## _wp1350 - Section IV deploys the model. First, it shows formally how SBTC in developed countries

## Source details

**Canonical URL:** [_wp1350 - Section IV deploys the model. First, it shows formally how SBTC in developed countries](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1350.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1350.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1350.pdf.json)

---

### I. Puzzle, scope, and empirical stylized facts
- Puzzle:
  - Both developed and developing countries experienced greater inequality alongside unprecedented increases in educational attainment and trade openness.
  - An increase in the supply of skilled labor should reduce its price, and trade liberalization in developing countries was supposed to have the same effect, yet inequality rose.
- Regional and political relevance:
  - MENA experienced exceptional expansion in education attainment, contributing nine of the world’s top 20 countries by this measure (Campante & Chor, 2012).
  - Higher inequality is linked to political instability and democratic reversals; inequality, unemployment and lack of opportunity are argued to have contributed to the uprisings beginning in 2011 (the Arab Spring).
- Stylized empirical patterns:
  - Capital deepening and Total Factor Productivity (TFP) growth were insufficient to account for observed changes in inequality.
  - The correlation of inequality and the quantity of skills is close to zero after allowing for exogenous trends.
  - Conditional correlation between trade and inequality: zero or positive in MICs; negative in LICs.
  - The standard HOSS mechanism alone makes a relatively limited contribution to the impact of trade on skill premia.

### II. Endogenous Skill-Biased Technical Change (SBTC): core mechanism and trade interaction
- Core mechanism:
  - Technological change can favor either skilled or unskilled workers; its bias is endogenous.
  - The decision to acquire a skill-biased technology depends on that technology’s relative profitability, which depends on the market for the machine, availability of skilled workers, and the price of the good produced with that machine.
- Trade interaction and heterogeneous impacts:
  - The price of skill-intensive goods can rise or fall after opening to international trade, depending on the economy’s skill endowment.
  - A rise in the supply of skilled labor can induce SBTC that increases the relative demand for skilled labor.
  - A MIC that is skill-abundant relative to the world and opening to trade would face a rise in the price of skill-intensive products and have an incentive to acquire skill-biased technologies.
  - Skill-scarce LICs would experience technological change that favors unskilled workers and see a fall in the skill premium after opening up to trade.
- Technology diffusion and global R&D:
  - Developing countries tend to acquire technologies from developed countries.
  - R&D in rich countries has been skill-biased, which affects technologies available to poorer countries and tends to induce acquisition of skill-biased technologies irrespective of local conditions.
  - Broader globalization—liberalization of trade in goods and access to foreign technology—plays an important role in how SBTC manifests.

### III. Empirical background and quantitative evidence (Section II.A and Appendix)
- Three documented patterns (linear projections in Table 1):
  - Inequality increased for all developing countries over time.
  - No correlation between inequality and the proportion of the skilled population.
  - Correlation between inequality and openness: negative for LICs, zero or positive for LMICs and UMICs.
- Selected coefficients and statistics (as reported in Table 1 and Appendix A1):
  - Skilled Proportion coefficients: -0.204*, 0.0607, -0.0577, 0.0668**, 0.027.
  - Trend coefficients (five-year increments 1980–2000): 0.109, 0.210***, 0.328**.
  - Trade/GDP share (Goods) coefficient examples: -6.312**, 0.0377, 1.898.
  - Open Dummy example: 0.0231, 0.207*.
  - R2 examples: 0.36, 0.23, 0.27, 0.18, 0.20.
  - N observations examples: 386, 260, 261; Countries counts examples: 15, 20, 15, 26, 17.
- Inequality over time (selected country numbers and EHII changes):
  - Of 40 countries with long-range data, 34 experienced increases in inequality; all the MENA countries did so; one country experienced no change.
  - Table 2 (MENA changes in EHII, 1980s to 2000s): Algeria 7.4; Egypt 9.2; Iran 9.7; Jordan 2.2; Syria 6.6; Tunisia 12.1.
  - EHII described as a GINI-like index between 0 and 100.
- Education and skill supply:
  - Median proportion completing primary school rose by approximately half; 39 out of 40 countries increased average years of schooling between 1965 and 2000.
  - Table 3 (MENA large rises in educational attainment, 1980–2010) changes in schooling years: Iran 35.3; Algeria 45.2; UAE 55.2; Bahrain 84.7; Jordan 94.7; Libya 104.6; Egypt 144.4; Saudi Arabia 184.1; Tunisia 194.1.
  - Expanded educational attainment has no robust association with household income inequality or wage premia in the observed data.
- Trade openness changes (Table 4, Goods Trade to GDP Ratio, MENA, 1980–2010):
  - Algeria 4; Egypt -16; Iran 19; Jordan 16; Kuwait -22; Lebanon -53; Libya 14; Mauritania 45; Morocco 22; Oman 1; Qatar -6; Saudi Arabia 0; Sudan 7; Syria 7; Tunisia 29; UAE 58.
- Robustness and data:
  - Inequality measure: EHII (0–100) from UTIP, 1963–2003; five-year averages matched to Barro & Lee schooling data for 1980–2000.
  - Trade data: IMF WEO (October 2012), Direction of Trade Statistics; five-year averages.
  - Skilled proportion: share completing at least primary school (Caselli & Coleman 2006 procedure).
  - Sample: data available for about 50 developing countries; skill premium regressions cross-section for 43 developing countries from 1985 onwards.
  - Estimation: OLS with robust standard errors; LAD yields similar results.

