## _wp04206

## Source details

**Canonical URL:** [_wp04206](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04206.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04206.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04206.pdf.json)

---

### Introduction and focus
- Examines the effect of regional trade agreements (RTAs) on foreign direct investment (FDI), focusing on RTAs among developing countries (South-South RTAs).
- Central hypothesis: RTA-created larger regional market size can increase FDI inflows; effects may differ for interregional versus intraregional FDI and across RTA members depending on locational advantages.
- Sample: 71 developing countries over the period 1980–99.

### Channels through which RTAs affect FDI
- Tariff and non-tariff barrier reduction and common external tariffs can:
  - Positively affect interregional FDI through larger market size.
  - Ambiguously affect intraregional FDI (depends on pre-existing investment structure and motives).
- Market size effect:
  - FDI involves non-negligible setup costs and increasing returns to scale; larger markets are more profitable for investors.
  - Firms with intangible assets (R&D-based) face stronger increasing returns to scale; market-seeking FDI benefits most.
- Vertical FDI:
  - May be stimulated if output is destined for other RTA members.
  - Lowering internal tariffs reduces tariff-jumping FDI incentives and facilitates vertical FDI that delocalizes production to countries with locational advantages.
- Efficiency gains from RTAs may raise growth and R&D, further stimulating interregional and intraregional FDI.
- Redistribution effect: FDI increases likely to concentrate in RTA members with the strongest locational advantages; some members may lose existing FDI.

### Empirical evidence from prior RTAs (selected findings)
- European Internal Market Programme (IMP):
  - Dunning (1997): IMP had positive effects on both interregional and intraregional FDI, more on interregional FDI.
  - Pain and Landsbury (1996): Increase of intra-regional FDI by the United Kingdom and Germany (1980–92); little support for pronounced intra-EC concentration, though some countries lost FDI (e.g., France and Belgium).
  - Girma (2001): With IMP implementation, European market size and unit labor costs relative to the EC became more important determinants of FDI than local market size (applies mostly to FDI for acquisitions).
- Mercosur (created 1991: Argentina, Brazil, Paraguay, Uruguay):
  - Strong increases in FDI inflows, particularly into Argentina and Brazil.
  - Macroeconomic stability appeared a pre-condition for FDI responses in Argentina and Brazil.
  - Large privatization programs likely contributed to FDI increases.
  - Inflows were uneven across members; Uruguay and Paraguay had mixed experiences.

### Data sources and measurement
- FDI data from World Investment Report (UNCTAD, 2000).
- FDI definition includes equity capital, reinvested earnings, intra-company loans; data often from IMF Balance of Payments Statistics Yearbook.
- FDI stocks are at book value or historical cost; often computed by accumulating flows or adding flows to benchmark estimates.
- Sample split into five-year periods: 1980–85, 1985–90, 1990–95, 1995–99.
- Beginning-of-period FDI stocks and annual average FDI inflows over each sub-period converted into 1995 U.S. dollars.

### Descriptive comparisons: RTA versus non-RTA (exact figures)
- 1980s:
  - RTA countries had somewhat smaller stocks of FDI than non-RTA countries.
- Since 1990:
  - RTA stocks of FDI outgrew those of non-RTA countries by 33 percent on average.
- 1990s:
  - Average stock of FDI for RTA countries: 24 percent of GDP.
  - Average stock of FDI for non-RTA countries: 18 percent of GDP.
- 1980s and 1990s combined:
  - RTAs’ stocks of FDI were on average 3 percentage point larger than non-RTAs’ stocks.
- FDI inflows (average FDI inflows in percent of GDP):
  - Differences between RTA and non-RTA countries are less marked than stocks, but data point to larger FDI inflows into RTA countries at the end of the period (1995–99).
- Regional variation examples (1999 and period averages):
  - ASEAN: average FDI stock of 44 percent of GDP in 1999; net FDI inflows averaged 4.5 percent of GDP over the 1980s and 1990s.
  - Western Hemisphere RTA countries: average FDI stock of 39 percent of GDP in 1999; net FDI inflows averaged 2.1 percent of GDP over the 1980s and 1990s, and 4.8 percent over the period 1995–99.
  - GCC and sub-Saharan RTA countries: average FDI stock of 25 percent of GDP in 1999 (equal to the average for all non-RTA countries in 1999).
  - Over 1980–99, net FDI inflows averaged 1.7 percent of GDP for the GCC and 1.3 percent of GDP for sub-Saharan RTA countries. In the 1990s, GCC net FDI inflows averaged 2.5 percent of GDP and sub-Saharan RTA countries averaged 1.5 percent of GDP.
- Evidence of FDI diversion: coefficients on variables measuring FDI received by RTA countries versus stand-alone countries are negative and significant or close to significant for stock measures.

