📈 economics

Minimum Wages, Firm Size Distribution, and the Labor Share

This paper identifies a "weighting effect" mechanism explaining the decline in the labor share, demonstrating that minimum wage hikes in China force labor-intensive small firms to exit and shift economic activity toward capital-intensive large firms, thereby mechanically depressing the macro labor share independent of market power.

Original authors: Jiyuan Lyu

Published 2026-07-22

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Jiyuan Lyu

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the economy as a giant, bustling kitchen where millions of chefs (firms) are cooking up goods and paying their helpers (workers). For decades, economists believed that the slice of the pie workers took home—their share of the total income—stayed roughly the same, like a steady heartbeat. This was a comforting idea, suggesting that as the kitchen got bigger and richer, everyone got a fair cut. But lately, that heartbeat has been skipping. In many places, the workers' slice is shrinking while the owners' slice is growing, and nobody is quite sure why. Is it because robots are taking over? Is it because companies are moving their kitchens to cheaper countries? Or is it because a few "superstar" chefs are hogging all the best ingredients and charging extra for their famous recipes? This paper dives into that mystery, looking at the hidden mechanics of how different-sized kitchens operate and how government rules about minimum pay can accidentally change the whole menu.

The authors of this study, Jiyuan Lyu, propose a new theory called the "weighting effect." Think of it like this: imagine two types of kitchens. The small, local cafes are very "hands-on," relying heavily on chefs and waiters to get things done. The massive, industrial food factories, on the other hand, are packed with expensive, capital-intensive machines that do most of the work, needing fewer humans per burger. The paper argues that large firms naturally use more machines and fewer people than small firms, even if they are all competing fairly without trying to cheat or raise prices.

Here is the twist: if the economy shifts so that more food is being made by the giant factories and less by the small cafes, the overall share of money going to workers will drop. It's not because the factories are paying their workers less; it's simply because the "weight" of the economy is moving toward the machine-heavy side. The authors call this a "technology-structure" channel, meaning the mix of firm sizes changes the math of who gets paid, independent of whether any single firm is being greedy or powerful.

To test this, the researchers looked at China between 1998 and 2007, a time when the government was raising the minimum wage. They treated these wage hikes like a natural experiment. The logic was simple: raising the minimum wage hurts small, labor-heavy cafes the most because they rely so much on cheap labor. Big factories, with their expensive machines, can absorb the cost better. So, when the minimum wage went up, many small cafes were forced to close their doors or shrink, while the big factories kept growing.

The results were striking. The study found that a standard increase in the minimum wage shock caused the overall labor share to drop by about 0.11 percentage points. While that sounds tiny, it represents roughly 3.5% of the average labor share in their data, which is a significant shift. The authors were very careful to prove this wasn't just because big companies were becoming more powerful or "superstars" that could charge higher prices. Even after controlling for how concentrated the market was, the effect remained. They showed that the minimum wage hike didn't just make firms swap workers for machines inside their own walls; it actually changed the structure of the market by pushing small firms out and letting big, machine-heavy firms take over their market share.

The paper also checked the "middle steps" of this story. They confirmed that when the minimum wage went up, small firms exited the market at higher rates, new small firms stopped entering, and the average size of the surviving big firms grew. This reallocation of output toward the capital-intensive giants is what mechanically lowered the total labor share.

Interestingly, the authors suggest that this mechanism works even in a perfectly fair market where no one is cheating. It's a structural side effect of how different-sized businesses are built. The study suggests that while minimum wages are good for protecting workers' basic rights, they can have an unintended consequence: by squeezing out small, labor-intensive businesses, they might inadvertently shift the economy toward a structure where capital (machines) gets a bigger slice of the pie than labor (people). The authors conclude that to protect the workers' share of income, policymakers might need to do more than just set wage standards; they might also need to help small businesses upgrade their technology so they can survive the cost shocks without disappearing, keeping the "kitchen" diverse and the workers' slice safe.

Technical Summary: Minimum Wages, Firm Size Distribution, and the Labor Share

1. Problem Statement and Motivation
The paper addresses the sustained global decline in the labor share of income since the 1980s, a trend that challenges Kaldor's classic "stylized facts" regarding the stability of factor shares. While existing literature attributes this decline to capital-biased technological change, globalization/outsourcing, or the "superstar firm" hypothesis (where market power allows large firms to capture markups), this paper proposes a distinct mechanism: a "weighting effect" driven by the firm size distribution.

The core theoretical proposition is that large firms systematically adopt more capital-intensive technologies than small firms, independent of market power. Consequently, a shift in economic activity toward larger firms mechanically depresses the aggregate labor share, even under perfect competition. The paper investigates whether labor market institutions, specifically minimum wage hikes, can act as an exogenous shock that alters the firm size distribution, thereby triggering this weighting effect.