### IV. Theoretical model structure (Sections III–IV): production, technology adoption, and key equations
- Production structure:
  - Final output Yt is CES of skilled intermediates ys t and unskilled intermediates yu t:
    - Yt = [ (ys t)^((-1)/) + (yu t)^((-1)/) ]^(/(-1)), with  > 0 the elasticity of substitution between intermediate inputs.
  - Intermediate production uses skilled/unskilled labor and Ts or Tu machine varieties (variety expansion, Romer 1990).
  - Intermediate outputs:
    - ys t = Ts t (Ls t)^(1-) Xt^
    - yu t = Tu t (Lu t)^(1-) Zt^
- Prices, wages, and elasticities:
  - ys/yu = (ps/pu)^(-); in a closed economy ps/pu = (ys/yu)^(-1/).
  - Define  =  +  - , the elasticity of substitution between skilled and unskilled labor.
  - Relative wage ratio: Wt = Tt p t (equation (8)), where Wt is skilled/unskilled wage ratio, Tt is relative availability of machines, pt is price ratio.
- Technology adoption and licensing:
  - Importing one machine unit costs 1 unit of Y; licence costs Cs for a skilled machine, Cu for unskilled machine, with Cs = (Ts/Rs)^k and Cu = (Tu/Rs)^k form described.
  - R is the skill-bias of the world’s available technologies; k governs how licence cost rises near the frontier.
  - In steady state with free entry, licences satisfy V s = C s and V u = C u leading to:
    - T = R (L p)^(1/k)  (equation 14).
    - W = R L^(1/k) p^((k+1)/k)  (equation 15).
  - Closed-economy closed-form relations (equation 17 forms):
    - T = L^((-1)/(k+1)) R^(k/(k+1))
    - p = R^(-k/(k+1)) L^(-(k+1)/(k+1))
    - W = L^((-2-k)/(k+1)) R^(k(-1)/(k+1))
  - Key parameter role: response of W to R is positive iff  > 1.

### V. Comparative statics and quantitative impacts (Section IV)
- A. SBTC in advanced countries:
  - A rise in R (skill-bias of world R&D) raises T in both open and closed economies.
  - In a small open economy dlnW/dlnR = 1; pervasive R increases affect all countries.
  - Feedback through international relative prices in closed economies: dlnW/dlnR = k(-1)/(k+1) (equation 18).
  - Opening to trade can be tantamount to a rise in effective R for previously closed economies.
- B. Trade and skill supply — Small Open Economy (SOE):
  - Define world ratio L0 and SOE with L > L0.
  - HOSS alone implies:
    - W_HOSS / W_closed = (p0 / p) = (L / L0)^(-(k+1)/(k+1)) > 1 (equation 19).
  - With endogenous SBTC, amplification occurs:
    - W_SOE / W_closed = (p0 / p)^((k+1)/k) = (L / L0)^(-(k+1)/(k+1) * (k+1)/k) > W_HOSS / W_closed (equation 20).
  - In SOE price-taker case:
    - W_SOE = R L^(1/k) p0^((k+1)/k); dlnW/dlnL = 1/k (equation 21) — unequivocally positive: increased L raises skill premium via induced SBTC.
- C. Trade and skill supply — Medium/Large symmetric developing countries opening
  - World price p_medium depends on endowments and technologies of both countries; amplification is present but smaller than SOE because of feedback to world prices.
  - W_medium / W_closed = ^(-(k+1)/(k+1)) (equation 23) with  defined from symmetric relations.
  - dlnW/dlnL = (-2-k)/(k+1) - (k+1)/(k+1) * @ln/@lnL (equation 24).
  - If all countries increase skill supply simultaneously,  unchanged and expression reduces to closed-economy case.
  - Closed-economy sign of dlnW/dlnL depends critically on  and k: induced SBTC can offset substitution; positive derivative requires  sufficiently large (text notes thresholds such as  > k+2 or at least 2 in some analyses).
- Empirical notes on  (elasticity of substitution):
  - Psacharopoulos & Hinchliffe (1972) estimate  ~ 2.1 to 2.5 for developing countries; some earlier studies reported very large values; micro studies often find lower values.
  - Little or no correlation between relative skill supply and skill premium can be consistent with endogenous SBTC countering substitution effects.