### Econometric approach and main empirical findings
- Objective: explain the stock of FDI received by a country (expressed in 1995 U.S. dollars); regional model replaces domestic market size with REGY (sum of domestic market size and market sizes of RTA partners) for RTA members.
- Main determinants included:
  - Host market size: ln(Y), gY; regional market size ln(REGY), REGgY.
  - Agglomeration: lagged ln(FDI); infrastructure proxied by tv (television sets per capita).
  - Labor cost/quality: educ (average schooling years over age 15).
  - Business climate: stabfin (financial risk index).
  - Openness: res(X/Y) (export-to-GDP ratio corrected for country size).
  - Time effects (ω) and country fixed effects (μ).
- Key results (competitive regional model, Table 5, column 3):
  - year90 coefficient = 0.27 (t-stat 5.20)**.
  - year95 coefficient = 0.59 (t-stat 7.85)**.
  - year99 coefficient = 0.72 (t-stat 6.05)**.
  - log(FDI) coefficient = 0.43 (t-stat 12.85)**.
  - log(REGY) coefficient = 0.60 (t-stat 5.63)**.
  - gY coefficient = 3.02 (t-stat 3.25)**.
  - log(educ) coefficient = 0.34 (t-stat 2.14)*.
  - log(GAPeduc) coefficient = 2.02 (t-stat 6.00)**.
  - log(stabfin) coefficient = 0.04 (t-stat 0.26) [not significant].
  - log(GAPstabfin) coefficient = 0.71 (t-stat 3.74)**.
  - res(X/Y) coefficient = 0.16 (t-stat 1.93) [near significance].
- Robustness and dynamic specifications:
  - First-difference estimations confirm main result: dlog(REGY) has a significantly positive effect on dlog(FDI+1).
  - Arellano-Bond instrumentation: RTA market size remains significant at the 10 percent level when instrumenting lagged dependent variable and potentially endogenous variables.
  - Sensitivity: RTA market size importance grew in the 1990s versus the 1980s; when splitting market size into GDP per capita and population, GDP per capita component of REGY matters, while domestic population positively affects FDI.

### Competition, diversion, and within-RTA determinants
- Relative locational advantages within RTAs matter:
  - Higher education relative to RTA partners (GAPeduc) increases FDI received.
  - Higher financial stability relative to RTA partners (GAPstabfin) increases FDI received.
- Negative partial correlations found between FDI received by RTA countries and stand-alone countries, indicating potential diversion effects across alternative FDI measures.

### Application to Maghreb countries (Algeria, Morocco, Tunisia)
- Historical FDI performance (selected exact figures):
  - Tunisia: average stock of FDI close to 80 percent of GDP over the 1980s and 1990s; FDI inflows averaged 2.7 percent of GDP in the 1980s and 1990s.
  - Morocco: stock of FDI averaged 7 percent of GDP in the 1980s and 17 percent of GDP in 1999; FDI inflows averaged 1.1 percent of GDP.
  - Algeria: stock of FDI averaged 4 percent of GDP in the 1980s and 3.5 percent in 1995–99; FDI inflows averaged about zero percent of GDP in the 1980s and 1990s.
- Model prediction accuracy (selected figures):
  - Predicted average FDI stock-to-GDP ratios over 1980–99: Algeria 4 percent (actual 4 percent); Morocco 8 percent (actual 9 percent); Tunisia 79 percent (actual 77 percent).
  - Without country fixed effects, predicted FDI/Y would be smaller: Morocco predicted 4 percent (instead of 9 percent); Tunisia predicted 15 percent (instead of 77 percent) in 1999.
- Predicted versus actual growth of FDI stock-to-GDP ratio (selected figures):
  - Algeria: actual growth 1985–90 = -0.24 percent; predicted = 7.41 percent. 1995–99 actual = -2.91 percent; predicted = 1.04 percent.
  - Morocco: 1985–90 actual = 9.64 percent; predicted = 11.31 percent. 1995–99 actual = 10.67 percent; predicted = 10.39 percent.
  - Tunisia: 1995–99 actual = -5.21 percent; predicted = 5.42 percent.
- Simulated FDI benefits from creating a Maghreb RTA (based on 1999 data and elasticities from competitive regional model, column 3 of Table 5):
  - Simulated increase in each country’s FDI stock: Algeria 62 percent, Morocco 85 percent, Tunisia 165 percent.
  - Decomposition of simulated total effect (selected entries):
    - Algeria: Total effect 110 percent; Market size effect 62 percent; Relative education effect 3 percent; Relative financial stability effect 35 percent.
    - Morocco: Total effect 78 percent; Market size effect 5 percent; Relative education effect -4 percent; Relative financial stability effect 44 percent.
    - Tunisia: Total effect 215 percent; Market size effect 165 percent; Relative education effect 19 percent; Relative financial stability effect 0 percent.
  - Caveats noted in the paper: magnitudes are large and must be interpreted with caution due to time-variation in REGY driven mostly by partner growth, model limitations for forecasting individual countries, and data limitations (e.g., education level in Morocco estimated from partial data).

