Candice Horbacz
Brussels, Belgium – August 31, 2026 – A recent legislative proposal by the Belgian government to adapt its corporate income tax system to the digital economy, put forth in May 2026, faces strong criticism from economists who argue it closely mirrors controversial Digital Services Taxes (DSTs) adopted across Europe. In written testimony submitted to the Belgian Finance and Budget Committee today, Economist Cristina Enache cautioned that the proposal, despite its framing as an income tax reform, fundamentally resembles a DST and risks generating substantial economic distortions, legal uncertainty, and trade frictions while yielding only modest revenue gains.
The Belgian law proposal seeks to modernize corporate taxation by establishing a Digital Permanent Establishment (DPE), attributing taxable revenues to Belgium based on user engagement and digital participation, implementing deemed profit allocation rules, and introducing a digital withholding tax intended to function as a minimum tax mechanism. These measures are designed to capture profits from digital companies operating in Belgium without a traditional physical presence, a challenge that has vexed tax authorities globally.
The Global Digital Tax Landscape and Belgium’s Unilateral Approach
The rapid expansion of the digital economy over the past two decades has profoundly reshaped business models, allowing companies like social media platforms, e-commerce marketplaces, and cloud service providers to generate significant revenue in countries where they have no physical headquarters or substantial tangible assets. This phenomenon has spurred an intense international debate on how and where these highly mobile digital businesses should be taxed. Many governments perceive the current international tax framework, largely designed for an industrial economy, as inadequate, leading to concerns about tax avoidance and an uneven playing field.
In response, a myriad of unilateral measures have emerged across the globe. European nations, in particular, have been at the forefront of introducing DSTs, which typically levy a tax on the gross revenues of large digital companies derived from certain digital services within their jurisdiction. These include taxes on online advertising, marketplace services, and the sale of user data. Such unilateral actions, however, have been a source of significant international friction, particularly with the United States, where many of the largest digital corporations are headquartered.
Recognizing the need for a coordinated solution, international bodies such as the Organisation for Economic Co-operation and Development (OECD), the United Nations (UN), and the European Union (EU) have been engaged in protracted negotiations to develop a consensus-based framework for taxing the digitalized economy. The OECD’s Inclusive Framework on Base Erosion and Profit Shifting (BEPS) project, specifically its "Pillar One" and "Pillar Two" initiatives, aims to reallocate taxing rights to market jurisdictions and establish a global minimum corporate tax rate. However, progress has been slow, and the complexity of reaching a universal agreement has often prompted individual nations, including Belgium, to pursue their own domestic solutions.
A DST in Disguise: Economic Substance vs. Legal Form
Enache’s testimony highlights that while the Belgian proposal is formally structured as an amendment to the existing corporate income tax (CIT) system, its core economic characteristics align closely with a DST. The distinction is crucial: a traditional corporate income tax is levied on a company’s net profits (revenue minus expenses), reflecting its actual economic performance. In contrast, DSTs typically tax gross revenues, regardless of profitability.
The Belgian proposal, according to Enache, taxes revenues indirectly rather than actual net profits. It relies on user-based allocation rules to determine what portion of a company’s global revenue is attributable to Belgian users, a hallmark of DSTs. It also targets a defined group of digital business models and establishes unilateral nexus standards that deviate from established international tax principles. Crucially, instead of directly applying a percentage tax on revenue, Belgium’s system would assign "deemed profit margins" to different business models (e.g., 25 percent for certain services, 15 percent for others, and 10 percent for still others). These deemed profits would then be subjected to Belgium’s 25 percent corporate income tax rate.
For example, a deemed profit margin of 25 percent would translate to an effective DST rate of 6.25 percent of revenue (25% of 25%). While these margins might appear modest, Enache warns they can lead to highly distortive effective tax rates. Many marketplace platforms, for instance, operate with actual profit margins below 15 percent. Under the proposed deemed 25 percent margin, such a firm could face an effective tax rate well above 40 percent, potentially reaching as high as 125 percent for a company with an actual profit margin of just 5 percent. This mechanism severely penalizes low-margin firms, discourages investment, and distorts business decisions, compelling companies to either absorb significant costs or pass them on. While companies may challenge these presumptions, the administrative burden and compliance costs associated with demonstrating lower actual profits would be substantial.