### VI. Policy-relevant findings, explanations, and recommendations (Section V)
- Core conclusions:
  - SBTC is an important contributor to rising inequality in developing countries.
  - Technical change in developing countries has tended to be skill-biased because advanced economies’ R&D is skill-biased and developing countries import technologies.
  - Trade liberalization raises relative prices of skill-intensive goods in MICs (skill-abundant), inducing adoption of skill-biased technologies; opposite in skill-scarce countries.
  - Increased skill supply can induce SBTC (skill supply creates its own demand), which can offset the expected reduction in skill premia from education expansion.
  - SBTC amplifies HOSS trade effects on wage inequality.
- Explaining empirical puzzles:
  - Why inequality rose widely despite increased educational attainment (e.g., MENA).
  - Negative correlation between trade and inequality in LICs vs zero/positive correlation in MICs.
- Policy implications:
  - Access to foreign technology, human capital expansion, and trade openness promote productivity and growth but may not be inclusive—particularly in MICs.
  - Equipping people with skills is reasonable, but education expansion may not by itself reduce inequality in the face of SBTC; Tinbergen’s “race between education and technology” applies—expanded education may not reduce inequality and does not help the unskilled.
  - Policies may be needed to protect the vulnerable and distribute gains from productivity.

*Source: Excerpt from IMF working paper content unit _wp1350 - Section IV deploys the model. First, it shows formally how SBTC in developed countries.*

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

### _wp1350 - References

### I. Introduction: scope and puzzle
- Over the past generation, both developed and developing countries experienced greater inequality alongside unprecedented increases in educational attainment and trade openness.
- The puzzle: an increase in the supply of skilled labor should reduce its price, and trade liberalization in developing countries was supposed to have the same effect, yet inequality rose.
- Regional focus: the Middle East and North Africa (MENA) experienced an exceptional expansion in education attainment, contributing nine of the world’s top 20 countries by this measure (Campante & Chor, 2012).
- Political relevance: higher inequality is linked to political instability and democratic reversals; inequality, unemployment and lack of opportunity are argued to have contributed to the uprisings beginning in 2011 (the Arab Spring).

### Stylized empirical patterns summarized in the text
- Capital deepening and Total Factor Productivity (TFP) growth were insufficient to account for observed changes in inequality (Berman & Machin, 2000; Acemoglu, 2002).
- The correlation of inequality and the quantity of skills is close to zero after allowing for exogenous trends.
- The conditional correlation between trade and inequality:
  - zero or positive in Middle-Income Countries (MICs)
  - negative in Low-Income Countries (LICs)
- The standard Heckscher Ohlin Stolper Samuelson (HOSS) mechanism alone makes a relatively limited contribution to the impact of trade on skill premia (Chusseau, Dumon & Hellier, 2008).

### Endogenous skill-biased technical change (SBTC) as proposed explanation
- Core mechanism:
  - Technological change can favor either skilled or unskilled workers; its bias is endogenous.
  - The decision to acquire a skill-biased technology depends on that technology’s relative profitability.
  - Profitability depends on the market for the machine, which depends on the availability of skilled workers and on the price of the good produced with that machine.
- Trade interaction:
  - The price of skill-intensive goods can rise or fall after opening to international trade, depending on the economy’s skill endowment.
  - A rise in the supply of skilled labor can induce SBTC that increases the relative demand for skilled labor.
  - A MIC that is skill-abundant relative to the world and opening to trade would face a rise in the price of skill-intensive products and have an incentive to acquire skill-biased technologies.
  - Skill-scarce LICs would experience technological change that favors unskilled workers and see a fall in the skill premium after opening up to trade.
- Technology diffusion and global R&D:
  - Developing countries tend to acquire technologies from developed countries.
  - R&D in rich countries has been skill-biased in recent times, which affects the skill-bias of technologies available to poorer countries and tends to induce acquisition of skill-biased technologies irrespective of local conditions (Berman & Machin, 2000).
  - Broader globalization—liberalization of trade in goods and access to foreign technology—plays an important role in how skill-biased technical change manifests.

### Modeling approach and implications
- The paper presents:
  - Section II: evidence on relationships between inequality, the skill endowment, and trade openness; evidence that the skill-bias of technical change is endogenous in developing countries.
  - Section III: a formal model—an endogenous growth model with two types of workers: skilled and unskilled.
- Key modeling features tailored to developing countries:
  - Instead of incurring R&D costs, firms pay for the monopoly licence to acquire a technology and import machines using that technology from abroad to sell domestically.
  - The amount of induced SBTC, and hence of inequality, is a function of the rate at which more advanced technologies get more expensive.
  - Many developing countries are small and have limited or no impact on the relative price of skilled and unskilled products; this affects quantitative implications for SBTC and wage inequality responses.