### Policy implications and recommendations
- Creation of an RTA enlarging market size tends to stimulate FDI inflows; RTA market size has a significantly positive effect on FDI across the sample of 71 developing countries (1980–99).
- Within RTAs, countries with relatively higher education and financial stability attract a larger share of FDI at the expense of other members—creating incentives for “virtuous competition.”
- To share in FDI benefits, each RTA member should improve the domestic business and investment environment to match the best available in the region (education, financial stability, infrastructure).
- Macroeconomic stability and domestic reforms (including privatization programs where applicable) can be pre-conditions or complementary factors to attract FDI after RTA formation.
- The paper does not conclude RTAs are necessarily desirable overall because RTAs may induce costs (e.g., trade diversion). However, dynamic FDI gains from enlarged market size may outweigh static trade-diversion costs in some cases.

### Key statistics and model diagnostics (selected exact figures)
- Sample: 71 developing countries; observations in Table 5: 259, 260, 268 across models; Number of countries = 71.
- First-difference (Table 6) competitive model:
  - dlog(REGY) coefficient (First difference) = 0.51 (t-stat 1.99)*; (2SLS) = 0.49 (t-stat 1.63).
  - dlog(FDI) coefficient = 0.17 (t-stat 5.33)**; (2SLS) = 0.27 (t-stat 3.93)**.
  - Observations = 183 (first-difference OLS), 118 (2SLS); Adjusted R-squared = 0.21 (OLS), 0.11 (2SLS).
- Summary statistics (Table A1, competitive regional model sample, 268 observations):
  - Mean FDI+1 = 11,488; Std. Dev. = 299,371; Min = 2; Max = 292,770.
  - Mean FDI = 6,845; Std. Dev. = 173,871; Min = 1; Max = 133,022.
  - Mean Y = 56,143; Std. Dev. = 107,811; Min = 368; Max = 703,960.
  - Mean REGY = 304,638; Std. Dev. = 478,799; Min = 265; Max = 15,201,514.
  - Mean gY = 0.03; Std. Dev. = 0.04; Min = -0.10; Max = 0.12.
  - Mean REGgY = 0.03; Std. Dev. = 0.03; Min = -0.10; Max = 0.12.
  - Mean educ = 4.7; Std. Dev. = 2.2; Min = 0.5; Max = 10.6.
  - Mean stabfin = 27.1; Std. Dev. = 8.3; Min = 6.8; Max = 48.0.
  - Mean tv = 107.5; Std. Dev. = 111.5; Min = 0.04; Max = 439.0.
  - Mean res (X/Y) = 0.5; Std. Dev. = 0.4; Min = 0.02; Max = 2.5.

*Source: _wp04206 (1999) — econometric analysis and simulations on RTAs and FDI (1980–99).*

### 1.       Ratio of Average Stock of FDI to GDP—Comparison RTA Versus Non-RTA ...........17

### 1.       Ratio of Average Stock of FDI to GDP—Comparison RTA Versus Non-RTA ...........17

### Introduction and focus
- The paper examines the effect of regional trade agreements (RTAs) on foreign direct investment (FDI), focusing on RTAs among developing countries (South-South RTAs).
- Central hypothesis: RTA-created larger regional market size can increase FDI inflows, but effects may differ for interregional versus intraregional FDI and across RTA members depending on locational advantages.
- Sample: 71 developing countries over the period 1980–99.

