigation. Use of water in the municipal sector is not discussed in this chapter. Replenishment of the Aral Sea as an alternative to irrigation is consistent with increased winter hydro generation, discussed below under “Unlocking the Benefits from Trade.”

This chapter looks at the energy-water nexus in Central Asia and the south Caucasus regions and the energy situation in Moldova. The central argument is that reform of domestic power (and water) tariffs is key to unlocking the potential for beneficial regional trade in both electric power and water and to stimulating the efficient use of resources. Of course, electric power reforms are linked to regional trade in other primary energy sources, particularly natural gas; where appropriate, these links will be drawn. When the Soviet Union collapsed, the cross-border price of primary energy traded between former republics suddenly increased; however, domestic power prices were not adjusted. For net energy importers this has meant a growing quasi-fiscal burden often associated with significant accumulation of foreign debt to energy-exporting countries. The cross-border payment problems are an important cause of irritation between countries in the region and have motivated a drive toward self-sufficiency in energy. The failure to adjust domestic energy prices has also limited the funds available for investment, and system degradation has resulted. Moreover, the low domestic energy price has had implications for the management of regional water resources. Because energy is not traded efficiently, upstream countries may be using more water for hydropower generation than is economically efficient. Moving toward cost-recovery tariffs in energy would create a basis for evaluating and realizing the significant benefits from trade (which are currently foregone); unlock necessary investments in the power systems of the region; and, if combined with the introduction of water charges, create incentives for a more efficient use not just of energy but also of water resources.

The chapter begins by describing the resource endowments of the region and the impact that the move toward market prices in primary energy trade between CIS countries had on quasi-fiscal deficits, inter-CIS debt, and on energy trade (following section). The section on end-user tariff reform examines the present level of domestic energy tariffs and makes the case for moving toward cost recovery. The section on energy price reform deals with the issue of affordability of cost-recovery tariffs for the poor in the region and the political acceptability of tariff reform. Broad popular opposition and concern about its social impacts have been the main obstacles to tariff reform. The section dealing with the benefits of trade examines the impact of moving to cost-reflective energy tariffs on regional power trade. The section on institutional reform discusses the overall energy sector reform framework and explains why tariff reform needs to be combined with institutional reform—regulatory reform in particular—to attract private investment. This section draws on case studies to show what has and has not worked in energy sector reforms in the transition countries.2 Concluding remarks follow.

Natural Links: Resources, Interdependence, and the Breakdown of Regional Trade

Energy Resources and Energy Trade

The CIS-7 countries and their neighbors have abundant energy resources. The Caspian states of Azerbaijan, Kazakhstan, and Turkmenistan have among the world’s largest hydrocarbon (oil and gas) reserves, estimated at around 6.4 billion tons of oil equivalent. Uzbekistan’s reserves are smaller but still amount to 1.8 billion tons of oil equivalent. The Kyrgyz Republic and Tajikistan—home to some 20,000 glaciers and at the source of the two main Central Asian rivers, the Amu Daria and the Syr Daria—have substantial hydroelectric potential, as do Uzbekistan, Georgia, and Armenia (see Table 11.1).

Table 11.1.Energy Resources in CIS-7 Countries and Their Neighbors
Natural Gas

(billion m3)
Oil

(billion tons)
Hydro

(MW)
Armenia1,010
Azerbaijan1200.9950
Georgia2,800
Kazakhstan1,8201.12,060
Kyrgyz Republic2,950
Moldova60
Russia47,6006.743,400
Turkmenistan2,8300.11
Tajikistan4,020
Uzbekistan1,8500.11,700
Sources: EBRD (2001) and U.S. Energy Information Administration.Note: Figures show proven reserves for oil and gas, and current capacity for hydro.
Sources: EBRD (2001) and U.S. Energy Information Administration.Note: Figures show proven reserves for oil and gas, and current capacity for hydro.

While these resources were not used effectively under central planning, the Soviet system took account of the heterogeneous resource endowments among republics. Primary energy resources, including crude oil, natural gas, and coal were shipped across the Soviet Union in a vast network of transportation links. The power networks in the Caucasus and Central Asia were integrated, and energy resources were exploited on a regional basis. In Central Asia, hydropower was supplied to the regional network by the Kyrgyz Republic and Tajikistan, while Kazakhstan, Uzbekistan, and Turkmenistan supplied thermal-generated (oil and gas) power. In the Caucasus, Armenia supplied nuclear and thermal power to the regional system, Azerbaijan supplied thermal power, and Georgia supplied hydro and thermal power.

Following the dissolution of the Soviet Union, primary energy prices increased substantially for cross-border trade between the newly independent states. For crude oil and coal, the breakdown of the Council for Mutual Economic Assistance (CMEA) meant access to the world markets and hence an almost immediate increase in opportunity costs, as both resources are globally traded. For natural gas, the situation has been more complicated, given limited storage capacity and the difficulty of transportation.3 Despite these bottlenecks, natural gas prices have also increased substantially in intra-CIS trade.

The move toward market prices in intra-regional primary energy trade has resulted in the emergence of significant payment problems by CIS importers, who are unable to generate the cash flow in their domestic energy systems that is required to cover higher import costs. Table 11.2 shows the extent of quasi-fiscal deficits in the energy sector in the CIS-7 countries, as well as estimates for the share of public debt due to the energy sector. By the end of the 1990s, quasi-fiscal deficits were still in the range of 4 to 8 percent of GDP in the energy-importing countries and energy-related public debt amounted to between 10 and 20 percent of GDP, much of this due to foreign creditors. In Azerbaijan and Uzbekistan, the two net energy exporters of the CIS-7, quasi-fiscal deficits were also large, but the burden was carried mainly by domestic energy producers rather than financed from abroad. Developments in regional energy trade are therefore closely linked to the foreign debt problem in the CIS-7, and debt sustainability will require these economies to tackle the root of the quasi-fiscal deficits in the sector, low domestic energy prices.