*Source: Excerpt from IMF working paper content unit _wp1350 - References.*

### Section IV deploys the model. First, it shows formally how SBTC in developed countries

### _wp1350 - Section IV deploys the model. First, it shows formally how SBTC in developed countries

### Section IV: Model deployment — overview and main mechanisms
- The model shows formally:
  - How SBTC in developed countries leads to SBTC and higher wage inequality in developing countries.
  - That HOSS effects on inequality in a skill-abundant small open economy (SOE) opening to trade are amplified by induced technical change.
  - In a SOE, a rise in skill supply unequivocally increases the skill premium.
  - When two large developing economies open to trade, effects on inequality are amplified relative to HOSS but are smaller than in the SOE case.
  - Increased skill supply leads to higher inequality if the elasticity of substitution between skilled and unskilled labor () is sufficiently high.
- Section V offers policy implications; emphasis is on a unified framework for endogenous SBTC in developing countries, explaining pervasive increases in inequality, the weak correlation with skill quantity, and differential trade effects in MICs and LICs.

### Empirical background — stylized patterns (Section II.A)
- Three documented patterns (linear projections in Table 1):
  - Inequality increased for all developing countries over time.
  - No correlation between inequality and the proportion of the skilled population.
  - Correlation between inequality and openness: negative for LICs, zero or positive for LMICs and UMICs.
- Table 1 (summary excerpted coefficients/notes):
  - Dependent variables: EHII (index of household income inequality) and Skill Premium.
  - Skilled Proportion coefficients (various columns): -0.204*, 0.0607, -0.0577, 0.0668**, 0.027 (standard errors shown in brackets in source).
  - Trend coefficients (five-year increments 1980–2000): 0.109, 0.210***, 0.328** (standard errors available in source).
  - Trade/GDP share (Goods) coefficient examples: -6.312**, 0.0377, 1.898 (standard errors in source).
  - Open Dummy example: 0.0231, 0.207* (standard errors in source).
  - R2 examples: 0.36, 0.23, 0.27, 0.18, 0.20; N observations examples: 386, 260, 261; Countries counts examples: 15, 20, 15, 26, 17 (exact tabulation in source).
- Inequality over time:
  - Of 40 countries with long-range data, 34 experienced increases in inequality over the period; all the MENA countries did so; one country experienced no change.
  - Table 2 (MENA changes in EHII, 1980s to 2000s): Algeria 7.4; Egypt 9.2; Iran 9.7; Jordan 2.2; Syria 6.6; Tunisia 12.1.
  - EHII described as a GINI-like index between 0 and 100.
- Inequality and schooling:
  - Median proportion completing primary school rose by approximately half; 39 out of 40 countries increased average years of schooling between 1965 and 2000 (Behar, 2011).
  - Table 3 (MENA large rises in educational attainment, 1980–2010) lists changes in schooling years: Iran 35.3; Algeria 45.2; UAE 55.2; Bahrain 84.7; Jordan 94.7; Libya 104.6; Egypt 144.4; Saudi Arabia 184.1; Tunisia 194.1 (source: Campante & Chor (2012), Barro-Lee dataset v1.2).
  - Table 1 suggests expanded educational attainment has no robust association with household income inequality or wage premia.
  - The absence of correlation raises concern: increased education does not necessarily reduce the skill premium because skill supply may create its own demand via induced SBTC.
- Inequality and trade:
  - World goods trade share rose dramatically since WWII; recent surges are South-South trade phenomena.
  - Table 4 (Change in Trade Openness, MENA, 1980–2010) — Goods Trade to GDP Ratio changes: Algeria 4; Egypt -16; Iran 19; Jordan 16; Kuwait -22; Lebanon -53; Libya 14; Mauritania 45; Morocco 22; Oman 1; Qatar -6; Saudi Arabia 0; Sudan 7; Syria 7; Tunisia 29; UAE 58 (source: IMF WEO).
  - Table 1 suggests LICs that opened had a smaller rise in inequality; LMIC exports show no relationship; UMICs that increased trade more tended to have larger rises in inequality (large standard errors).
  - Modified HOSS with multiple skill abundance levels can reconcile some heterogeneity: UMICs may be skill-abundant relative to LMICs/LICs but not to advanced countries.
  - Case evidence: Latin America trade liberalizations associated with increased skill premia; Tunisia’s increased trade associated with a fall in inequality (Ben Salha, 2012).
- Empirical implication: trade and technology are linked; trade liberalization can change relative prices and profitability of producing skill-biased technologies (Acemoglu, 2003).

### Drivers of Skill-biased Technical Change (SBTC) (Section II.B)
- Two views:
  - Intrinsic SBTC: machines/complements or new technologies inherently favor skills (Griliches, 1969; Nelson & Phelps, 1966).
  - Endogenous SBTC: R&D and technology adoption respond to profit incentives; the choice to develop or adopt skill-biased technologies depends on relative profitability (Hicks, Schmookler, Habbakuk).
- Determinants from developing-country perspective:
  - Relative cost of technology types depends on the skill-bias of R&D in developed countries.
  - Relative revenues depend on the local quantity of skilled labor and relative price of skill-intensive goods.
  - Empirical findings: top five patent sources account for 84 percent of patents worldwide; developing countries account for 3 percent (Behar, 2011). Developing countries rely largely on foreign technology transfer (Savvides & Zachariadis, 2005).
  - Local absorption matters: case evidence of adaptation and substitution (Knight, 1979; Fransman, 1985); Esposito & Stehrer (2009) find positive relation between initial relative skilled quantity and subsequent SBTC.
  - Trade affects embodied technology prices and composition; the skill-bias of imported machines varies with local conditions (Caselli & Wilson, 2004); multinationals tailor technologies to subsidiaries (Acemoglu & Zilibotti, 2001).
  - Evidence: increased openness in Asia led to SBTC in MICs but not LICs (Almeida, 2009); Gourdon (2011) finds SBTC more in South-South trade sectors than South-North.
- Synthesis: advanced-country SBTC and the tendency for developing countries to import these technologies imply that the skill-bias in developing countries reflects both global R&D bias and local endowments/absorption.