### Channels through which RTAs affect FDI
- Reduction/elimination of tariff and non-tariff barriers between RTA members and imposition of a common external tariff can:
  - Positively affect interregional FDI through larger market size.
  - Ambiguously affect intraregional FDI (depends on pre-existing investment structure and motives).
- Market size effect:
  - FDI involves non-negligible setup costs and increasing returns to scale; larger markets are more profitable for investors.
  - Firms with intangible assets (R&D-based) face stronger increasing returns to scale; market-seeking FDI benefits most.
- Vertical FDI may be stimulated if output is destined for other RTA members.
- Lowering internal tariffs:
  - Reduces incentives for tariff-jumping FDI.
  - Facilitates vertical FDI that delocalizes production to countries with locational advantages.
  - Should not negatively affect intraregional FDI motivated by intangible assets.
- Efficiency gains from RTAs may raise growth and R&D, further stimulating interregional and intraregional FDI.
- Redistribution effect: FDI increases are likely to concentrate in RTA members with the strongest locational advantages; some members may lose existing FDI.

### Empirical evidence from prior cases (literature)
- European Internal Market Programme (IMP):
  - Dunning (1997): IMP had positive effects on both interregional and intraregional FDI, more on interregional FDI.
  - Pain and Landsbury (1996): Increase of intra-regional FDI by the United Kingdom and Germany (1980–92); little support for pronounced intra-EC concentration, though some countries lost FDI (e.g., France and Belgium).
  - Girma (2001): With IMP implementation, European market size and unit labor costs relative to the EC became more important determinants of FDI than local market size (applies mostly to FDI for acquisitions).
- Mercosur (created 1991: Argentina, Brazil, Paraguay, Uruguay):
  - Strong increases in FDI inflows, particularly into Argentina and Brazil.
  - Macroeconomic stability appeared a pre-condition for FDI responses in Argentina and Brazil.
  - Large privatization programs likely contributed to FDI increases.
  - Inflows were uneven across members; Uruguay and Paraguay had mixed experiences.

### Data sources and measurement
- FDI data from World Investment Report (UNCTAD, 2000).
- FDI definition: “an investment involving a long-term relationship and reflecting a lasting interest and control of a resident entity in one economy in an enterprise resident in an economy other than that of the foreign direct investor.”
- Components of FDI flows: equity capital, reinvested earnings, intra-company loans.
- For most countries, data originate from IMF Balance of Payments Statistics Yearbook.
- FDI stocks are at book value or historical cost; often computed by accumulating flows or adding flows to benchmark estimates.
- Sample split into five-year periods: 1980–85, 1985–90, 1990–95, 1995–99.
- Beginning-of-period FDI stocks and annual average FDI inflows over each sub-period converted into 1995 U.S. dollars.

### Descriptive comparisons: RTA versus non-RTA (key statistics preserved)
- 1980s:
  - RTA countries had somewhat smaller stocks of FDI than non-RTA countries.
- Since 1990:
  - RTA stocks of FDI outgrew those of non-RTA countries by 33 percent on average.
- 1990s:
  - Average stock of FDI for RTA countries: 24 percent of GDP.
  - Average stock of FDI for non-RTA countries: 18 percent of GDP.
- 1980s and 1990s combined:
  - RTAs’ stocks of FDI were on average 3 percentage point larger than non-RTAs’ stocks.
- FDI inflows (average FDI inflows in percent of GDP):
  - Differences between RTA and non-RTA countries are less marked than stocks, but data point to larger FDI inflows into RTA countries at the end of the period (1995–99).
- Regional variation example:
  - ASEAN countries received the most FDI, with an average FDI stock of 44 percent of GDP (period coverage implied from tables referenced).

### Econometric approach and main empirical findings (as described)
- The paper develops econometric models to isolate regional market size effects and control for other FDI determinants.
- Main findings reported in the paper:
  - Significantly positive effect of RTA market size on FDI.
  - Domestic population size also matters, likely via labor supply availability.
  - Relative locational advantages within the RTA are important determinants of FDI received:
    - Higher educational level relative to other RTA members increases FDI received.
    - Higher financial stability relative to other RTA members increases FDI received.
  - Evidence of a negative partial correlation between FDI received by RTA countries and that received by non-RTA countries, possibly indicating diversion effects.

