## _wp11267

## Source details

**Canonical URL:** [_wp11267](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11267.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11267.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11267.pdf.json)

---

### I. Introduction and research questions
- Research focus: business cycle transmission and spillovers from systemically important countries (BRICs: Brazil, Russia, India, China) to low-income countries (LICs).
- Motivations:
  - Rapid growth in trade and financial relations between LICs and BRICs over the past decade.
  - Possible direct channels: trade, FDI, development assistance.
  - Possible indirect channels: global goods and financial markets, commodity prices, cost of financing.
- Main questions:
  - To what extent and through what channels does growth from BRICs spill over to LICs?
  - Do BRICs’ growth and commodity-price effects indirectly impact LIC growth?
  - Did BRICs’ resilience during the global financial crisis help LICs weather the crisis?

### II. Methodology
- Model: global vector autoregression (GVAR) model (Pesaran, Schuermann, and Weiner lineage) for cross-country spillover analysis.
- Complementary model: structural VAR (SVAR) to identify BRIC shocks’ effects on global variables.
- Data/sample:
  - 29 countries: Sub-Saharan Africa 12; Asia 10; Middle East and Europe 4; Latin America 3.
  - Annual data 1970–2009.
  - LIC domestic variables: Growth (log of real GDP, PPP-based, 2000=100), Trade (log of LIC-BRIC trade volume), FDI (BRIC share × total FDI), Inflation (log CPI, 2000=100).
  - BRIC foreign variables: BRIC aggregate GDP (log, PPP-weighted, 2000=100), BRIC aggregate demand (log, weighted average import volume, 2000=100), BRIC real effective exchange rate index (log, 2000=100).
  - Global factors: International crude oil price (US$ per barrel), World commodity price index (2000=100), U.S. Fed Fund rate (percent).
- Estimation strategy (high level):
  - Country-specific VAR(1,1) models with time-varying trade weights w_ij.
  - Identification of cointegration/error-correction relations where relevant.
  - Stack country models into a Global VAR for generalized impulse response functions (GIRFs) and generalized variance decomposition.
  - Use SVAR to identify BRIC effects on global factors; combine SVAR and GVAR results to compute indirect spillovers.
  - Lag length limited to 1 given data constraints.

### III. Principal findings and quantitative results
- Overall conclusion:
  - Overall spillovers from BRICs to LICs are considerable and persistent in the long run.
- Main channels (ranked by strength reported):
  - Trade shocks: strongest channel; account for around 60 percent of the impact of BRICs on LIC growth.
  - Real exchange rate shocks: next most important; BRIC real appreciation improves LIC growth via higher exports to BRICs.
  - FDI from BRICs: limited impact to date.
- Sectoral and regional variation:
  - Spillovers strongest in commodity-exporting LICs.
  - Asia and MNCA (Middle East, North Africa, Central Asia) show largest responses; African LICs also show strong trade responses.
- Crisis-period contribution:
  - BRICs’ resilience during the global crisis may have added 0.3-1.1 percentage points to LIC growth compared with a scenario in which BRIC GDP declined at the same pace as advanced economies.

### IV. Stylized facts on BRICs’ global role and LIC-BRIC linkages
- BRICs’ shares in world totals (period breakdowns given as percent of world total):
  - Population: 44.7, 43.6, 42.8, 41.7 (1991–94, 2000–04, 2005–09, 2015 projections).
  - Labor Force: 47.0, 45.8, 45.4, 44.0 (same period breakdown).
  - GDP (at market exchange rates): 5.8, 8.5, 13.1, 20.7 (same period breakdown).
  - Exports: 4.2, 7.9, 12.4, 18.8 (same period breakdown).
  - Imports: 4.0, 7.0, 10.5, 17.5 (same period breakdown).
- Drivers of LIC–BRIC ties:
  - Economic complementarity (China and India’s demand for natural resources; LICs’ resource endowments and low-cost manufactures).
  - Improvements in macroeconomic management and business climate in many LICs.
  - Global commodity booms partially resulting from BRIC growth.

### V. Empirical correlations and stylized facts (selected numeric excerpts)
- Linear fits reported in Figure 1 (2000–07 average):
  - y = 0.9064x + 4.7524
  - y = 0.0689x + 18.666
  - y = 0.199x + 5.5146
- Appendix Table 1 (unweighted 2000–07 averages; selected country examples):
  - Angola: FDI Inflows 11.5; Trade with BRIC 22.4; Real GDP Growth 11.8
  - Bangladesh: FDI Inflows 0.2; Trade with BRIC 16.9; Real GDP Growth 5.8
  - Mongolia: FDI Inflows 11.6; Trade with BRIC 58.4; Real GDP Growth 6.6
  - Myanmar: FDI Inflows 0.5; Trade with BRIC 24.7; Real GDP Growth 12.9
  - Nigeria: FDI Inflows 11.2; Trade with BRIC 16.0; Real GDP Growth 9.3

