The redistributive effects of real-time electricity pricing

May 25, 2022 Energy

According to the Observatoire National de la Précarité Énergétique, 20% of the French population declare to have suffered from the cold during the winter of 2020-2021, more than a third of them for financial reasons. What would happen if, like in Spain, the price they paid for electricity fluctuated permanently according to supply and demand?

The market and the tariff shield

On February 1, 2022, three weeks before Russia's aggression against Ukraine, the retail electricity tariff in France should have increased by an average of 40% before tax by application of the cost-stacking rule that frames the calculations of the Energy Regulatory Commission. But faced with the devastating effects of such an increase, especially during an election period, the government used the "tariff shield" provided for in Article 181.VI of the 2022 Finance Law to limit the increase to 4%. In fact, this is more of a cover than a shield because the price difference will be passed on to future semesters. This disconnect between the retail and wholesale price shows how difficult it is for our energy-hungry societies to adapt to the convulsions of the energy markets.

In theory, marginal cost pricing - selling a product at a price equal to the cost of the last unit delivered - is a sign of efficient resource allocation. In practice, marginal cost pricing may be unworkable. In the electricity industry, in the absence of large-scale storage, the marginal cost of energy delivered at a given time is that of the most expensive energy produced at that time: solar, wind, thermal, hydro or nuclear (see our previous post). Therefore, marginal cost pricing is a real time pricing. Without the advantages of volume amortization through storage, the price varies constantly according to the availability of production and the level of demand. The variations are significant from hour to hour, and can be considerable from day to day and between summer and winter. In France, during the month of December 2021, the price of 1 MWh on the wholesale market reached €620 on the 21st at 5pm and dropped to €15 on the 28th at 4am, i.e. 40 times less.

Regressive redistribution

In addition to being difficult to apply, retail pricing based on price fluctuations on the wholesale electricity market is undesirable for the poorest consumers. This is confirmed by a recent study coordinated by economists Natalia Fabra and Mar Reguant using Spanish data. Spanish electricity consumers are the only ones in the world to be subject to real-time pricing by default since 2015. Using hourly consumption data from 1 million households over 18 months, the authors show that real-time pricing penalizes low-income households.

Their argument is that in a system where the energy part of the retail price does not vary over time, to balance the accounts of the various operators this part is an average, so it is higher than the wholesale price at some times (off-peak hours), and lower at others (peak hours). Thus, compared to the cost of the system reflected by the wholesale price, when the retail price is invariant, households that consume a lot during off-peak hours subsidize those that consume a lot during peak hours. Is this redistribution progressive (from high income to low income) or, on the contrary, regressive? To answer this question, we need to identify the relationship between daily consumption patterns and income. This is done by using data on the regional distribution of income by zip code and by using proven statistical methods to allocate each household to a particular income distribution.


Air conditioning and electric heating

The results show a regressive effect of real-time pricing, and thus an increase in inequality. However, this effect is not very significant. The explanation lies in the low variability of intraday wholesale prices during the study period (January 2016 - May 2017), whereas price differences are much more pronounced between summer and winter. However, the temporal shifts in consumption to take advantage of variable prices are quite limited. For example, we can shift the start of the washing machine by a few hours, or even a day, but this is not what will result in the greatest savings. In contrast, it would be profitable to shift the start of the electric heating and the air conditioning, but the comfort felt would be strongly affected. Now, note that air conditioning is mainly used by high-income households and of course in the summer when prices are low, while electricity is the source of heating for many low-income households during the winter when high demand pushes up wholesale prices. Thus, with the generalized switch from an annual fixed price tariff to real-time pricing, low-income earners lose and high-income earners gain. This negative redistributive effect is partially mitigated by a positive intra-day redistribution: on a daily basis the rich consume more electricity than the poor, and this excess is relatively larger at peak hours. Although the intra-daily effect of the rate switch is the opposite of the inter-seasonal effect, the balance of the two effects remains to the disadvantage of low-income households.


*     *

The efficiency of real-time pricing should not obscure the redistributive effects of any change in pricing, which may exacerbate inequality. It may be objected that the average increase in bills for low-income Spanish households estimated by the authors of the study remains small, around 2 percent. Nevertheless, it could be higher in the future because the study was carried out during a period when price volatility was low, whereas for several months now, prices at peak times have been reaching very high levels. On the other hand, the very idea that this increase will benefit wealthy households whose bills have fallen is politically unbearable. Another objection is that variable price contracts are not mandatory: a household can opt for a fixed price contract if it so requests. But the " if it so requests" clause is prohibitive when it overestimates the expertise and contractual optimization capacity of households.