Research for this report involved the gathering of information on compensation measures implemented in the Nordic countries as a response to sharp energy and transport price increases from 2021–2023. We also compiled theoretical and analytical frameworks to enable evaluation of a selection of these compensation measures, and to identify alternative measures. These frameworks were especially important for the evaluation since the work was completed during 2023, prior to the availability of any definitive data or research studies. The evaluation focused on distributional effects, climate, and environmental impacts, as well as coherence with existing policies. Out of a large set of compensation measures (presented in Appendix A1 & A2), we selected three Swedish, three Danish, two Norwegian, and two Finnish measures for evaluation.
8.1 Overall evaluation of the measures
Jointly, the evaluation of the selected measures revealed some similarities as well as disparities between the countries. All countries acted relatively quickly (considering the amount of money redistributed). However, the speed led to implementation problems and targeting errors. Undesirable (dis)incentives also occurred.
Furthermore, almost all measures aimed at electricity bills were set up so that those with high electricity consumption received the most compensation. But since low-income and high-income households in the Nordic countries differ with respect to whether electricity/energy costs constitute a high share of the disposable income, there is no consistent result on distributional effects of these measures. The electricity bill measures also differed as to whether they were implemented as payback measures (retrospective subsidies), tax reductions, or deferral measures. Among the selected measures, only one Danish and one Norwegian measure targeted vulnerable low-income groups. These were also the only ones that implemented an increase in investment subsidies for green technology solutions (i.e., with potentially positive long-term effects on emissions of greenhouse gases). But, as shown in Appendix Table A1 and Table A2, there were more measures introduced that targeted low-income groups. Most of the measures set up also directly affected the electricity price. The literature does, in general, propose that measures directly targeting the price are the least efficient and have the largest conflicts with green transition goals. The Finnish deferral measure for households and businesses is one example of a measure to alleviate the financial burden of the high electricity bills without distorting the price signal or incentives to save energy.
Given the limited data and studies available, the only distributional effect on businesses we can identify is a difference between businesses in different regions. Both Sweden and Norway had a structure that gave explicit or implicit support to certain regions. For Sweden, businesses situated in electricity price zones 3 and 4 were given electricity cost support, whilst northern businesses received nothing in the first round. One consequence of the support ceiling was that large and electricity-intensive businesses received support for a lower share of their total electricity consumption, compared to those who consumed within the ceiling. For Norway, the outcome was that mainly companies in the Southwestern part of Norway received support.
Electricity savings due to price increases, have a positive environmental effect. However, the producers are constrained by emission caps under the EU ETS. Hence, it can be argued that the electricity saving measures (or as here price increase) have little to no effect on climate and environment. In the long term, it is feasible that consumption measures taken today can affect future cap levels through political derogations, and several studies advocate using both demand and supply side measures for the green transition (e.g., ESABCC, 2023b). Of the measures evaluated, the Danish increase in subsidies for disconnecting gas-powered heating, and the Norwegian electricity support for energy savings in businesses, both have the direct potential for long-term reductions in greenhouse gas emissions. As for distributional effects, the literature proposes that adverse effects on climate and environment are best avoided if the price signal remains undisturbed.
None of the evaluations identified had any significant incoherencies with existing policies, but a number of other aspects should be highlighted. Firstly, an important factor for long term effects is whether households and businesses will anticipate similar compensation measures if future price shocks occur. If so, it is perceived that the actors will fail to take action to reduce their vulnerability. At the time of writing, we have not identified any clear signal from the Nordic governments on whether the market interventions taken during 2021–2023 were one-time events, or if they will re-occur, so there is a risk that households and businesses will, anticipate compensation from future price shocks (which has more or less been expressed). As a consequence, households may not be eager to change their behaviour. For Norway, the electricity support to households has been extended through 2024, possibly already creating expectations of a more permanent arrangement.
