Candice Horbacz
For years, economists, journalists, and policymakers have pointed to a specific labor share of income series from the Bureau of Labor Statistics (BLS) as a primary indicator that capital is capturing an ever-increasing portion of the economic pie. This widely cited BLS series depicts labor’s share plummeting from nearly two-thirds in the 1950s to approximately half today, igniting alarmist headlines such as “US workers’ share of national income falls to a new low.” Such pronouncements have fueled widespread concerns about income inequality, the erosion of workers’ bargaining power, and the potential for technological advancements, particularly artificial intelligence, to further displace human labor and exacerbate these trends. The narrative suggests a fundamental shift in the distribution of economic gains, with implications for social stability, consumer demand, and long-term economic growth. However, a meticulous examination of the broader national income accounts reveals a significantly different and more nuanced picture, suggesting that labor’s share is both higher and notably more stable than the BLS series implies.
Challenging the Prevailing Narrative: A Misinterpretation of "Capital Income"
The prevailing narrative, heavily influenced by the BLS series, often simplifies the complex allocation of national income into two broad categories: labor and capital. Most recently, for the second quarter of 2026 (preliminary data), the BLS series reported that 53 percent of income accrued to labor, leaving a substantial 47 percent to what commentators frequently label as "capital" or "owner" income. This seemingly stark division has become a cornerstone of arguments positing an aggressive shift of wealth from workers to asset owners. Yet, in much of this public commentary, the precise definitions of income and the underlying assumptions employed by the BLS to categorize and split these components often remain unclarified, leading to potential misinterpretations.
To truly untangle these complexities, it is essential to build directly from the comprehensive national accounts, which offer a more holistic view of how total U.S. income is divided between capital and labor, and then contrast this with the more narrowly defined BLS series. Gross Domestic Income (GDI), a measure of the total income earned from the production of goods and services in the U.S. economy, stood at approximately $32.2 trillion at an annual rate in the second quarter of 2026. Of every dollar of this GDI, a clear 50.4 cents was paid directly to workers as compensation, comprising 41.5 cents in wages and salaries and an additional 8.9 cents in benefits. This portion is unequivocally labor income. The crucial point of divergence, however, arises when interpreting the remaining 49.6 cents of "nonlabor" income. It does not automatically follow that this entire amount accrues to capital.
Deconstructing Capital’s True Share: A Far Smaller Slice
A granular breakdown of the national accounts reveals that the income unambiguously attributable to capital includes corporate profits after corporate tax, interest income, and rents. When aggregated, these components amounted to approximately 17 cents of every dollar of gross income. Within this, about 3.6 cents represents "imputed rent," which is an estimate of what homeowners would hypothetically pay to rent their own homes. While an important accounting convention for national statistics, imputed rent is not actual cash flow collected by individuals and does not align with the common understanding of "capital income" received by investors. (It is also worth noting that the national accounts include the "current surplus of government enterprises," which typically operate at a loss and thus contribute a small, negative value, usually excluded from these analyses.)
Further complicating the simple labor-capital dichotomy is the income generated by proprietorships and partnerships, which accounted for another 6.7 cents of every dollar. This category represents a blend of compensation for the owners’ active work and a return on their invested capital. Even if one were to take the most extreme position and attribute this entire 6.7 cents solely to return on investment – thereby maximizing capital’s share – capital would still only earn a total of 24 cents per dollar of gross income. This figure is starkly lower than the nearly half attributed to "nonlabor" income by the BLS series and frequently portrayed in media headlines.
The Illusion of Gross Income: What Doesn’t Accrue to Anyone
The significant discrepancy arises because the remaining categories within "gross income" do not actually represent income that accrues to any individual or entity in the same way wages or profits do. These components are essential for understanding the economy but are often miscategorized when discussing income distribution.
