## _wp0716 - References

## Source details

**Canonical URL:** [_wp0716 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0716.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0716.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0716.pdf.json)

---

### I. Analyzing trade-income inequality linkages: lacunae in the literature
- Research question: Is there any cause, sui generis, for trade per se reducing inequality across nations?
- Core conceptual findings:
  - Long-term sustained productivity growth and distributional outcomes depend on human capital formation, structural reforms, trade reforms, and sound macroeconomic policies (Global Economic Prospects 2004, p.239).
  - Empirical literature often finds that trade helps nations move to a higher growth path and hence affects income distribution (Grossman and Helpman 1991; Dollar and Kraay 2002; Frankel and Romer 1999; Richardson 1995; Helpman 2002).
  - Trade liberalization tends to alleviate poverty over the medium to long term (Winters, McCulloch, and McKay 2004), but distributional responses determine poverty outcomes (Deininger and Squire 1996; Bourguignon 2004).
  - Technology and knowledge diffusion (ICT, bio-technology, nanotechnology) drive productivity and growth and thus poverty reduction (Millennium Project Report 2005; Bussolo and O’Connor 2002).
  - Trade is an important conduit for technology spillovers from richer partners to recipients (Keller 2004; Coe and Helpman 1995; Bayoumi, Coe, and Helpman 1999).
- Analytical contribution of the paper:
  - Presents a heuristic model linking exogenous technological change, adoption, productivity improvements, and their effects on income inequality, poverty, and welfare.
  - Uses an augmented GTAP global CGE model (78 regions × 57 sectors) to study multi-sectoral, multi-regional linkages and international spillovers.
  - Formal objectives:
    - (a) Estimate the impact of technical changes and trans-border propagation on income inequality under different trade policy configurations.
    - (b) Assess the role of human capital and socio-institutional features in reducing income inequality.
- Policy-relevant synthesis:
  - Trade reform alone is insufficient; complementary socio-institutional factors (human capital, governance, R&D, social acceptance) and absorptive capacity are critical to capture trade- and technology-induced benefits and to reduce inequality and poverty.
  - Failure to establish appropriate national institutional and policy frameworks hinders realization of trade-induced benefits (Gunter, Cohen, and Lofgren 2005).

### II. Trade, technology and income inequality: background quantitative evidence
- Data and measures:
  - GTAP global database Version 5.4 with income split between two labor types: skilled and unskilled (Dimaranan and McDougall 2002 and 2003).
  - National Gini coefficients from World Bank's World Development Indicators (2003, 2005).
  - Proportion of skilled personnel (Ps) proxied by tertiary enrolment ratio (percentage of population age 15-64 having tertiary education or more) from World Bank WDI.
  - GTAP skilled-unskilled income shares based on average years of education (Barro and Lee 1996).
- Methodology:
  - Yitzhaki (2002) Gini decomposition adapted to compute inter-skill Gini indexes for skilled (s) and unskilled (u) cohorts using equations reported in the source (Equations (1)–(4) in text).
- Empirical pattern:
  - Computed inter-skill Gini results (Table 1) indicate that across almost all regions the computed Gini coefficient is lower for the unskilled labor category than for the skilled labor category.

### III. Interoccupational inequality and trade openness: empirical support
- Trade orientation measure: OPEN = (Export (X) + Import (M)) / GDP.
- Regression linking trade openness to GDP (78 GTAP regions):
  - ln GDP (r) = α + β ln OPEN (r) + ε(r)  (Equation (5))
  - Estimated: ln GDP (r) = - 0.0605  + 1.035 ln OPEN (r), R2 =0.91, S.E. = 0.55      (5a)
  - t-values shown: (-0.155) for intercept, (29.04) for slope.
  - Interpretation: GDP growth is positively related to trade openness; slope 1.035 and R2 =0.91 are highlighted.
- Regression linking trade openness to Gini coefficients (elasticities):
  - ln Gyi (r) = γ + δ ln OPEN(r) + μ (r), with i ∈{o, s, u}  (Equation (6))
  - Estimates reported exactly:
    - ln Gyo (r) = 4.128 +  (-) 0.121   ln OPEN(r), R2 =0.08, S.E. =0.234      (6a)
      - t-values: (18.20) for intercept, (-2.29) for slope.
    - ln Gys (r) = 4.437 +  (-) 0.165   ln OPEN(r), R2 =0.05, S.E. =0.426      (6b)
      - t-values: (10.77) for intercept, (-1.73) for slope.
    - ln Gyu (r) = 4.726+  (-) 0.349   ln OPEN(r), R2 =0.08, S.E. =0.643      (6c)
      - t-values: (6.923) for intercept, (-2.190) for slope.
- Empirical takeaways (numeric highlights preserved):
  - Trade openness strongly positively associated with GDP: coefficient 1.035 in (5a) with R2 =0.91, S.E. = 0.55.
  - ln Gyo: slope = (-) 0.121, R2 =0.08, S.E. =0.234 (6a).
  - ln Gys: slope = (-) 0.165, R2 =0.05, S.E. =0.426 (6b).
  - ln Gyu: slope = (-) 0.349, R2 =0.08, S.E. =0.643 (6c).
  - Point estimates interpreted: a 1 percent increase in trade openness tends to lower inequality indexes by 0.12% (skilled), 0.17% (unskilled), and 0.35% (overall) as discussed in the chapter.

### IV. Technology, institutions, absorption capacity and inequality (model indices)
- Key indices and exact definitions preserved:
  - Absorption capacity (AC): region-specific with 0 ≤ AC ≤ 1; depends on education and schooling years.
  - Governance parameter (GP): bilateral GP_rs = min[1, GP_s / GP_r] with 0 ≤ GP_rs ≤ 1 (equation (8)).
  - Technological congruence (TC): TC_rs = min[1, TC_s / TC_r] with TC_rs ∈ [0,1] (equation (9)).
  - Social acceptance (SA): SA_s = min[1, SA_s / SA_threshold] with SA_s ∈ [0,1] (equation (10)).
  - Institutional-structural congruence (SC): SC_rs = GP_rs · TC_rs (equation (13)).
  - Technology appropriation / capture parameter (θ_s or TAP): θ_s = AC_s · SA_s · SC_rs (equation (14)); θ_s ∈ [0,1].
- Technology Achievement Index (TAI) values (selected exact entries from Table 4):
  - Bangladesh: 0.04
  - Canada: 0.59
  - China: 0.30
  - EU: 0.51
  - India: 0.20
  - Japan: 0.70
  - United States: 0.73
- Spillover coefficient stylized functional form (preserved):
  - γ_s = 1 − (E_rs)^(1−θ_s) (equation (15)) with properties γ_s(0) = 0, γ_s(1) = 1, γ'_s > 0, γ''_s < 0.

### V. Stylized PROGRESS model linking capture, spillovers, and inequality
- Conceptual mapping:
  - Higher capture parameter θ increases realized spillovers γ and reduces converted Gini G_convert.
  - Boundary behavior: θ = 1 → G → 0; θ = 0 → G → 1 (stylized).
- Functional specification and parameters (preserve definitions and values used):
  - PROGRESS scalar: PROGRESS = g × γ × η.
  - Parameter settings cited:
    - g = average proportional rate of decrease of Gini due to growth in trade dependency ratio, empirical value taken as 0.2%.
    - η = elasticity of poverty with respect to trade liberalization, implementations set η(r) uniformly at 2.0.
  - Comparative statics preserved:
    - ∂G_convert / ∂θ < 0.
    - With γ → ∞, G_convert → 0.
    - With γ → 0 and η ∈ [2, 2.5], G_convert → G_initial.
- Poverty gap linkage formulas (preserved forms):
  - PG_r = G_convert(r) − μ(r) × ln(θ_r) (equation (23)).
  - PG(s) = G_convert(r) − μ(s) × ln(θ_rs) for recipient s with unique source r (equation (24)).
  - Comparative statics: ∂PG_r / ∂θ < 0 for θ, μ > 0.

### VI. Model implementation framework, data, and parameterization
- Modeling approach:
  - Modified multi-regional, multi-sectoral CGE based on GTAP Version 5.4 / Version 6 data; GTAP disaggregates world into 78 regions and 57 sectors; reduced 14×12 aggregation used for calibration.
  - GEMPACK software suite used to conduct simulations.
- Augmented parameter-sets beyond standard GTAP:
  - Skill-induced AC index, Governance parameter GP, Social acceptance SA, Technological proximity TC.
- Representative parameter orderings and exact values reported (selected from source tables):
  - AC ordering: ACUSA > ACHTW > ACEUFTA > ACJAPAN > ACMERCOSUR > ACSAM > ACCAN > ACCHINA > ACINDIA > ACSOUTHASIA.
  - GP example magnitudes:
    - India: GP magnitude 0.65
    - Sri Lanka: GP magnitude 0.69
    - China: GP binary parameter 0.32
    - HTW: GP binary parameter 0.74
- Capture and congruence construction follow equations (8)–(14) as specified in the source.