### Simulation and policy implication example
- A simulation of potential FDI benefits from creating an RTA between Algeria, Morocco, and Tunisia:
  - Illustrates that to share in FDI benefits, each RTA member must improve the domestic business and investment environment to match the best available in the region.
- Policy implications emphasized by the analysis:
  - Creation of RTAs can raise FDI, but distribution of gains depends on member countries’ relative locational attributes (education, financial stability, macroeconomic stability).
  - Macroeconomic stability and domestic reforms (including privatization programs where applicable) can be pre-conditions or complementary factors to attract FDI after RTA formation.
  - Members with weaker locational advantages risk losing existing FDI unless they improve competitiveness.

*Source: _wp04206 - 1.       Ratio of Average Stock of FDI to GDP—Comparison RTA Versus Non-RTA ...........17 (IMF working paper content provided).*

### 1999. Their net FDI inflows averaged 4.5 percent of GDP over the 1980s and 1990s. The

### _wp04206 - 1999. Their net FDI inflows averaged 4.5 percent of GDP over the 1980s and 1990s. The

### Major findings on RTAs and FDI
- The most attractive RTA countries (ASEAN) had an average FDI stock of 44 percent of GDP in 1999 and net FDI inflows averaged 4.5 percent of GDP over the 1980s and 1990s.
- Western Hemisphere RTA countries had an average FDI stock of 39 percent of GDP in 1999; their net FDI inflows averaged 2.1 percent of GDP over the 1980s and 1990s, and 4.8 percent over the period 1995–99.
- GCC and sub-Saharan RTA countries had an average FDI stock of 25 percent of GDP in 1999, equal to the average for all non-RTA countries in 1999.
- Over 1980–99, net FDI inflows averaged 1.7 percent of GDP for the GCC and 1.3 percent of GDP for sub-Saharan RTA countries. In the 1990s, GCC net FDI inflows averaged 2.5 percent of GDP and sub-Saharan RTA countries averaged 1.5 percent of GDP.
- Evidence of FDI diversion: coefficients on variables measuring FDI received by RTA countries versus stand-alone countries are negative and significant or close to significant for stock measures, indicating a negative partial correlation between FDI received by RTA countries and stand-alone countries.

### Econometric model (structure and variables)
- Objective: explain the stock of FDI received by a country (expressed in 1995 U.S. dollars).
- Rationale: stock of FDI provides a better measure of actual “inflow of FDI” than increment in the FDI stock.
- Main determinants included:
  - Size of the host market: domestic market size proxied by real GDP (Y, in 1995 U.S. dollars) and expected growth proxied by past real GDP growth (gY). Regional market size (REGY) replaces domestic market size for countries in RTAs; REGY is the sum of domestic market size and market sizes of RTA partners.
  - Agglomeration effects: lagged FDI stock; infrastructure proxied by number of television sets per capita (tv).
  - Labor cost and quality: average schooling years in the population over age 15 (educ) used as proxy for labor quality and inverse proxy for labor cost (data from Barro and Lee (2000)).
  - Business/investment climate: financial risk index (stabfin) of the Political Risk Services Group used (among three risk indices, financial risk index remains significant when introduced simultaneously).
  - Openness: trade openness measured by export-to-GDP ratio corrected for country size (res(X/Y); trade data from Direction of Trade Statistics (IMF, 2000)).
  - Time- and country-specific fixed effects: time effects (ω) and country fixed effects (μ).
- Domestic model specification (abbreviated): ln(FDI) related to ln(Y), gY, ln(educ), ln(stabfin), ln(tv), res(X/Y), lagged ln(FDI), μ, ω, and error term.
- Regional model extends domestic model by adding ln(REGY), REGgY, and GAP variables (GAPeduc, GAPstabfin, GAPtv) measuring locational advantage relative to RTA partners.

### Key econometric results
- Sample: 71 developing countries over 1980–99; time intervals are five-year periods: 1980–85, 1985–90, 1990–95, 1995–99. About 65 percent of observations are from South-South RTA members.
- Domestic model (Table 5, column 1): positive and statistically significant effects on FDI from lagged FDI, education, financial stability, and openness. Infrastructure proxy not significant.
- Regional model (Tables 5 columns 2–3):
  - ln(REGY) (regional market size) has a positive and significant effect on FDI; regional market size shows a robust positive partial correlation with FDI whereas domestic market size shows zero or negative partial correlation (Figures 1 and 2).
  - Competition within RTAs: positive and significant coefficients on GAPeduc and GAPstabfin indicate that higher domestic education and financial stability relative to RTA partners increase a country’s share of FDI within the RTA.
  - REGgY (regional average market growth) is not significant in levels; domestic market growth (gY) is positive and significant.
- First-difference estimation and instrumentation (Table 6):
  - Main result confirmed: dlog(REGY) has a significantly positive effect on dlog(FDI+1) in first-difference estimations.
  - Arellano-Bond instrumentation: coefficient on RTA market size remains significant at the 10 percent level when instrumenting lagged dependent variable and potentially endogenous variables.
- Sensitivity tests (Table 7):
  - When splitting market size into GDP per capita and population, the component of RTA market size that matters is GDP per capita; domestic population positively affects FDI (possible labor-supply effect).
  - Interaction with decade dummies: RTA market size plays a more important role in the 1990s than in the 1980s; domestic market size coefficient changes from positive in the 1980s to negative in the 1990s (estimates imprecise).
  - Competition/diversion tests find negative partial correlations between FDI received by RTA countries and stand-alone countries across alternative FDI measures (Measure 1: FDI inflow in percent of GDP; Measure 2: end-of-period stock; Measure 3: beginning-of-period stock).

### Application to Maghreb countries (Algeria, Morocco, Tunisia)
- Historical FDI performance (Table 8):
  - Tunisia: average stock of FDI close to 80 percent of GDP over the 1980s and 1990s; FDI inflows averaged 2.7 percent of GDP in the 1980s and 1990s.
  - Morocco: stock of FDI averaged 7 percent of GDP in the 1980s and 17 percent of GDP in 1999; FDI inflows averaged 1.1 percent of GDP.
  - Algeria: stock of FDI averaged 4 percent of GDP in the 1980s and 3.5 percent in 1995–99; FDI inflows averaged about zero percent of GDP in the 1980s and 1990s.
- Model prediction accuracy (Table 9):
  - Predicted average FDI stock-to-GDP ratios over 1980–99: Algeria 4 percent (actual 4 percent), Morocco 8 percent (actual 9 percent), Tunisia 79 percent (actual 77 percent).
  - Without country fixed effects, predicted FDI/Y would be smaller: Morocco predicted 4 percent (instead of 9 percent); Tunisia predicted 15 percent (instead of 77 percent) in 1999.
- Predicted vs actual growth of FDI stock-to-GDP ratio (Table 10):
  - Algeria: actual growth 1985–90 = -0.24 percent; predicted = 7.41 percent. 1995–99 actual = -2.91 percent; predicted = 1.04 percent.
  - Morocco: 1985–90 actual = 9.64 percent; predicted = 11.31 percent. 1995–99 actual = 10.67 percent; predicted = 10.39 percent.
  - Tunisia: 1995–99 actual = -5.21 percent; predicted = 5.42 percent.
- Simulated FDI benefits from creating a Maghreb regional market (Table 11, based on 1999 data and elasticities from competitive regional model, column 3 of Table 5):
  - Simulated increase in each country’s FDI stock: Algeria 62 percent, Morocco 85 percent, Tunisia 165 percent.
  - Decomposition of simulated total effect (Table 11):
    - Algeria: Total effect 110 percent; Market size effect 62 percent; Relative education effect 3 percent; Relative financial stability effect 35 percent; other components implied by table.
    - Morocco: Total effect 78 percent; Market size effect 5 percent; Relative education effect -4 percent; Relative financial stability effect 44 percent; other components implied by table.
    - Tunisia: Total effect 215 percent; Market size effect 165 percent; Relative education effect 19 percent; Relative financial stability effect 0 percent.
  - Caveats: magnitudes are large and must be interpreted with caution due to (i) time-variation in REGY in regressions driven mostly by partner growth rather than RTA composition changes, (ii) model limitations for forecasting individual countries, (iii) data limitations (e.g., education level in Morocco estimated from partial data).
  - Policy implication for Morocco: securing simulated FDI gains would benefit from further improvement in education level.

### Policy implications and recommendations
- Creation of an RTA enlarging market size tends to stimulate FDI inflows; RTA market size has a significantly positive effect on FDI across the sample of 71 developing countries (1980–99).
- Within RTAs, countries with relatively higher education and financial stability attract a larger share of FDI at the expense of other members—creating incentives for “virtuous competition.”
- If an RTA is to be created, each member country should strive to bring the quality of its business environment to the best available in the region (education, financial stability, infrastructure) to secure its share of FDI benefits.
- The paper does not conclude that RTAs are necessarily desirable overall because RTAs may induce costs (e.g., trade diversion). However, the dynamic FDI gains from enlarged market size may outweigh static trade-diversion costs in some cases.

### Key statistics and model diagnostics (selected exact figures preserved)
- Sample: 71 developing countries; observations in Table 5: 259, 260, 268 across models; Number of countries = 71.
- Selected coefficients and significance (Table 5, competitive regional model, column 3):
  - year90 coefficient = 0.27 (t-stat 5.20)**.
  - year95 coefficient = 0.59 (t-stat 7.85)**.
  - year99 coefficient = 0.72 (t-stat 6.05)**.
  - log(FDI) coefficient = 0.43 (t-stat 12.85)**.
  - log(REGY) coefficient = 0.60 (t-stat 5.63)**.
  - gY coefficient = 3.02 (t-stat 3.25)**.
  - log(educ) coefficient = 0.34 (t-stat 2.14)*.
  - log(GAPeduc) coefficient = 2.02 (t-stat 6.00)**.
  - log(stabfin) coefficient = 0.04 (t-stat 0.26) [not significant].
  - log(GAPstabfin) coefficient = 0.71 (t-stat 3.74)**.
  - res(X/Y) coefficient = 0.16 (t-stat 1.93) [near significance].
- First-difference (Table 6) competitive model:
  - dlog(REGY) coefficient (First difference) = 0.51 (t-stat 1.99)*; (2SLS) = 0.49 (t-stat 1.63).
  - dlog(FDI) coefficient = 0.17 (t-stat 5.33)**; (2SLS) = 0.27 (t-stat 3.93)**.
  - Observations = 183 (first-difference OLS), 118 (2SLS); Adjusted R-squared = 0.21 (OLS), 0.11 (2SLS).
- Summary statistics (Table A1, competitive regional model sample, 268 observations):
  - Mean FDI+1 = 11,488; Std. Dev. = 299,371; Min = 2; Max = 292,770.
  - Mean FDI = 6,845; Std. Dev. = 173,871; Min = 1; Max = 133,022.
  - Mean Y = 56,143; Std. Dev. = 107,811; Min = 368; Max = 703,960.
  - Mean REGY = 304,638; Std. Dev. = 478,799; Min = 265; Max = 15,201,514.
  - Mean gY = 0.03; Std. Dev. = 0.04; Min = -0.10; Max = 0.12.
  - Mean REGgY = 0.03; Std. Dev. = 0.03; Min = -0.10; Max = 0.12.
  - Mean educ = 4.7; Std. Dev. = 2.2; Min = 0.5; Max = 10.6.
  - Mean stabfin = 27.1; Std. Dev. = 8.3; Min = 6.8; Max = 48.0.
  - Mean tv = 107.5; Std. Dev. = 111.5; Min = 0.04; Max = 439.0.
  - Mean res (X/Y) = 0.5; Std. Dev. = 0.4; Min = 0.02; Max = 2.5.

*Source: _wp04206 (1999) — econometric analysis and simulations on RTAs and FDI (1980–99).*

### References

### References

### Methodology and Econometrics
- Arellano, M., and S. Bond, 1991, “Some Tests of Specification for Panel Data: Monte-Carlo Evidence and an Application to Employment Equations,” Review of Economic Studies, Vol. 58, pp.277–97.
- Sevestre P, and A. Trognon, 1996, “Dynamic Linear Models” in The Econometrics of Panel Data: A Handbook of the Theory with Applications, ed. by L. Matyas and P. Sevestre (Norwell, Mass.: Kluwer Academic Publishers), pp,120–44.
- Caves, R.E. 1996, “Multinational Enterprise and Economic Analysis,” Cambridge Surveys of Economic Literature, Cambridge University Press.