### Theory — model setup (Section III): production, prices, technology adoption
- Production:
  - Final output Yt is a CES aggregate of skilled intermediates ys t and unskilled intermediates yu t:
    - Yt = [ (ys t)^( (-1)/ ) + (yu t)^( (-1)/ ) ]^( /(-1) )
    -  > 0 is the elasticity of substitution between intermediate inputs.
  - Intermediate production uses skilled/unskilled labor and Ts or Tu machine varieties (variety expansion model, Romer 1990).
  - Firm-level demands and economy-wide demands derived lead to expressions:
    - Xt = ^( 1/(1-) ) (ps t)^( 1/(1-) ) Ls t  (equation (2) form)
    - Zt = A ^( 1/(1-) ) (pu t)^( 1/(1-) ) Lu t  (equation (3) form)
  - Intermediate outputs:
    - ys t = Ts t (Ls t)^(1-) Xt^
    - yu t = Tu t (Lu t)^(1-) Zt^  (equations 4a, 4b)
- Prices and wages:
  - Final output producers choose inputs so ys/yu = (ps/pu)^(-) (equation (5)).
  - In a closed economy, relative price satisfies ps/pu = (ys/yu)^(-1/) (equation (6)).
  - Combining intermediates and demands yields (equation (7)) relating relative price to Ts/Tu and Ls/Lu with elasticity parameters; defines  =  +  - , the elasticity of substitution between skilled and unskilled labor.
  - In a small open economy, prices are international (price taker), so changes in domestic supply need not adjust domestic relative prices.
  - Relative wage ratio:
    - Wt = Tt p t  (equation (8)) where Wt is skilled/unskilled wage ratio, Tt is relative availability of machines, and pt is price ratio.
- Modeling technology adoption:
  - Importing one machine unit costs 1 unit of Y; licence costs Cs for a skilled machine, Cu for unskilled machine, with Cs = (Ts/Rs)^k and Cu = (Tu/Rs)^k form described (equation (9) representation).
  - R is the skill-bias of the world’s available technologies; k governs how licence cost rises near the frontier.
  - Value of a licence equals present value of future profits; per-period profits for skilled licence involve (ps)^(1/(1-)) Ls; analogous for unskilled (equations 10, 11).
  - In steady state with free entry V s = C s and V u = C u, the relative value V = L p (equation 12) and V = C (equation 13).
  - Number of technologies:
    - T = R (L p)^(1/k)  (equation 14).
  - Combining with wages:
    - W = R L^(1/k) p^((k+1)/k)  (equation 15).
  - Closed-economy substitutions yield:
    - T = L^((-1)/(k+1)) R^(k/(k+1))  (equation 16 form)
    - p = R^(-k/(k+1)) L^(-(k+1)/(k+1))
    - W = L^((-2-k)/(k+1)) R^(k(-1)/(k+1))  (equation 17)
  - Key parameter role: elasticity of substitution  critical; response of W to R is positive iff  > 1.

### Comparative statics (Section IV)
- A. SBTC in advanced countries:
  - A rise in R (skill-bias of world R&D) raises T in both open and closed economies (equations 14 and 16).
  - In a small open economy dlnW/dlnR = 1; pervasive R increases affect all countries.
  - Feedback through international relative price of skilled goods attenuates/affects W in closed economies:
    - dlnW/dlnR = k(-1)/(k+1) (equation 18).
  - Opening to trade can be tantamount to a rise in effective R for previously closed economies.
- B. Trade and skill supply — Small Open Economy (SOE):
  - Define world ratio L0; SOE has L > L0; other parameters same.
  - Absent induced technical change, HOSS implies:
    - W_HOSS / W_closed = (p0 / p) = (L / L0)^(-(k+1)/(k+1)) > 1  (equation 19).
  - Allowing for endogenous SBTC, T rises with higher relative prices; wage amplification:
    - W_SOE / W_closed = (p0 / p)^((k+1)/k) = (L / L0)^(-(k+1)/(k+1) * (k+1)/k) > W_HOSS / W_closed  (equation 20).
  - In SOE price-taker case:
    - W_SOE = R L^(1/k) p0^((k+1)/k); dlnW/dlnL = 1/k (equation 21) — unequivocally positive.
  - Mechanism: increased L raises value of skill-biased technologies → induced SBTC → higher skill premium; magnitude depends on k.
- C. Trade and skill supply — Medium/Large economies (two large symmetric developing countries opening)
  - World price p_medium depends on endowments and technologies of both countries (equation 22).
  - With T = R(L p_medium)^(1/k) and symmetric relations, define asymmetry parameter  and derive:
    - W_medium / W_closed = ^(-(k+1)/(k+1))  (equation 23).
  - W_medium > W_closed and amplification is present but smaller than SOE due to feedback to world prices.
  - W response to L:
    - dlnW/dlnL = (-2-k)/(k+1) - (k+1)/(k+1) * @ln/@lnL  (equation 24).
  - If all countries increase skill supply simultaneously,  unchanged and expression reduces to closed-economy case.
  - In closed economy, increase in skill supply can either raise W (via induced SBTC) or lower it (standard substitution); sign depends critically on  and k:
    - For positive derivative in closed economy need  sufficiently large (text notes threshold:  > k+2 or at least 2 in some analyses).
  - Empirical notes on  estimates:
    - Psacharopoulos & Hinchliffe (1972) estimate  ~ 2.1 to 2.5 for developing countries; some estimates as high as 1000 in older studies; micro studies often lower.
    - Behar (2009) notes little or no correlation between relative skill supply and skill premium can be consistent with endogenous SBTC countering substitution effects.

