Environmental taxation and biofuel production in the EU: Heterogeneous effects across producer scales

Is environmental taxation associated with faster or slower biofuel production in the European Union? We ex­ amine this question on a balanced panel of the 27 EU member states over 2010–2023, combining Eurostat environmental-tax accounts with the OPEC Reference Basket as a globally exogenous oil-price control. Across cluster-robust LSDV, Driscoll–Kraay HAC, and System-GMM estimators, the pooled relationship between the en­ vironmental tax-to-GDP ratio and biofuel capacity is statistically flat. A non-parametric Hansen (1999) threshold test fails to identify any regime break (𝑝 = 0.85), and the flat result is robust to two- and three-year lags and to two-way (country and year) fixed effects; in a distributed-lag specification the cumulative effect over a three-year ∑3 horizon is a modest negative one ( ℎ=0 𝛽̂ℎ = −0.71, 𝑝 = 0.036), consistent with slow-acting cost pass-through rather than a non-linear response. The pooled null, however, is an artefact of aggregation: a joint 𝐹 -test re­ jects coefficient homogeneity across producer-size terciles (𝐹 = 3.12, 𝑝 = 0.032; robust to year fixed effects and to alternative, outcome-independent classifications), and large export-oriented producers display a marginally significant and suggestive inverted-U pattern while smaller producers do not. Biofuel production is also highly per­ sistent (System-GMM 𝜌̂ ≈ 0.96), implying that the fiscal lever moves slowly. Because the design does not exploit an exogenous policy shock, we interpret these estimates as conditional associations rather than causal effects, and discuss their implications for differentiated environmental fiscal policy under the post-RED II framework.