The Problem of Tax Pyramiding and Disincentives to Specialization
A significant criticism leveled against DSTs, and consequently against the Belgian proposal, is the risk of "tax pyramiding" or cascading taxation. Modern digital value chains are highly specialized, involving multiple layers of service providers—from search platforms and advertising networks to cloud infrastructure providers and payment processors. Each layer contributes to the final service or product delivered to the end-user.
Unlike value-added tax (VAT) systems, which typically include credit mechanisms to prevent multiple taxation at each stage of production, the Belgian proposal lacks such a safeguard. This means that a single commercial transaction could generate taxable revenue allocations at several different levels of the value chain, leading to the same economic activity being taxed multiple times. This cascading effect, akin to turnover taxes, disproportionately harms low-margin firms and distorts economic behavior.
Enache cites a 2026 Tax Foundation study demonstrating how DSTs effectively penalize specialized digital activities. By taxing each intermediary layer, such taxes discourage outsourcing, stifle innovation, and encourage vertical integration (where companies try to bring more functions in-house to avoid multiple tax points). This ultimately reduces overall economic efficiency, a consequence Belgium should anticipate given its proposal’s focus on specialized digital intermediaries.
The Digital Permanent Establishment: A Solution or a New Problem?
The proposal introduces a Digital Permanent Establishment (DPE) concept, departing from the traditional OECD Model Tax Convention, which requires a physical presence for a permanent establishment to exist. Belgium’s DPE would be triggered by quantitative thresholds related to user numbers, connection counts, or revenue generated within the country during a tax year.
However, Enache references an International Monetary Fund (IMF) paper which argues that even if a DPE nexus is established, attributing profits under the arm’s-length principle (the standard for transfer pricing) remains challenging when significant functions, assets, and risks are absent in the jurisdiction. The Belgian proposal implicitly acknowledges this difficulty by substituting actual profit attribution with formulaic deemed margins, effectively confirming the IMF’s critique rather than resolving it. This approach moves away from the core principle of taxing actual economic activity.
Increased Risk of Double Taxation and Legal Disputes
One of the most immediate and severe consequences of unilateral digital tax measures is the heightened risk of double taxation. When Belgium unilaterally redefines a permanent establishment through its DPE rule, it expands its taxing rights over income that other countries might still claim under their traditional tax treaties or domestic laws. This creates a situation where the same income (or revenue) could be taxed twice: once in Belgium and again in the company’s residence jurisdiction or other market jurisdictions.
The testimony points to a 2020 Tax Foundation report illustrating how expanding the definition of permanent establishment to include digital activity thresholds can lead to more than 100 percent of a company’s income being subjected to taxation. This risk is exacerbated when, as in Belgium’s case, the tax is imposed on deemed profits derived from revenue, rather than actual profits.
Furthermore, the proposal potentially violates Article 24(3) of the OECD Model Tax Convention, which mandates that permanent establishments be taxed in the same manner as resident companies. By taxing DPEs on deemed profits (derived from revenue), Belgium would be treating them differently from ordinary resident companies, which are taxed on actual net profits. This divergence from treaty norms could lead to disputes under mutual agreement procedures, increased litigation, and significant taxpayer uncertainty, ultimately undermining international tax cooperation and investment.
The Belgian proposal itself acknowledges this risk, suggesting that if combined taxes limit a company’s profit margin too much, it might increase prices. The proposal frames this as an intended effect to counteract "artificially low" prices due to tax optimization. However, Enache argues that if such effects arise from just two jurisdictions, they would be amplified significantly if multiple countries adopted similar unilateral measures.