Table 11.2.Quasi-fiscal Deficits and Total Debt of the Energy Sector, CIS-7 Countries
CountryQuasi-Fiscal

Deficit

(in percent of GDP)
Energy Sector

Debt for 2000

(in percent of GDP)
Comments
Armenia0.4112 (or $224 million}Flow deficit greatly reduced since 1998, bur total Financing in 2000 still 3.9 percent of GDP; debt adjusted for debt-equity swap with Gazprom in 2000.
Azerbaijan271Deficit relates to total energy sector including oil, gas, and power; at least $200 million in foreign debt was contracted for the power sector.
Georgia62181 (or $555 million)Deficit falling since 1999; cash collection in particular was up in 2002.
Kyrgyz Republic73Deficit in all utilities; power sector alone was 4 percent of GDP.
Moldova18.5Debt includes debt to Gazprom for gas deliveries, but adjusted for debt equity-swap.
Tajikistan7.3111Deficit for power and gas sector includes technical losses; debt is total accumulated payables.
Uzbekistan303Uzbek power prices are significantly below long-run marginal cost; however, gas prices were raised roughly to import parity level from 1995 onward (although at the official exchange rate).
Sources: IMF (Uzbekistan), World Bank (other countries).

2000.

1999.

2001.

Sources: IMF (Uzbekistan), World Bank (other countries).

2000.

1999.

2001.

Partly in response to regional payment problems, and partly out of a political desire for resource independence, the CIS-7 countries have moved toward a policy of power sector self-sufficiency. In the Caucasus, trade has fallen drastically, with limited imports to Georgia from Armenia and Russia, and no trade between Azerbaijan and the rest of the system. Energy trade in the South Caucasus is also affected by the dispute between Armenia and Azerbaijan over Nagorno Karabakh. There are currently no official trade relations between the two countries.

In Central Asia, hydropower that had been exported to meet peak demand in Soviet times was substituted by domestically generated thermal power. Table 11.3 shows the change in power trade in Central Asia between 1990 and 2000. It is notable that exports of hydropower to Kazakhstan and Uzbekistan from Tajikistan and, to a lesser extent, the Kyrgyz Republic fell substantially over this period. Likewise, imports by Tajikistan and Uzbekistan of thermal-generated power fell significantly. This represents reduced system integration as opposed to the more general decline in power production between 1990 and 2000: the ratio of power trade to total output in Central Asia fell 80 percent over this period.

Table 11.3.Central Asia Power Trade, 1990 and 2000(In GWh)
19902000
KazakhstanExports
Imports9,0641,269
Kyrgyz RepublicExports3,9782,376
Imports318.7
TajikistanExports5,700333.8
Imports6,9001,681.2
TurkmenistanExports6,0501,060
Imports
UzbekistanExports13,000848.7
Imports12,5001,349.8
Source: EBRD consultants’ report.
Source: EBRD consultants’ report.

Hydro Power and Water Management: The Case of Central Asia

Changes in hydropower generation patterns and the drive of the Kyrgyz Republic, and to a lesser extent Tajikistan, toward energy self-sufficiency have led to growing tensions over water allocation among Central Asian states and underlie the unsustainable nature of water management in Central Asia. The inefficiency and environmental costs of water management in Central Asia and the Aral Sea basin in particular are well documented,4 but excessive water use has also caused problems in Armenia, where the water table of Lake Sevan has been reduced as a consequence of water overuse.

Much of the pattern of Central Asian water use is a direct legacy of the old Soviet system. Starting in the 1920s and culminating in the 1950s, the Soviet Union began to put in place the infrastructure and institutions for the large-scale use of Central Asia’s water resources. The primary focus was on irrigation and the production of cotton, and to a lesser extent other crops. Other concerns, such as hydropower production and flood protection, were clearly subordinated to the primary objective of maximizing the output from irrigated agriculture. The massive Toktogul and Nurek cascades on the Syr Daria and Amu Daria, respectively, were operated in irrigation mode, with water release peaking in summer to coincide with the vegetation period. The summer hydropower generated in the process was mostly consumed in Uzbekistan and Kazakhstan. The two upstream republics were compensated with Uzbek gas and Kazakh coal, which allowed them to meet their winter electricity peak.5

The breakup of the Soviet Union has in many ways exacerbated existing problems with this system. The Central Asian water and energy infrastructure was designed as an integrated, centrally managed system. With the emergence of five independent states this structure fell apart, and what was once a technocratic exercise in central planning became a problem of intergovernmental coordination. The downstream states insisted on a minimum level of water release in summer as part of their riparian rights, but the poorer upstream states found it increasingly difficult to find the hard currency to pay for winter fuel. In this situation the five Central Asian governments resorted to annual barter arrangements that in essence emulate the old Soviet system. Thus, in the case of the Syr Daria, the Toktogul cascade is operated in irrigation mode, allowing Uzbekistan and Kazakhstan to maintain their extensive agricultural programs. The Kyrgyz Republic also exports summer hydropower for which it is paid in Uzbek and Kazakh coal and gas. Institutionally, the system is supported by a new regional water management body, the Interstate Commission for Water Coordination (ICWC), which was set up in 1992 in Tashkent and made responsible for the management of annual water allocations and the schedule for the operation of reservoirs.