One of the largest such categories is depreciation, which consumes nearly 17 cents of every dollar of gross income. Depreciation represents the cost of replacing worn-out buildings, machinery, equipment, and software. It is an economic reality – the capital stock needs constant replenishment simply to maintain its existing productive capacity. This spending merely returns the capital stock to its starting point each year and, critically, never translates into a paycheck for a worker or a deposit in a brokerage account for a capital owner. Counting depreciation as "income accruing to capital" fundamentally misunderstands its economic function as a cost of doing business rather than a return on investment.
Similarly, taxes on production and imports (TOPI), totaling 7.0 cents, and corporate income taxes, accounting for 2.8 cents, are collected before income ever reaches a household or directly benefits capital owners. TOPI includes various levies such as sales and property taxes, federal excise taxes, and customs duties (net of subsidies). These are costs borne by businesses that are passed on to consumers or reduce overall business profitability before any distribution to labor or capital owners. Attributing these taxes as "capital income" is particularly problematic. Consider tariffs, for instance: every dollar of tariff revenue mechanically increases the "nonlabor" share of income and is thus implicitly categorized as "capital income" under this broad definition. However, analysts’ interpretations of tariff incidence vary widely, but no common economic theory considers tariff revenue to be a direct return to capital owners. These taxes represent revenues for the government, not income for private capital.
Using a gross measure of income, therefore, conflates these essential economic costs and government revenues with actual capital returns, distorting the true picture of income distribution. This methodological choice significantly inflates the perceived share of income accruing to capital.
A Clearer Picture: Labor’s Share of Net Income Within Historical Norms
To gain a more accurate understanding of how income is truly distributed, it is necessary to move beyond gross income and focus on net income, which is calculated by removing depreciation and taxes. In the second quarter of 2026, this adjusted measure indicated roughly $23.7 trillion of private sector income that was actually paid out to people.
A leading body of economic literature on the labor share, as highlighted in a prominent literature survey, distinguishes between three critical categories of net income for a clearer analysis:
- Unambiguous labor income: This primarily consists of employee compensation (wages, salaries, and benefits).
- Unambiguous capital income: This includes corporate profits, interest, and rents.
- Ambiguous remainder: This largely comprises proprietors’ income, after excluding taxes, whose nature is a mix of labor and capital returns.
Tracking these three categories as shares of net income since 1947 presents a much cleaner and substantially different narrative compared to the often-cited BLS trend. Unambiguous labor income, far from being at a "never-before-seen" low, has experienced a "round trip" over the post-war era. It constituted approximately 69 percent of net income in the late 1940s, rose to about 75 percent in the 1970s, and currently stands at 68.3 percent. This indicates that labor’s share, when properly measured using net income, is within historically precedented levels, not at an unprecedented trough.
Conversely, unambiguous capital income has seen an increase, rising from about 13 percent of net income in the late 1940s to 22.6 percent today. More than half of this rise has occurred since 2000, when it was around 17 percent, with a notable acceleration during the pandemic years of 2020 and 2021, predating the widespread public awareness and adoption of advanced AI tools. This suggests that while capital’s share has grown, it remains significantly below the near-half mark often suggested by gross income analyses.
The "ambiguous proprietors’ slice" has also seen significant shifts. It fell from about 18 percent of net income in the 1940s to roughly 10 percent by 1970, a trend largely driven by the structural decline in the agricultural sector and the rise of the corporate economy. It reached a low of approximately 6.7 percent in 1982, subsequently recovering to hover between 9 percent and 10 percent, most recently measuring 9.1 percent.
The inherent ambiguity of proprietors’ income means its allocation significantly impacts the final calculation of labor versus capital share. If this income is entirely assigned to labor, capital’s share remains at 22.6 percent of net income. If, conversely, it is entirely attributed to capital – an approach that generates the highest possible capital share from this dataset – capital’s share rises to 31.7 percent of net income. This figure, while higher, is still considerably less than half. Recent research, however, strongly suggests that proprietors’ income is predominantly labor income, implying that even the 31.7 percent estimate is likely an overstatement of capital’s true share.
Crucially, because proprietors’ income has maintained a relatively stable share since the late 1980s, its trends cannot be solely responsible for driving recent shifts in the BLS labor share. While a time-varying allocation of proprietors’ income can influence overall trends, it’s one of several factors contributing to the BLS’s reported decline.
