As is often the case, it all started with a good idea: to put the French on the path to sobriety. But here it is. The method followed by France, which is based on energy saving certificates (CEE), is the subject of a vitriolic report from the Court of AuditorsLawyers are calling for a major overhaul of this mechanism, which is considered costly for households and ineffective.
Created in 2005, the system requires – in exchange for certificates – electricity, gas or fuel suppliers to support energy saving actions. On paper, the principle is simple. But over time, the system has become more complex. To the point of including an unfathomable forest of measures: building insulation, changing boilers, trainingindustrial heat recovery…
A disguised tax
“To find funding sources without releasing new budget lines, the State has continued to expand the system,” adds Matthieu Glachant, economist and professor at Mines ParisTech. Since 2018, more than 280 regulatory texts have changed its outlines… As a result, the CEEs are now based “on multiple and unstable rules and mechanisms”, differentiated according to energy or suppliers, without providing proof of “the reality of the savings obtained”, criticizes the Court of Auditors.
According to the government, the actions carried out within the framework of the CEE would have made it possible to reduce energy consumption in France by 6.5% between 2014 and 2020. However, for the Court of Auditors, the energy savings announced for the year 2022-2023 would be overestimated by 30%. The authors point out several dysfunctions. A large proportion of the certificates have no direct link with energy savings; the way of accounting for these reductions encourages approximations; the actions listed by the State to issue these certificates are based on their supposed effects, without any ex post control being able to attest to this… Finally, when the work is well done, it sometimes leads to a significant rebound effect, i.e. an increase in energy consumption.
Ineffective, the system nevertheless generates a lot of money. Between 2022 and 2023, the total cost of the mechanism approached 6 billion euros. A sum that is actually paid by households and companies in the tertiary sector, because energy suppliers pass on the cost of the certificates in their sales prices, which de facto represents an energy tax. “In 2023, each household has thus, by paying their energy bills, financed the system to the tune of €164 on average”, underlines the Court of Auditors. “At a time of calls for moderation, this represents an easy way for the State to avoid increasing energy taxes”, notes Matthieu Glachant.
Avoiding fraud
To correct the system, a major reform is necessary, notes the study. If the State wishes to keep the system, it will be imperative to simplify it and make it truly measurable. “The measurement and publication of actual results, freed from the artifices of the current unit of account, would be an essential first step forward in this regard,” the report suggests. A complementary solution would be to drastically tighten the target of the system by concentrating it on professional beneficiaries. “This would support the needs of economic players and ensure that real energy savings are made since these players generally monitor their consumption,” write the authors. This would also have the benefit of avoiding a large part of certificate fraud, which is mainly encountered in the private sector.
The Court of Auditors also calls for more transparency to be integrated into the system. To this end, it advocates that Parliament play a role in guiding the system, in particular by integrating its objectives into the multiannual energy programming (PPE). “Determining consumption reduction objectives via the CEE in a framework law such as the PPE is the right solution,” stresses Matthieu Glachant, “because they are administrative tools that ensure a certain stability in the system.” But if the anomalies persist, the elimination of the system could be considered, notes the Court. This is also the option chosen by Denmark.
.