The new arrangements have not prevented the dispute over water rights from growing. The built-in tension of this system is that it favors heavily the cotton producers downstream, who consume most of the water (see Table 11.4), while the mountainous states shoulder much of the burden in terms of maintaining and operating the upstream reservoirs. Not surprisingly, following their independence the Kyrgyz Republic and Tajikistan have started to press for more generous water quotas that reflect their abundant resources. They increasingly ignore the tough winter release limits imposed on them under the old water management system, and run the cascades on hydropower mode. The change in reservoir operations is causing considerable problems downstream, where water is lacking in summer and the dilapidated infrastructure cannot cope with the additional volumes in winter.

Table 11.4.Water Allocation in Central Asia, October 1996–October 1997(In percent of water flow)
Amu DaryaSyr DaryaTotal
Downstream
Kazakhstan3010
Turkmenistan3624
Uzbekistan364238
Upstream
Kyrgyz Republic<1<1<1
Tajikistan13711
Source: Micklin (2001).Note: Residual amounts are allocated to the Aral Sea region.
Source: Micklin (2001).Note: Residual amounts are allocated to the Aral Sea region.

Moreover, the lack of transparency in the annual bilateral barter arrangements negotiated between the different upstream and downstream parties is a cause of significant friction. Even if the system were run in irrigation mode, the downstream states object that, because they are forced to buy summer power, they are in essence paying for water resources that under riparian rights should be theirs for free. Yet the barter arrangements make it impossible to assign a monetary value to the different energy flows and establish what implicit water charges may be contained in the agreement, although generally implicit water charges would appear to be negligible. Also, because of varying water flows and irrigation water needs that depend on climatic conditions, there is significant uncertainty over the water take off requirements of downstream countries in any one year. In this situation, the incentives for the upstream countries to stick to the bilateral arrangements and operate the system in irrigation mode are not very high.

The region’s water management problems are further exacerbated by the unreformed nature of much of Central Asian agriculture, which creates an artificially high water demand. As a consequence of the state’s procurement of cash crops at below market prices and the massive subsidisation of water inputs through the system of water allocation, the efficiency of water use is extremely low. It has been estimated that as much as a third of the water delivered to farms is not used by crops, which is almost three times more than in a well-managed system. At 12,000 m3 per hectare, water use in the region is more than double that of Israel, the world’s most efficient water manager.6

The lack of reform has affected the ability and willingness of farmers and the state to maintain the extensive irrigation infrastructure. As a consequence, these assets, which were of poor quality to start with, have begun to deteriorate. The deterioration in the hydraulic infrastructure and embankments has increased the risk of flooding and has reduced the carrying capacity of the rivers during the high water flows of winter. The problem extends beyond agricultural infrastructure to the large dams and reservoirs upstream, and the safety of many of these is now seriously impaired. The total investments needed for rehabilitation and repair have been estimated at around $10–20 billion. In the water and power sectors alike, there is a need to adjust tariffs both to provide an incentive for the efficient use of these resources and to raise cash for the rehabilitation and maintenance of systems.

Getting the Prices Right: The Crucial Role of End-User Tariff Reform

In the Soviet era, power tariffs were low relative to long-run costs in order to provide abundant supplies for use in production and consumption. This situation was sustainable, given the implicit and explicit subsidies in the form of low primary energy and water prices and budgetary transfers to the power industry. As the previous section showed, however, the implicit subsidies have been eroded as the price of primary energy has risen following the end of communism. In addition, supply costs have increased due to the loss in efficiency caused by the breakdown in regional trade. The inability of budget-constrained governments to increase support and the failure to increase tariffs to cost-recovery levels have undermined the financial viability and sustainability of power sectors—and power trade—in the region.

The extent of power sector underpricing is illustrated in Table 11.5. The optimal pricing rule that is applied to transition economies where demand is stagnant and excess capacity often exists, sets price somewhere between marginal operating cost and long-run marginal cost (LRMC). Marginal operating cost is defined as the cost of producing an extra unit of output using the existing capital stock. Long-run marginal cost is marginal operating cost plus the cost of additional capacity required to increase output. The price could be expected to rise above marginal operating cost and toward LRMC as demand picks up and investments are undertaken.

Table 11.5.Power Tariffs 2001(In U.S. cents per kWh)
CountryResidential TariffIndustrial Tariff
Armenia4.42.9
Azerbaijan2.16.3
Georgia4.23.3
Kyrgyz Republic0.61.4
Moldova5.25.2
Tajikistan0.21.1
Uzbekistan1.01.5
Source: EBRD survey of regulatory authorities.
Source: EBRD survey of regulatory authorities.

Further work is required to establish the LRMC for the CIS-7 countries. LRMC is specific to each (national or regional) power network and depends on factors such as technology, fuel prices, transmission links, and so on. For example, LRMC is likely to differ according to whether power generation is based on hydro, gas, or coal. Similarly, the LRMC for a hydro system operating in a regional system that includes thermal capacity is likely to be different from the LRMC for the same hydro system operating in isolation. The LRMC is around US¢8 per kWh in the United States, and slightly higher in Western Europe. It is likely to be lower in the CIS because of lower economic fuel prices, although in practice the price might have to rise above LRMC to recoup past debt. This depends on a number of factors, such as rents-to-existing-capital under LRMC pricing and terms of debt rescheduling, and is not discussed here in detail. Notwithstanding this, LRMC pricing would ensure the viability of power sectors on a forward-looking basis. As a rough estimate, we assume LRMC to be of the order US¢5 per kWh, which suggests that further price increases will be required in all countries, with the highest raises needed in Central Asia. For example, various consultant reports for Uzbekistan put the LRMC in that country at US¢3.5–7 per kWh.

