No link turn---we have studies, they don’t---national energy policies, threshold effects, and trade linkage spillovers prevent solvency.

Policy makers and experts have been arguing that the transition away from fossil fuels could support price stability while also fighting climate change. Panetta (2022) calls that possible outcome a “divine coincidence”. While the divine coincidence hypothesis is widely discussed, we are not aware of published empirical research studying how renewable energy affects the sensitivity of inflation to changes in fossil energy prices.

Our empirical results are inconsistent with the divine coincidence hypothesis. That result is robust to sub-periods, country sub-samples and alternative metrics of inflation, fossil prices and renewable energy. This result seems counter-intuitive at first. Indeed, one would expect to see a positive relationship between a country’s consumption of a commodity and that commodity’s impact on the country’s inflation. Three reasons could explain why the empirical results are inconsistent with a dampening effect of renewable energy on the fossil energy price-inflation relationship in this study: national energy policies, threshold effects induced by marginal pricing and trade-linkage spillovers.

National energy policies

One possibility could be for the divine coincidence effect to be drowned in countries’ varying energy policies, which did not fully include in the regression for data-availability reasons. While we did include the level of fossil fuel subsidies in the regression – and found it to shield domestic energy inflation from fossil energy price fluctuations – we lack a structured data set on the broad range of other energy policies that can affect the transmission of international fossil energy prices to retail energy prices. These policies include: market structure and liberalization, long-term price agreements on imports, or price controls with no budgetary impact.

As an illustration, in June 2022, the Spanish and Portuguese governments implemented a mechanism that came to be know as the “Iberian exception” (Corbeau et al., 2023). The objective of that policy was to decouple domestic electricity prices from international gas prices, which were skyrocketing in the wake of Russia’s invasion of Ukraine. This was achieved by a price cap on gas used in electricity generation, which was funded by a general levy on utilities and therefore had only a small impact on fiscal expenses. This is just one example of an energy policy affecting the relationship between international fossil energy prices and retail energy prices but would not be fully captured by official statistics on fossil fuel subsidies. Not accounting for these policies in the regression could lead to the divine coincidence signal to be drowned in noise.

Marginal pricing

A second reason for the empirical results being inconsistent with the divine coincidence hypothesis could be the existence of threshold effects in the moderating effect renewable energy has on the co-movement of inflation and fossil-fuel prices.

Indeed, a number of final energy goods that make up the energy consumption basket are commodities, the properties of which do not depend on whether they were produced with fossil or renewable primary energy sources. This is the case for electricity, where a kilowatt-hour (kWh) from a wind turbine is indistinguishable from a kWh produced in a coal-fired power plant. It is largely true also of biomethane or biodiesel.

In the absence of transportation costs, commodity markets tend to converge to a single price (Ielpo, 2018). This is the case in domestic wholesale power markets. Indeed, if the domestic wholesale power market is liberalized, the marginal producer sets the price received by all power plants, even if they have far lower marginal production costs as is the case for renewables.24 If, as illustrated on Figure 11, that marginal producer is a gas-fired power plant, the entire domestic wholesale power price is linked to the international gas price and fluctuations of that will feed through to domestic energy inflation.

Graph omitted

In that situation, renewable energy producers to the very left of the supply curve on Figure 11 would be paid a much higher price than their marginal production costs, earning a windfall profit. Consequently however, the divine coincidence hypothesis – a gradual decoupling of fossil energy prices and energy inflation when the share of renewables increases – does not hold. Indeed, as long as the marginal producer is a gas fired power plant, it does not matter whether 5% or 75% of electricity production comes from renewables, power prices and gas prices move in sync.

This market structure implies a threshold effect for decoupling energy inflation from international fossil energy prices. Once the share of renewables in electricity production increases towards 100%, the number of hours where renewables are the marginal producer and gas-fired power plants are “pushed out” will increase. Only above that threshold would the divine coincidence hypothesis materialize.

To test this threshold effect, a regression focusing specifically on electricity prices was run. Eurostat publishes granular inflation for individual items in the consumer price index for 36 EU member states and accession candidates. It is therefore possible to estimate the relationship shown in equation (6) but using electricity price inflation as the dependent variable (π elec i,t in equation 8 below) and the share of renewables in electricity production as the modulating variable (X elec i,t in the equation below).

Equation Omitted

More precisely, we estimated the relationship using the continuous modulating variable (the share of renewables in electricity production) as well as dummy variables, indicating whether that share was above certain thresholds (50%, 60%, etc. up to 85%25).

Table 6 presents the results. Column (1) shows the results of the continuous regression which are similar to previous results: the interaction terms are not economically or statistically significant. This means that any marginal increase in the fraction of renewables in electricity production is not associated with a weaker co-movement of electricity prices and international fossil-fuel prices

However, Column (6) holds an interesting result. The contemporaneous and one-lag ∆E terms and the one-lag interaction term are both statistically and economically significant. This indicates that the electricity prices do rise in the year that international fossil energy prices rise, but for countries with very high shares of renewables in electricity production (≥85%) the electricity prices no longer rise in the second year while they do so for other countries.

This preliminary result might be interpreted as indication of the divine coincidence hypothesis in its threshold formulation. The shielding effect is relatively weak, materializing with a one-year delay only and is limited to one of the six items in the basket of the energy consumer price index. A direct shielding effect on ECPI inflation let alone on headline inflation is not supported by these results. Furthermore this result is obtained only on a sub-sample of European countries. Further research on this threshold effect is needed.

Trade linkages and spillovers

A third possibility could be that the divine coincidence hypothesis simply does not hold at the country level, because of trade linkages and spillovers. A country covering all of its domestic energy consumption with renewable energy would see the international demand for its (renewable) electricity (Gugler et al., 2018), heating biomass and biofuels increase significantly in the face of an oil price shock as its fossil energy-dependent trading partners look for cheaper sources of energy. This would lead domestic energy prices to increase, which would eventually feed through to other prices and increase inflation. In this case, the divine coincidence hypothesis would hold at global level only, but not at the level of individual countries.