The Digital Withholding Tax: An Additional Layer of Complexity
The proposed digital withholding tax mechanism adds another layer of complexity and potential distortion. Operating as a non-refundable levy and a minimum tax, it effectively functions as an additional gross-basis charge. Under the proposal, a digital withholding tax is imposed in the subsequent tax year at 80 percent of the prior year’s digital corporate income tax liability. This levy is generally non-refundable unless it exceeds 80 percent of the final digital corporate income tax liability. This provision is particularly problematic for loss-making companies, as it imposes a cash-flow burden even when little or no digital corporate income tax is ultimately due, further complicating financial planning and potentially penalizing struggling businesses.
Treaty Reinterpretation: A Legal Minefield
A core issue with the Belgian proposal is its reliance on a reinterpretation of existing international tax treaties, particularly concerning the concept of a permanent establishment. Enache delves into Article 31 of the Vienna Convention on the Law of Treaties (VCLT), which governs treaty interpretation, emphasizing the ordinary meaning of terms in context and in light of the treaty’s object and purpose. Supplementary means of interpretation (Article 32) are typically reserved for clarifying ambiguities, not for fundamentally altering the original intent of treaty negotiators.
Enache argues that the proponents’ reasoning for reinterpreting treaty provisions to encompass digital businesses is circular. They claim Article 31 leads to an "absurd" result (failure to tax digital businesses) and then invoke Article 32 to justify a broader, teleological interpretation that Article 31 alone cannot support. This approach, Enache contends, risks departing from the original intentions of the treaty parties and undermining the principle of good faith. The ongoing international debates at the OECD, EU, and UN on new frameworks for the digitalized economy demonstrate a global consensus that traditional permanent establishment concepts cannot easily capture businesses operating without a physical presence. If existing treaties could be so readily reinterpreted, years of international negotiations would have been unnecessary.
Geopolitical Ramifications: Trade Frictions and Retaliation Risks
Unilateral digital taxes have historically ignited international trade disputes, particularly with the United States. Many large multinational digital companies targeted by DSTs are US-based, leading to perceptions of discriminatory taxation. The US government has consistently opposed DSTs, previously threatening Section 301 investigations under the Trump administration and, more recently, considering Section 899 retaliatory taxes. Although Section 899 was removed from a recent legislative act, the underlying tensions remain.
Enache warns that Belgium’s proposal, by placing a disproportionate burden on large multinational digital companies (many of which are US-headquartered), risks reigniting these transatlantic disputes. The United States is Belgium’s fourth-largest export market, with significant trade in goods and services. While overall trade is substantial, Belgium is a net importer of Information and Communication Technology (ICT) services from the US. Unilateral measures targeting US digital firms could destabilize this important economic relationship, leading to legal uncertainty, strained trade relations, and potentially retaliatory actions that would be detrimental to all parties involved.
The Burden on Belgian Consumers and SMEs
Economists generally agree that the economic incidence of a DST is more akin to an excise tax than a corporate income tax. While corporate income tax is largely borne by shareholders, excise taxes are often passed on to consumers through higher prices, making them potentially regressive as lower-income households consume a larger share of their income.
Evidence from other countries that have implemented DSTs supports this. Apple, Amazon, and Google (Alphabet) all passed on the UK’s 2 percent DST to their users. Google, for instance, explicitly adds a surcharge for DSTs in countries where ads are accessed. Recent research by Dominika Langenmayr and Rohit Reddy Muddasani suggests that attempts to target large digital platforms often result in the cost primarily falling on European consumers. The IMF has also linked DST adoption to lower imports of digital services.
Enache concludes that Belgium should expect similar outcomes: the actual burden of its digital tax would likely fall on Belgian consumers through higher prices for digital services, and on domestic businesses—especially small and medium-sized enterprises (SMEs)—that rely heavily on digital platforms for advertising, sales, and operations. This means the tax, rather than effectively targeting foreign tech giants, could end up harming Belgium’s own economy.
