The Impact of Financial Support to SMEs to Improve Energy Efficiency: A Pan-European RCT

The EENergy project, funded by the EU’s Single Market Program (SMP), supports small and medium enterprises (SMEs) in improving energy efficiency and sustainability, contributing to the EU’s goal of achieving climate neutrality by 2050. Implemented by the Enterprise Europe Network (EEN), the program offers advisory services to 1,964 SMEs, with 986 of them being offered up to €10,000 in direct financial support. This funding is earmarked for investments in renewable technologies, energy efficiency consultancy, and employee training. All eligible SMEs must collaborate with EEN’s sustainability advisors to create tailored action plans targeting reduction in energy consumption.

2024-07-21

2026-03-31

energy efficiency, energy resilience, business performance

1. First Stage
a) Received funds from implementation partner

2. Investment and Composition of Investment:
Index of any investment
a) Any energy efficient investment
b) Any renewable investment

Index of expenditures comprising
a) Expenditure on training for EE and RE investments
b) Expenditure on capital for EE and RE investments
c) Expenditure on consulting for EE and RE investments
d) Index of type of expenditure (e.g. solar PV investment, type of EE investment such as lighting, heat capture etc.), also to be reported separately

3. Energy Use, Intensity, and Cost:

a) Total energy expenditure (logged), as well as sourcewise expenditure (heating, electricity, other)
b) Emissions intensity: log (revenue divided by energy expenditure) and log(profits divided by energy expenditure)
c) If emissions factors are available, emissions by source-wise expenditure

Downstream firm-level outcomes, satisfaction with energy advisor, and support for mitigation policies

1. Final firm outcomes:
Index comprising:
a) Total number of FTE (full-time employees)
b) Value of total capital stock
c) Employees working on sustainability-related tasks
d) Formal sustainability practices: Adoption of sustainability targets (binary); Regular monitoring of energy usage (binary); Obtaining energy performance certificates (binary); Assignment of sustainability responsibilities (binary).
e) Revenues
f) Profits


2. Intentions for future energy investments

3. Satisfaction with energy advice: index comprising

a) How closely did they follow the baseline report’s advice
b) Agreement with statement “the EEN advisor(s) spent enough time to support the establishment's implementation of energy efficiency plans”

4. Support for climate mitigation policy:

Index comprising agreement with the following statements:

a) Climate change will have a serious impact on the quality of life of people in the EU during my lifetime
b) Human activities are a significant cause of climate change
c) On balance, the transition to renewable energy solar from fossil fuels like coal and gas will be beneficial for the EU economy in the next 10 years
d) The EU should do more to reduce greenhouse gas emissions
e) European companies should do more to reduce greenhouse gas emissions

The design is a firm-level randomized trial, implemented over two calls. The experiment was divided into two calls to manage operational capacity. Call 1 began with baseline surveys in February-May 2024, with 1,406 eligible firms participating. Call 2 began in December 2024 with 558 additional eligible firms.

The randomization was performed after the eligibility checks were completed. For Call 1, 1,406 firms were randomized, with 707 firms (50.3%) in the treatment group and 699 firms (49.7%) in the control group. Randomization was stratified by consortium and advisor characteristics to ensure balance. For Call 2, 558 firms were randomized into treatment vs. control (279 in each group).

Not available

The allocation of firms that applied for the EENErgy call into the two groups — the Cash Group and the Standard Beneficiary Group — was performed in two rounds (per call) using STATA.

Specifications:
1. Treatment vs control
2. Test for spillovers: compare control firms amongst high-intensity advisors (exploring exogenous variation) with other control firms
3. Test for spillovers amongst treated firms: compare treated firms amongst high-intensity advisors with other treated firms
4. Advisor elasticity of effort, prices: regress prices, effort of advisors on high intensity dummy
Pool across calls with calls and strata dummies if specification comparable. PDS lasso to improve precision using baseline controls.
Correct for differential attrition using weighted regressions and check for differential attrition by baseline characteristics.

Main: Firm

No

Randomization is at the firm level, but we considered strata of EEN advisory service providers/ partners.

1,964 SMEs.

Treatment Group: 986 SMEs,.
Control Group: 978 SMEs.

Power calculations use the full experimental sample (N=1,964) with one endline survey, 1:1 balanced assignment, two-sided tests at α=0.05, compliance/take-up of 75% (so ITT = 0.75×ATE), and attrition scenarios of 15% and 30%. For continuous outcomes, we compute MDEs under an ANCOVA specification using the control-group SD (benchmarking 0.10/0.15/0.20 SD), and for binary extensive-margin outcomes we use two-sample proportion tests with the baseline control-group mean as (p_0). Under these assumptions, we are powered to detect ITT effects of about 0.20 SD for grant attainment and about 0.15 SD for energy and other continuous outcomes (power roughly 0.85–0.94 across attrition scenarios).