The HHI-index is a measure of concentration that is often used to give a rough indication of the degree of competition in a market. It takes into account the relative size distribution of the firms. It approaches zero when a market is occupied by a large number of firms of relatively equal size and reaches its maximum of 10,000 points when a single firm controls a market. The index is derived from a model where it is assumed that firms compete in quantities, i.e. Cournot-competition – see for instance Matsumoto et al. (2012). This implies that the measure is somewhat less informative when firms compete on prices. Despite this, it is generally accepted that the lower concentration, the more competitive the market. This follows from the assumption that firms with large market shares have considerable market power.
Competition agencies normally consider HHI below 1,500 as low concentration, between 1,500 and 2,500 as moderate concentration and above 2,500 as high concentration – and assumes that the market is more competitive, the lower concentration. This is also intuitive as low concentration requires a relatively large number of firms, of equivalent size, that compete for the consumers. Such a market would normally imply well-functioning competition.
The HHI-index should, however, be interpreted with caution, and it is critical to apply the right geographical market. For instance, if the competition is local, HHI measures at national level will normally not be informative. HHI measured at a national level will for instance be low in a situation where there are many local monopolies. In electricity markets, suppliers with local affiliation may have loyal customers in their respective areas – and thus market power. Therefore, HHI may underestimate the rivalry in spite of most suppliers having national offers.
In Iceland, the practical barrier to entry is, however, significantly higher than in the other countries as there is no well-functioning wholesale market and because the market is much smaller than the other markets. Furthermore, our assessment indicates that the retail market in Iceland suffers significantly from the lack of a well-functioning wholesale market, as this facilitates market power for the retailers with integrated production. Due to low consumption, the Icelandic consumers also have weaker incentives to participate actively in the market, for instance to switch retailer as a response to higher price. Based on this, we conclude that the competition in the Icelandic retail market is significantly weaker than indicated by the concentration measure.
In markets with low barriers to entry, low concentration is to be expected. In Norway and Sweden, the concentration is particularly low. However, the concentration measure does not capture that in some areas, retailers with a local affiliation have high ‘market’ shares and that consumers are loyal to their local supplier. Despite this, the consumers in all areas apart from Iceland can choose from a large number of retailers. Furthermore, none of the retailers enjoys a significant cost advantage for supply of physical electricity, as they all trade at equal terms in the wholesale market. A large number of suppliers that compete on equal terms will normally translate into efficient competition. For supply of fixed price contracts, however, large retailers, and in particular those with integrated production, seem to enjoy an advantage due to imperfections in the hedging markets. Perhaps not from the electricity price itself, but from the availability of long-term power agreements, transaction costs associated with trading, and also credit costs if they are considered more solvent. On the supply side, the markets in Denmark, Finland, Sweden and Norway seem to support a well-functioning market, perhaps with the exception of fixed price contracts.
Despite favourable conditions on the supply side, conditions on the demand side have historically led to imperfections in all countries’ retail markets, where search and switching costs and asymmetric information are the most prominent issues. It follows from economic theory that search and switching costs are an important source of market power. The conditions on the supply side have probably led to higher mark-ups than what theoretically was to be expected, given homogenous commodities, low fixed costs and few expansion barriers. High mark-ups may in turn explain why the retail markets have supported a rather high number of active suppliers.
There are some indications that the asymmetric information has had the most adverse impact on the Norwegian retail market. There are probably compounded reasons for this. First, an average Norwegian customer uses more electricity than an average customer does in the other countries. Second, wholesale prices have historically been so low that electricity has remained a low-interest commodity, despite relatively high mark-ups and high consumption. Combined, this has made it very profitable to win customers, which has incentivized heavy and aggressive marketing of contracts that have not been in the consumers’ best interest. In the other countries, the incentives to invest in such marketing practices may have been lower due to lower consumption and higher wholesale prices.
Variable contracts have been common in all Nordic countries, probably due to meters that had to be read manually. Fixed price contracts also work well with manual reading, and such contracts exist in large numbers, especially in the Finnish market. Norway has on the other hand stood out with a very low share of fixed price contracts, and hence they have probably had the highest share of variable contracts.