### VI. GVAR specification and identification (condensed)
- Individual LIC VAR(1,1) (notation as in source):
  - X_it = c_i + Φ_i X_it-1 + a_i X*_it + Λ_i1 X*_it-1 + Γ_i0 D_t + Γ_i1 D_t-1 + u_it
  - Error-correction form:
    - ΔX_it = c_i + α_i β_i′ (X_it-1) + Λ_i0 ΔX*_it + Ψ_i1 ΔZ_it-1 + Γ_i0 D_t + Γ_i1 D_t-1 + u_it
- Foreign variables:
  - X*_it = Σ_{j=1}^{N+M} w_ij X_jt with country-specific time-varying weights w_ij.
- Identification of impacts:
  - Direct impacts: from GVAR GIRFs of BRIC shocks on LIC fundamentals.
  - Indirect impacts: (a) SVAR to trace BRIC → global variables; (b) combine SVAR responses with GVAR LIC responses to global variables.
  - Total impact = direct + indirect.

### VII. Direct spillovers: quantitative region/channel excerpts
- Trade shocks (one-standard error increase in volume of total BRIC imports from LICs, in logs):
  - Trade shocks account for around 60 percent of BRICs’ impact on LIC growth.
  - Table 2 generalized impulse responses (selected "Overall" values):
    - Africa: to BRIC overall trade shocks 0.93; technology shocks 0.09; real exchange rate shocks 0.47; FDI shocks 0.37.
    - Asia: to BRIC overall trade shocks 1.25; technology shocks 0.81; real exchange rate shocks 0.65; FDI shocks 0.09.
    - Oil exporter countries: to BRIC overall trade shocks 1.51; technology shocks 0.36; real exchange rate shocks 1.08; FDI shocks 0.44.
- Real exchange rate shocks:
  - Positive BRIC appreciation generally associated with positive LIC growth effects, especially for oil exporters; magnitudes smaller than trade shocks.
- Technology shocks:
  - Generally significant; lead to higher long-run LIC growth with sometimes negative short-term responses.
- FDI shocks:
  - Modest impact relative to other channels; potential for larger impacts in LICs receiving larger BRIC FDI.

### VIII. Indirect spillovers via global variables
- Global variables considered: world oil prices, other commodity prices, US Fed Fund rate, global demand.
- Key numeric findings:
  - Roughly one-third of changes in world oil prices can be attributed to shocks originating in BRICs.
  - Less than 10 percent of the change in the U.S. Fed Fund rates is explained by shocks from BRICs.
- Table 3 (average contributions, 10 years ahead; selected "Overall" entries in percent):
  - Global Demand: 20.4
  - World Oil Prices: 33.5
  - Other Global Commodity Prices: 17.6
  - US Fed Fund Rates: 8.1
- Indirect impacts on LICs (Table 4; selected "Overall" entries):
  - Africa: Total indirect 0.09; World Oil Prices contribution 0.05; Other Commodity Prices 0.04.
  - Asia: Total indirect 0.04.
  - Oil Exporting Countries: Total indirect 0.12; World Oil Prices 0.05; Other Commodity Prices 0.07.
- Interpretation:
  - Indirect spillovers are positive and meaningful but mostly smaller than direct spillovers.
  - Indirect effects largest via world oil and other commodity prices in the short run.
  - African LICs and oil exporters receive stronger indirect spillovers.

### IX. Total spillovers (direct + indirect) — illustrative numeric results
- Overall impact of a 1 percentage point increase in BRICs’ demand and productivity:
  - Cumulative increase in LICs’ output: 0.7 percentage point over 3 years and 1.2 percentage point over 5 years.
  - Pre-2007 comparison: 0.5 percentage point over 3 years and 0.6 percentage point over 5 years.
- Region-specific totals (Table 5; "Overall" entries):
  - Africa: Total impact 0.17; Direct 0.09; Indirect 0.09.
  - Asia: Total impact 0.85; Direct 0.81; Indirect 0.04.
  - Europe and Middle-East: Total impact 0.78; Direct 0.79; Indirect -0.01.
  - Latin America: Total impact 0.45; Direct 0.38; Indirect 0.07.
  - Oil Exporting Countries: Total impact 0.48; Direct 0.36; Indirect 0.12.
  - Other Commodity Exporters: Total impact 0.27; Direct 0.24; Indirect 0.04.
- Note: totals refer to demand and technology shocks in BRICs and derive from impulse response functions; methodological distinctions mean tables are not mechanically summable.

### X. BRICs’ role during the global financial crisis
- Pre-crisis: BRICs explained roughly 20 percent to 30 percent of changes in LICs’ output growth (generalized variance decomposition).
- During crisis: BRIC contributions to LIC growth increased substantially across regions; largest increases in African and Asian LICs and in commodity exporters.
- Counterfactual simulation:
  - "LIC growth would have been 0.3 percentage point to 1.1 percentage points lower during the crisis if BRICs’ GDP growth had declined at the same pace as advanced economies."
  - Method: panel growth regression of LIC growth on short-run determinants including trade-weighted trading-partner growth; scenario holds BRIC growth declining like Advanced Economies.

