food processing firms input quality upgrading
play

Food Processing Firms, Input Quality Upgrading Eric Tseng and Ian - PowerPoint PPT Presentation

Food Processing Firms, Input Quality Upgrading Eric Tseng and Ian Sheldon and Trade Virginia Tech AAEC Seminar, April 17th, 2015 Motivation - Quality Matters Quality an important determinant of trade flows (Linder 1961) Schott (2004)


  1. Food Processing Firms, Input Quality Upgrading Eric Tseng and Ian Sheldon and Trade Virginia Tech AAEC Seminar, April 17th, 2015

  2. Motivation - Quality Matters ❖ Quality an important determinant of trade flows (Linder 1961) ❖ Schott (2004) and Hummels & Klenow (2005) link exporter GDP per capita and product quality ❖ Hallack (2006) links product quality to importer GDP per capita ❖ Manova & Zhang (2012) show successful exporting firms in China use higher-quality intermediate inputs to produce higher-quality goods and firms vary quality of produces across destination markets ❖ Vertical product differentiation matters and should be modeled

  3. Motivation - Food Markets ❖ Food markets no longer characterized by homogenous products (Sexton 2013) ❖ Food quality matters, and firms in food industry use vertical product differentiation strategies ❖ Sunk costs related to production capacity and product quality matter ❖ Curzi, Raimondi & Olper (2014) investigate impact of trade liberalization on food product-quality ❖ Trade liberalization in exporting countries leads to faster upgrading of product quality for products closer to technology frontier ❖ On average, EU voluntary food-quality standards have positive effect on rate of quality upgrading

  4. Goals of Analysis ❖ Use modified heterogenous-firms framework allowing for intermediate input markets (Kugler & Verhoogen, 2012) ❖ Extend to focus on food quality and quality of agricultural inputs (Sexton, 2013), the impact of trade liberalization on food product-quality (Curzi et al, 2014) ❖ Examine closely relationship between food product- quality, trade liberalization, and ability of firms to upgrade quality of final goods

  5. 
 
 Model - Consumers ❖ Consumers: ❖ Utility 
 σ ⎡ σ − 1 ⎤ σ − 1 ( ) ∫ (1) U = q ( ω ) x ( ω ) σ d ω ⎢ ⎥ ⎣ ⎦ ω ∈ Ω ❖ Demand 
 − σ p O ( ω ) ⎛ ⎞ (2) x ( ω ) = Xq ( ω ) σ − 1 ⎜ ⎟ ⎝ ⎠ P

  6. 
 Model - Firms ❖ Firms ❖ Intermediate agricultural good produced via I production function: 
 I ( A , c ) = A (3) F c ❖ Since intermediate input market is perfectly p I ( c ) = c competitive, then

  7. Model - Firms ❖ Food processors (final good producers) require fixed λ λ investment cost to obtain capability , where is 
 f e 
 G ( λ ) = 1 − λ m k ⎛ ⎞ ⎜ ⎟ drawn from Pareto distribution with , 
 ⎝ ⎠ λ 0 < λ m ≤ λ ❖ Firms must pay fixed costs to enter market, , and also f incur fixed cost of exporting in all periods f x

  8. 
 
 
 
 
 Model - Firms ❖ Firms use inputs of capability, intermediate agricultural input and φ composite input of specific quality φ ❖ : additional tangible input that affects firm quality choice, i.e., capital equipment required to ensure quality control ❖ Production function for final good is: 
 F ( n ) = n λ a φ (4) MC = φ p I ( c ) λ a MC X = τφ p I ( c ) λ a

  9. 
 Model - Firms ❖ Food processors constrained by quality choice ❖ Inputs as complements in determining quality of good (Kremer, 1993; Kugler & Verhoogen, 2012) 
 β + 1 β + 1 ⎡ ⎤ ( ) ( ) ( ) (5) q = 1 β 3 λ b 3 φ 3 3 c 3 ⎢ ⎥ ⎣ ⎦

  10. 
 Model - Firms ❖ The importance of b ❖ is the scope of product-quality differentiation, b approximating fixed costs of investment required to translate capability into product quality ❖ Additional channel impacting firms’ quality choices, where lower restricts available quality choices, b while higher has higher available quality choices 
 b

  11. 
 
 
 
 
 
 
 
 
 
 
 Equilibrium ❖ Profit maximization yields following: 
 b (7a) c * ( λ ) = p I * ( λ ) = λ 3 b (7b) φ * ( λ ) = λ 3 (7c) q * ( λ ) = λ b σ ⎛ ⎞ 2 b 3 − a * ( λ ) = ⎟ λ ⎜ (7d) p O ⎝ ⎠ σ − 1 σ ⎛ ⎞ 2 b 3 − a * ( λ ) = ⎟ τλ ⎜ p O , X ⎝ ⎠ σ − 1 σ − 1 ) σ − 1 ⎛ ⎞ ⎡ ⎤ ( ) b ( λ η XP σ , η ≡ σ − 1 (7e) r * ( λ ) = 1 + Z τ 1 − σ 3 + a ⎜ ⎟ ⎢ ⎥ ⎝ ⎠ σ ⎣ ⎦

  12. 
 