### Policy-relevant findings and implications (from Conclusion, Section V)
- Core conclusions:
  - SBTC is an important contributor to rising inequality in developing countries.
  - Technical change in developing countries has tended to be skill-biased because advanced economies’ R&D is skill-biased and developing countries import technologies.
  - Trade liberalization raises relative prices of skill-intensive goods in MICs (skill-abundant), inducing adoption of skill-biased technologies; opposite in skill-scarce countries.
  - Increased skill supply can induce SBTC (skill supply creates its own demand), which can offset the expected reduction in skill premia from education expansion.
  - SBTC amplifies HOSS trade effects on wage inequality.
- Explains empirical puzzles:
  - Why inequality rose widely despite increased educational attainment (e.g., MENA).
  - Negative correlation between trade and inequality in LICs vs zero/positive correlation in MICs.
- Policy implications highlighted:
  - Access to foreign technology, human capital expansion, and trade openness promote productivity and growth but may not be inclusive—particularly in MICs.
  - Policies may be needed to protect the vulnerable and distribute gains from productivity: equipping people with skills is reasonable, but education expansion may not by itself reduce inequality in the face of SBTC; Tinbergen’s “race between education and technology” applies — expanded education may not reduce inequality and does not help the unskilled.

### Appendix — data and methodology (summary)
- Inequality measure: EHII (0–100) from University of Texas Inequality Project (UTIP), 1963–2003; five-year averages matched to educational attainment data for 1980, 1985, 1990, 1995, 2000 (Barro & Lee 2001).
- Trade data: IMF World Economic Outlook (October 2012), Direction of Trade Statistics; five-year averages used.
- Skilled proportion: constructed as share completing at least primary school (Caselli & Coleman 2006 procedure); unskilled proportion aggregated below primary completion.
- Sample: data available for about 50 developing countries; panel regressions include country fixed effects exploiting within-country variation.
- Skill premium regressions: cross-section for 43 developing countries from 1985 onwards; skill premium constructed using Mincerian earnings estimates from Behar (2009); openness robustness uses Sachs & Warner (1999) dummy.
- Estimation: Ordinary Least Squares with robust standard errors; Least Absolute Deviations yields similar results.
- Table A1 (robustness) presents alternative trade openness measures and coefficients (selected preserved values shown in body):
  - Skilled Proportion examples: -0.179*, 0.0602, -0.0634, -0.236*, 0.0574, -0.0564, -0.221* (standard errors reported in source).
  - Trend examples: 0.12, 0.21***, 0.327**, 0.139, 0.213***, 0.337**, 0.179**, 0.212***, 0.328** (standard errors in source).
  - Trade/GDP (Goods, Services) coefficients examples: -4.619**, 0.0783, 2.15 (standard errors in source).
  - Export/GDP (Goods) coefficients examples: -5.905*, 0.39, 1.172 (standard errors in source).
  - Export/GDP (Goods, Services) example: -5.686*, 0.279, 0.727 (standard errors in source).
  - R2 examples: 0.15, 0.27, 0.25, 0.22, 0.27, 0.24, 0.18, 0.27, 0.24; N and country counts consistent with main text.
- Notes on openness dummy (Sachs & Warner): country counted as not open if any of five conditions hold (tariffs > 40 percent; nontariff barriers cover > 40 percent of imports; socialist economic system; black market premium > 20 percent; exports controlled by state monopoly).