### XI. Policy implications and recommendations
- Growing LIC–BRIC linkages:
  - Can alter short-run volatility and contribute to higher sustainable growth in the long run.
  - As BRIC cycles are not fully synchronized with advanced economies, ties with BRICs may help dampen LIC growth volatility.
- For LIC policymakers:
  - Pay increased attention to macroeconomic developments in BRICs and other dynamic EMEs when assessing macro policy stance and growth potential.
  - Recognize heterogeneous bilateral relations; country-specific analysis required for policy design.

### XII. Caveats, limitations, and robustness notes
- Results should be treated as broad orders of magnitude rather than definitive point estimates.
- Limitations include limited time series length, frequent structural breaks in LIC and BRIC economies, and heterogeneity across countries.
- The approach in counterfactuals focuses on BRIC growth changes and does not endogenously account for associated commodity price, interest rate, or capital flow adjustments.

*Source — _wp11267 (IMF working paper chapter excerpts).*

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

### _wp11267 - References

### I. Introduction and Research Questions
- Research context: business cycle transmission and spillovers from systemically important countries to the rest of the world, with limited attention to transmission to low-income countries (LICs) from major emerging market economies such as BRICs (Brazil, Russia, India, and China).
- Motivations:
  - Rapid growth in trade and financial relations between LICs and BRICs over the past decade.
  - Potential for BRICs to affect LIC growth directly (trade, FDI, development assistance) and indirectly (global goods and financial markets, commodity prices, cost of financing).
- Main questions examined:
  - To what extent and through what channels does growth from BRICs spill over to LICs?
  - Given LICs’ high dependence on global demand and world commodity prices, does growth in BRICs have any indirect impact on LIC growth?
  - To what extent, if any, did BRICs’ resilience during the global financial crisis help LICs weather the crisis?

### II. Methodology
- Model employed: global vector autoregression (GVAR) model to examine growth spillovers from BRICs to LICs.

### III. Principal Findings and Quantitative Results
- Overall conclusions:
  - Overall spillovers from BRICs to LICs are considerable and persistent in the long run.
- Channel-specific impacts (in order of strength reported):
  - Trade shocks from BRICs exert the strongest effect on growth in LICs.
  - Exchange rate shocks are the next most important, with real appreciation of BRIC currencies improving LIC growth through higher exports to BRICs.
  - BRIC FDI to LICs appears to have limited impact on LIC growth thus far.
- Sectoral and regional variations:
  - BRIC spillovers are strongest in commodity-exporting LICs, reflecting the importance of commodities in LIC-BRIC trade relations.
- Indirect/global effects:
  - Demand shocks from BRICs exert significant influence over global commodity prices and global demand.
  - Global oil price and demand are the most affected global variables by shocks originating in BRICs, translating into significant spillovers to growth in many LICs.
- Crisis-period contribution:
  - The resilience of the BRIC economies during the global crisis may have added 0.3-1.1 percentage points to LIC growth compared with a scenario in which BRIC GDP had declined at the same pace as advanced economies.

### IV. Stylized Facts on BRICs’ Role and LIC-BRIC Linkages
- BRICs’ global economic weight (period averages and projections cited):
  - Population: BRICs 44.7, 43.6, 42.8, 41.7 (1991–94, 2000–04, 2005–09, 2015 projections respectively; in percent of world total).
  - Labor Force: BRICs 47.0, 45.8, 45.4, 44.0 (same period breakdown).
  - GDP (at market exchange rates): BRICs 5.8, 8.5, 13.1, 20.7 (same period breakdown; in percent of world total).
  - Exports: BRICs 4.2, 7.9, 12.4, 18.8 (same period breakdown; in percent of world total).
  - Imports: BRICs 4.0, 7.0, 10.5, 17.5 (same period breakdown; in percent of world total).
- Key points on BRICs in the world economy:
  - With a combined labor force of more than 1 billion people, BRICs have potential to be key global players.
  - Since the early 1990s, BRICs have more than doubled their share in global output and are now the third largest after the United States and the Euro Area.
  - BRICs’ GDP (based on market exchange rates) is projected to surpass that of the Euro area before 2015 (as per IMF projections cited).
  - BRICs’ share in world exports nearly tripled over the past two decades, overtaking that of the United States and catching up with the Euro area.
- Drivers of growing LIC-BRIC ties:
  - Economic complementarity (China and India’s demand for natural resources; LICs’ resource endowments and supply of competitively priced manufactures).
  - Improvements in macroeconomic management and business climate in many LICs.
  - Global commodity booms partially resulting from BRIC growth.

