Research Paper : Why Manufacturing plants are smaller in India

 The sources present a compelling hypothesis, supported by extensive evidence from India, that the reason manufacturing plants are typically smaller in developing countries is primarily driven by greater consumer demand for lower quality goods in poorer regions. This demand structure interacts with the technology required to efficiently produce goods of varying quality.

The Core Hypothesis: Demand for Quality and Production Scale

The fundamental argument posits that the size-income relationship—where poorer regions have a higher share of employment in small establishments—is a natural consequence of low income levels, rather than solely caused by distortionary policies.

1. Demand Side (Consumer Preferences): Poorer households exhibit high demand for low quality products. This is modeled through non-homothetic preferences with respect to quality, where richer households are significantly more likely to choose higher quality levels. As a region develops and income levels increase, demand naturally shifts toward higher quality products.
2. Supply Side (Technology and Scale): Production of higher quality goods requires a larger scale due to the need for larger sunk and fixed costs (such as greater fixed investments, R&D, and skilled labor). Conversely, lower quality goods can be produced efficiently in small establishments because they require small fixed investments. The requirement for higher fixed costs means that producers of high quality goods must be larger on average to recover their costs, according to the free entry condition in the model.

This relationship ensures that poor countries or states, where low quality demand dominates, will have a size distribution characterized by many small plants.

Empirical Evidence from India

The authors provide extensive empirical evidence using Indian consumer and producer surveys to support the links between income, quality, price, and plant size.

1. Richer Households Buy Higher Quality (Higher Price) Goods

Using consumer expenditure surveys, the data confirms the non-homotheticity of preferences:

• Richer households pay a higher unit price for similar goods in the cross-section.
• The elasticity of price with respect to per-capita consumption (a proxy for income) is robustly positive, estimated between 11 and 17 percent. For example, the average price paid by a household at the 95th percentile of per-capita expenditure is 23 to 39 percent more than that paid by a household at the 5th percentile.
• This positive relationship between income and price paid for the same product holds even when controlling for product, state, and urban-rural fixed effects. It is also demonstrated using proxies for household income, such as education level and the number of dependents.

2. Larger Plants Produce Higher Quality (Higher Price) Goods

Combining data from the formal (Annual Survey of Industries, ASI) and informal (Survey of Unorganized Manufacturing, SUM) sectors, the evidence shows that plant size is strongly correlated with quality proxies:

• Output Price: Larger plants charge a higher unit price for similar goods than smaller plants, a relationship that holds across formal and informal plants. For instance, a plant employing 500 people charges a price approximately 62.9 percent more than a plant employing 5 workers when combining both surveys.
• International Quality Markers: Larger firms are more likely to be certified with international quality standards (like the ISO 14000 series certification) and are more likely to be exporters.
• Input Use and Investment: Consistent with higher quality production requiring more sophisticated inputs and capital:
  • Larger plants use more expensive material inputs.
  • Larger plants have a higher capital stock to output ratio and a higher capital investment flow to output ratio.
  • Firms with a higher capital-to-labor ratio charge higher prices for their goods.
• Skilled Labor: Larger firms hire more skilled workers. In establishments of size greater than 20 workers, 22 percent have graduated high school, compared to only 3 percent in establishments of size five or less. This suggests that high-quality production is skill-intensive.

Quantitative Explanatory Power

Using a general equilibrium model calibrated to match these micro-level facts (the price-income elasticity and the price-size relationship), the authors simulate changes in income levels (by varying productivity and the skill level of the population).

• Cross-State Variation: The model shows that the demand shift towards higher quality (and thus larger plants) as income rises explains a significant portion of the observed differences in plant size across Indian states. The model accounts for about 43 percent of the cross-state variation in the share of employment in plants of size five or less.
• Over-Time Variation: When applied to India's development over time, the model predicts 65 percent of the reduction in the share of employment in small plants observed between 1989 and 2009.

Contrast with Alternative Explanations

The sources acknowledge other leading explanations for the small size of manufacturing plants in developing countries, such as size-dependent regulation and institutional/structural factors:

• Regulatory Policies: A leading view attributes the prevalence of small plants to size-dependent regulations (like licensing rules). However, the sources note that these policies cannot be the only factor, as previous research suggests they leave a large portion of the size differences unexplained, and many developing countries have a large share of small plants irrespective of such policies.
• Structural Factors: Other complementary factors include technological design that favors small-scale production, low skill availability, weak selection pressures, and constraints like credit-market imperfections and limited managerial delegation.

The paper's conclusion is that the variation in plant size seen across countries and states is substantially a "natural consequence of the low levels of income" in developing countries, suggesting that the presence of small plants is not necessarily due to policy failures.