*Source: _wp1350 - Section IV deploys the model. First, it shows formally how SBTC in developed countries (IMF PDF content provided).*

### REFERENCES

### _wp1350 - REFERENCES

### Technical change, skill bias, and labor markets
- Acemoglu, D (1998). Why Do New Technologies Complement Skills? Directed Technical Change and Wage Inequality.Quarterly Journal of Economics, 113(4):1055-1089.
- Acemoglu, D (2002a). Technical Change, Inequality and the Labor Market.Journal of Economic Literature, 40(1):7-72.
- Acemoglu, D (2002b). Directed Technical Change.Review of Economic Studies, 69(4):781-809.
- Acemoglu, D (2003). Patterns of Skill Premia.Review of Economic Studies, 70(2): 199-230.
- Atkinson. A & J Stiglitz (1969). A New View of Technological Change.Economic Journal, 79:573-578.
- Griliches, Z (1969). Capital Skill Complementarity.Review of Economics and Statistics, 51(4):465-468.
- Hicks, J (1963).The Theory of Wages(2nd edition). MacMillan & Co Ltd.
- Katz, L & D Autor (1999). Changes in the Wage Structure and Earnings Inequality in O Ashenfelter & D Card (eds.)Handbook of Labor EconomicsVol 3A, North-Holland.
- Krusell, P, L Ohanian, J Rios-Rull & G Violante (2000). Capital-skill Complementarity and Inequality: a Macroeconomic Analysis.Econometrica, 68(5):1029-1054.
- Machin, S & A Manning (1997). Can Supply Create its Own Demand? Implications for Rising Skill Differentials.European Economic Review, 41(3-5):507-516.
- Martins, P & P Pereira (2004). Does Education Reduce Wage Inequality? Quantile Regression Evidence From 16 Countries.Labor Economics, 11(3):355-371.
- Psacharopoulos, G & H Hinchliffe (1972). Further Evidence on the Elasticity of Substitution Among Different Types of Educated Labor.Journal of Political Economy, 80(4):786-792.
- Behar, A (2007).Are Skilled and Unskilled Labor Complements or Substitutes?Doctoral Thesis Published by Verlag Dr. Muller.
- Behar, A (2009). Directed Technical change, the Elasticity of Substitution and Wage Inequality in Developing Economies.Oxford Department of Economics Discussion Paper,467.
- Behar, A (2011). Skill-biased Technology Imports, Increased Schooling Access, and Income Inequality in Developing Countries.Journal of Globalization and Development, 2(2):1-29.
- Esposito, P & R Stehrer (2009). The Sector Bias of Skill-biased Technical Change and the Rising Skill Premium in Transition Economies.Empirica, 36(3):351-364.
- Tinbergen, J (1975).Income Distribution Analysis and Policies. North-Holland Publishing Co, Amsterdam.
- Thoenig, M & T Verdier (2003). A Theory of Defensive Skill-Biased Innovation and Globalization. American Economic Review, 93(3):709-728.

### Trade, openness, and technology diffusion
- Anderson, E (2005). Openness and Inequality in Developing Countries: A Review of Theory and Recent Evidence.World Development, 33:1045-1063.
- Almeida, R (2009). Openness and Technological Innovation in East Asia: Have They Increased the Demand for Skills?IZA Working Paper, 4474.
- Berman, E, J Bound & S Machin (1998). Implications of Skill-biased Technical Change: International Evidence.Quarterly Journal of Economics, 113(4):1245-1279.
- Berman, E & S Machin (2000). Skill-biased Technology Transfer: Evidence of Factor-biased Technical Change in Developing Countries. mimeo.
- Burstein, A, J Cravino & J Vogel (2012). Importing Skill-Biased Technology. mimeo.
- Bustos, P (2011). Trade Liberalization, Exports, and Technology Upgrading: Evidence on the Impact of MERCOSUR on Argentinian Firms.American Economic Review, 101(1): 304-40.
- Caselli, F & D Wilson (2004). Importing Technology.Journal of Monetary Economics, 51(1):1-32.
- Eaton J & S Kortum (2001). Trade in Capital Goods.European Economic Review, 45(7):1195-1235.
- Feenstra, R & G Hanson (1997). Foreign Direct Investment and Relative Wages: Evidence from Mexico’s Maquilladoras.Journal of International Economics, 42:371-393.
- Fajnzylber, P. and Ana Fernandes, 2004. “International Economic Activities and the Demand for Skilled Labor: Evidence from Brazil and China.”World Bank Policy Research Working Paper, 3426.
- Goldberg, P & N Pavcnik (2007). Distributional Effects of Globalization in Developing Countries. Journal of Economic Literature, 45(1):39-82.
- Gourdon, J (2011). Wage Inequality in Developing Countries: South–South Trade Matters. International Review of Economics,58:1-25.
- Matsuyama, K (2007). Beyond Icebergs: Towards a Theory of Biased Globalization.Review of Economic Studies, 74(1):237-253.
- Meschi, E & M Vivarelli (2009). Trade and Income Inequality in Developing Countries.World Development, 37(2):287-302.
- Wood, A (1994).North-South Trade. Employment and Inequality. Oxford University Press, Oxford, United Kingdom.
- Wood, A (1997). Openness and Wage Inequality in Developing Countries: The Latin American Challenge to East Asian Conventional Wisdom.World Bank Economic Review, 11(1):33-57.
- Wood, A & J Mayer (2011). Has China De-industrialised Other Developing Countries?.Review of World Economics, 147(2):325-350.
- Matsuyama, K (2007). Beyond Icebergs: Towards a Theory of Biased Globalization.Review of Economic Studies, 74(1):237-253.
- Savvides, A & M Zachariadis (2005). International Technology Diffusion and the Growth of Total Factor Productivity in the Manufacturing Sector of Developing Economies.Review of Development Economics, 9(4):482-501.
- Chusseau, N, M Dumont & J Hellier (2008). Explaining Rising Inequality: Skill-biased Technical Change and North-South Trade.Journal of Economic Surveys, 22(3):409-457.
- Conte, A & M Vivarelli (2011). Imported Skill-biased Technological Change in Developing Countries.The Developing Economies, 49(1):36-65.
- Burstein, A, J Cravino & J Vogel (2012). Importing Skill-Biased Technology. mimeo.