Significant Administrative Complexity and Modest Revenue
The Belgian proposal would impose extensive compliance and reporting obligations on affected companies. Firms would need to track global revenues, user numbers, connection counts, allocate users at a country level, perform GDP-weighting calculations, and adhere to detailed reporting requirements. Challenging default allocation methodologies would demand robust evidence, further increasing administrative costs for both businesses and the tax administration. This complexity is a common criticism of DSTs and digital nexus rules, creating significant challenges in determining user location, attributing revenues, and ensuring compliance, especially in the absence of clear guidance.
Despite these significant burdens, the revenue generated by DSTs in other European countries has been relatively modest. Data from Austria, France, Italy, Spain, Turkey, and the UK shows annual DST revenues ranging from €137 million (Austria) to €1.04 billion (UK), typically representing less than 0.1 percent of total government revenue. Even in the highest case (Turkey), it reached only about 0.24 percent.
Tax Foundation modeling, based on a 3 percent Belgian digital tax comparable to the May 2026 proposal, projects annual revenues of approximately €148 million. This represents less than 0.06 percent of Belgium’s total tax revenues, indicating a limited source of additional government income.
Critically, the economic costs are projected to significantly outweigh these modest revenue gains. The tax is estimated to reduce Belgian GDP by approximately 0.056 percent (equivalent to about €342 million annually), with investment declining by 0.073 percent, and wage levels and employment each falling by 0.03 percent. The estimated reduction in economic output (€342 million) is approximately 2.3 times larger than the projected annual revenue collection (€148 million). This implies that the broader contraction of the tax base could substantially offset the revenue generated, potentially resulting in a net negative fiscal effect once lower collections from other taxes are considered.
A Superior Alternative: Reforming VAT
Enache strongly advocates for reforming the value-added tax (VAT) system as a superior alternative if Belgium’s objective is to raise more revenue from digital services. A destination-based VAT is considered the most coherent and least distortionary instrument for taxing cross-border digital services. VAT systems are already equipped to tax streaming, online advertising, cloud computing, marketplace services, and software subscriptions without the need for specialized, distorting digital taxes.
The EU has already made significant strides in adapting its VAT rules to the digital economy. Reforms requiring non-EU businesses to register and remit VAT in the consumer’s Member State have proven highly successful, with EU VAT revenues from these measures increasing tenfold from €3 billion in 2015 to over €33 billion in 2024.
Applying Belgium’s standard 21 percent VAT rate to all imports from information industries could generate approximately €8.15 billion in tax revenue, equivalent to about 3.3 percent of Belgium’s total tax revenues—a far greater sum than any digital tax could deliver. Furthermore, Belgium’s actionable VAT policy gap—the additional revenue that could be collected by eliminating reduced rates and certain exemptions—was 27.6 percent in 2024. Broadening the VAT base could generate up to €26.9 billion in additional national revenue, representing 10.76 percent of Belgium’s 2023 total tax revenue.
Compared to digital taxes, VAT offers clear advantages: neutrality, a broad tax base, avoidance of tax pyramiding, international consistency, higher revenue generation, and the absence of trade-related risks.
Conclusion
In her testimony, Cristina Enache concludes that the Belgian 2026 proposal, while presented as a modernization of corporate income taxation, fundamentally replicates the problematic features of Digital Services Taxes. By combining DST-style user-based market taxation, unilateral digital nexus rules, and gross-basis withholding and minimum taxation mechanisms, the proposal introduces a highly complex and distortionary tax regime.
These instruments are proven to generate limited revenues while imposing substantial economic costs through tax pyramiding, risks of double taxation, increased compliance burdens, trade tensions, greater international tax disputes, reduced tax neutrality, and the pass-through of costs to domestic businesses and consumers. Rather than representing a genuine improvement, the Belgian proposal amplifies many fundamental shortcomings by integrating multiple distortionary mechanisms.
A more coherent and economically sound policy approach for Belgium would focus on strengthening destination-based VAT systems and actively pursuing coordinated international solutions, rather than introducing yet another unilateral digital tax framework that risks undermining economic growth and international tax harmony.