### V. Stylized Facts and Correlations between LICs and BRICs
- FDI:
  - FDI inflows from BRICs to LICs appear to be positively correlated with LIC growth.
  - Correlation is stronger for most Asian LICs and Sub-Saharan African countries, particularly resource-rich countries such as Angola, Nigeria, Zambia, and the Republic of Congo.
  - A large part of BRIC FDI (mostly China) to LICs is concentrated in natural resources and infrastructure.
- Trade:
  - LIC trade with BRICs appears to be positively correlated with LIC growth.
  - The correlation between BRIC-LIC trade and LIC growth is strongest for many Asian LICs, followed by some Sub-Saharan African countries, including Angola, Nigeria, Ethiopia, Uganda, and Tanzania.
- FDI–Trade relationship:
  - The correlation between FDI from BRICs and LIC trade with BRICs is insignificant.
  - Interpretation: likely reflects the early stage of BRIC FDI in LICs and the dominance of BRIC economic growth (rather than FDI) in driving trade; expected spillover channels from FDI to exports may not yet be well established.

### VI. Policy Implications
- BRICs can alter the volatility of LIC economies in the short run and contribute to their sustainable growth rates in the long run.
- Linkages with dynamic emerging market economies (EMEs) similar to BRICs could have comparable impacts on LICs.
- Recommendation for LIC macroeconomic assessment:
  - Greater attention should be paid to LICs’ linkages with BRICs and other EMEs, via both direct and indirect channels, when assessing macroeconomic policy stance and growth potential.

*Italic: Source — _wp11267 - References (excerpt provided). *

### 4. BRIC-LIC ties have significant regional dimensions as well as reflect the relative size

### 4. BRIC-LIC ties have significant regional dimensions as well as reflect the relative size of individual BRICs.

### Regional patterns of BRIC-LIC linkages
- BRIC trade ties with Africa and Asian LICs are far stronger than with Latin American and Eastern Europe LICs.
- Among BRICs:
  - China is by far the largest trade partner for LICs in Africa, Asia, Middle East and North Africa.
  - Brazil is a dominant partner in Latin America.
  - Russia’s prominence is seen in Eastern Europe.

### Empirical evidence from figures (2000–08, 2000–07 averages)
- Figure 1 (2000–07 average) relates:
  - "Trade with BRIC, as % of the total trade" to "Real GDP growth rate, %".
  - "FDI inflows from BRIC, as % of the country's GDP" to "Trade with BRIC, as % of the total trade".
  - "FDI inflows from BRIC, as % of the country's GDP" to "Real GDP Growth Rate, %".
- Reported linear fits shown in Figure 1:
  - y = 0.9064x + 4.7524
  - y = 0.0689x + 18.666
  - y = 0.199x + 5.5146
- Figure 2. LICs’ Trade with BRICs by Region, 2000–08 (In millions of US dollars) presents time series of exports and imports (2000–2008) for:
  - Asia: Export and Imports — Brazil, Russia, India, China (scale up to 120,000)
  - Africa: Export and Imports — Brazil, Russia, India, China (scale up to 100,000)
  - Middle East, North Africa, and Eastern Europe: Export and Imports — Brazil, Russia, India, China (scale up to 35,000)
  - Latin America: Export and Imports — Brazil, Russia, India, China (scale up to 12,000)
- Figure 3. LICs’ Trade with BRICs by Type of Exporters, 2000–08 (In millions of US dollars) presents time series (2000–2008) for:
  - Oil Exporters: Export and Imports — Brazil, Russia, India, China (scale up to 90,000)
  - Commodity Exporters: Export and Imports — Brazil, Russia, India, China (scale up to 12,000)
  - Non-Commodity Exporters: Export and Imports — Brazil, Russia, India, China (scale up to 160,000)
  - All Low-income countries: Export and Imports — Brazil, Russia, India, China (scale up to 300,000)

### The GVAR model: rationale and advantages
- Motivation:
  - Need for a consistent identification technique to model international spillovers of shocks.
  - Limitations of alternatives: panel data (one-size-fits-all, endogeneity), single-country VARs (parameter proliferation), panel VARs (limited cross-country heterogeneity control), large-scale macro models (many behavioral equations), dynamic factor models (atheoretical, lack structural identification, residual cross-country interdependencies remain).
- Choice:
  - Use a Global Vector Autoregression (GVAR) model (Pesaran, Schuermann, and Weiner (PSW) lineage) to analyze BRIC spillovers to LICs.
- Key features of the GVAR approach:
  - Multivariate, multicountry framework investigating cross-country and regional interdependency.
  - Links individual countries/regions via observed country-specific foreign variables (trade-weighted foreign variables).
  - Minimizes parameter requirements while covering many geographical areas.
  - Allows backward (elasticities) and forward (impulse response functions and variance decomposition) analysis.
  - Allows: (i) domestic variables related to trade-weighted foreign variables; (ii) non-zero pair-wise correlation in residuals between countries/equations; (iii) introduction of common observed shocks; (iv) control of country idiosyncratic factors.

### GVAR specification (as presented)
- Notation and components:
  - Consider N LICs and M BRIC countries (M = 4).
  - X_it denotes a k_i x 1 vector of LIC domestic (endogenous) variables.
  - X*_it denotes a k*_it x 1 vector of BRIC variables (country-specific foreign variables).
  - D_t denotes an m x 1 vector of common global factor variables.
  - X*_it and D_t are assumed weakly exogenous.
- Individual LIC country VAR(1,1) model (equation (1)):
  - X_it = c_i + Φ_i X_it-1 + a_i X*_it + Λ_i1 X*_it-1 + Γ_i0 D_t + Γ_i1 D_t-1 + u_it
  - Where:
    - c_i is a k_i x 1 vector of fixed intercepts;
    - Φ_i is a k_i x k_i matrix of coefficients on lagged domestic variables;
    - Λ_i0 and Λ_i1 are k_i x k*_i matrices of coefficients on contemporaneous and lagged foreign variables, X*_it and X*_it-1;
    - Γ_i0 and Γ_i1 are k_i x m matrices of coefficients on contemporaneous and lagged global factor variables, D_t and D_t-1;
    - u_it is a k_i x 1 vector of country-specific shocks, serially uncorrelated with zero mean and non-singular covariance matrix Σ_ii; u_it ~ i.i.d.(0, Σ_ii).
  - Idiosyncratic shocks u_it are allowed to be correlated across countries/regions to a limited degree, i.e., for t = t′, E[u_it u_jt′′′] = Σ_ij; for t ≠ t′, 0.
- Error-correction representation (equation (2)):
  - ΔX_it = c_i + α_i β_i′ (X_it-1) + Λ_i0 ΔX*_it + Ψ_i1 ΔZ_it-1 + Γ_i0 D_t + Γ_i1 D_t-1 + u_it
  - Where β_i is a (k*_i x r_i) matrix of cointegrating vectors, rank r_i, etc.
- Weights and foreign variables:
  - Country-specific weights w_ij are used (PSW approach) instead of common weights to construct cross-country averages and control for the relative importance of country j for country i.
  - Foreign variable construction formula (PSW-style):
    - X*_it = Σ_{j=1}^{N+M} w_ij X_jt  (time-varying weights)

### Estimation strategy and steps
- Lag length choice:
  - Given data limitations and parameter count, lag length is limited to 1.
- Econometric modeling steps:
  - Step 1: Estimate country-specific small-dimensional models (VAR(1,1)) for each LIC to obtain direct spillover estimates from BRICs and global factors.
  - Step 2: Identify long-run (cointegrating) and error correction relations for selected countries.
  - Step 3: Stack and solve all estimated coefficients from country-specific models in one large system (Global VAR) for generalized impulse response function (GIRF) analysis.
  - Step 4: Estimate an additional structural VAR (SVAR) including BRIC country-specific foreign variables (GDP, trade, real exchange rates, FDI) and identified global factors (global demand, international oil prices, global commodity prices, U.S. Fed Fund rates) to pin down direct impacts of BRICs on global factors and thus indirect spillovers to LICs.
- Identification of direct, indirect, and total impacts:
  - Direct impacts: Obtained from GVAR GIRFs of identified shocks in BRICs on LIC fundamentals (GDP, trade, inflation, real exchange rates).
  - Indirect impacts: Two-step approach:
    - (a) Use SVAR to identify how BRIC shocks affect global factors (e.g., world oil price).
    - (b) Combine SVAR impulse responses with LIC responses to global factors from the GVAR to compute indirect effects.
  - Total impact: Sum of direct and indirect effects.
- Assessing BRICs’ alleviating effects during the global financial crisis:
  - Compute Generalized Variance Decomposition of the GVAR in-sample (1972–2009, including a global financial crisis period dummy) and out-of-sample (1972–2007, excluding crisis period).
  - Calculate relative contributions of (i) domestic variables, (ii) BRIC spillover variables, and (iii) global factor variables to changes in LIC output.
  - Complement with simulation from a model linking LIC growth with trading partner growth to estimate the extent BRICs may have alleviated the global financial crisis impact.

### Data and sample
- GVAR estimated for 29 countries:
  - Sub-Saharan Africa: 12
  - Asia: 10
  - Middle East and Europe: 4
  - Latin America: 3
- Annual data covering 1970 through 2009.
- Individual LIC variables constructed:
  1. Growth: log of real GDP, PPP-based, 2000=100
  2. Trade: log of LIC-BRIC trade volume
  3. FDI: FDI from BRICs proxied by time-varying shares of FDI from BRICs to LICs multiplied by total FDI received by each country
  4. Inflation: log of the CPI, 2000=100
- BRIC (country-specific foreign) variables constructed:
  1. BRIC aggregate GDP (log), weighted by PPP-based average, 2000=100
  2. BRIC aggregate demand (log), weighted average import volume (2000=100)
  3. BRIC real effective exchange rate index (log), 2000=100
- Global factors:
  1. International crude oil price (US$ per barrel)
  2. World commodity price index (2000=100)
  3. U.S. Fed Fund rate (percent)

*Source: IMF working paper chapter titled "4. BRIC-LIC ties have significant regional dimensions as well as reflect the relative size of individual BRICs."*

### 4. World import volume (log), excluding BRIC imports, 2000=100

### 4. World import volume (log), excluding BRIC imports, 2000=100

### Variable selection and identification
- Technology: proxied by shocks to BRICs’ GDP per capita. The Ricardian model link: technology disturbances affect the marginal product of factors, investment opportunities, and trade patterns.
- Trade: Heckscher-Ohlin-Samuelson and Stolper-Samuelson frameworks imply comparative advantage from factor endowments; trade linkages can generate demand- and supply-side spillovers and affect output synchronization.
- FDI: shifts capital across countries; LIC–BRIC ties could reduce investment correlations and raise LIC investment growth, but empirical evidence on FDI growth effects is inconclusive and may involve threshold effects (Kose et al., 2009b; Dabla-Norris et al., 2010).
- Real exchange rates: movements in BRICs’ real exchange rates affect relative prices of tradables vs nontradables and thereby LIC exports, trade patterns, and business synchronization.
- Financial linkages: proxied by the U.S. Fed Fund Rate. Rising financial linkages can increase business cycle co-movement via cost of capital and wealth effects, but also could reduce cross-country output correlations through reallocation consistent with comparative advantage.
- World global demand proxy: world imports (excluding BRIC imports). A VAR is run replacing the U.S. Fed Fund rate with the global demand variable to preserve degrees of freedom.

### Direct spillovers (BRICs → LICs via four channels)
- Overall: Direct spillovers from BRICs generally produce significant improvements in LIC output over time, with variation across channels and regions.
- Trade shocks:
  - Trade shocks dominate among channels and account for around 60 percent of the impact of BRICs on LIC growth.
  - Trade shocks lead to positive short- and long-run growth effects for most regions; exceptions: European and Middle East LICs show negative short-run responses (possible third-market competition).
  - Greatest responses in LICs in Asia and MNCA (Middle East, North Africa, and Central Asia); African LICs show a strong response as well.
  - Overall impact on oil exporters is almost twice as large as on non-oil exporters.
  - Note: trade shock defined as a one-standard error increase in the volume of total BRIC imports (in logs) from LICs.
- Real exchange rate shocks:
  - Positive BRIC real exchange rate shock (appreciation) is associated with positive growth effects on LICs, especially oil exporters.
  - Short-run reduced growth associated with BRIC appreciation in LTNC LICs (Latin America and the Caribbean), possibly reflecting limited BRIC-destined exports and higher import costs.
  - Statistically significant but magnitude small compared with trade shocks.
- Technology shocks:
  - Generally significant and lead to higher long-run growth in LICs; short-term responses sometimes negative but turn positive over longer horizon.
  - Asian LICs and MNCA benefit particularly; non-oil commodity exporters also benefit.
- FDI shocks:
  - Impact is modest relative to other channels even in the long run, possibly reflecting relatively small volumes of BRIC FDI in LICs and the need for threshold conditions for larger growth effects.
  - Impact could be larger in individual LICs receiving larger BRIC FDI volumes.

- Quantitative excerpts from Table 2 (selected values; generalized impulse responses of LICs' Real GDP Growth to identified (+1 s.e.) shocks):
  - Africa: to BRIC overall trade shocks 0.93 (Overall), to BRIC technology shocks 0.09 (Overall), to BRIC real exchange rate shocks 0.47 (Overall), to FDI shocks from BRIC 0.37 (Overall).
  - Asia: to BRIC overall trade shocks 1.25 (Overall), to BRIC technology shocks 0.81 (Overall), to BRIC real exchange rate shocks 0.65 (Overall), to FDI shocks from BRIC 0.09 (Overall).
  - Oil exporter countries: to BRIC overall trade shocks 1.51 (Overall), to BRIC technology shocks 0.36 (Overall), to BRIC real exchange rate shocks 1.08 (Overall), to FDI shocks from BRIC 0.44 (Overall).
  - Europe and Middle-East, Latin America, Other commodity exporters: see Table 2 for region-specific numbers.