In most of the Nordic countries, variable contracts appear to be the least competitive type of contract as the price of electricity is then under the electricity retailers’ control and changes regularly. The retailer may thus exploit the combination of low awareness and switching costs to charge high mark-ups to existing customers. Hence, in contrast to fixed price contracts, the retailer may charge a low price on variable contracts to get customers on-board, for subsequently to increase the price. Spot contracts are on the other hand much more transparent, as the mark-up in contrast to a variable price contract is directly observable to the customer – making it more difficult to increase the mark-up without it being noticed by the customers. The variable price contracts that are most common in Sweden are an exception, as these contracts are based on the average spot prices in the previous month. Such contracts can be beneficial for the consumers if the timing of their electricity consumption corresponds with the time when spot prices are high. These contracts stem from the implementation of smart meters in Sweden, as the first smart meters were only read on a monthly basis.
Finland has traditionally had, and still has, a very high share of fixed price contracts. However, a higher share of fixed price contracts may have contributed positively to competition in Denmark and Sweden as well. However, today Norway has the lowest share of variable contracts and few, if any, retailers offer variable contracts. There are at least two reasons for this. First, smart meters became mandatory and spot contracts fit very well to smart meters, as they measure the exact consumption at every time. Thus, consumers may save money by adapting to price signals. This is especially valuable to the high share of the population that has electrical cars. Furthermore, spot contracts have been heavily promoted over a long time by for instance the Consumer Council. However, this cannot be the full explanation for the lack of fixed price contracts, as the retailers in the other countries face the same challenges related to hedging as in Norway. Smart meters, similar to the Norwegian, are also widespread in both Denmark and Finland. Thus, some of the explanation must be related to differences between the countries on the demand side.
In contrast to the other countries, electricity is the most common source of heating in Norway. This implies a higher consumption, which may make the consumers more price sensitive and aware. Furthermore, the prevalence of electrical cars is also much higher in Norway than in the other countries. Households and SMEs that use electrical cars can be expected to have a higher consumption of electricity than counterparts without such cars do. They also have incentives to charge the cars in low price periods, which in turn may contribute to more awareness and higher demand for contracts, which expose the consumers to the price variations.
In addition, in Norway different actors have warned against variable price contracts over time. The Consumer Council, which operates the official price comparison tool, have also made changes in the tool such that it is not tempting to choose a variable price contract. The price support scheme for households, which the Norwegian government introduced as a response to the energy crises, worked in practice like a price ceiling. This further reduced the households’ demand for fixed-price contracts, which always have been low in Norway. Consequently, the supply of other than spot price contracts to households have more or less stopped. Data from Statistics Norway show for instance that the share of households with variable contracts decreased from 22 to 4 percent between the fourth quarter of 2021 and the third quarter of 2023, while the share with spot price contracts increased from 75 to 93 percent.
Spot price contracts have a simple price structure, consisting of a fixed mark-up on the wholesale price and a fixed monthly fee. This makes it easy for the consumers to compare contracts and identify changes in a running contract, at least compared to variable contracts. This has probably alleviated the information problem, which historically has given scope for market power in the Norwegian market. Price comparison tools and smart meters have probably also contributed to lower search and switching costs. Thus, the current conditions in the Norwegian market may facilitate a healthy and competitive retail market.
In the Finnish and Danish retail markets, there has also been a development towards more spot contracts, something that we expect will continue. We expect to see the same development when smart meters, which can report real-time hourly-based electricity consumption, become common in Sweden. Thus, we also expect these markets to become more competitive in the future. However, as demand for fixed price contracts have historically been higher in these markets, retailers that are vertically integrated with production may have more scope for market power in these markets than in Norway, if the imperfections in the markets for hedging are not resolved and the consumers continue to prefer fixed price contracts.