Unpacking the BLS Approach: Methodological Assumptions and Their Impact
Every published estimate of labor share necessitates making specific assumptions to divide income into labor and capital categories. While the corporate sector allows for a relatively clear split between wages and profits, the noncorporate sector, where income is a blend of labor and capital, requires estimation. This is where different methodologies yield different results. Many analyses, for instance, restrict their scope to the corporate sector to avoid these complexities, while others exclude government, nonprofits, farms, and imputed rent.
The BLS measure, which is often cited as the definitive source, employs a specific imputation strategy to split the noncorporate business sector’s income into labor and capital components, resulting in a time-varying allocation. It covers only the nonfarm business sector, representing approximately three-quarters of the U.S. economy. This limited scope contrasts sharply with the national accounts approach, which strives to keep all income under one consistent convention.
The BLS’s imputation strategy for proprietor income is particularly influential. It assumes that proprietors "pay themselves" the average hourly compensation of employees within their sector, multiplied by their hours worked. Any remaining income is then treated as capital income. Since hours per worker tend to move slowly, this method essentially compares two averages. This approach has led to a significant shift in the inferred capital share of proprietors’ income: estimates suggest it rose from less than a fifth in 1990 to about half today, a trend well-documented by researchers like Elsby, Hobijn, and Şahin through 2012. This methodological assumption alone can dramatically alter the perception of capital’s growing share.
Furthermore, the BLS measure excludes several significant parts of the economy. Government and nonprofits, which collectively represent roughly 15 percent of the economy, are omitted. As BLS economists themselves acknowledge, including these sectors, which are heavily labor-intensive, would likely increase the estimated labor share. Farms are also excluded; while their impact is minimal today, this exclusion renders historical comparisons anachronistic, given that farm proprietors’ income constituted 6 percent of the total in the late 1940s.
Finally, the BLS ratio calculates labor share by dividing income-side compensation by product-side output. This means that the statistical discrepancy, or the bookkeeping gap between Gross Domestic Product (GDP) and Gross Domestic Income (GDI), can introduce noise and affect the reported trend. In contrast, approaches that consistently use the income accounts ensure that all components sum to total income by construction, avoiding such discrepancies.
Implications for Policy and Public Discourse
The pervasive description of the labor share as "unprecedented" or at a "record low," the notion of a "close to 50-50 split" between labor and capital, and the widely circulated figures depicting a consistent downward trend since the 1940s are, therefore, demonstrably misleading. While labor earns approximately half of every dollar of gross income, it is crucial to understand that the remainder should not be indiscriminately attributed to capital. A significant portion of what is counted in gross income, such as depreciation and various taxes, does not accrue to anyone as disposable income or return on investment.
A more accurate and economically meaningful measure uses net income. Under this refined lens, the share of net income accruing to capital ranges between 22.6 percent and 31.7 percent, depending on how proprietors’ income is allocated. This is considerably less than half. The labor share, far from being at a "never-before-seen" low, is actually back to a historically precedented level. Furthermore, the simplistic "downward trend throughout" the post-war era needs to be replaced with a more accurate understanding: the labor share rose, then fell, completing a "round trip" rather than a consistent decline from its starting point.
These distinctions are not merely academic; they carry significant implications for policy debates surrounding income inequality, taxation, and economic regulation. If policymakers operate under the assumption that capital is taking an ever-larger and unprecedented slice of the pie, they might advocate for policies aimed at drastically redistributing wealth or curtailing capital accumulation. However, if the true picture suggests a more stable distribution, with capital’s share significantly lower and within historical bounds, then the focus of policy interventions might shift towards other drivers of inequality or specific labor market issues, rather than a generalized capital-labor conflict. Correcting the headline narrative with a more rigorous, empirically sound understanding of labor’s share is paramount for informed economic policymaking and for fostering a more accurate public discourse about the state of the U.S. economy and the well-being of its workforce.