2. Theoretical Framework
The authors construct a parsimonious theoretical model linking firm size distribution to the macro labor share:

Assumptions: Firm output follows a truncated Pareto distribution. Crucially, the model assumes a "size gradient in technology choices": the output elasticity of capital demand ( $\gamma$ ) exceeds that of labor demand ( $\beta$ ), implying that as firm size expands, the capital-labor ratio rises ("capital deepening").
The Weighting Effect: The firm-level labor share is approximated as a linear function of log size with a negative slope ( $\delta < 0$ ). The macro labor share is the output-weighted average of firm-level shares. The model derives a reduced-form equation:
$LS_{macro} = LS + \delta \cdot \Phi(\xi, r)$
where $\xi$ is the Pareto exponent (measuring the equality of the size distribution) and $\Phi$ is a strictly decreasing weighting factor.
Mechanism: A decline in $\xi$ (indicating a concentration of activity toward large firms) reduces the macro labor share because the weighting factor shifts output toward firms with lower labor shares ( $\delta < 0$ ).
The Shock: Minimum wage increases are modeled as an exogenous force that raises the exit probability of labor-intensive small firms, reducing $\xi$ and shifting output toward large, capital-intensive firms, thereby depressing the aggregate labor share.

3. Data and Identification Strategy

Data: The study utilizes the Chinese Annual Survey of Industrial Firms (CASIF) from 1998–2007, covering approximately 1.6 million firm-year observations across 30 two-digit manufacturing industries.
Unit of Analysis: "City-industry-year" market units.
Instrumental Variable (Bartik Strategy): To address endogeneity (e.g., labor costs causing exit), the authors employ a Bartik-style instrument.
- Initial Exposure ( $S_{cj}$ ): The share of small firms (output below the median) in a city-industry cell during the pre-sample period (1998–2000).
- Exogenous Shock ( $G_{pt}$ ): The annual growth rate of the provincial minimum wage.
- Instrument ( $B_{cjt}$ ): The interaction $S_{cj} \times G_{pt}$ .
- Logic: City-industry cells with a high initial share of small firms in a province experiencing rapid minimum wage growth face greater "pressure," leading to higher small-firm exit and a shift in the size distribution.
Secondary Instrument: To test the interaction with the technology gradient, the authors interact the Bartik instrument with the industry-level initial technology gradient ( $\delta^0_j$ ), estimated from the 1998–2000 data.

4. Key Empirical Results

Baseline Effect: A one-standard-deviation increase in the Bartik instrument (minimum wage shock) reduces the market-level labor share by approximately 0.11 percentage points, representing about 3.5% of the sample mean.
Independence from Market Power: The effect remains robust and virtually unchanged after controlling for market concentration (HHI and CR4), indicating the mechanism operates independently of the "superstar firm" market power channel.
Mechanism Validation: Direct tests confirm the causal chain:
- The shock increases firm exit rates and decreases entry rates.
- The number and share of small firms decline significantly.
- The average output of surviving large firms expands, confirming the reallocation of output toward capital-intensive producers.
Causal Link between Size Distribution and Labor Share: Supplementary IV estimates regressing the labor share on the Pareto exponent (instrumented by the Bartik shock) yield a coefficient of 0.4046, which is roughly three times larger than the OLS estimate (0.1435). This suggests OLS suffers from attenuation bias and confirms that a more equal firm size distribution causally raises the labor share.
Separation from Factor Substitution: When controlling for industry-average capital-labor ratios (to isolate within-firm substitution), the Bartik coefficient declines by roughly half but remains statistically significant. This indicates that the reallocation (weighting) effect accounts for a substantial portion of the total impact, distinct from direct capital-for-labor substitution within firms.

5. Robustness and Placebo Tests
The findings are robust to:

Alternative labor share measures (output-based vs. value-added).
Alternative size distribution measures (Gini coefficient).
Excluding first-tier cities or specific years (e.g., 2007).
Province-specific linear time trends.
Randomization Inference: Permutation tests where provincial minimum wage growth rates are randomly reassigned yield a mean coefficient near zero, with the true estimate lying far outside the distribution ( $p=0.000$ ).
Fake Exposure Placebo: Using industry capital intensity (unrelated to small-firm labor sensitivity) as the exposure variable yields a negligible coefficient, confirming that the effect is specific to the exposure of labor-intensive small firms.

6. Significance and Contributions
The paper claims three primary contributions:

Theoretical: It extends the Houthakker-type aggregation framework by endogenizing the relationship between firm size and technology choice. It formalizes the "weighting effect," demonstrating that a shift toward large firms depresses the labor share even without market power or markup behavior.
Empirical: It establishes a causal chain from labor market institutions (minimum wages) to macro distributional outcomes via the firm size structure. It complements the superstar firm hypothesis by identifying a parallel mechanism where labor regulations, rather than product market power, drive the reallocation of activity toward large, low-labor-share firms.
Policy: The findings suggest that safeguarding the labor share requires more than just wage standards. Policymakers must consider the structural impact of minimum wages on the firm size distribution. Specifically, measures should support the technological upgrading and capital access of small and medium-sized enterprises (SMEs) to prevent their forced exit, thereby maintaining a more balanced firm size structure that supports a higher aggregate labor share.

The paper concludes that while minimum wages are essential for worker protection, their general equilibrium consequences—specifically the asymmetric pressure on small, labor-intensive firms—can inadvertently reshape the market structure in a way that reduces the overall labor share.

More like this