Nuclear Capacity Auctions.

AuthorFridolfsson, Sven-Olof

1. INTRODUCTION

The decision of the Swedish Parliament in 2010 to open up for new nuclear power marked a u-turn in the country's nuclear policy. The previous 30 years the official policy had been full abandonment. Reactors were to be phased out as fast as the energy system permitted, bearing in mind the consequences for employment and economic welfare. The fundamental role played by nuclear power--on average it accounted for 43.5% of Swedish annual electricity production from 2003 to 2012--can help explain why only two out of twelve reactors have been shut down. (1)

The Fukushima meltdowns following a massive earthquake and tsunami brought renewed attention to the complexity and dangers of operating nuclear power plants. Furthermore, the profitability of investing in nuclear power has declined, not least owing to the decrease in wholesale electricity prices resulting from renewable electricity support schemes. These developments have caused observers to doubt the political and economic viability of nuclear power (e.g. Davis, 2012), and trigger two fundamental policy questions: How much new nuclear power should there be from a welfare point of view? How will investments, if socially desirable, come about? (2)

Prior to liberalization, investment incentives were shaped by regulation. A drawback was that policy makers lacked information about the socially optimal level of capacity. The difficulty with creating appropriate incentives by means of detailed regulations was one of the key arguments in favor of liberalizing the Swedish electricity market (Nutek, 1991). (3)

In a liberalized electricity market, investment decisions are delegated to the market participants who will invest in nuclear capacity if and only if doing so is privately profitable. Market external effects can be corrected by an appropriate menu of taxes. However, it is unlikely that investment decisions would be optimal even if the owners could be impelled to internalize the full social cost of nuclear power. Because of their size, every new reactor lowers market prices and decreases the profitability of other production. Three of the largest companies in the Nordic market, E.ON, Fortum and Vattenfall, share the ownership of all three Swedish nuclear plants and jointly decide about investment. Market concentration implies that the nuclear owners are likely to take the surplus reduction on existing generation into account in their investment decision. Exercise of such long-run market power leads to underinvestment and excessive electricity prices.

Long-run market power is usually curtailed by imports or by new producers entering the market. But import capacity is limited by bottlenecks in the transmission network. And entry barriers are significant, as incumbent producers in practice control nuclear investments even under the new Swedish legislation: At most ten new reactors can be built, one for each of the reactors currently in operation; a new reactor cannot be set into operation until an old one permanently shuts down; and all new reactors must be located at the three current nuclear sites owned by the incumbent nuclear producers.

We propose nuclear capacity auctions as the key to unlocking the market for nuclear investment. In a nuclear capacity auction, the seller, say a government agency, auctions off a license to build and operate a nuclear reactor. The winner commits to constructing and operating the reactor according to specifications. Compared to a situation where nuclear investment is delegated to incumbents, the auction mitigates long-run market power by introducing competition at the investment stage. Thereby the license may be allocated to a more efficient bidder - either in terms of lower investment costs or because the bidder expects to be able to produce more efficiently than its competitors. The bids also reveal information about the economic viability of nuclear power. In particular, the license remains unsold and no new nuclear power is built if bids are too low (they could even be negative).

An auction is likely to produce a more efficient result the larger the set of bidders because the expected minimum investment cost is lower and bidding competition is fiercer the more bidders are active in the auction. And the mere threat of entry mitigates incumbents' incentives to bid for market power. Still, producers usually fail to account for the effect of the investment on consumer surplus. A bidding consortium of producers and industrial consumers would partly align consumer and producer interests in the bidding process. Thus we recommend encouraging as many bidders as possible to participate in the auction, not only entrants but also incumbents and energy intensive industries, in bidding consortia for nuclear capacity. (4) Joint ownership by incumbent producers exacerbates underinvestment because the perceived opportunity cost of new nuclear power increases. We thus also recommend to avoid the participation of more than one incumbent producer in each bidding consortium, if possible.