### VII. Simulation design and policy experiments (focus and shocks)
- Regions of interest: India and China (together contain "totally 2.4 billion out of global 6.2 billion").
- Poverty headcount references (World Bank / Cline, 2004) quoted exactly:
  - China: 53.7% (income less than $2 per day)
  - India: 86.2% (income less than $2 per day)
  - (Cline (2004) alternative estimates: India 48.4% and China 39.2%)
- Shock types:
  1. Technology shocks: TFP augmentation in a high-technology sector in USA transmitted to other nations.
  2. Trade policy shocks: trade liberalization under Hub-and-Spokes (HAS) FTA configurations between USA vis-à-vis India and China; subsequent Indo-China FTA between spokes.
- Technology shock specifics:
  - Keller (1999) reports average annual multifactor productivity growth in composite hi-tech sector = 3.2% during 1970-1991.
  - Authors extrapolate over 6 years and implement a 4% shock to the Hicks-Neutral technological coefficient in the hi-tech sector of the USA.

### VIII. Simulation results: macroeconomic repercussions of 4% Hi-tech TFP shock in the USA
- Regional TFP growth (percent changes) from Table 6 (preserve exact values):
  - US: 3.6
  - Canada: 2.2
  - EU: 0.76
  - China: 0.11
  - South America: 1.65
  - India: 0.10
  - MERCOSUR: 0.30
  - Bangladesh: 0.12
  - Sri Lanka: 0.16
- Real GDP at Factor Cost (percent changes) mirror TFP column (exact values preserved):
  - US: 3.6
  - Canada: 2.2
  - EU: 0.76
  - China: 0.11
  - South America: 1.65
  - India: 0.10
  - MERCOSUR: 0.30
  - Bangladesh: 0.12
  - Sri Lanka: 0.16
- Table 7 selected base-period parameter values (preserve exact tuples for each region):
  - USA: Embodiment Index 0.62, Spillover Coefficient 0.82, Capture-Parameter 0.71, ACs 0.71, GPrs 1.00, TCr s 1.00, SCrs 1.00, SAs 1.00
  - Canada: 0.38, 0.51, 0.32, 0.40, 1.00, 0.80, 0.80, 1.00
  - China: 0.02, 0.03, 0.03, 0.23, 0.32, 0.41, 0.13, 0.95
  - India: 0.02, 0.03, 0.03, 0.21, 0.65, 0.27, 0.18, 0.74
  - MERCOSUR: 0.05, 0.06, 0.08, 0.44, 0.46, 0.46, 0.21, 0.90
  - Bangladesh: 0.03, 0.02, 0.002, 0.28, 0.22, 0.05, 0.01, 0.62
  - Sri Lanka: 0.03, 0.03, 0.05, 0.27, 0.69, 0.28, 0.19, 0.90
  - EU: 0.04, 0.18, 0.47, 0.67, 1.00, 0.69, 0.69, 1.00
  - HTW: 0.09, 0.20, 0.33, 0.71, 0.74, 0.62, 0.46, 0.99
  - Japan: 0.02, 0.10, 0.48, 0.62, 0.82, 0.95, 0.78, 1.00
- Table 8 selected simulated regional aggregate changes (exact values preserved):
  - Terms-of-trade:
    - USA: 0.29; Canada: 0.13; China: 0.03; India: -0.24; MERCOSUR: -0.42; South America: 0.12; Bangladesh: -0.08; Sri Lanka: 0.24
  - Aggregate export price index:
    - USA: -1.26; Canada: -1.27; China: -1.55; India: -1.79; MERCOSUR: -2.00; South America: -1.34; Bangladesh: -1.66; Sri Lanka: -1.35
  - Aggregate import price index:
    - USA: -1.54; Canada: -1.39; China: -1.58; India: -1.55; MERCOSUR: -1.57; South America: -1.46; Bangladesh: -1.59; Sri Lanka: -1.58
  - Real value of exports:
    - USA: 0.53; Canada: 1.75; China: 1.36; India: 2.32; MERCOSUR: 3.11; South America: 1.58; Bangladesh: 2.56; Sri Lanka: 0.84
  - Real value of imports:
    - USA: 4.17; Canada: 2.53; China: 0.15; India: -0.58; MERCOSUR: -0.86; South America: 1.85; Bangladesh: 0.13; Sri Lanka: 0.32
  - Regional Household Income:
    - USA: 4.01; Canada: 2.50; China: 0.15; India: 0.07; MERCOSUR: 0.22; South America: 1.91; Bangladesh: 0.12; Sri Lanka: 0.31
  - Change in trade balance (in US $ million):
    - USA: -41848.0; Canada: -1317.3; China: +2572.9; India: +1272.8; MERCOSUR: +4066.7; South America: -352.7; Bangladesh: +150.1; Sri Lanka: +43.23

### IX. Welfare, trade balances, and heterogeneity in TFP spillovers
- Welfare (EV) examples (exact values preserved from reported rows):
  - USA: 293090.9 (Joint HAS), 295569.9 (Indo-China FTA)
  - China: 5702.7 (Joint HAS), 2322.3 (Indo-China FTA)
  - India: 773.62 (Joint HAS), 721.39 (Indo-China FTA)
  - South America: 13381.02 (Joint HAS), 14032.1 (Indo-China FTA)
  - Canada: 13111.5 (Joint HAS), 14199.4 (Indo-China FTA)
- Contribution of TFP to EV (in million US$) (exact examples preserved):
  - USA: 285401.7 (Joint HAS), 292745.3 (Indo-China FTA)
  - China: 1000.1 (Joint HAS), 1246.1 (Indo-China FTA)
  - India: 503.58 (Joint HAS), 710.30 (Indo-China FTA)
  - South America: 12267.8 (Joint HAS), 12158.7 (Indo-China FTA)
  - Canada: 11923.47 (Joint HAS), 11926.1 (Indo-China FTA)
- Key heterogeneity findings:
  - TFP-induced spillover growth is of lower order for India, China, Bangladesh, and Sri Lanka compared with NAFTA, EUFTA, and Latin American regions due to lower trade-embodiment indexes and lower capture-parameter values.
  - The ordering of spillover coefficients matches the ordering of real GDP responses in the simulations.

### X. Income distribution, Gini outcomes, and sectoral impacts
- Gini coefficient responses (Table 9 exact values preserved):
  - BGD: Initial 0.3451 → Converted Eq.20 0.345 → Converted Eq.22 0.1724
  - CHN: Initial 0.447 → Converted Eq.20 0.446 → Converted Eq.22 0.2234
  - EUFTA: Initial 0.3124 → 0.3119 → 0.1557
  - HTW: Initial 0.3651 → 0.364 → 0.1818
  - India: Initial 0.325 → 0.324 → 0.1624
  - OSA: Initial 0.3406 → 0.3404 → 0.1701
  - SAM: Initial 0.502 → 0.5001 → 0.2491
  - SEA: Initial 0.4043 → 0.403 → 0.2018
  - SLK: Initial 0.4171 → 0.417 → 0.2084
  - CAN: Initial 0.331 → 0.329 → 0.1638
  - JPN: Initial 0.359 → 0.358 → 0.1791
  - MERCOSUR: Initial 0.5317 → 0.5314 → 0.2655
  - US: Initial 0.408 → 0.405 → 0.2007
- Interpretation:
  - Both PROGRESS specifications (Eq.20 inverse logistic and Eq.22 multiplicative exponential) produce falls in ex post Gini relative to initial values; the exponential specification yields sharper falls.
  - Regions with higher socio-institutional congruence (and higher θ) show larger declines in Gini.
- Sectoral spillover, TFP growth and output (Table 11 selected entries preserved exactly):
  - United States:
    - LMnfcs: spillover coeff 0.91 → Sectoral TFP Growth 3.65 → Sectoral Output 3.45
    - HMnfcs: 0.91 → 3.63 → 2.92
    - HiTech: 0.88 → 4.00 → 2.48
  - CAN:
    - LMnfcs: 0.63 → 2.48 → 2.44
    - HMnfcs: 0.59 → 2.34 → 2.23
    - HiTech: 0.66 → 2.62 → 2.15
  - China:
    - LMnfcs: 0.02 → 0.10 → 0.95
    - HMnfcs: 0.03 → 0.13 → 0.39
    - HiTech: 0.04 → 0.22 → 0.25
  - India:
    - LMnfcs: 0.02 → 0.12 → 0.82
    - HMnfcs: 0.02 → 0.12 → 0.13
    - HiTech: 0.05 → 0.32 → 0.11
  - South America:
    - LMnfcs: 0.47 → 1.85 → 2.05
    - HMnfcs: 0.48 → 1.90 → 1.74
    - HiTech: 0.50 → 1.99 → 2.15
- Interpretation:
  - Higher capture parameters and embodiment indexes in the USA, Canada, EU, and South America lead to larger sectoral TFP responses and output changes.
  - China and India register much lower sectoral TFP growth due to low θr magnitudes.