### Foreign Direct Investment and Regional Integration
- Baldwin, R.E., and T. Venables,1995, “Regional Economic Integration,” in Handbook of International Economics, edited by G. Grossman and K. Rogoff, Vol. 3 (New York: North-Holland).
- Blomström, M., S. Globerman, and A. Kokko, 2000, “The Determinants of Host Country Spillovers From Foreign Direct Investment,” CEPR Working Paper No 2350.
- Blomström, M., and A. Kokko, 1997, “Regional Integration and Foreign Direct Investment”, NBER Working Paper No 6019.
- Borenzstein, E., J. De Gregorio, and J.W. Lee, 1995, “How Does Foreign Direct Investment Affect Economic Growth?,” NBER Working Paper No 5057.
- Dunning, J. 1997, “The European Internal Market Programmed and Inbound Foreign Direct Investment,” Journal of Common Market Studies, Vol. 35, Issue 2, June, pp.189–223.
- Michalet, C.A., 1997, “Strategies of Multinationals and Competition for Foreign Direct Investment: The Opening of Central and Eastern Europe,” Foreign Investment Advisory Service Occasional Paper 10 (Washington: The World Bank).
- Pain, N., 1997, “Continental Drift: European Integration and the Location of U.K. Foreign Direct Investment,” The Manchester School of Economic and Social Studies, Vol. 65, Issue 0, Supplement, pp. 94–117.
- Pain, N., and M. Landsbury, 1996, “The Impact of the Internal Market on the Evolution of European Direct Investment,” (unpublished; London National Institute of Economic and Social Research).
- Pain, N., 1997, “Regional Economic Integration and Foreign Direct Investment: The Case of German Investment in Europe,” National Institute Economic Review, Vol. 0, Issue 160, April, pp. 87–99.
- Torrisi, C.R., 1985, “The Determinants of Direct Foreign Investment in a Small LDC,” Journal of Economic Development, Vol.10 No. 2, pp. 29–45.
- Wheeler, D., and A. Mody, 1992, “International Investment Location Decisions, The Case of U.S. Firms,” Journal of International Economics, Vol. 33, Issue 1-2, August, pp. 57–76.
- Chudnovsky D., A. Lopez, and F. Porta, 1995, “New Foreign Direct Investment in Argentina: Privatization, the Domestic Market, and Regional Integration,” in Foreign Direct Investment in Latin America, ed. by M.R. Agosin (Washington: Inter American Development Bank), distributed by John Hopkins University Press, Baltimore.
- Girma, S., 2001, “The Process of European Integration and the Determinants of Entry by Non-EU Multinationals in UK manufacturing”, GEP Research Paper 2001/20, (Nottingham, England: University of Nottingham), September.
- Borenzstein, E., J. De Gregorio, and J.W. Lee, 1995, “How Does Foreign Direct Investment Affect Economic Growth?,” NBER Working Paper No 5057.

### Human Capital, Growth, and Related Determinants
- Barro, R.J., and J.W. Lee, 2000, “International Data on Educational Attainment Updates and Implications,” NBER Working Paper No. 7911 (Cambridge, Mass: National Bureau of Economic Research).
- Barro, R.J., and X. Sala-i-Martin, 1999, “Determinants of Economic Growth: Implications of the Global Evidence for Chile,” Cuadernos de Economia (Chile) 36, No. 107, pp. 443–78.
- Lim, E.G., 2001, “Determinants of, and the Relation Between, Foreign Direct Investment and Growth: A Summary of the Recent Literature,” IMF Working Paper 01/175 (Washington: International Monetary Fund) November.
- Singh H., and K.W. Jun, 1995, “Some New Evidence on Determinants of Foreign Direct Investment in Developing Countries,” World Bank Policy Research Working Paper 1531, (Washington: The World Bank).

### Data Sources and Institutional Reports
- International Monetary Fund, Balance of Payments Statistics Yearbook (Washington: various years).
- ——–, Direction of Trade Statistics (Washington: various years).
- ——–, World Economic Outlook (Washington: various years).
- UNCTAD, 2000, World Investment Report 2000, (New York: United Nations).
- World Bank, 2001, World Development Indicators (Washington).

*Content derived from the document titled "_wp04206 - References".*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04206.pdf_