In addition to the widespread problem of underpricing, the difference in prices between consumer groups can also be problematic. Typically, residential power tariffs are lower than those for industry. On average, industrial tariffs are 2.1 times higher than residential tariffs across the CIS-7 countries. This contrasts sharply with Western Europe, where industrial tariffs are on average two-thirds of the price charged to households, reflecting the relative costs of supplying these two customer categories.

A further distortion relates to price differentials according to time of day and time of year. In general, LRMC will be specific to times of day and year, given that electricity demand will fluctuate substantially. Consequently, demand during a peak period generally will incur operating costs plus capital costs, while offpeak demand generally will incur only marginal operating costs. This is because capacity exists for peak demand. Additional demand in the peak therefore requires additional capacity. Additional off peak demand can be satisfied using existing capacity. There has been very little progress as regards time-of-day pricing in CIS-7 countries, the exception here being Armenia. Prices tend to be uniform across time of year, and, typically, charges for large customers are small relative to the underlying cost. Clearly there is significant scope for further tariff rebalancing to pricing arrangements throughout the region.

The collection of payments plays a critical role in power pricing in the transition economies (Table 11.6). Cash collection (the percentage of total cash collected relative to amount billed) and revenue collection (cash collection plus barter payment) are far below 100 percent, and commercial losses (defined as nonbilled consumption) are substantially above OECD levels. In Western Europe and the United States, cash collection is typically close to 100 percent, and commercial losses are typically close to zero. In the CIS-7 countries, cash collection averages only 46 percent and commercial losses 20 percent. Cash collection is particularly low for industrial consumers, and consequently incentives for industrial restructuring are limited.

Table 11.6.Power Cash Collection and Commercial Losses(In percent)
CountryCash CollectionCommercial Losses
Armenia871
Azerbaijan30115
Georgia32228
Kyrgyz Republic45217
Moldova55228
Tajikistan14
Uzbekistan251
Source: EBRD survey of regulatory authorities.

2001.

2000.

Source: EBRD survey of regulatory authorities.

2001.

2000.

The same issues that applied to power tariffs—low prices, price variation between consumer groups, and low collection rates—are also present in the district heating sector. Heat prices are often close to zero in the CIS-7 countries. This can be compared with the LRMC of a freestanding boiler—the closest alternative to district heat—of around US¢3 per kWh in Central and Eastern Europe, the Baltics, and Southeastern Europe, and US¢2 per kWh in the CIS, where gas is available more cheaply. An LRMC benchmark for district heat is difficult to establish because systems differ widely, and there are few comparison networks outside the region. As in the case of power, industry tends to be charged higher rates than residential consumers, despite the higher cost of serving the latter. Anecdotal evidence suggests that collection is also problematic, particularly from public sector consumers, but no systematic data are available.

There is little disagreement in principle over the main challenges facing the energy sector: to increase prices, to reduce the cross subsidy between customer categories, and to improve payments discipline. Meeting these challenges will reduce energy sector liabilities for governments and thus reduce pressure in the areas of budget and external debt. It will make investments economically viable without the need for a sovereign guarantee. Finance will become available for new power and heat-generating stations and the upgrading of transmission and distribution networks, resulting in declining losses and improvements in system security. Residents will have incentives to regulate heat and power consumption. Industry will have incentives to improve energy efficiency and to move away from energy-intensive production methods. Sector cash flows will be sufficient to support trade, and this, together with water tariff reform, may in turn facilitate the move toward more efficient water management as well (see section below).

Affordability and Energy Price Reform: How to Protect the Poor

Price reform does not come without costs; and for elected governments, not the least of these might be political, because tariff increases are unpopular with voters. Thus, tariff increases must be sold on the basis that they will result in improved sector performance. Gradual tariff increases may ease political concerns, though this would not be practical where investment needs are urgent. It would be preferable to introduce subsidies to benefit all groups, while moving away from the blanket subsidy implicit in low prices. This can be done through a lifeline tariff, as discussed below.

From a social point of view, the impact of tariff reform on poorer groups is of particular concern. Electric power and heat are important basic services. People without access to power may suffer genuine hardship. In the climate of the CIS-7 countries, people without access to heat would be putting their health at significant risk during the winter. Household incomes have already fallen in many transition economies, particularly in the CIS-7 countries, and there is widespread poverty. This has important implications for how price adjustments can be introduced and highlights the importance of effective compensation mechanisms.

The affordability of price increases is a concern for all those households living at or below subsistence level. A significant share of the population in most CIS countries would have difficulty affording large price increases for energy. The poor typically comprise between 25 and 50 percent of all households. While a significant number of these households lack connections to utilities, particularly in rural areas, the majority of the urban poor would be greatly affected by any price increases. In addition to welfare concerns, the political viability of price reform remains a key issue. Attention needs to be paid therefore to the level of poverty in the population and to the burden of utility payments in current household budgets.

How is it possible to make sure that all those eligible to receive energy subsidies really benefit while minimizing the subsidy to those who could afford to pay cost-recovery prices? And what are the implications for public finances and the financial performance of the utility? Broadly speaking, there are four types of subsidies for the transition economies: continued supply to nonpaying customers or across-the-board price subsidies; lifeline tariffs, where consumers receive an initial block of energy for free or at a low price, with consumption of additional blocks charged at higher prices; targeted subsidies; and nontargeted subsidies or general income support. In evaluating the benefits of these approaches, it is important to consider certain conflicting aims. On the one hand, the effectiveness of the subsidy scheme depends on the extent to which it covers all potentially eligible groups. On the other hand, its efficiency depends on reaching only those consumers who need it. Additional objectives are to minimize the scheme’s costs for the government or the utility; to make it fair and cost efficient to administer; and to minimize distortions arising from the scheme. Clearly a country must also have the institutional capacity to operate the chosen scheme.

Providing across-the-board price subsidies or continuing to provide power to nonpaying customers essentially would mean an extension of the status quo and would not achieve most of these aims. While such a scheme typically covers all poor people with utility connections (which may in some cases be a minority of the poor), it lacks targeting. As a result, the costs are high either for the government or for the utilities. Incentives to economize on energy consumption are also absent under this scheme.

The benefits of the other subsidy mechanisms are more difficult to evaluate and, to a large extent, depend on the level of poverty, its frequency and severity, and on the capacity of the state to administer social transfer schemes. Lifeline tariffs are most appropriate when the number of people with difficulty paying is relatively large and state capacity is limited. Lifeline tariffs can be implemented only when consumption can be adequately metered, which applies to electricity but not necessarily to heating services, where metering is often possible only at the apartment block level. Apartment size is sometimes used as a proxy for heat consumption. Moreover, the costs of lifeline tariffs could be quite substantial compared with more directly targeted subsidies. On the plus side, lifeline tariffs benefit all groups in the population and can therefore ease political resistance to tariff increases.7 A lifeline tariff with a generous free consumption block declining over time could therefore be advantageous from a political point of view, though this would have to be tailored according to government budget constraints.

Targeted subsidies can achieve better efficiency if the subsidy is directly related to household incomes. These subsidies are appropriate for situations where there is no metering and therefore are a possible means of subsidizing heat consumption. Russia and Ukraine have at various times used a form of targeted subsidy known as a burden limit. This is defined as a percentage of expenditure on utilities above which a household would receive a subsidy. Such burden limits often fail to reach a significant proportion of the poor, however, who may be forced to forego energy consumption in order to pay for food and shelter. Measurement of household incomes is particularly difficult in many transition countries, where large informal economies exist. The introduction of targeted subsidies has to take into account the state’s administrative capacity. These subsidies will typically work best when combined with an already existing system of income support, as is the case in many Central European countries. In many transition economies, however, the coverage of targeted subsidies is often less than perfect, leading to a trade-off between the higher coverage achieved by a lifeline tariff and the better targeting and potential cost savings of a targeted scheme. For instance, targeted subsidies were adopted in conjunction with a recent EBRD investment in the power sector in Georgia, but considerable technical assistance was needed to identify the poor.

While there may be difficulties implementing either lifeline tariffs or targeted subsidies in CIS-7 countries, there is an urgent need to move away from blanket subsidies in the form of low prices and payment collection. Depending on institutional capacity and metering—both of which should improve over time—either lifeline tariffs or targeted subsidies could be implemented efficiently and help to reduce political resistance and the adverse social impacts of tariff reform.

Unlocking the Benefits from Trade

The resurrection of trade in Central Asia and the Caucasus could limit the tariff increases required to reach cost recovery. When resource endowments across countries within a region are heterogeneous, there is the potential for substantial system cost reductions through trade. At the same time, without domestic tariff adjustments the benefits from regional trade will remain unrealized because cash flows within the system are insufficient to cover import costs.

The benefits from regional trade in energy are particularly large in regions where countries are in different time zones and thus have noncoincidental peak demand, in which case trade opportunities for a given installed capacity increase. In addition, costs associated with reserve capacity to meet unexpected upswings in demand or to compensate for units that are unexpectedly unable to supply energy are lower in an integrated system.

In the case of Central Asia, the installed capacity would suggest that opportunities for profitable trade still exist (Table 11.7). The Kyrgyz Republic and Tajikistan have hydro capacity far in excess of peak demand and thus have the potential to export. It is intuitive that these countries should provide power to the system in peak periods. Looking only at the power sector, the optimal solution would usually be to generate hydropower during the winter peak. This pattern may change, however, if wider considerations are taken into account. Much of the thermal capacity in, for example, the Kyrgyz Republic comes from combined heat and power stations (CHPs), which are run most effectively in winter when the synergy between the two outputs (heat and power) is strongest. Similarly, the value of the by-product of hydro generation, water release, is highest in summer, suggesting that the traditional operation of the system in irrigation mode might in fact have some economic justification. A recent World Bank study found that the marginal costs of Kyrgyz power generation in winter are in fact lower than the marginal costs of Uzbek summer production, especially once the value of irrigation water is taken into account.8 This suggests that it may be advantageous to produce hydropower in summer, when it replaces expensive Uzbek capacity, rather than in winter, when it would replace cheaper Kyrgyz capacity.

Table 11.7.Installed Capacity and Peak Load of the Energy System in Central Asia(In megawatts)
CountryHydro CapacityThermal CapacityPeak Load
Kazakhstan2,0578,722
Kyrgyz Republic2,9503,7422,622
Tajikistan4,0213002,723
Turkmenistan<402,6031,581
Uzbekistan1,7059,8737,571
Source: EBRD.
Source: EBRD.

In addition to the benefits from trading thermal and hydropower, there may be an additional benefit from trading thermal power, to the extent that efficiency (both thermal and operational) differs between Kazakhstan, Uzbekistan, and Turkmenistan. Without presenting comprehensive data here, it is likely that there are differences in thermal efficiency, given the different mix in plant fuel type, technology, and age.

The key to unlocking benefits from trade is in pricing policy. A necessary condition for power trade, whether this is centralized at the level of countries or at the level of generators and large consumers, is that sectors are financially viable and thus able to pay for imports. If a decentralized market in power is to develop, end-user tariffs must reflect the costs for each customer category. Development of this type of market is typically one objective of a power sector reform program.

In addition to cost reflectivity at the end-user level, trading prices should also be cost reflective, to ensure that trade is efficient. The limited trade that currently takes place in Central Asia is characterized by price distortions. It is governed by bilateral and multilateral agreements that cover barter exchange of electricity, water, and primary energy (see also preceding section on end-user tariff reform).

Under economic pricing, in order to encourage efficient trade, resources would be valued at opportunity cost, that is, their value in alternative uses. There is a link here between establishing the conditions for efficient trade and introducing an appropriate charge for water. Despite the overall economic benefits of running the hydropower infrastructure in the upstream countries in irrigation mode, the value of additional power to the downstream countries in summer may be below the opportunity cost of foregone power generation capacity upstream in winter.9 In this case, power trade would not take place without the introduction of an additional charge for water. Indeed, given the overall economic benefits, water should at least be priced at a level that reflects winter (thermal) power generation cost to allow efficient trade to take place. Studies suggest that the downstream value of water for use in irrigation far exceeds this minimum price. The introduction of water charges would signal to Uzbekistan the appropriate level of water to use for irrigation. It is likely that higher prices would lead to some reduction in downstream water usage, with increased hydropower generation in the winter peak by the Kyrgyz Republic and Tajikistan.

Given the strong insistence of the downstream countries on their riparian water rights, the introduction of water charges may not be easy to accomplish politically. One alternative, which has recently been proposed by Kazakhstan and the Kyrgyz Republic for the Kambarata hydropower station, is for the downstream countries to invest in the upstream infrastructure, thereby allowing upstream countries to better regulate the water flow while at the same time increasing hydropower generation capacity. It remains to demonstrated, however, whether hydro capacity forms part of the sector's least-cost expansion plan, whether large investments in hydro are bankable, and whether they can meet the stringent social and environmental conditions imposed by most financial institutions and the World Commission on Dams. Internationally, the case of Lesotho and South Africa provides an example of a multiyear agreement whereby South Africa pays Lesotho water royalties for an agreed off-take of drinking water resources. The agreement also includes significant upstream investments by South Africa and resulting hydropower benefits for Lesotho. Lesotho in turn bears all environmental and social costs of the investment and loses all rights to its water resources.10

Moving to technical capacity for regional trading, investments would be required in the areas of transmission network substation rehabilitation and installation of metering and data communication equipment. The costs associated with these investments are small relative to the associated benefits in terms of cost reductions associated with trade. In addition, investments in institutional capacity would be required if a market is to work. The institutional arrangements to support a market are discussed in the following section.

In the Caucasus, installed capacity, as shown in Table 11.8, suggests that there is scope for gains from regional trade, given the heterogeneity of resources. The intuitive dispatch pattern would be for nuclear and thermal generation to meet base demand, with hydro generation supplying peak demand. This can be compared to the present situation, where Georgia generates hydropower to meet base demand. Thus benefits would accrue if Georgia were to export hydropower to Armenia in the peak period, and to import thermal-generated power from Armenia, Azerbaijan, or Russia to meet base demand.

Table 11.8.Installed Capacity in the Power Sector in the Caucasus(In megawatts)
CountryThermalHydroNuclear
Armenia1,4001,000300
Azerbaijan4,000900
Georgia2,1002,800
Source: EBRD.
Source: EBRD.

Currently, the incentives for trade are limited due to price distortions. For example, the price of power in Georgia is too low to provide an incentive to import power in base periods. If power were to be priced at opportunity cost, that is, if hydropower were priced at the level of the Armenian thermal-generating plant, then the correct incentives for trade would be in place. In this situation, demand for imports to Georgia would increase—the price of hydro would be higher than imported base generation, nuclear or thermal—and there would be demand for Georgian hydropower in peak periods at a price up to the peak generating cost in Armenia.

There would be a need for investments in network rehabilitation, metering, and data communication to support trade, but associated costs would probably be small relative to benefits. There is also the possibility of building new network connections to link Azerbaijan, Georgia, and Turkey. In this case, Georgia could export hydropower to Turkey, while Azerbaijan could export some of the newfound gas reserves in the Caspian Sea in the form of power. Preliminary consultants' reports suggest that this interconnection would be economically viable.

Making Market Prices Stick: The Need for Institutional Reform

To make tariff adjustments work in the energy sector institutional reform is critically needed alongside price adjustments in two areas. First, of private investment and management expertise must be attracted to improve collection rates and the technical efficiency of the power system. Improvements in efficiency resulting from private participation may in turn reduce the price increases necessary to ensure the viability of the power sector. The introduction of the private sector would also help to mobilize finance and increase the possibility of further investments in the sector. Second, institutional reforms are needed to support private investment through a system of credible and effective regulation. The power sector is a market characterized by significant network externalities in transmission, and full competition can only emerge in a well-regulated system. Private investment will only be forthcoming if network access is guaranteed and tariffs are set in the context of transparent regulation.

There are powerful reasons for accelerating private sector involvement when payments discipline is low, as it is in the CIS-7 countries. A private firm owned or managed by a foreign strategic investor will have a stronger incentive to enforce payment discipline than elected or underpaid officials. It will also have the technical knowledge and finance required for essential re-metering programs, computerization of billing, and other measures that can help improve payments performance. Effective metering is necessary for improved collection. Experience shows that people are prepared to pay for what they can be shown to consume. Furthermore, effective metering is required in order to reduce commercial losses. For example, there is evidence to demonstrate that commercial losses fall after Soviet-style meters are replaced with modern tamper-proof meters. Experience to date suggests that in cases where the private sector has entered power distribution, payment collection has gone up. Table 11.9 outlines developments in Kazakhstan (Almaty and Karaganda), Georgia (Telasi) and Moldova, where there have been major improvements in payment discipline.

Table 11.9.Cash Collection Rates1 Pre-and Postprivatizarion(In percent)
PreprivatizationYear 1Year 2Year 3
Almaty, Kazakhstan1517080
Karaganda, Kazakhstan10253546
Telasi, Georgia8142955
Moldova2658
Source: EBRD.