### Empirical evidence, country studies, and regional analyses
- Behar, A (2011). Skill-biased Technology Imports, Increased Schooling Access, and Income Inequality in Developing Countries.Journal of Globalization and Development, 2(2):1-29.
- Ben Salha, O (2012). Economic Globalization. Wages and Wage Inequality in Tunisia: Evidence from the ARDL Cointegration Approach. mimeo.
- Bustos, P (2011). Trade Liberalization, Exports, and Technology Upgrading: Evidence on the Impact of MERCOSUR on Argentinian Firms.American Economic Review, 101(1): 304-40.
- De Gregorio, J & J Lee (2002). Education and Income Inequality: New Evidence From Cross-Country Data.Review of Income and Wealth, 48(3):395-416.
- Fedderke, J Kayemba, S Henderson, M Marrioti & P Vaze (2001). Changing Factor Market Conditions in South Africa: the Capital Market - a Sectoral Description.Development Southern Africa, 18(4):493-511.
- Fransman, M (1985). Conceptualising Technical Change in the Third World in the 1980s: An Interpretive Survey.Journal of Development Studies, 21(4):572-652.
- Hanson, G (2012). The Rise of Middle Kingdoms: Emerging Economies in Global Trade.Journal of Economic Perspectives, 26(2):41-64.
- Campante, F & D Chor (2012). Why Was the Arab World Poised for Revolution? Schooling, Economic Opportunities, and the Arab Spring.Journal of Economic Perspectives, 26(2):167-188.
- Robbins, D. (1996). HOS Hits Facts: Facts Win; Evidence on Trade and Wages in the Developing World.Harvard Institute for International Development Discussion Paper,557.
- Robbins, D. (2003). The Impact of Trade Liberalization Upon Inequality in Developing Countries - A Review of Theory and Evidence.ILO Working Paper,13.
- Esposito, P & R Stehrer (2009). The Sector Bias of Skill-biased Technical Change and the Rising Skill Premium in Transition Economies.Empirica, 36(3):351-364.
- Fajnzylber, P. and Ana Fernandes, 2004. “International Economic Activities and the Demand for Skilled Labor: Evidence from Brazil and China.”World Bank Policy Research Working Paper, 3426.

### Theory, history, and mechanisms of innovation and diffusion
- Nelson, R & E Phelps (1966). Investment in Humans, Technological Diffusion and Economic Growth.American Economic Review, 56(2):67-75.
- Romer, P (1990). Endogenous Technological Change.Journal of Political Economy, 98(5):S71-S102.
- Rosenberg, N (1969). The Direction of Technological Change: Inducement Mechanisms and Focusing Devices.Economic Development and Cultural Change, 18(1):1-24.
- Schmookler, J (1966).Invention and Economic Growth. Harvard University Press, Cambirdge, MA.
- Mokyr, J (1990).The Levers of Riches: Technological Creativity and Economic Progress. NY: Oxford University Press, Oxford, United Kingdom.
- Habbakuk, J (1962).American and British Technology in the Nineteeth Century: Search for Labor Saving Inventions. Cambridge University Press, Cambridge, United Kingdom.
- Hayami, Y & V Ruttan (1970). Factor Prices and Technical Change in Agricultural Development: The United States and Japan:1880-1960.Journal of Political Economy, 78(5):1115-1141.
- Rosenberg, N (1969). The Direction of Technological Change: Inducement Mechanisms and Focusing Devices.Economic Development and Cultural Change, 18(1):1-24.
- Caselli, F & J Coleman (2006). The World Technology Frontier.American Economic Review, 96(3):499-522.
- Habbakuk, J (1962).American and British Technology in the Nineteeth Century: Search for Labor Saving Inventions. Cambridge University Press, Cambridge, United Kingdom.
- Hayami, Y & V Ruttan (1970). Factor Prices and Technical Change in Agricultural Development: The United States and Japan:1880-1960.Journal of Political Economy, 78(5):1115-1141.

*Reference list as provided in _wp1350 - REFERENCES.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1350.pdf_