The competitive landscape in Åland is quite different compared to the Nordic countries, and the competition is in practice non-existing. New electricity retailers can enter the market, but currently there are two electricity retailers in Åland, who are also integrated DSOs. According to the interviews, one third of the customers are living in the grid area of one of the retailers, and the remaining two thirds are in the grid area of the second retailer. The switching rate between these two retailers appears to be almost zero, indicating almost full correlation between the customers’ electricity supplier and their DSO. The organization structure of the retailers impacts their market behaviour. The two suppliers in Åland are either owned by the municipality or operate as an economic cooperative owned by the customers, which can mean that the companies might not focus on financial targets, but rather on giving customers a stable and reliable electricity supply. Neither of the companies seem to actively try to win customers from the other. For instance, one of the retailers only offers fixed price contracts to the customers located in their DSO area. In addition to the general requirements of setting up a supplier company in Åland, the fact that the legal framework is based on the Finnish regulation while the price area belongs to Sweden SE3, makes it more difficult for electricity retailers from either Finland or Sweden to enter the market. Although new retailers can enter the market, the functioning of the market appears to be similar to the electricity retail markets in Finland and Sweden before the deregulation (Finland 1995 Sweden 1996).
3.2.2 Impacts of energy crisis
The energy crisis has increased awareness and knowledge of the electricity markets in all countries. The experience with high and volatile prices may also have increased the interest in spot contracts in terms of consumers being able to adjust their energy consumption to the prices throughout the day. The prevalence of spot contracts has increased in Denmark, Finland and to some extent Sweden. In addition, Norwegian retailers have stopped offering variable price contracts. Part of the explanation is probably linked to that variable price contracts require more hedging than spot price contracts and that the energy crisis illustrated the advantage of being able to move consumption to low price periods. The demand is also low as it poses a risk for consumers to enter into variable price contracts when the electricity support is based on spot prices. In Sweden, there has also been challenges regarding large differences between bidding zones for hedging possibilities, leading to large price differences between the areas. Further, hedging possibilities have also been affected by increased risks when electricity production is located in other bidding zones.
The development on the demand side may contribute to more active and informed consumers, and thereby to more well-functioning markets. However, for the retailers, it has become significantly more difficult and expensive to hedge, which has led to a reduction in offering and more expensive fixed price contracts. In Sweden, contrary to Denmark, Finland, and Norway, many electricity meters are not yet on an hourly level, which can impose a challenge regarding customers who want to be more active. However, consumers in Sweden can choose an hourly electricity contract, which requires hourly metering without any additional cost, and the DSO must replace the meter if necessary even if most meters can be reconfigured at a distance.
Another consequence of the energy crisis was some bankruptcies among the retailers. There were not a lot of bankruptcies, but the bankruptcies that did occur may have made the consumers more aware of the risks associated with entering into contracts with smaller retailers. Thus, the consumers’ preferences for solid retailers may have increased, especially related to fixed price contracts or contracts with pre-payments. At least in Sweden, there has been a significant diversion from small to large retailers, which is compatible with such a hypothesis. If customers have preferences for large retailers, this may allow these to charge higher mark-ups and also to increase the practical barriers for entry.
However, we consider increased consumer knowledge and awareness as the most prominent effect of the energy crisis. Unaware customers are perhaps the biggest obstacle for a well-functioning electrical retail market, as the existence of such customers may reduce the electricity retailers’ incentives to compete on price. Thus, the crisis has probably alleviated the information problem, which can be expected to improve competition. However, it is too early to say anything about the duration of the effect on awareness.
3.2.3 Availability of fixed price contracts, or contracts with fixed priced elements
In Norway, fixed price contracts are offered to SMEs, and only a few fixed price contracts are available to household customers. The explanation is probably imperfections in the hedging markets combined with the price support scheme for the households. The lack of fixed price contracts may in particular be a challenge to SMEs that are not protected from high prices by the price support scheme. In Denmark, Finland, and Sweden, fixed price contracts are still offered, but the realized demand is slightly lower than before the crisis due to unusual high mark-ups. As explained above, the competition in the segment may be adversely impacted by advantages for large retailers with integrated production. In Åland, two contract types are available, and most household customers have open-ended variable price contracts, with fixed price elements, similar to Finnish variable price contracts.