Defined as the ratio of cash revenue to total amount billed.

Source: EBRD.

Defined as the ratio of cash revenue to total amount billed.

Apart from improving payment discipline, there is only limited evidence from transition economies to suggest that private sector participation improves operational performance. Experience in infrastructure reform in the United Kingdom and the widespread private participation in infrastructure around the world, however, suggest that the introduction of the private sector in a well-regulated and liberalized environment results in performance improvements.

A Blueprint for Reform

Introducing the private sector into a well-regulated and, where possible, liberalized environment is an important objective, but the order in which reform policies are implemented is also critically important. There are no formulaic solutions, and the appropriate reforms will tend to vary from country to country. For example, an institutionally more-advanced country may adopt more sophisticated trading arrangements, while recourse for investors in the event of a regulatory dispute may differ according to the level of independence and integrity of the local judiciary. The following steps—based on the successful experience in England and Wales and lessons from transition countries where restructuring and privatization have been undertaken—provide some broad parameters for power and heat reform.11

  • Corporatization and commercialization of the industry: The first step is to set up a joint-stock company wholly owned by the state, with the separation of accounts for different parts of the business. The next step is to unbundle the company, dividing it into subsidiaries into which private activity can be introduced and, ultimately, where market liberalization can occur.

  • Regulatory authority: A regulatory authority that is free from day-to-day political interference needs to be set up. This agency should establish tariffs for those parts of the industry that remain a monopoly. This is likely to include retail tariffs as well as access charges. The agency should also develop and implement rules for network access. There will also be a need to enforce environmental, health, and safety standards.

  • Entry of the private sector: Before a full framework for private participation is in place (i.e., before regulatory and market rules are fully developed), a limited number of concessions can be granted when there is an urgent need for rehabilitation of power generation assets.12 This is with a view to limiting the amount of capacity tied up under long-term contracts at the time of market liberalization. Once the institutional framework is in place, assets may be sold outright or tendered under management contracts or concessions, and free entry of private companies into the power generation sector may be permitted.

  • Tariff reform: Tariffs should be raised to cover long-term costs and to reduce any cross-subsidy element, while taking care to ensure that customers can afford to pay.

  • Market liberalization: Typically, this can be achieved by allowing third-party network access. Generators would compete with each other for bilateral contracts with large consumers for the supply of power. In countries with strong institutional capacity and where cash collection is not a problem, it may be feasible to introduce power pools, whereby a central body takes bids to supply from generators and determines the cheapest plant configuration to meet power demand. It subsequently requests this plant to supply the network, collecting revenues from distribution companies and large consumers and making payments to generators accordingly. There are substantial, and often prohibitive, technical requirements (for example, data, communications, and software equipment) and institutional requirements to operate a pool successfully.

With respect to the best order of implementation for these measures, it is crucial that a sound regulatory framework be in place prior to privatization and that tariffs have been increased to at least cover operating costs. Furthermore, when privatization/private sector participation occurs, it should involve strategic investors in order to maximize privatization revenues, to secure financing for necessary investments and to strengthen incentives for improved efficiency. Privatization of the distribution network should occur not later than privatization of power generation facilities when payment discipline is a problem. This is because privatization of power generation when there is low cash collection is likely to produce low revenues from sales and may not support necessary investments. This could in turn lead to increasing political objections to such changes. Regarding liberalization, this should be the last step, undertaken after industry restructuring, the setting up of a regulator, and the introduction of the private sector. The industry structure at the end of the reform process is represented in Figure 11.1.

Figure 11.1.Reformed Structure of Electricity Sector

Source: EBRD.

Reform Progress in CIS-7 Countries

What reform progress has been achieved to date in the CIS-7, and what lessons can be drawn from this? In Georgia and Moldova, industry restructuring has taken place, independent sector regulators have been introduced, and the private sector has been introduced to distribution (in both countries), and to generation (in Georgia). The challenge in both countries is to consolidate reform through the privatization of remaining state-owned distribution companies, thus securing financing to cover ongoing operation and capital costs. The key lesson from these experiences is that reforms cannot stop halfway. Unless incentives for efficiency improvements and better cash collection are introduced along the whole chain from production to distribution, the financial viability of the sector will continue to be impaired.

Armenia embarked on a path similar to Georgia and Moldova, with unbundling and the creation of an independent regulator. Two attempts to privatize distribution companies—one through a tender and the other through negotiation with a strategic investor—failed. Recently, distribution networks were sold to a private investor with limited power sector experience. The challenge now is for the regulator to support this investor, providing a stable environment for the turnaround of the distribution company. In time, the challenge will be to introduce the private sector to the ownership and/or management of thermal generating plants.

Uzbekistan and the Kyrgyz Republic have started to restructure their power industries. In Uzbekistan, most entities in the power sector have been corporatized as separate entities. In the Kyrgyz Republic, four distribution companies, a generation company, and a transmission company have been set up. There has been limited progress in both countries as regards regulatory reform; strengthening of the regulatory framework therefore remains a key challenge. There are important questions of how assets will be bundled for private sector participation, and the form that this will take (whether through asset sales, management contracts, or concessions). In addition, decisions must be made on the type of market models that will be introduced.

In Tajikistan, reform progress has been made through corporatization of the country's power utilities, though further steps toward commercialization have yet to be taken. Prices are still the furthest away from cost recovery among the CIS-7 countries, and quasi-fiscal deficits in the energy sector are a very considerable drain on public resources.