### XI. Trade-policy configurations: HAS and Indo-China FTA results
- HAS scenario: USA liberalizes simultaneously with India and China (Hub-and-Spokes); subsequent Indo-China FTA between spokes evaluated.
- Selected comparative results from Table 10 (exact values preserved):
  - Terms-of-trade (examples):
    - USA: 0.64 (Joint HAS), 0.19 (Indo-China FTA)
    - China: 0.92 (Joint HAS), 0.24 (Indo-China FTA)
    - India: 0.07 (Joint HAS), -0.27 (Indo-China FTA)
  - Regional Household Income (examples):
    - USA: 4.06 (Joint HAS), 3.97 (Indo-China FTA)
    - China: 0.74 (Joint HAS), 0.30 (Indo-China FTA)
    - India: 0.21 (Joint HAS), 0.20 (Indo-China FTA)
  - Change in trade balance (US $ million) (examples):
    - USA: -44010.22 (Joint HAS), -38978.11 (Indo-China FTA)
    - China: -374.9 (Joint HAS), 2370.9 (Indo-China FTA)
    - India: 581.33 (Joint HAS), 882.83 (Indo-China FTA)
- Observed policy-relevant outcomes:
  - Welfare increases are driven predominantly by technical change.
  - Indo-China FTA produces slightly larger falls in income inequality relative to HAS due to indirect spillover under full-fledged liberalization between India and China.
  - Export diversion between spokes is present but limited; productivity and spillovers still improve welfare.

### XII. Concluding synthesis (paper’s main findings)
- The augmented global CGE simulations indicate:
  - Considerable welfare gains in recipient regions receiving technological spillovers embodied in trade flows.
  - Reduction in income inequality is enabled by higher technology appropriation—driven by social acceptance, institutional congruence, governance, human capital, and human development—leading to productivity growth and inequality convergence.
  - Trade openness raises GDP (coefficient 1.035, R2 =0.91 in the reported regression) and, in the presence of sufficient absorptive capacity and socio-institutional support, can reduce inequality and poverty.
  - Policy implication: Reducing trade barriers is necessary but not sufficient; trade must be embedded in broader development strategies including macroeconomic policy, infrastructure, education, health, governance, R&D, and social acceptance for trade- and technology-induced gains to be pro-poor.

*Source: _wp0716 - References (author's calculations using GTAP database, World Bank WDI data as described in the source content).*

### References..............................................................................................................

### _wp0716 - References..............................................................................................................

### I. ANALYZING TRADE-INCOME INEQUALITY LINKAGES: LACUNAE IN THE LITERATURE
- Research question: Is there any cause, sui generis, for trade per se reducing inequality across nations?
- Core assertions and conceptual findings:
  - Long-term sustained productivity growth and distributional outcomes are contingent on human capital formation, structural reforms, trade reforms, and sound macroeconomic policies (Global Economic Prospects 2004, p.239).
  - Empirical literature often finds that trade helps nations move to a higher growth path and hence affects income distribution (Grossman and Helpman 1991; Dollar and Kraay 2002; Frankel and Romer 1999; Richardson 1995; Helpman 2002).
  - Trade liberalization tends to alleviate poverty over the medium to long term (Winters, McCulloch, and McKay 2004), but the outcome depends on distributional changes: if growth worsens distribution, poverty can increase (Deininger and Squire 1996; Bourguignon 2004).
  - For growth to be pro-poor, the poor need access to modern science and technology, and an adequate socio-institutional set-up including human capital, governance, education, and infrastructure (Bourguignon 2004; Sen 2004).
  - Technology and knowledge diffusion (ICT, bio-technology, nanotechnology) are important drivers of productivity and growth and thus poverty reduction (Millennium Project Report 2005; Bussolo and O’Connor 2002).
  - Trade is an important conduit for technology spillovers from richer partners to recipients (Keller 2004; Coe and Helpman 1995; Bayoumi, Coe, and Helpman 1999).
- Analytical contribution of the paper:
  - Offers a model that heuristically links exogenous technological change, adoption, productivity improvements, and their effects on income inequality, poverty, and welfare.
  - Uses an augmented GTAP global CGE model (78 regions × 57 sectors) to study multi-sectoral, multi-regional linkages and international spillovers.
  - Formal objectives:
    - (a) Estimate the impact of technical changes and trans-border propagation on income inequality under different trade policy configurations.
    - (b) Assess the role of human capital and socio-institutional features in reducing income inequality.
- Policy-relevant synthesis:
  - Trade reform alone is insufficient; complementary socio-institutional factors (human capital, governance, R&D, social acceptance) and absorptive capacity are critical to capture trade- and technology-induced benefits and to reduce inequality and poverty.
  - Failure to establish appropriate national institutional and policy frameworks hinders realization of trade-induced benefits (Gunter, Cohen, and Lofgren 2005).

### II. TRADE, TECHNOLOGY AND INCOME INEQUALITY: BACKGROUND QUANTITATIVE EVIDENCE
A. Incidence of Income Inequality Across Skill Cohorts: Macro Lens
- Data sources and measures:
  - GTAP global database Version 5.4 with income split between two labor types: skilled and unskilled (Dimaranan and McDougall 2002 and 2003).
  - National Gini coefficients from World Bank's World Development Indicators (2003, 2005).
  - Proportion of skilled personnel (Ps) proxied by tertiary enrolment ratio (percentage of population age 15-64 having tertiary education or more) from World Bank WDI.
  - GTAP skilled-unskilled income shares based on average years of education (Barro and Lee 1996).
- Methodology:
  - Adopt Yitzhaki (2002) Gini decomposition adapted to compute inter-skill Gini indexes for skilled (s) and unskilled (u) cohorts:
    - Gyo = P s S ys G ys + P u S yu G yu + G between     (Equation (1))
    - G between = P u − S yu                                      (Equation (2))
    - Assumption: G between = |G ys − G yu |                     (Equation (3))
    - Algebraic result used to compute G ys (Equation (4) in text).
- Empirical pattern (from computed results reported in Table 1):
  - "From Table 1, it is evident that across almost all regions, compared to the Gini index for the skilled labor force, the computed Gini coefficient is lower for the unskilled labor category."
  - The analysis relies on GTAP skill composition and Yitzhaki’s method applied to the GTAP global database.
  - Note: alternative proxies for Ppoor were explored (percentage below national poverty line, 50% of median income) but GTAP’s skill decomposition based on educational attainment was adopted for consistency.

B. Interoccupational Inequality and Trade Openness: Empirical Support
- Trade orientation measure (OPEN): regional (Export (X) plus Import (M)) normalized by regional GDP, i.e., (X+M)/GDP. Regions indexed by r.
- Regression linking trade openness to GDP (log-linear OLS across 78 GTAP regions):
  - ln GDP (r) = α + β ln OPEN (r) + ε(r)             (Equation (5))
  - Estimated equation reported:
    - ln GDP (r) = - 0.0605  + 1.035 ln OPEN (r), R2 =0.91, S.E. = 0.55      (5a)
    - t-values shown: (-0.155) for intercept, (29.04) for slope (as printed).
  - Interpretation: GDP growth is positively related to trade openness and statistically significant. Pathways cited: imported inputs embodying advanced technology, learning effects, and efficiency gains.
- Regression linking trade openness to Gini coefficients (elasticities estimated separately for overall, skilled, and unskilled Gini):
  - ln Gyi (r) = γ + δ ln OPEN(r) + μ (r)      with i ∈{o, s, u}      (Equation (6))
  - Estimated equations:
    - ln Gyo (r) = 4.128 +  (-) 0.121   ln OPEN(r), R2 =0.08, S.E. =0.234      (6a)
      - t-values: (18.20) for intercept, (-2.29) for slope (as printed).
    - ln Gys (r) = 4.437 +  (-) 0.165   ln OPEN(r), R2 =0.05, S.E. =0.426      (6b)
      - t-values: (10.77) for intercept, (-1.73) for slope (as printed).
    - ln Gyu (r) = 4.726+  (-) 0.349   ln OPEN(r), R2 =0.08, S.E. =0.643      (6c)
      - t-values: (6.923) for intercept, (-2.190) for slope (as printed).
  - Empirical findings:
    - Estimated slopes δ are negative for overall, skilled, and unskilled Gini regressions, indicating an inverse relationship between trade openness and Gini measures in these specifications.
    - R2 values are low (0.05–0.08) indicating limited explanatory power of ln OPEN alone for Gini variation, but slope coefficients are reported as statistically significant in some cases (t-statistics shown).
- Visualization:
  - Figures 1–3 (Line Fit Plots) illustrate fitted relationships for the overall, skilled, and unskilled Gini regressions versus log Trade/GDP ratio.

### Key empirical takeaways and policy-relevant points
- Trade openness is strongly positively associated with GDP across GTAP regions: coefficient 1.035 in (5a) with R2 =0.91, S.E. = 0.55.
- In the log-linear regressions reported:
  - ln Gyo: slope = (-) 0.121, R2 =0.08, S.E. =0.234 (6a).
  - ln Gys: slope = (-) 0.165, R2 =0.05, S.E. =0.426 (6b).
  - ln Gyu: slope = (-) 0.349, R2 =0.08, S.E. =0.643 (6c).
- Computed inter-skill Gini results (Table 1) indicate that in almost all regions the computed Gini coefficient is lower for the unskilled labor category than for the skilled labor category.
- Policy implication emphasized throughout:
  - Reducing trade barriers is necessary but not sufficient; trade must be embedded in a broader development strategy that includes macroeconomic policy, infrastructure, education, health, governance, R&D, and social acceptance to ensure technology spillovers and growth are pro-poor.
  - Socio-institutional and absorptive capacity factors determine whether trade- and technology-induced gains reduce inequality and poverty.