Furthermore, the analysis is hampered by the politicisation of the measures. For example, the Swedish measure to pause the increased ambition of the fuel reduction mandate was announced as politically desirable by the parties currently governing Sweden long before the 2021–2023 price hike (Kinnunen et.al., 2020). It is therefore possible that the implementation of this measure was only secondarily a crisis handling measure. In addition, the interest in actively reducing income disparities between low- and high-income households is a traditional difference between left/right political parties. The political inclination of the currently governing party could correspondingly affect how compensation measures are redistributing incomes. This may also be part of the explanation for the degree to which redistribution was made, and how targeted it was towards low-income households.
One interesting aspect that had to be omitted from this analysis was the impact of the compensation measures on national debt. Some of the Nordic countries already had high national debt. With further borrowing comes an increased cost of repayment for the national debt, with potential negative down-stream effects on both income redistribution efforts and environmental efforts. Part of the Swedish support was, however, financed by congestion-based income resulting from high demand and restricted opportunities to transport electricity between geographical areas. This support has no impact the national debt. This aspect would be interesting to investigate further but is beyond the scope of the current report.
Another omitted aspect, which was outside of scope this project, is the effect of the ‘do-nothing’ option. What would have happened if governments did not engage in any compensation measures at all? One can only speculate, but since governmental bailouts have been hypothesised to have contributed to post-covid inflation in some countries, it can be suspected that the ‘do-nothing’ option would have contributed to lower inflation in the Nordic countries and lower governmental debt. From lower inflation, one can speculate a slower speed of central bank interest rate hikes, with lower pressure on heavily mortgaged (Swedish) homeowners. On the flip side, the ‘do-nothing’ option, could also have contributed to a hike in unemployment rates.
8.2 Alternative measures
From the literature it is possible to identify a number of alternative ways to implement price spike compensation measures, primarily to compensate for household electricity bills. One option is that each household get the same flat support rate, i.e., that each household is paid the same amount regardless of their electricity consumption. This measure can be seen as a support measure that only redistributes the extra earnings from the power companies to the people. Such a measure has been recommended by a number of Norwegian economists (e.g., Skonhoft, 2022), and may be considered more fair as most power companies are government-owned (municipalities, counties and the state). Households that have a lower-than-average electricity consumption would gain more from this measure than from the current measure, which is advantageous in terms of the distributional profile of the measure (progressive). At the same time, this electricity subsidy also provides a better incentive to save on electricity, since consumption is independent of support and the price signal for electricity scarcity is maintained through the market price. There is a small positive income effect, but increased income will be spent on many things, not just electricity.
Depending on political orientation, such a measure can be modified to become more or less progressive, e.g., such that lower income households are paid more. Some of the applied Danish measures had such characteristics – in being decoupled from energy consumption, but typically targeting specific user groups (e.g., people with children), rather than income based. Both of these approaches are, as such, recommendable as they redistribute according to political orientation, but maintain the price signal fully, and thereby allow for a more efficient market and, if otherwise addressed sufficiently, the long-term societal goals of a green transition.
The challenge of ensuring that compensation measures target the vulnerable population has been highlighted by several other evaluations, for example the international literature cited in section 1.3. If a flat rate support measure is not possible to implement (for whatever reason), an alternative is to amend the current measure so that the support is decoupled from current electricity consumption. This has for example been suggested by Hoel & Golombek (2022). They suggest that the electricity subsidy should be linked to historical electricity consumption, not to current electricity consumption. Their proposal suggests that the sum of the electricity subsidy for all relevant subscribers be calculated as it is currently, monthly. However, instead of distributing the subsidy to the subscribers according to their actual consumption for the month, the amount should be distributed according to the subscribers' consumption for the corresponding calendar month for the previous year. For the vast majority, this will give approximately the same amount of support as the current measure. If, for example, all subscribers in October 2022 use 10 percent less electricity than in October 2021, each subscriber will receive exactly the same amount of support with the revised measure as with the current measure. With this proposal, it will be more profitable to reduce electricity use, since the subscriber will now save the full market price for every kWh that is not used. In practice, there will be a few electricity customers who have very different consumption in 2021 and 2022 (among other factors, due to new subscribers), but it should be possible to create special rules for this group. A requirement for such a measure to work is that it is a one-time event. If it is recurring, it will negatively affect households’ incentives to invest in otherwise profitable energy saving measures, cf. the discussion above.