Contents

Not Every Dividend Is Capital

Is everything that the financial and economic system calls a dividend actually the same thing? The question may sound rhetorical, yet it is not: not everything labeled as “profit” or “dividend” necessarily corresponds to the same economic phenomenon. That being the case, the tax debate begins in a deeper place—that of concept.

This is not, strictly speaking, an article about increasing the tax burden, nor is it a piece on the political expediency of resuming the taxation of dividends in Brazil. The issue raised here precedes both. To discuss the return of dividend taxation, one must first examine the nature of what is intended to be taxed; the question is more conceptual and historical than revenue-driven.

This doubt becomes especially relevant when examining professional partnerships. For a long time in Brazil, lawyers, physicians, engineers, accountants, and other professionals organized under corporate structures have received a central portion of their compensation through profit distributions. Tax language grew accustomed to treating this outflow as profit or dividend. Yet giving different things the same name does not make them the same thing. What truly matters is whether the economic reality behind such distributions actually aligns with what is traditionally understood as a dividend.

In the case of a classic dividend, the answer seems relatively straightforward. A business corporation (sociedade anônima), for example, by legal definition, is a company whose capital is divided into shares, with shareholders’ liability limited to the issue price of the shares subscribed or acquired. Within this framework, the dividend appears as a clear expression of the partner’s position as a holder of invested capital. In standard economic terms, it represents the return on investment. Capital enters the company, assumes risk, and, if distributable positive results are generated, receives its share. That is the mental model typically attached to the word “dividend.”

That, however, is not necessarily the reality of professional partnerships. In the legal profession, for instance, Brazilian regulatory tradition moved in a different direction. As early as 1965, the Brazilian Bar Association (OAB) referred to a “civil labor partnership” (sociedade civil de trabalho), designed to govern the practice and the financial results earned from providing legal services. That is no trivial phrase; it reveals that, at the institutional origin of this type of organization, the central focus was never the productive autonomy of capital, but rather the association of professionals for the coordinated exercise of a personal, technical, and intellectual activity.

This approach was preserved, under different terminology, by the Brazilian Bar Statute. The law permits lawyers to associate in simple partnerships for the provision of legal services or to form single-member law firms, establishing that an attorney’s professional services are, by their very nature, technical and singular. The wording changed, yet the underlying premise remained the same. This is not an ordinary commercial enterprise whose decisive factor is the impersonal organization of means of production. Rather, it is an activity whose primary source of value continues to reside in the individual professional, in their technical qualification, and, above all, in their personal liability.

It is precisely here that the equivalence begins to unravel. When a partner in a professional partnership receives distributed earnings at the end of a fiscal year, they are not necessarily receiving a return on capital in the same sense that a corporate shareholder receives dividends. Calling it a dividend may be an accounting convenience; describing it as pure capital income, however, is a misreading. In practice, what reaches the partner’s pocket is not the yield of invested capital, but the sweat of a legal thesis, a court hearing, and a personal responsibility that cannot be delegated to a machine or a financial contribution.

In other words, what is being distributed is the product of intellectual labor performed by the partners themselves, channeled through a corporate structure.

Brazilian law itself, moreover, has already recognized in other areas that professional partnerships possess a distinct identity. Under Theme 918, the Supreme Federal Court (STF) ruled that municipal laws creating obstacles to the application of the fixed or per capita tax regime provided for in Decree-Law No. 406/1968 to professional law partnerships are unconstitutional. More recently, the Superior Court of Justice (STJ) reaffirmed, under Theme 1,323, that differentiated Municipal Service Tax (ISS) treatment depends on the personal rendering of services, individual technical liability, and the absence of an entrepreneurial structure that undermines the strictly personal nature of the activity—regardless of whether the entity is organized as a limited liability company (sociedade limitada).

Obviously, this is not a matter of automatically transposing the ISS framework into Income Tax law. Yet the systemic takeaway is significant: the legal order has long recognized the distinction between a capital company and a labor partnership.

Even the Brazilian Federal Revenue Service (Receita Federal), when examining single-member law firms, acknowledged something that weakens overly simplistic interpretations. In Cosit Private Letter Ruling No. 79/2021, it established that when the owner makes withdrawals, at least part of those amounts may constitute executive compensation (pró-labore), carrying the corresponding social security implications. While this does not, on its own, settle the debate over the taxation of so-called dividends, it demonstrates that even for the tax administration, a professional partner’s withdrawal cannot be automatically and entirely equated with passive capital income. There is, at minimum, an area in which the system recognizes the structural presence of labor in the composition of that compensation.

It is against this backdrop that Law No. 15,270/2025 ought to be read. The statute instituted a minimum tax for high-income individuals and established that, starting in January 2026, profits and dividends paid by a single legal entity to a single individual resident in Brazil exceeding BRL 50,000 in the same month are subject to a 10% withholding tax. The bill that gave rise to the law was introduced with that stated purpose: to establish a minimum tax for high-income individuals. The official framing, therefore, centers on high personal income and correcting the regressivity of the tax system.

What happens, however, is that Law No. 15,270/2025 operates like a terminological steamroller. Under the current drive to tax “high income,” it throws into the exact same bucket the rentier heir and the liberal professional who sustains a fixed-cost structure through their own labor. The legislature overlooked substantive differences in favor of revenue-raising convenience.

Intellectual honesty is warranted here, in any event. The final statute contains no express exception for single-profession partnerships. Nor would it be rigorous to claim that the legislature had a positive intent to exclude them from taxation. What can be stated with greater certainty is something more direct: the official justification for the measure was crafted to address minimum taxation on high incomes and the issue of earnings sheltered in the form of profits and dividends, without examining in sufficient depth the difference between capital dividends and profit distributions in labor partnerships.

Indeed, Senate deliberations recorded proposed amendments seeking to exempt from withholding tax the profits and dividends distributed by legal entities providing professional services subject to oversight by a professional regulatory council—a clear sign that the inadequacy of this equivalence was noticed during the legislative debate, even though it was not incorporated into the final text.

For that reason, in the case of professional partnerships, the most important critique may not immediately concern the size of the tax burden. It precedes it. It is a matter of questioning whether the system is calling things that are not alike by the same name. The classic dividend remunerates invested capital. A professional partner’s withdrawal, by contrast, is far closer to a sharing of proceeds from a labor partnership than to a passive return on capital.

Taxing intellectual effort as if it were capital gain is not, strictly speaking, an act of tax fairness. I would call it, rather, an error of legal taxonomy. The debate is not about how much to pay, but about what, exactly, we are paying on. In tax law, errors of taxonomy tend to generate the fiercest disputes.

Perhaps the greatest shortcoming of this debate lies in imagining that the problem can be solved by choosing between two ready-made labels: either it is a dividend, or it is pró-labore; either it is capital, or it is labor. Yet the reality of professional partnerships has never fit comfortably within such dichotomies. The professional partner is not a passive investor, because the wealth they receive stems from their own technical labor. Nor are they merely an employee of themselves, because they bear the costs, risks, and infrastructure that make the practice possible. The Federal Revenue Service itself, in stating that not everything can be treated as a dividend, acknowledges the hybrid nature of the phenomenon. What remains lacking is granting this hybridity an appropriate tax treatment. Until that happens, we will continue to witness the same mistake: calling what is not merely capital a dividend, calling what is not merely labor pró-labore, and taxing what ought not to be taxed.