*Source: _wp0716 - References (author's calculations using GTAP database, World Bank WDI data as described in the source content).*

### Chapter 2: The Debate over Trade Liberalization.  Trade Liberalization and Poverty: A Handbook, CEPR.

### Chapter 2: The Debate over Trade Liberalization

### A. Empirical findings on trade and income inequality
- Regression results indicate a statistically significant inverse relationship between trade openness and income inequality for overall, skilled, and unskilled categories.
- Point estimates: a 1 percent increase in a region’s trade openness index (trade to GDP ratio) tends to lower the relevant inequality indexes by:
  - 0.12% for the skilled group
  - 0.17% for the unskilled group
  - 0.35% overall
- Interpretation:
  - Trade lowers inequality primarily via a growth effect: trade raises income levels and subsequent growth, which compresses within-group income dispersion.
  - The elasticity of the Gini coefficient is lower for the skilled group than for the unskilled group. Because within-group income inequality is initially higher among skilled workers in GTAP regions, trade-induced growth reduces inequality faster among the relatively more egalitarian unskilled groups.
  - Trade can reduce or at least not increase wage inequality between skilled and unskilled labor; Lawrence and Slaughter (1993) find trade has a dampening effect on wage inequality.

### B. Growth in trade: global, regional and sectoral patterns (GTAP 1965–1998)
- Sectoral average annual growth rates for global trade in manufactures (Table 2):
  - Light manufactures: 13
  - Heavy manufactures: 12
  - Hi-technology cluster: 13
  - (Source: Author's calculations based on GTAP database, Version 6.)
- Regional average annual growth rates in light, heavy, and hi-technology manufactures (Table 3):
  - Canada: Light 0.09; Heavy 0.12; Hi-technology 0.12
  - China: Light 0.21; Heavy 0.17; Hi-technology 0.25
  - EU (FTA): Light 0.11; Heavy 0.10; Hi-technology 0.11
  - Hong-Kong SAR: Light 0.14; Heavy 0.17; Hi-technology 0.19
  - India: Light 0.17; Heavy 0.16; Hi-technology 0.17
  - Japan: Light 0.07; Heavy 0.12; Hi-technology 0.15
  - MERCOSUR: Light 0.19; Heavy 0.16; Hi-technology 0.16
  - South America: Light 0.15; Heavy 0.10; Hi-technology 0.24
  - South East Asia: Light 0.18; Heavy 0.14; Hi-technology 0.29
  - United States: Light 0.10; Heavy 0.10; Hi-technology 0.11
  - (Source: Author's calculations based on GTAP database, Version 6.)
- Observations:
  - Hi-technology registered the highest growth rate in most regions (except MERCOSUR).
  - Heavy manufactures grew faster than light manufactures.
  - China, Japan, India and South East Asia show much higher average annual growth rates than the United States and Canada.
  - Compositional change toward skill-intensive products implies potential impacts on returns to skilled versus unskilled labor; however, if trade-led growth does not worsen intra-group inequality, average income rises will not necessarily accentuate wage inequality and poverty.
- Regional growth and poverty reduction example:
  - East Asian nations: GDP growth was 6.7% in 2002.
  - The number of people living below $2 a day fell from 50% to 34% between 1999 and 2002 (The World Bank, November 2004).

### C. Technology, institutions, absorption capacity and inequality
- Key concepts and indices:
  - Absorption capacity (AC): depends on education and schooling years (Barro and Lee 1996; Cohen and Levinthal 1989, 1990; Pack and Westphal 1984; Nelson 1990). AC is region-specific with 0 ≤ AC ≤ 1.
  - Governance parameter (GP): bilateral GP_rs = min[1, GP_s / GP_r] with 0 ≤ GP_rs ≤ 1 (equation (8)). GP_rs = 1 if destination’s governance exceeds source’s.
  - Technological congruence (TC): TC_rs = min[1, TC_s / TC_r] with TC_rs ∈ [0,1] (equation (9)). TC_rs = 1 implies maximal proximity to the invention frontier.
  - Social acceptance (SA): SA_s = min[1, SA_s / SA_threshold] with SA_s ∈ [0,1] (equation (10)). Captures human development threshold effects on technology adoption.
  - Institutional-structural congruence (SC): SC_rs = GP_rs · TC_rs (equation (13)).
  - Technology appropriation / capture parameter (θ_s or TAP): θ_s = AC_s · SA_s · SC_rs (equation (14)); θ_s ∈ [0,1], with θ_s = 1 implying full appropriation.
- Technology Achievement Index (TAI) values for selected GTAP regions (Table 4, Human Development Report 2001):
  - Bangladesh: 0.04
  - Canada: 0.59
  - China: 0.30
  - EU: 0.51
  - HongKong SAR–Taiwan Province of China: 0.46
  - India: 0.20
  - Japan: 0.70
  - MERCOSUR: 0.34
  - Other South Asia: 0.12
  - Rest-of-the-World: 0.25
  - South America: 0.32
  - South East Asia: 0.36
  - Sri Lanka: 0.20
  - United States: 0.73
  - (Source: Human Development Report 2001, Table A2.1. Composite regions use group averages.)
- Theory and mechanism:
  - Trade-mediated technology transfer occurs via intermediates; embodiment indexes E_irjs = F_irjs / M_ijs (equation (7)).
  - Spillover coefficient γ_s depends on embodied trade intensity E_rs and capture θ_s:
    - γ_s has properties γ_s(0) = 0, γ_s(1) = 1, γ'_s > 0, γ''_s < 0 (see equations (11) and related text).
    - Stylized functional form: γ_s = 1 − (E_rs)^(1−θ_s) (equation (15)) for specified parameter ranges.
  - TFP transmission: ava(j, s) = (E_sijrs)^(θ_s − 1) · ava(i, r) for i ≠ j, r ≠ s (equation (18)).

### D. A stylized model linking capture, spillovers, and inequality (PROGRESS)
- Conceptual mapping:
  - Higher capture parameter θ (function of AC, SA, GP, TC) increases realized spillovers γ and thus reduces income inequality (Gini index G).
  - When θ = 1 (perfect capture), inequality can vanish (G → 0); when θ = 0 (no capture), inequality can be extreme (G → 1).
- Functional specification:
  - Transformed Gini: G_convert = F(G_initial, θ) (equation (19)), with F(G_initial,0) → 1 and F(G_initial,1) → 0.
  - Inverse logistic specification (equation (20)):
    - G_convert(s) = [1 + exp(−g(s) · γ(s) · η(s) · Y_initial(s))]^(−1) · [1 − G_initial(s)] + G_initial(s) ??? (the chapter provides a stylized inverse logistic relation in equation (20) with the PROGRESS scalar; preserve the role of parameters rather than inventing exact collapsed form)
  - Definitions:
    - g = average proportional rate of decrease of Gini coefficient due to growth in trade dependency ratio (trade as % of GDP). Empirical value taken as 0.2%.
    - γ(s) = regional spillover coefficient parameter.
    - η(s) = elasticity of poverty with respect to trade liberalization. Implementations set η(r) uniformly at 2.0.
    - Y_initial(s) = initial poverty threshold income level (relative), normalized around unity.
  - PROGRESS = g × γ × η (the scalar product is called the ‘Poverty Reduction on GRowth and Embodied Spillover Synergy’ parameter).
- Comparative statics:
  - ∂G_convert / ∂θ < 0: higher θ reduces converted Gini.
  - With γ → ∞, G_convert → 0 (inequality vanishes asymptotically).
  - With γ → 0 and η ∈ [2, 2.5], G_convert → G_initial (no improvement).
  - The multiplicative exponential variant (equation (22)) produces sensitivity to η: with high η (> 2) converted Gini can “explode” numerically in that specification.
- Poverty gap linkage:
  - Poverty gap PG_r related to converted Gini and mean income μ(r) (μ(r) = mean absolute income as ratio to $1.08/day poverty line):
    - PG_r = G_convert(r) − μ(r) × ln(θ_r) (equation (23))
  - For a recipient s with unique source r:
    - PG(s) = G_convert(r) − μ(s) × ln(θ_rs) (equation (24))
  - Comparative statics: ∂PG_r / ∂θ < 0 for θ, μ > 0: higher capture reduces the poverty gap.

### E. Model implementation framework and data
- Modeling approach:
  - A modified multi-regional, multi-sectoral computable general equilibrium (CGE) model based on GTAP is used (GTAP Version 5.4 / Version 6 data referenced).
  - The GTAP database details bilateral trade flows, production, consumption and intermediate flows; Version 5.4 disaggregates the world into 78 regions and 57 sectors. A reduced 14×12 aggregation is used for calibration.
  - GEMPACK software suite is used to conduct simulations (Harrison and Pearson 1994, 1996).
- Notes on interpretation:
  - The model emphasizes trade as a vehicle for embodied technology transfers via intermediates; role of FDI is acknowledged in literature but the implementation focuses on trade flows in the medium-run.
  - Human Development Index (HDI) or related indices are used as proxies for social acceptance (SA) and for broader capability effects in poverty reduction modeling.

*Source: Chapter 2: The Debate over Trade Liberalization. Trade Liberalization and Poverty: A Handbook, CEPR. (excerpt from _wp0716 - Chapter 2: The Debate over Trade Liberalization. Trade Liberalization and Poverty: A Handbook, CEPR.)*

### 1. Pdr [Paddy, Rice] 1.USA [United States]

### 1. Pdr [Paddy, Rice] 1.USA [United States]

### Regions and Sector Aggregation
- Regions (author’s aggregation based on GTAP V5.4 database; sectoral composition not reported for parsimony):
  - 1. USA (United States)
  - 2. CAN (Canada)
  - 3. Ind (India)
  - 4. Chn (China)
  - 5. Jpn (Japan)
  - 6. Sam (Latin American Countries)
  - 7. MERCOSUR (Argentina, Brazil, Uruguay, Paraguay)
  - 8. Sea (South East Asia)
  - 9. Osa (Other South Asia)
  - 10. Slk (Sri Lanka)
  - 11. HTW (Hong Kong and Taiwan)
  - 12. Bgd (Bangladesh)
  - 13. EUFTA (European Union)
  - 14. Rest of the World [ROW]

### Parameter Settings — Overview
- Augmented theoretical model incorporates four added parameter-sets beyond standard GTAP:
  - Skill-induced AC index (absorption capacity parameter)
  - Governance parameter GP
  - Social acceptance parameter SA
  - Technological proximity parameter TC
- AC calculation:
  - Regional ratios of skilled to unskilled labor payment shares proxy AC.
  - Calculated ordering of AC-values: ACUSA > ACHTW > ACEUFTA > ACJAPAN > ACMERCOSUR > ACSAM > ACCAN > ACCHINA > ACINDIA > ACSOUTHASIA.
  - Composite regions (HTW, Sea, Sam) register high AC figures relative to many developing regions.
- GP (governance parameter) construction:
  - Uses World Bank six-dimensional governance indicators (Kaufmann et al. (2003); Kaufmann (2004)): Voice and accountability, Political stability, Government effectiveness, Regulatory quality, Rule of law, Control of corruption.
  - Composite governance indicator = simple arithmetic average of six components, bounded between -2.5 and +2.5 in original data; transformed to binary indexes via Equation (8) and scaled relative to USA (benchmark).
  - Interpreted values:
    - USA, EU, Japan and Canada are more institutionally homogeneous.
    - MERCOSUR and other Latin American bloc show quite low governance indicator.
    - India: GP magnitude 0.65
    - Sri Lanka: GP magnitude 0.69
    - China: GP binary parameter 0.32
    - Hong Kong-Taiwan (HTW): GP binary parameter 0.74
- TC (technological proximity) proxies:
  - Use Technology Achievement Index (TAI) or R&D expenditure as percentage of GDP; TAI values from Human Development Report (2001) and R&D from Human Development Report (2003).
  - For missing TAI, assign shares of hi- and medium-technology exports as indicators.
  - USA, Canada, EU and Japan have binary TC values ≈ unity.
- SA (social acceptance) proxy:
  - Human Development Index (Human Development Report, 2003).
  - Threshold: East and South East Asia composite human development index.
  - Perfect acceptance level assigned for USA, Canada, Hong Kong, Japan and EU.
- Model solution:
  - Solved using customized Windows program Gempack (based on GEMPACK software suite).

### Simulation Design and Policy Experiments
- Focus regions of interest: India and China (contain more than 33% of world population: "totally 2.4 billion out of global 6.2 billion").
- Poverty headcount references (World Bank / Cline, 2004):
  - China: 53.7% (headcount fraction of population in poverty, income less than $2 per day)
  - India: 86.2% (same measure)
  - Cline (2004) estimated poverty rates: India 48.4% and China 39.2%
- Two generic shock types studied:
  1. Technology shocks: TFP augmentation in a high-technology sector in USA transmitted to other nations.
  2. Trade policy shocks: trade liberalization under Hub-and-Spokes (HAS) FTA configurations between USA vis-à-vis India and China.
- Hub-and-spokes scenario rationale:
  - Consideration of USA as hub with India and China as major spokes, examining poverty-reducing potential of trade linkages.

### Technology Spillover and Inequality Scenario: Pure Productivity Shock
- Source/assumptions for TFP shock:
  - Hi-tech sector in USA is source of innovation.
  - Keller (1999) reports average annual multifactor productivity growth in composite hi-tech sector = 3.2% during 1970-1991.
  - Authors extrapolate growth rates linearly over 6 years; use extrapolated growth rate of 4 (approximation to 3.86) per cent per annum as the TFP shock.
  - Implemented as a 4% shock to the Hicks-Neutral technological coefficient in the hi-tech sector of the USA.
  - Closure: standard GTAP macroeconomic closure (see Hertel 1997; McDougall 2003, GTAP version 6.2).
- Focused sectors in analysis: hi-tech, heavy manufacturing, light manufacturing for regions US, Canada, China, India, other Asian countries and MERCOSUR.

### Results — Macroeconomic Repercussions of 4% Hi-tech TFP Shock in the US
- Table 6: Simulated Regional Effects of 4% TFP Shock in the Hi-tech Sector in the US on Selected Macroeconomic Variables (percent changes)
  - Region-wide index of TFP growth:
    - US: 3.6
    - Canada: 2.2
    - EU: 0.76
    - China: 0.11
    - South America: 1.65
    - India: 0.10
    - MERCOSUR: 0.30
    - Bangladesh: 0.12
    - Sri Lanka: 0.16
  - Real GDP at Factor Cost (percent changes):
    - US: 3.6
    - Canada: 2.2
    - EU: 0.76
    - China: 0.11
    - South America: 1.65
    - India: 0.10
    - MERCOSUR: 0.30
    - Bangladesh: 0.12
    - Sri Lanka: 0.16

- Key mechanisms and interpretation:
  - Technology capture depends on economy-wide and sectoral embodiment indexes and spillover coefficients in destination regions relative to USA.
  - USA has higher aggregate embodiment index and higher capture-parameter (θr), hence reaps maximum spillover.
  - Recipients with higher capture parameters, embodiment indexes, and favorable socio-institutional constellations (GP, TC, SA) capture more technology (e.g., Canada, EU, South America).
  - Lower TC and SC values reduce capture (e.g., Bangladesh), but higher AC and trade-embodiment can still facilitate technology-led growth in some developing regions.

- Table 7: Values of Economy-Wide Embodiment-Indexes, Spillover Coefficients, Parameters for Governance, Social Acceptance, Capture and Congruence (base-period pre-shock levels)
  - Columns correspond to: (1) Embodiment Index (Eirs/Eir), (2) Spillover Coefficient (γirs/γir), (3) Capture-Parameter (θr), (4) ACs, (5) GPrs, (6) TCr s, (7) SCrs = TCr s × GPrs, (8) SAs
  - Selected region entries (values preserved exactly):
    - USA: 0.62, 0.82, 0.71, 0.71, 1.00, 1.00, 1.00, 1.00
    - Canada: 0.38, 0.51, 0.32, 0.40, 1.00, 0.80, 0.80, 1.00
    - China: 0.02, 0.03, 0.03, 0.23, 0.32, 0.41, 0.13, 0.95
    - India: 0.02, 0.03, 0.03, 0.21, 0.65, 0.27, 0.18, 0.74
    - MERCOSUR: 0.05, 0.06, 0.08, 0.44, 0.46, 0.46, 0.21, 0.90
    - South America: 0.38, 0.38, 0.01, 0.40, 0.04, 0.43, 0.02, 0.90
    - Bangladesh: 0.03, 0.02, 0.002, 0.28, 0.22, 0.05, 0.01, 0.62
    - Sri Lanka: 0.03, 0.03, 0.05, 0.27, 0.69, 0.28, 0.19, 0.90
    - EU: 0.04, 0.18, 0.47, 0.67, 1.00, 0.69, 0.69, 1.00
    - HTW: 0.09, 0.20, 0.33, 0.71, 0.74, 0.62, 0.46, 0.99
    - Japan: 0.02, 0.10, 0.48, 0.62, 0.82, 0.95, 0.78, 1.00

- Table 8: Simulated Regional Effects on Aggregate Performance (selected changes)
  - Changes reported (exact values preserved):
    1. Terms-of-trade:
       - USA: 0.29
       - Canada: 0.13
       - China: 0.03
       - India: -0.24
       - MERCOSUR: -0.42
       - South America: 0.12
       - Bangladesh: -0.08
       - Sri Lanka: 0.24
    2. Aggregate export price index:
       - USA: -1.26
       - Canada: -1.27
       - China: -1.55
       - India: -1.79
       - MERCOSUR: -2.00
       - South America: -1.34
       - Bangladesh: -1.66
       - Sri Lanka: -1.35
    3. Aggregate import price index:
       - USA: -1.54
       - Canada: -1.39
       - China: -1.58
       - India: -1.55
       - MERCOSUR: -1.57
       - South America: -1.46
       - Bangladesh: -1.59
       - Sri Lanka: -1.58
    4. Real value of exports:
       - USA: 0.53
       - Canada: 1.75
       - China: 1.36
       - India: 2.32
       - MERCOSUR: 3.11
       - South America: 1.58
       - Bangladesh: 2.56
       - Sri Lanka: 0.84
    5. Real value of imports:
       - USA: 4.17
       - Canada: 2.53
       - China: 0.15
       - India: -0.58
       - MERCOSUR: -0.86
       - South America: 1.85
       - Bangladesh: 0.13
       - Sri Lanka: 0.32
    6. Regional Household Income:
       - USA: 4.01
       - Canada: 2.50
       - China: 0.15
       - India: 0.07
       - MERCOSUR: 0.22
       - South America: 1.91
       - Bangladesh: 0.12
       - Sri Lanka: 0.31
    7. Change in trade balance (in US $ million):
       - USA: -41848.0
       - Canada: -1317.3
       - China: +2572.9
       - India: +1272.8
       - MERCOSUR: +4066.7
       - South America: -352.7
       - Bangladesh: +150.1
       - Sri Lanka: +43.23

### Interpretive Findings from Simulations
- USA, as source of innovation, records the highest overall technological progress and corresponding welfare gains.
- Recipients with higher embodiment indexes, capture parameters, and favorable socio-institutional characteristics (GP, TC, SA) — notably Canada, EU, South America — receive larger TFP improvements.
- India, Sri Lanka, China and other South Asian regions are relatively backward in capture capacity; lower TC and SC values reduce capture. Bangladesh registers particularly low capture-parameter (θr = 0.002) and low SC but may benefit if AC and trade-embodiment are higher.
- Trade flows (e.g., NAFTA linkages) inflate embodiment indexes for regions like Canada and South America, increasing their technology appropriation.
- TFP-neutral shock (Hicks-neutral 4% in US hi-tech) translates approximately into equivalent increases in real GDP at factor cost in regions, though magnitudes vary across regions.

*Source: Simulations and text from the provided IMF working paper excerpt.*

### 8. Welfare (EV) (in million US $)    289291.2    14191.3  1149.5    244.68    2226.4 14169.8  48.5 42.5

### 8. Welfare (EV) (in million US $)    289291.2    14191.3  1149.5    244.68    2226.4 14169.8  48.5 42.5

### Welfare and trade balance outcomes
- Table note: First 6 rows report results in Percentage changes. Reported welfare and trade balance changes in rows 7 and 8 are in levels ($ million).
- Post-simulation aggregate export volumes:
  - Increase in principal beneficiaries of TFP changes.
- Real imports:
  - Increase for all regions except India and MERCOSUR.
- Trade balance changes:
  - Regions for which export prices have fallen more than aggregate import prices (MERCOSUR, India, Bangladesh) register positive changes in their trade balance due to technology transmission and capture.
- Welfare (EV) impacts (examples from Table 10, rows reproduced as reported):
  - USA: 293090.9 (Joint HAS), 295569.9 (Indo-China FTA)
  - China: 5702.7 (Joint HAS), 2322.3 (Indo-China FTA)
  - India: 773.62 (Joint HAS), 721.39 (Indo-China FTA)
  - South America: 13381.02 (Joint HAS), 14032.1 (Indo-China FTA)
  - Canada: 13111.5 (Joint HAS), 14199.4 (Indo-China FTA)
- Contribution of TFP to EV (in million US$) (Table 10, row 7 examples):
  - USA: 285401.7 (Joint HAS), 292745.3 (Indo-China FTA)
  - China: 1000.1 (Joint HAS), 1246.1 (Indo-China FTA)
  - India: 503.58 (Joint HAS), 710.30 (Indo-China FTA)
  - South America: 12267.8 (Joint HAS), 12158.7 (Indo-China FTA)
  - Canada: 11923.47 (Joint HAS), 11926.1 (Indo-China FTA)

### TFP spillovers, capture, and regional heterogeneity
- Relative magnitudes:
  - Growth in TFP induced by spillover is of lower order of magnitude for India, China, Bangladesh, and Sri Lanka compared with NAFTA, EUFTA, and Latin American regions.
- Reasons cited:
  - These South Asian countries "do not have much trade-mediated spillover capture from USA."
  - Lower values of trade-embodiment indexes imply lower spillover coefficients for South Asian nations relative to EU, Canada, and South America.
  - Capture parameter values are relatively lower in India, China, Bangladesh, and Sri Lanka compared to Japan, Canada, EU, and Hong Kong and Taiwan (see column 4 reference).
  - Socio-institutional congruence with the US is lower for these regions (Column 8 reference), reducing transmitted benefit capture relative to institutionally more similar regions like Canada or EU.
- Ordering and consistency:
  - The ordering of the spillover coefficient in column 3 of Table 7 matches the ordering of real GDP in row 2 of Table 6.

### Income, welfare distribution, and inequality effects
- Regional income and welfare:
  - Technological spillover causes changes in regional income and welfare and affects factor income distribution.
  - Comparing India and China: regional income and welfare changes are much higher for China.
  - Within the South Asian bloc: India is ahead of Sri Lanka and Bangladesh in income and welfare changes (rows 6 and 8 comparisons).
  - Rationale: higher doses of spillover from USA into India relative to neighboring countries.
- Gini coefficients (Table 9):
  - Under both specifications (via Equations 20 and 22), ex post Gini index values fall relative to initial values.
  - Exponential specification of the 'PROGRESS' parameters produces sharper falls in Gini values.
  - Regions with higher socio-institutional congruence with the US (and higher technology appropriation) are capable of reducing income inequality more (compare Column 1 with Column 3).
- Reported Gini values (Table 9 reproduced exactly):
  - BGD: Initial Gini 0.3451 → Converted via Equation 20 0.345 → Converted via Equation 22 0.1724
  - CHN: Initial Gini 0.447 → Converted via Equation 20 0.446 → Converted via Equation 22 0.2234
  - EUFTA: Initial Gini 0.3124 → Converted via Equation 20 0.3119 → Converted via Equation 22 0.1557
  - HTW: Initial Gini 0.3651 → Converted via Equation 20 0.364 → Converted via Equation 22 0.1818
  - India: Initial Gini 0.325 → Converted via Equation 20 0.324 → Converted via Equation 22 0.1624
  - OSA: Initial Gini 0.3406 → Converted via Equation 20 0.3404 → Converted via Equation 22 0.1701
  - SAM: Initial Gini 0.502 → Converted via Equation 20 0.5001 → Converted via Equation 22 0.2491
  - SEA: Initial Gini 0.4043 → Converted via Equation 20 0.403 → Converted via Equation 22 0.2018
  - SLK: Initial Gini 0.4171 → Converted via Equation 20 0.417 → Converted via Equation 22 0.2084
  - CAN: Initial Gini 0.331 → Converted via Equation 20 0.329 → Converted via Equation 22 0.1638
  - JPN: Initial Gini 0.359 → Converted via Equation 20 0.358 → Converted via Equation 22 0.1791
  - MERCOSUR: Initial Gini 0.5317 → Converted via Equation 20 0.5314 → Converted via Equation 22 0.2655
  - US: Initial Gini 0.408 → Converted via Equation 20 0.405 → Converted via Equation 22 0.2007
- Interpretation:
  - Higher TAP parameter magnitudes (via Equation 14) enable reductions in income inequality and income poverty through conducive growth climates and enabling factors such as good governance, technological achievement, and human development.
  - When socio-institutional parameters are optimal and spillover coefficient is higher, Gini measures deflate (via Equation 22).

### Trade-policy configurations and scenario results (HAS and Indo-China FTA)
- Scenario definitions:
  - HAS configuration: USA simultaneously liberalizes trade with spokes India and China (non-sequential), establishing China-U.S. FTA and India-U.S. FTA simultaneously.
  - Subsequent simulation: intra-spoke liberalization between India and China (Indo-China FTA) following HAS.
  - Policy experiment assumptions: immediate, complete, and preferential removal of relevant tariffs and quantifiable non-tariff barriers (no phasing-in, no excluded sectors, no partial liberalization, no liberalization with non-members).
- Effects observed under combined TFP shock plus joint HAS (Table 10 highlights):
  - Terms-of-trade movements preserve ranking and order of magnitude except China and USA, which register relatively higher TOT improvement due to preferential market access and trade rises.
  - Welfare increases considerably driven predominantly by technical change (rows 6 and 7).
  - China and India register improved trade balances due to trade creation in the HAS phase.
- Indo-China FTA (intra-spoke liberalization) outcomes:
  - Terms-of-Trade (TOT) falls in India in the Indo-China FTA scenario while other regions maintain same sign.
  - Export diversion between the two spokes occurs but is not substantial; technology transfer still improves welfare.
  - Productivity mechanism and induced spillover reduce income inequality slightly; values are almost the same as in Table 9.
  - Relative to HAS, the Indo-China FTA yields a bit more fall in income inequality due to indirect spillover under full-fledged liberalization between India and China.
- Comparative numerical examples from Table 10 (Changes in percentage and levels reproduced as reported):
  - Terms-of-trade (row 1):
    - USA: 0.64 (Joint HAS, column 1), 0.19 (Indo-China FTA, column 2)
    - China: 0.92 (Joint HAS), 0.24 (Indo-China FTA)
    - India: 0.07 (Joint HAS), -0.27 (Indo-China FTA)
    - South America: -0.16 (Joint HAS), 0.14 (Indo-China FTA)
    - Canada: -0.32 (Joint HAS), 0.15 (Indo-China FTA)
  - Aggregate export price index (row 2) and import price index (row 3) values are reported in Table 10 exactly as presented.
  - Regional Household Income (row 4) examples:
    - USA: 4.06 (Joint HAS), 3.97 (Indo-China FTA)
    - China: 0.74 (Joint HAS), 0.30 (Indo-China FTA)
    - India: 0.21 (Joint HAS), 0.20 (Indo-China FTA)
    - South America: 1.81 (Joint HAS), 1.89 (Indo-China FTA)
    - Canada: 2.31 (Joint HAS), 2.48 (Indo-China FTA)
  - Change in trade balance (in million US$) (row 5) examples:
    - USA: -44010.22 (Joint HAS), -38978.11 (Indo-China FTA)
    - China: -374.9 (Joint HAS), 2370.9 (Indo-China FTA)
    - India: 581.33 (Joint HAS), 882.83 (Indo-China FTA)
    - South America: 179.36 (Joint HAS), -327.09 (Indo-China FTA)
    - Canada: -1011.20 (Joint HAS), -1246.50 (Indo-China FTA)

### Sectoral impacts and spillover coefficients
- Sectoral spillover and TFP growth (Table 11 excerpts; columns 3–5 reproduced as reported):
  - United States (spillover coefficient base period):
    - LMnfcs: 0.91 → Sectoral TFP Growth 3.65 → Sectoral Output 3.45
    - HMnfcs: 0.91 → Sectoral TFP Growth 3.63 → Sectoral Output 2.92
    - HiTech: 0.88 → Sectoral TFP Growth 4.00 → Sectoral Output 2.48
  - CAN:
    - LMnfcs: 0.63 → Sectoral TFP Growth 2.48 → Sectoral Output 2.44
    - HMnfcs: 0.59 → Sectoral TFP Growth 2.34 → Sectoral Output 2.23
    - HiTech: 0.66 → Sectoral TFP Growth 2.62 → Sectoral Output 2.15
  - China:
    - LMnfcs: 0.02 → Sectoral TFP Growth 0.10 → Sectoral Output 0.95
    - HMnfcs: 0.03 → Sectoral TFP Growth 0.13 → Sectoral Output 0.39
    - HiTech: 0.04 → Sectoral TFP Growth 0.22 → Sectoral Output 0.25
  - India:
    - LMnfcs: 0.02 → Sectoral TFP Growth 0.12 → Sectoral Output 0.82
    - HMnfcs: 0.02 → Sectoral TFP Growth 0.12 → Sectoral Output 0.13
    - HiTech: 0.05 → Sectoral TFP Growth 0.32 → Sectoral Output 0.11
  - South America:
    - LMnfcs: 0.47 → Sectoral TFP Growth 1.85 → Sectoral Output 2.05
    - HMnfcs: 0.48 → Sectoral TFP Growth 1.90 → Sectoral Output 1.74
    - HiTech: 0.50 → Sectoral TFP Growth 1.99 → Sectoral Output 2.15
  - MERCOSUR:
    - LMnfcs: 0.04 → Sectoral TFP Growth 0.16 → Sectoral Output 0.68
    - HMnfcs: 0.08 → Sectoral TFP Growth 0.32 → Sectoral Output 0.64
    - HiTech: 0.14 → Sectoral TFP Growth 0.56 → Sectoral Output 0.43
- Interpretation:
  - USA shows highest sectoral TFP growth across sectors, magnified by its high capture parameter.
  - Canada, EU, and South America also exhibit notable sectoral TFP responses.
  - China and India, with low θr magnitudes, register very low sectoral TFP growth.
  - TFP improvements raise percentage output across regions; cost savings and declines in supply prices follow from lower composite value-added prices.
  - In full-fledged FTA scenario, India registers larger price falls than China, enabling market access gains at China’s expense.

### Concluding synthesis (as stated)
- The paper simulates an augmented global CGE model to examine distributional consequences of trade-mediated technology spillovers from an innovating country (USA) to developing regions (India, South America, China).
- Main findings summarized:
  - Considerable welfare gains in each recipient region receiving technological spillovers embodied in trade flows.
  - Mechanism of inequality reduction: higher technology appropriation—enabled by social acceptance, institutional congruence, governance, human capital, and human development—leads to productivity growth and inequality convergence.
  - Reduction in income inequality leads to poverty alleviation, but requires enabling factors; without these, growth from higher trade volumes might not reduce inequality.
- Comparisons with prior work:
  - Findings align broadly with Cline (2004) on static and dynamic gains from trade openness.

*Source: Simulation by author.*

### REFERENCES

### _wp0716 - REFERENCES

### Trade, Growth, and Integration
- Amiti, M., and S-J Wei (October 2004). "Fear of Service Outsourcing: Is it Justified?" IMF Working Paper, Research Department, WP/04/186. pp. 1-42.
- Armington, P.A. (1969). A Theory of Demand for Products Distinguished by Place of Production. IMF Staff Papers 16:159-78.
- Bhagwati, J. (1993), Regionalism and Multilateralism: An Overview, J. de Melo and A. Panagariya (eds), New Dimensions in Regional Integration (NY: Cambridge University Press).
- Cline, W. R. (2004). "Trade Policy and Global Poverty." Centre for Global Development, Institute for International Economics. Washington DC, USA.
- Dollar, David and A. Kraay (2002). "Growth is Good for the Poor." Journal of Economic Growth, 7, pp. 195-225.
- ---------- (2003). "Institutions, Trade, and Growth." Journal of Monetary Economics, 50(1), pp. 133-162.
- Eaton, J. and S. Kortum (1996). Trade in Ideas: Patenting and Productivity in the OECD. Journal of International Economics 40: 251-278.
- Frankel, J. A., and D. romer. (1999). "Does Trade Cause Growth?" American Economic Review (June), pp. 379-399.
- Groot, H. L. F. de., G-J Linders, P. Rietvelda, and U. Subramanian (2004). The Institutional Determinants of Bilateral Trade Patterns. Kyklos 57 (February):103-124.
- Meijl, H. van, and F. W. van Tongeren (1998). Trade, Technology Spillovers, and Food Production in China. Weltwirtschaftliches Archiv 134 (3):443-449.
- Milanovic, B., and L. Squire (January 2005). "Does Tariff Liberalization Increase Wage Inequality? Some Empirical Evidence." NBER Working Paper 11046, Massachusetts, Cambridge, USA. Pp. 1-65.
- Winters, A. L., N. McCulloch, and A. McKay (March 2004). "Trade Liberalization and Poverty: The Evidence So Far." Journal of Economic Literature, Volume XLII, Number 1, pp. 72—115.
- McCulloch, N., L. Alan Winters, and X. Cirera (2001). "Trade Liberalization and Poverty: A Handbook." CEPR and Department for International Development., U.K.
- Hertel, T. W., ed. (1997). Global Trade Analysis: Modeling and Applications. Cambridge, MA, Cambridge University Press.
- Francois, J. F. and C. R. Shiells (1994), “ AGE Models of North American Free Trade’ in J. F. Francois and C. R. Shiells, (eds) Modeling Trade Policy, Applied General Equilibrium Assessments of North American Free Trade, (Cambridge: Cambridge University Press).
- De Rosa, D. A. (1998). “Regional Integration Arrangements: Static Economic Theory, Quantitative Findings, and Policy Guidelines", background paper for a World Bank Policy Research Report, Regionalism and Development, Virginia: ADR International Ltd.

### Technology, R&D, and Diffusion
- Bayoumi, T., D. T. Coe, and E. Helpman (1999). R&D Spillovers and Global Growth. Journal of International Economics 47:399-428.
- Coe, D., and E. Helpman (1995). International R&D Spillovers. European Economic Review 39 (5): 859-87.
- Coe, D. T., E. Helpman, and A. W. Hoffmaister (1997). North-South R&D Spillovers. The Economic Journal 107 (January): 134-49.
- Cohen, W. M., and D. A. Levinthal (1989). Innovation and Learning: The Two Faces of R&D. The Economic Journal 99(September): 569-596.
- ⎯-⎯-and ⎯-⎯⎯(1990). Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly 35(1): 128-52.
- Eaton, J. and S. Kortum (1996). Trade in Ideas: Patenting and Productivity in the OECD. Journal of International Economics 40: 251-278.
- Keller, W. (1997). Trade and the Transmission of Technology. NBER Working Paper, 6113, Cambridge, MA, July, 1-33 + Appendix.
- _____ (2000). .Do Trade Patterns and Technology Flows affect Productivity Growth? World Bank Economic Review 14: 17-47.
- _____ (2001). The Geography and Channels of Diffusion at the World’s Technology Frontier. NBER Working Paper, 8150, Cambridge, MA, 1-28 + Appendix.
- _____ (2004). .International Technology Diffusion. Journal of Economic Literature Vol. XLII (September 2004):752-782.
- Das, Gouranga G. (2002). “Trade, Technology and Human Capital: Stylised Facts and Quantitative Evidence”⎯ The World Economy, Volume 25, Issue 2, Blackwell Publishers, Oxford (U.K.) & Boston (U.S.A.).
- De Ferranti, D., G. E. Perry, I. S. Gill, J. L. Guasch, W. F. Maloney, C. S. Paramo, and N. Schady (2003b). Closing the Gap in Education and Technology. The World Bank, Washington.
- De, Janvry, A., G. Graff, E. Sadoulet and D. Zilberman (2000), “Technological change in Agriculture and Poverty Reduction,” A Concept Paper for the World Development Report on Poverty And Development, 2000/01, World Bank, Washington DC, processed.
- Huang, J. and S. Rozelle (1996), “Technological Change: Rediscovering the Engine of Productivity Growth in China’s Rural Economy.” Journal of Development Economics, 49.
- Huang, J., R. Hu, H. van Meijl, and F. van Tongeren (2004). "Biotechnology boosts to crop productivity in China: trade and welfare implications." Journal of development Economics, 75, pp. 27-54.
- Pack, H., and L. E. Westphal (1986). "Industrial Strategy and Technological Change: Theory versus Reality." Journal of Development Economics, 87, pp. 87-128.
- Nelson, R.R. (1990). On Technological Capabilities and their Acquisition. In R.E. Evenson and G. Ranis (eds.), Science and Technology: Lessons for Development Policy, Westview Press.
- Bartel, A., S. Lach, and N. Sicherman (February 2005). “Outsourcing and Technological Change.” Working Paper # 1158, NBER, pp. 1-37.
- Autor, David. (November, 2001). “Why do Temporary Help Firms Provide Free General Skills Training?” Quarterly Journal of Economics.
- Autor, David. (2002). “Outsourcing at Will: Unjust Dismissal Doctrine and the Growth of Temporary Help Employment.” Journal of Labor Economics.

### Poverty, Inequality, and Human Development
- Anand, Sudhir and Amartya Sen (1997). “Concepts of Human Development and Poverty: A Multidimensional Perspective.” In Poverty and Human Development, Human Development Papers. Human Development Report Office. The United Nations Development Programme, NY, USA.
- Bleaney, Michael and Akira Nishiyama (2004). “Some international evidence on Poverty Alleviation and Economic Growth.” The European Journal of Development Research. 16 (4), pp. 827-844.
- Bourguignon, Francois (March 2004). "The Poverty-Growth-Inequality Triangle." Working Paper No. 125, Indian council for Research on International Economic Relations, New Delhi, India.
- -----------------(2003). "The Growth Elasticity of Poverty Reduction: Explaining Heterogeneity across Countries and Time Periods." In T. Eicher and S. Turnovsky, eds. Inequality and Growth: Theory and Policy Implications. Cambridge: MIT Press.
- Bussolo, Maurizio and David O’Connor (2002). “Technology and Poverty: Mapping the Connections.” In Jorge Braga de Macedo and Tadao Chino (eds.), Technology and Poverty Reduction in Asia and the Pacific. Development Centre Seminars, OECD.
- Chakrabarti, A. (2000). "Does Trade Cause Inequality?" Journal of Economic Development, 25(2).
- Chen, S., and M. Ravallion. (2004). "How Have the World's Poorest Fared since the Early 1980s?" The World Bank Research Observer, 19 (2), World Bank, pp. 141-169.
- Deininger, K., and L. Squire (1996). "A New Dataset measuring Income Inequality," The World Bank Economic Review 10 (3), pp. 565-591.
- Dollar, David and A. Kraay (2002). "Growth is Good for the Poor." Journal of Economic Growth, 7, pp. 195-225.
- Ravallion, M. (1994). Poverty Comparisons. Harwood Academic Books.
- ------------(1995). "Growth and Poverty: Evidence for Developing Countries in the 1980s." Economics Letters, 48, pp. 411-417.
- ------------(2001). "Growth, Inequality and Poverty: Looking Beyond Averages." World Development. 29(11), pp. 1803-1815.
- ------------(2004). "Looking Beyond Averages in the Trade and Poverty Debate." World Bank Policy Research Working Paper 3461. Washington D.C.: World Bank.
- Richardson, J. D. (1995). "Income Inequality and Trade: How to Think, what to Conclude." Journal of Economic Perspectives, 9, pp.33-55.
- Robinson, S., and H. Lofgren (2005). " Macro Models and Poverty Analysis: Theoretical Tensions and Empirical Practice. Development Policy Review, 23 (3), pp. 267-283.
- Sachs et al. (2004). “Ending Africa’s Poverty Trap.” Brookings Institution.
- Sachs, Jeffrey D. (2005) “The End of Poverty: Economic Possibilities of Our Time.” The Penguin Press, New York.
- Sen, A. (1993). "Capability and Well-being", in The Quality of Life, (eds.) Martha Nussbaum and A. Sen. Oxford: Clarendon Press.
- Sen, A. (2004). "Development as Freedom", Oxford University Press.
- Yitzhaki, Shlomo (2002). "Do we need a Separate Poverty Measurement?" European Journal of Political Economy, 18, pp. 61-85.
- Berg, A., and A. Krueger (2003). "Trade, Growth, and Poverty—A Selective Survey," Annual World Bank Conference on Development Economics 2003, World Bank.
- Gunter, B. G., M. J. Cohen, and H. Lofgren (2005). "Analysing Macro-poverty Linkages: An Overview." Development Policy Review, 23 (3), pp. 243-265.

### Institutions, Governance, and Policy Analysis
- Kaufmann, D. A. Kraay, and M. Mastruzzi (2003). Governance Matters III: Governance Indicators for 1996–2002. World Bank Policy Research Working Paper 3106. Dataset at: www.worldbank.org/wbi/governance/govdata2002/.
- Kaufmann, D (2004). Governance Redux: The Empirical Challenge. Chapter in Global Competitiveness Report 2003-2004, World Economic Forum.
- Gordon, James and Poonam Gupta (2004). Understanding India’s Services Revolution. IMF Working Paper, WP/04/171.
- Hertel, T. W., and J. J. Reimer (November 2004). "Predicting the Poverty Impact of Trade Reform." World Bank Policy Research Working Paper 3444, World Bank.
- De Ferranti, D., G. E. Perry, I. S. Gill, J. L. Guasch, W. F. Maloney, C. S. Paramo, and N. Schady (2003b). Closing the Gap in Education and Technology. The World Bank, Washington.
- Dyke, Nancy Berg (2001) in Aspen Institute Conference Report on International Peace, Security and Prosperity (2001), “Attacking Global Poverty: Technology for Economic and Social Uplift.” August 18-20, 2000, Aspen, Colorado.

### Methods, Data, and Modeling Tools
- Barro, R. J. and J. W. Lee (1996). "International Measures of Schooling Years and Schooling Quality." Papers and Proceedings of American Economic Review. 86 (2), pp. 218-223.
- Harrison, W. J. and K. R. Pearson (1996) “Computing Solutions for Large General Equilibrium Models Using GEMPACK.” Computational Economics, 9: pp. 83-127.
- Dimaranan, B., and R. A. A. Mc Dougall (eds.) (2003). Global Trade, Assistance, and Protection: The GTAP 5 Database, Center for Global Trade Analysis, Purdue University, USA.
- McDougall, R. (October 2003). Release Notes for GTAP.Tab 6.2. GTAP Center, Purdue University, USA.
- Hertel, T. W., ed. (1997). Global Trade Analysis: Modeling and Applications. Cambridge, MA, Cambridge University Press.
- Bayoumi, T., D. T. Coe, and E. Helpman (1999). R&D Spillovers and Global Growth. Journal of International Economics 47:399-428.
- Coe, D., and E. Helpman (1995). International R&D Spillovers. European Economic Review 39 (5): 859-87.

### International Reports and World Bank / UN Publications
- World Bank (1999). Knowledge for Development. World Development Report 1998/9, New York, Oxford University Press.
- World Bank (2000). Attacking Poverty. World Development Report 2000/2001, New York, Oxford University Press.
- World Bank (2004a). Making Services work for Poor People. World Development Report 2004, New York, Oxford University Press.
- World Bank (2004b). Global Economic Prospects 2004: Realizing the Development Promise of the Doha Agenda. Washington D.C.: World Bank.
- World Bank (2004c). World Development Indicators. World Bank.
- World Bank (2005). A Better Investment Climate for Everyone. World Development Report 2005, New York, Oxford University Press.
- UNDP. Human Development Report 2003: Millenium Development Goals: A Compact among Nations to end Human Poverty. New York, Oxford University Press.
- Human Development Report 2001: New York, Oxford University Press.

*Reference list as presented in the source PDF: _wp0716 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0716.pdf_