The Regional Dimension

In addition to institutional arrangements that would facilitate private sector investment, other reform challenges relate to the institutional framework for the development of regional energy markets. If, as argued above, regional energy trade would unlock significant efficiency gains and thus reduce the extent to which prices need to rise to cover production costs, then institutional reforms should pay attention to the conditions necessary for such trade to take place.

A prerequisite for any power market is a grid code (technical terms and conditions for market participants); without this system integrity is jeopardized. In a regional market context, national grid codes should be at least mutually consistent, although ideally there would be a regional grid code. The point here is that national authorities should not develop grid codes in isolation. Coordination might be achieved here through a regional regulators association, possibly supported by international financial institutions (IFIs) and bilateral donors.

A second prerequisite for any market is a cost-reflective transmission tariff methodology—transmission tariffs should cover costs and, in some circumstances, should reflect geographical differences in costs. In a regional context, questions arise over how system operators in transit countries should be recompensed for transmission costs occurring due to international trade. Charging mechanisms here should be cost based, that is, based on the underlying flows of electricity (as opposed, for example, to distance based or based on the number of countries between trading parties).

Another important area is whether transmission costs are levied on generators or consumers: if in one country generators are charged while in another consumers are charged, the result could be double charging for trading parties, something that would undermine economic trade. A harmonized approach is required here: all countries should agree, for example, that transmission charges are to be levied on generators. As in the case of grid codes, a regional regulators association, supported by IFIs and bilateral donors, could be an appropriate forum to achieve harmonization.

Mechanisms for coordinating investments in a regional context should be in place, given that these may be both substitutes and complements; for example, a transmission investment strengthening links between countries might substitute a generation investment in an importing country, and might complement a generation investment in an exporting country. It should also be noted that investments may require sovereign involvement (debt or guarantee), particularly in the medium term before industry reform has been fully implemented. Given that sovereign debt capacity is typically limited, these investments should be prioritized: countries may have to choose, for example, between domestic generation investments and regional transmission investments. The mechanisms for these choices should be good information—a regional system study—and cooperation between governments. In the medium term, coordination can be achieved through harmonization of regulatory rules by national regulators.

Regarding market rules, bilateral contracts are likely to be the most appropriate model for liberalization, given the cash collection problems in CIS-7 countries. For noncontracted demand, a balancing market13 could help to ensure that this is met at minimum cost. A regional balancing market would support impartial dispatch (i.e., no favoring of generators on the basis of nationality). To the extent that this is not politically feasible, coordination between national markets would be required.

Conclusions and Key Policy Recommendations

The CIS-7 countries currently do not make the best use of their rich natural resource endowments and of the potential for mutually beneficial regional trade in energy and water that results from the distribution of these endowments across countries. This paper has argued that a key reason is the absence of cost-reflective energy and water pricing, both on the domestic market and in existing trade between countries.

The failure to raise energy prices on the domestic market to cost-recovery levels has led to huge quasi-fiscal deficits in the energy sector, which have drained scarce public resources and in many cases contributed significantly to these countries' current debt problems. Low domestic cash flows have discouraged investment and regional trade, and have led to an inefficient drive toward self-sufficiency in energy supply.

The lack of cost-reflective prices for primary energy, water, and power in bilateral trade results in the wrong incentives as regards the mix between thermal and hydro generation and the use of water for irrigation. The low price of water for downstream countries encourages overuse in agriculture, while at the same time encouraging upstream countries to restrict exports, thus creating tension. Energy prices below cost-recovery levels distort the incentives for private power and thereby also preserve the inefficient regional allocation of water resources.

Any solution to the present problems will therefore need to start with domestic energy price adjustments. The CIS-7 countries vary in the extent to which price reform has progressed, with the Caucasus countries and Moldova being the most advanced, and the Kyrgyz Republic, Tajikistan, and Uzbekistan lagging the farthest behind. In parallel, efforts should be increased to move toward cost-reflective pricing in the trade of primary energy and water resources between CIS states. In this regard, the water and energy nexus in Central Asia represents an urgent, if politically sensitive, challenge.

In addition to tariff reform, restored financial viability will require improved payment discipline, which in turn will require private sector participation in distribution. Evidence suggests that the institutional framework for private participation is key and that a strong regulatory framework is required to provide confidence to investors. In this respect, there is scope for strengthening regulatory institutions in all CIS-7 countries, as a tradition of independent regulation has not yet been established. Tariff reform will require complementary changes in social support for power to remain affordable for poorer groups. Again, progress has been limited here, and there exists scope for improvement through the introduction of lifeline tariffs or targeted subsidies. Changed end user prices would provide a basis for the mobilization of investment financing and for trade between countries. Immediate price increases, complemented by progressive strengthening of regulation, would provide a good signal to investors and would go a long way to providing the cash flows necessary to support trade.

Price reforms, both domestically and in regional energy and water trade, are politically difficult. Governments need to be sensitive to social concerns if price increases are to be sustainable. At the same time, in many CIS-7 countries the deterioration of the infrastructure has proceeded to a point where failure to solve the financial crisis in the energy sector presents a significant political risk in its own right. The looming crisis may impart a sense of urgency to governments to move forward rapidly with price adjustments, and IFIs should stand ready to support the process with targeted assistance to the poor, as well as financing for critical investments.

Finally, as the experience of the region shows, there are clear benefits to choosing an institutional arrangement that is commensurate with implementation capacity. When the tradition of independent regulation is weak, it may be better to introduce the private sector gradually through management contracts and aim to attract fresh private capital only once a track record of regulation has been established and the energy sector has begun to generate positive operational cash flows.

Cheap energy was one of the motors of socialist industrialization. Overcoming this legacy is one of the central challenges of transition. Failing to address the issue could jeopardize the prospects for sustainable growth for years to come.

    Other Resources Citing This Publication