 
 
 
 
 
 
 Comparative Statics ❖ Comparative statics regarding effects of parameters on firm size and final good quality choice 
 ( ) Z τ − σ ( ) Z τ − σ (8a) ∂ ln r * = 1 − σ 2 < 0, and ∂ ln q * = b 1 − σ 2 < 0 ( ) ( ) ∂ τ ∂ τ 1 + Z τ 1 − σ η 1 + Z τ 1 − σ ( ) = σ − 1 (8b) ∂ ln r * ln λ > 0, and ∂ ln q * = ln λ > 0 ∂ b ∂ b 3 b (8c) c * ( λ ) = φ * ( λ ) = λ 3

  13. Comparative Statics ❖ Comparative statics imply: ❖ Firms’ size (i.e., revenue) and the quality choice of final good increase with falling trade costs ❖ A firm that is better able to translate capability into quality produces higher-quality goods and is larger ❖ Trade costs negatively impact quality choice

  14. Comparative Statics

  15. Comparative Statics ❖ Comparative statics examining impact of trade liberalization and ability to translate capability on export entry cutoff point k − η ( ) − η ( ) k 1 − σ ⎛ ⎞ ∂ ln τ = k 1 − σ (11a) ∂ ln λ * f η η f λ m τ η < 0 ⎜ ⎟ ( ) η δ f e k − η ⎝ ⎠ f x 1 σ − η − 1 (11b) ∂ ln λ x ∂ ln τ = σ − 1 ⎛ ⎞ * η f x λ * τ η > 0 ⎜ ⎟ ⎝ ⎠ η f ❖ (11) states falling trade costs induce most productive non-exporting firms to enter export market, and least productive firms forced out of market, as exporting firms now capture larger market share ❖ Classic heterogenous-firms result (see Melitz, 2003)

  16. Comparative Statics ❖ Comparative statics examining impact of trade liberalization and ability to translate capability on market entry cutoff point ⎡ ⎤ ⎛ k − η ⎞ k ⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞ − k η η ⎢ f f ⎥ f ( ) ln τ ⎜ ⎟ ⎟ − σ − 1 Λ − ρ + τ 3 a + b ln ⎜ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎢ ⎥ − k ⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎜ ⎟ ⎝ ⎠ f X f X f X ⎝ ⎠ ⎛ ⎞ (12a) ∂ λ * λ m f ⎢ ⎥ b ⎣ ⎦ 3 + a ∂ b = 3 k τ ⎜ ⎟ δ f e ρ Λ 2 ⎝ ⎠ ⎡ ⎤ 1 σ − 1 ⎛ ⎞ ⎛ ⎞ ⎛ ⎞ (12b) ∂ λ x ∂ b = − λ * σ − 1 * η f f ⎢ ⎥ ( ) ln τ τ η ⎟ + σ − 1 ln ⎜ ⎟ ⎜ ⎜ ⎟ ⎢ ⎥ 3 η 2 ⎝ ⎠ ⎝ ⎠ ⎝ ⎠ f X f X ⎣ ⎦ ⎛ ⎞ f ( ) , Λ = 3 η − k ( ) ρ = ⎟ 3 a + b ⎜ ⎝ ⎠ f X ❖ These results are ambiguous in sign, due to other parameters

  17. Comparative Statics ❖ (12a) is dependent on this condition: (13a) ∂ λ * ∂ b < 0 when k < η + γ , and vice versa ⎛ ⎞ ⎛ η − k ⎞ ⎛ ⎞ ⎛ ⎞ − k ⎛ ⎞ η η f f ( ) ln τ ⎜ ⎟ ⎜ ⎟ γ = − 1 + τ ⎟ − σ − 1 ⎟ > 0 3 a + b ln ⎜ ⎟ ⎜ ⎜ ⎟ σ − 1 ⎜ ⎝ ⎠ ⎝ ⎠ ⎜ ⎟ ⎝ ⎠ f X f X ⎝ ⎠ ⎝ ⎠ ❖ The impact of depends on shape of the distribution of b firms, , i.e., market structure k

  18. Comparative Statics

  19. Comparative Statics ❖ (12b) is dependent on this condition: ⎛ ⎞ (13b) ∂ λ X * f ( ) ln τ > 0, and vice versa ∂ b < 0 when ln ⎟ + σ − 1 ⎜ ⎝ ⎠ f X f X > f f → f X ❖ The impact of depends on extent that . If , b then export rents outweigh fixed costs of exporting f X >> f given an increased . If , then fixed costs of b exporting outweigh export rents, leading to export exit.

  20. Data ❖ Source: Chile’s Encuesta Nacional Industrial Anual (ENIA), an unbalanced panel data set. Industry-level tariff rates from TRAINS database (WITS). ❖ Sample years: 2001-2007. ❖ Sample size: 11,196 observations, approximately 1,600 food-processing firms per year in the sample.

  21. Data

  22. Data

  23. Empirical Specifications ❖ Export Entry: ( ) = α + β 1 ⋅ c + β 2 ⋅ φ + γ ⋅ b + δ ⋅Δ τ + µ ⋅ λ + κ ⋅ X + ε (14) Pr Export i , t = 1 Export i , t − 1 = 0 ❖ Market Exit: ❖ Specification forthcoming: depends on how ENIA tracks firm exit ❖ Quality Choice: (15) q = α + β 1 ⋅ c + β 2 ⋅ φ + γ ⋅ b + δ ⋅Δ τ + µ ⋅ λ + κ ⋅ X + ε ❖ Changes in Size: 
 (16) Δ Size = α + β 1 ⋅ c + β 2 ⋅ φ + γ ⋅ b + δ ⋅Δ τ + µ ⋅ λ + κ ⋅ X + ε

  24. Results

  25. Conclusion ❖ Theoretical model adapts heterogenous-firms framework to food industry context ❖ Firms that remain in the market select higher quality and are larger given falling trade costs and increased ability to upgrade quality, and use concurrently higher-quality inputs ❖ Trade liberalization forces the least productive firms out of the market while most productive non-exporters enter the export market ❖ Impact of ability to upgrade quality dependent on the market structure: distribution of firms in the market and structure of fixed costs matter ❖ Empirical analysis currently provides mixed evidence: results cast doubt on quality constraint, but generally support the impact of on firm b characteristics and market structure

Recommend


More recommend