1.1 Economic and Legal Basis for the Exclusion of Intangible Assets
The concept of the “substance trap,” formalized by STEELDY in their research on the interaction between Pillar Two and environmental tokenization, refers to a situation where an investment vehicle holding highly intangible assets is structurally unable to reduce its tax base through the SBIE due to the lack of eligible tangible assets and personnel. This phenomenon is not a technical artifact but the logical consequence of a deliberate tax policy choice: the GloBE rules were specifically designed to target income generated by mobile, relocatable assets that are disconnected from substantial economic activity—a category in which carbon credit tokens fall.
The legal basis for this exclusion lies in Article 5.3.2 of the GloBE rules, which defines “Excluded Substance-Based Elements” as not including intangible assets, inventory, or trade receivables. The characterization of TCCs as intangible assets stems from their legal nature as digital rights representing environmental claims, affirmed by French tax doctrine (BOFiP), which equates digital assets to intangible fixed assets within the meaning of Article 381 of the General Tax Code. This classification is further supported by the position of the French Accounting Standards Authority (ANC) on the accounting treatment of digital tokens, which categorizes them as intangible assets.
1.2 Quantifying Tax Erosion: Monte Carlo Simulations
The Monte Carlo simulations conducted by STEELLDY precisely quantify the extent of tax erosion induced by the substance trap. In an extreme stress scenario deemed “total tax neutrality,” where the Pillar Two mechanism completely eliminates the differential tax advantage between jurisdictions, the portfolio value erosion reaches 57% over the 2025-2035 projection horizon. Breaking down this 57% loss reveals that the portion attributable to the top-up tax mechanism itself accounts for approximately 35% of the depreciation, while the remaining 22% results from the contagion effect on liquidity premiums and the adjustment of valuation multiples in the secondary token market.
These results are obtained through 10,000 Monte Carlo iterations using a three-level input variable architecture: the simulated top-up tax rate (TUT_t), modeled by a normal distribution centered at 15% with a standard deviation of 2.5%; the CCQI trajectory (CCQI_t), modeled by an Ornstein-Uhlenbeck diffusion process with a long-term value of 72, a mean reversion rate of 0.35, and a volatility of 15%; and the base return of the TCC portfolio (R_base_t), modeled by a log-normal distribution with an expected value of 6.5% and a volatility of 22%. The dependence among these three variables is captured by a Student’s t-copula with three degrees of freedom, allowing for the modeling of tail dependencies observed during joint market-regulatory stress episodes.