### Indirect spillovers (BRICs → global variables → LICs)
- Approach: (i) use an SVAR to estimate BRIC shocks’ impacts on global variables (technology and demand shocks); (ii) feed those impacts into the GVAR to assess LIC spillovers.
- Four global variables considered: world oil prices, other commodity prices, global interest rates (U.S. Federal Fund Rate), and global demand.
- Key findings:
  - BRIC demand shocks have larger impacts on global variables than BRIC technology shocks.
  - Effects of technology shocks tend to die down in the long run; demand shocks remain strong in the long run.
  - Largest impacts on global variables in the short run are on world oil and other commodity prices; impacts on global demand and interest rates are generally small or negligible.
  - Roughly one-third of changes in world oil prices can be attributed to shocks originating in BRICs.
  - Less than 10 percent of the change in the U.S. Fed Fund rates is explained by shocks from BRICs.
- Quantitative contributions of BRICs to global variables (Table 3, Average, in percent, 10 years ahead; Total BRIC imports of goods and services; Total World imports excluding BRIC imports):
  - Global Demand: 20.4 (Overall), 2.3, 18.1, 79.6 (of which: global demand component breakdown shown).
  - World Oil Prices: 33.5 (Overall), 14.2, 19.3, 66.5.
  - Other Global Commodity Prices: 17.6 (Overall), 6.2, 11.4, 82.4.
  - US Fed Fund Rates: 8.1 (Overall), 2.7, 5.4, 91.9.
- Indirect impacts on LICs (Table 4, Estimated Indirect Response of LIC Outputs to Shocks from BRIC):
  - Africa: Total indirect (through) 0.09 (Overall), World Oil Prices contribution 0.05 (Overall), Other Commodity Prices 0.04 (Overall).
  - Asia: Total indirect (through) 0.04 (Overall).
  - Oil Exporting Countries: Total indirect (through) 0.12 (Overall), World Oil Prices 0.05 (Overall), Other Commodity Prices 0.07 (Overall).
  - Other regions have region- and channel-specific indirect impacts; see Table 4 for detailed numbers.

- Interpretation:
  - Indirect spillovers are generally positive and meaningful but mostly smaller than direct spillovers.
  - Indirect spillovers through world oil and other commodity prices are largest in the short run.
  - African LICs and oil exporters receive stronger indirect spillovers.
  - In some cases indirect impacts are comparable to direct impacts (e.g., Africa for technology shocks; oil exporters: indirect technology impact ≈ one-third of direct).

### Total spillovers (direct + indirect; BRIC demand and productivity shocks)
- Overall impact of a 1 percentage point increase in BRICs’ demand and productivity:
  - Cumulative increase in LICs’ output: 0.7 percentage point over 3 years and 1.2 percentage point over 5 years.
  - Pre-2007 period comparison: a 1 percentage point increase in BRICs’ demand and productivity would change LICs’ output by about 0.5 percentage point over 3 years and 0.6 percentage point over 5 years.
- Region-specific totals (Table 5; Overall, Short run, Long run):
  - Africa: Total impact 0.17 (Overall); Direct 0.09 (Overall); Indirect 0.09 (Overall).
  - Asia: Total impact 0.85 (Overall); Direct 0.81 (Overall); Indirect 0.04 (Overall).
  - Europe and Middle-East: Total impact 0.78 (Overall); Direct 0.79 (Overall); Indirect -0.01 (Overall).
  - Latin America: Total impact 0.45 (Overall); Direct 0.38 (Overall); Indirect 0.07 (Overall).
  - Oil Exporting Countries: Total impact 0.48 (Overall); Direct 0.36 (Overall); Indirect 0.12 (Overall).
  - Other Commodity Exporters: Total impact 0.27 (Overall); Direct 0.24 (Overall); Indirect 0.04 (Overall).

- Notes:
  - Total impacts reported are confined to demand and technology shocks in BRICs and derived from impulse response functions.
  - Differences exist between these totals and direct-spillover estimates due to methodological distinctions; sums across tables should not be mechanically aggregated.

### BRICs’ role during the crisis
- BRIC contributions to LIC growth during the global financial crisis were significant and rose relative to pre-crisis levels.
- Before the crisis, BRICs explained roughly 20 percent to 30 percent of changes in LICs’ output growth (generalized variance decomposition).
- During the crisis, BRIC contributions increased substantially across regions, with largest increases in African and Asian LICs, and in oil and other commodity exporters.
- Domestic factors also contributed to GDP growth in African and Asian LICs during the crisis, reflecting improved macroeconomic management and policy buffers built prior to the crisis.
- Counterfactual analysis setup (Figure 6): WEO Baseline (Actual & Projections) versus alternative scenario assumes that BRICs' growth declines at the same rate as that of Advanced Economies (2.5 percent in 2008,...).

*Source: Authors' calculations (from the IMF working paper chapter "4. World import volume (log), excluding BRIC imports, 2000=100").*

### 3.5 p ercen t in  2009).

### _wp11267 - 3.5 p ercen t in  2009).

### Key findings
- BRICs’ rapid growth and integration into the world economy have substantially strengthened their trade and financial ties with LICs.
- Direct spillovers from BRICs to LICs are significant and persistent, with bilateral trade identified as the most powerful channel of transmission.
- Exchange rate movements, productivity innovations in BRICs, and FDI flows from BRICs also transmit spillovers, albeit more modestly.
- Indirect spillovers (through global demand, international commodity prices, and world interest rates) are significant but generally much smaller than direct spillovers.
- Commodity-exporting LICs and LICs in Asia experience the strongest direct impacts from BRIC shocks; African LICs also show strong responses reflecting rapid expansion of trade with BRICs in commodities.
- BRICs’ mild slowdown and quick recovery during the global financial crisis helped cushion LICs’ growth; growing bilateral trade played a key role.
- Average LICs growth was only about 7 percent in 2008-09; the estimated BRIC contribution materially mitigated downward pressure on LIC growth in those two years.

### Simulation exercise and quantitative results
- A counterfactual simulation indicates: "LIC growth would have been 0.3 percentage point to 1.1 percentage points lower during the crisis if BRICs’ GDP growth had declined at the same pace as advanced economies."
- The scenario used a panel growth regression of growth in LICs on short-run determinants including external demand measured as the trade-weighted growth of trading partners.
- Caveat noted in the source: the approach considers only the impact of slower growth in BRICs and does not account for associated changes such as commodity prices, interest rates, or capital flows.

### Model framework and methods
- Employed a GVAR model together with a structural VAR (SVAR) model to assess spillovers from BRICs to LICs.
- Used generalized variance decomposition techniques to analyze changing roles of BRICs in LIC growth and output fluctuations.
- Conducted generalized impulse response and variance decomposition analyses to quantify direct and indirect impacts of BRIC productivity and demand shocks on LIC output.

### Channels of transmission (as characterized)
- Direct channels:
  - Bilateral trade (primary channel).
  - Exchange rate movements.
  - Productivity innovations in BRICs.
  - FDI flows from BRICs.
- Indirect channels:
  - Global demand.
  - International commodity prices (notably world oil prices).
  - World interest rates (generally small or negligible impacts).

### Appendix summary: selected numeric indicators and results excerpts
- Table excerpts and statistics preserved in source include:
  - Variance decomposition and contribution tables covering periods 1972–2009 and 1972–2007.
  - Appendix Table 1: Unweighted average in 2000–07 for inward FDI flows to GDP ratio and LIC trade with BRICs as percent of total trade for many LICs (examples from the table as presented):
    - Angola: FDI Inflows 11.5; Trade with BRIC 22.4; Real GDP Growth 11.8
    - Bangladesh: FDI Inflows 0.2; Trade with BRIC 16.9; Real GDP Growth 5.8
    - Mongolia: FDI Inflows 11.6; Trade with BRIC 58.4; Real GDP Growth 6.6
    - Myanmar: FDI Inflows 0.5; Trade with BRIC 24.7; Real GDP Growth 12.9
    - Nigeria: FDI Inflows 11.2; Trade with BRIC 16.0; Real GDP Growth 9.3
  - Appendix Table 3 (SVAR impulse responses of global factors) reports numeric responses over 10 years ahead for:
    - World oil prices, world commodity prices, US Fed rate, global demand (overall, short-run, long-run impacts; example entries: World oil prices overall impact 0.06; Standard deviation 0.07; US Fed rate overall impact 0.00; Standard deviation 0.35).
  - Appendix Table 4 (BRIC shocks to LICs: Indirect Impacts) reports generalized impulse response of LICs' output to a one-percentage shock with region- and sample-specific entries (sample cell examples as presented):
    - Africa, First year, In-sample 0.0089; Out-of-sample 0.0099
    - Asia, Third year, In-sample 0.0270; Out-of-sample 0.0290
    - Europe & Middle-East, Third year, In-sample 0.0841; Out-of-sample 0.0832
    - Latin America, First year, In-sample 0.0582; Out-of-sample 0.0647
    - Oil exporters, Third year, In-sample 0.0225; Out-of-sample 0.0281
  - Appendix Table 5: Variance decomposition reported forecast error variance of LICs' real GDP growth explained by domestic factors, BRIC demand and technology, and global factors (detailed numeric tables in source).

### Policy implications and forward-looking points
- Increasing LIC-BRIC trade and financial ties will further strengthen business cycle synchronization over time.
- As long as BRIC cycles are not fully synchronized with advanced economies, growing ties with BRICs should help dampen LIC growth volatility.
- LIC policymakers should pay increased attention to macroeconomic developments in BRICs when assessing domestic macroeconomic policy stance.
- Continued expansion of BRIC economies could alter LICs’ long-run growth potential, serving as new drivers of growth for LICs.

### Caveats and limitations highlighted
- Estimated spillovers should be treated as broad orders of magnitude rather than definitive estimates.
- Limitations include limited length of time series and frequent structural breaks in LIC and BRIC economies, reducing robustness.
- Heterogeneity: neither BRICs nor LICs are homogeneous; bilateral relations vary greatly across countries, requiring country-specific examination for policy implications.

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11267.pdf*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11267.pdf_