Incumbent producers may be willing to pay a premium on the license for the opportunity to exercise short-run market power. Incumbents bidding for short-run market power distort the auction. The standard remedy is to modify the auction (e.g. Jehiel et al., 1996). We propose a simpler solution which requires the licensees to sell a significant share of their capacity as virtual power plant (VPP) contracts. A VPP contract is an option which gives the holder the right to purchase the contracted amount of electricity from the producer at marginal production cost. VPP contracts effectively delegate the production decision to the buyers of the contracts and thereby mitigate short-run market power and the incentives to bid for it. (5)

The profitability of nuclear investment depends not only on market conditions, but also on current and expected taxes. One problem is that policy makers have an incentive to increase taxes once the plant is in operation and the investment costs are sunk. Swedish authorities have for instance increased the tax on installed nuclear capacity several times over the years. A novel finding is that investors may protect themselves against tax expropriation by selling long term supply contracts at nuclear marginal production cost prior to setting the plant into operation. Long term contracts help investors secure financing of the power plant and simultaneously reduce the producer surplus susceptible to expropriation.

2. NUCLEAR POWER IN THE NORDIC COUNTRIES

Sweden is part of the Nordic electricity market together with Denmark, Finland, Norway and the Baltic countries. (6) Bottlenecks in the transmission grid regularly divide the Nordic market into price areas. Over the last years, Finland and Sweden often have formed a joint price area against the other markets. Nuclear power accounts for roughly 22% of installed capacity in this Finnish-Swedish price area, whereas hydro power and thermal capacity other than nuclear account for approximately 35% each. The rest is predominantly Swedish wind power (all data are from NordREG, 2013).

Table 1 identifies the companies with ownership shares above 10% in the five nuclear plants currently operating in the Nordic market, along with the net capacity of each plant (the number of reactors is in parenthesis). All three Swedish plants are owned jointly by two or more large generation companies. This is not the case in Finland, where Fortum owns Loviisa on its own and Olkiluoto jointly with smaller energy companies and the energy intensive industry. Pohjolan Voima is controlled by the pulp and paper manufacturers United Paper Mills and Stora Enso.

All reactors in Table 1 were phased in between 1971 and 1985. Estimates of their lifespan range between 40 and 60 years. Several of them are therefore likely to be phased out by 2030. The two reactors at the Barseback plant were shut down in 1999 and 2005 as a consequence of the Swedish decision to abandon nuclear power. Finland has remained generally positive to nuclear power and has instead decided to expand nuclear production. A third reactor is under construction at the Olkiluoto site, and the Finnish government has since then authorized the construction of two additional reactors.

Operating performance displays substantial variation between the different plants. Figure 1 displays the aggregate annual capacity utilization of the ten active Swedish reactors compared to the two Finnish plants. Swedish nuclear power has systematically underperformed relative to both Finnish plants from the outset, even predating market liberalization in 1996. From 1981 and onwards, annual capacity utilization in the Swedish plants on average was 16.68 percentage points lower than in Olkiluoto. This amounts to a full reactor of the size currently under construction at Olkiluoto, assuming a capacity utilization of 87%. Note also that Olkiluoto has outperformed Loviisa for most of the period since 1981. The average difference in capacity utilization between the two plants is 3.38%. (7)

3. THE INCENTIVES FOR INVESTING IN NUCLEAR POWER

Nuclear power represents the archetypical base load generation. The fixed cost constitutes the major portion of the total cost and is made up of the construction cost plus incremental capital costs, discounted fixed O&M expenditures and decommissioning costs. The construction cost itself is usually calculated as an overnight cost which accounts also for the interest and equity payments that accrue during the construction of the plant. The variable cost consists of fuel costs, variable O&M and waste funds and is comparatively small. (8) The cost structure of nuclear power implies that profitable investment requires wholesale electricity prices substantially and consistently above the marginal (i.e. variable unit) cost.

Figure 2 illustrates the market effects of an investment in new nuclear power. The demand and supply schedules are...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT