## 11. List of countries

## Source details

**Canonical URL:** [11. List of countries](https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020248-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2020/english/wpiea2020248-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2020/english/wpiea2020248-print-pdf.pdf.json)

---

### Introduction and objective
- Political stability fosters sustainable growth and social well-being; import food price shocks can trigger socio-political instability.
- Paper objective: assess whether fiscal policy and remittances mitigate the impact of import food price shocks on socio-political instability.
- Sample: 101 low- and middle-income countries over 1980-2012.

### Data, variables, and empirical approach
- Food price shocks:
  - Constructed using IMF WEO prices and FAO trade quantities; country-specific weighted price index across wheat, sugar, soybeans, soybean oil, maize, and rice.
  - Main shock variable: residual from country-by-country regression of ln(price index) on trend and first and second lags (equation (3)).
  - Alternative shock: frequency of positive residuals within non-overlapping four-year periods.
- Fiscal policy:
  - Two measures:
    - (i) annual growth rate of government consumption expenditure as share of GDP;
    - (ii) fiscal stimuli proxy based on residuals from country regressions of government consumption on cyclical GDP growth (Cycle_GDPgrowth, Hodrick-Prescott λ = 6.25), inflation, and inflation squared, then a 4-year rolling standard deviation dummy.
- Socio-political instability (dependent variables):
  - Civil conflict: dummy = 1 if intrastate conflict with more than 25 battle deaths in a year.
  - Conflict intensity: number of conflicts with more than 25 battle-related deaths in a year.
  - Expropriations: dummy = 1 if systemic elimination of political opposition (jailing/execution) in a year.
  - Government crisis: dummy = 1 for rapidly evolving contexts threatening regime downfall (excluding revolt aimed at overthrow).
- Controls:
  - GDP per capita growth (first lag instrumented), IMF-World Bank poverty reduction program dummy, macro-financial crises (Systemic Banking Crisis, Currency Crisis, Sovereign Debt Crisis, Sovereign Debt Restructuring).
- Estimators:
  - Probit for binary outcomes; maximum likelihood Tobit for conflict counts.
  - IV Probit and IV Tobit (instruments: lagged remittances and lagged GDP per capita growth) and two-stage least squares for robustness.
  - Standard errors clustered at country level; Probit standard errors bootstrapped (250 replications) for some checks.

### Key empirical findings
- Main effects of food price shocks:
  - Positive food price shocks significantly increase the probability of civil conflict, government crises, and expropriations; increase conflict intensity.
  - Quantitative example: "a deviation of the food price of one unit from its long run trend leads to an increase of 0.8 to 4 % in the likelihood of civil conflict".
  - Baseline coefficient estimates (selected from Table 2) include shock coefficients: 0.604**, -1.288, 4.425***, 0.696, 5.152***, 8.035*** (t-statistics reported in table).
- Role of remittances:
  - Remittances dampen the adverse effect of food price shocks on the likelihood of civil conflict and expropriations.
  - Example magnitude: "an additional 1% of received remittances (in share GDP) leads to a decrease in likelihood of civil conflict of 0.09."
  - Interaction Shock*Remittances coefficients (Table 2): -0.483**, -0.509***, -0.487***, -0.324**, -0.123 (with t-statistics shown).
- Role of fiscal policy:
  - Evidence is mixed across specifications and fiscal measures.
  - Government consumption growth alone generally not significant in reducing conflict likelihood.
  - Fiscal stimuli interaction with shocks sometimes significantly negative: Shock*Fiscal Stimuli = -2.721** (Table 2) in one specification, indicating fiscal stimuli episodes can mitigate shock effects in some models.
  - Overall: mitigating role of fiscal policy is described as "barely significant" across full sample; stronger mitigation in countries with low private credit and low income.
- Robustness and additional results:
  - Using frequency of positive shocks (Positive_Shock): positive and significant association with civil conflict. Positive_Shock coefficients: 0.235***, 0.342***, 0.311***, 0.202***, 0.337***, 0.205** (Table 6).
  - Positive_Shock*Remittances interactions robustly negative in multiple specifications (e.g., -0.0479***, -0.0388***, -0.0384**).
  - Results robust to IV methods, Tobit, and when controlling for regional dummies and additional controls.
  - Structural reforms (IMF-World Bank program) reduce conflict intensity in some models (e.g., coefficient -0.866** in Table 2, column 5).

### Heterogeneity results
- By income level:
  - Median GDP per capita in sample = $ 35,599.5 (constant base 2011).
  - Food price shocks increase civil conflict likelihood in both lower- and higher-income subsamples, but coefficient in lower-income sample is more than double that in higher-income sample (Table 4).
  - Government crises increase significantly with shocks mainly in lower-income subsample.
- By access to private credit:
  - Median private credit in sample = 22% of GDP.
  - Shock effects on conflict and government crisis stronger and more statistically significant in countries with lower access to private credit.
  - Remittances reduce civil conflict likelihood more clearly in higher credit access countries in some specifications; results on remittances and government crisis are mixed.
- By region:
  - Continental dummies included; some findings unchanged.
  - African countries in sample found less likely to experience government crises than other continents in one specification, with caution advised due to limited representation of highly unstable cases.

### Robustness checks and sensitivity
- Alternative shock measure (frequency of positive shocks) confirms main conclusions on remittances damping effect.
- Using expropriations as dependent variable:
  - Food price shocks generally increase expropriation likelihood; remittances generally mitigate shock effects; fiscal policy mitigation not consistently observed.
- Lagged dependent variables and additional lags of shock variables tested; cumulative shocks increase expropriation likelihood.
- Additional IV regressions treating both GDP growth and remittances as endogenous yield similar conclusions; two-stage least squares show GDP growth and IMF-World Bank program often significantly and negatively associated with instability.

### Policy implications and recommendations
- Findings underline adverse effects of import food price shocks on political instability, particularly in lower-income countries and those with limited private credit access.
- Remittances play a vital damping role on the probability of civil conflict and expropriations during positive import food price shocks.
  - Policy recommendation: measures to favor remittance inflows, including reduction of transfer costs.
- Fiscal policy recommendation: fiscal authorities should adopt more countercyclical measures targeting low-income households and vulnerable groups during import food price shocks.
- General recommendation: policies that reduce developing countries’ vulnerability to import food price shocks are highly recommended.

### Appendix — Figures (selected visuals and reported values)
- Figure 3: Patterns of Food Commodity Prices from 1980 to 2012 (World prices of the six staple foods considered).
- Figure 4: Socio-political instability with Fiscal stimuli versus no Fiscal Stimuli — yearly averages shown:
  - Civil Conflict: 0.70 (No Fiscal Stimuli), 0.17 (Fiscal Stimuli)
  - Conflict Intensity, Likelihood of Civ conflict or Gov Crisis: 1.32 (No Fiscal Stimuli), 0.76 (Fiscal Stimuli)
  - Govt Crisis: 0.20 (No Fiscal Stimuli), 1.25 (Fiscal Stimuli)
  - Chart labels shown: 0, .5, 1, 1.5 on x-axis
- Figure 5: Remittances and Food Price Shocks — residuals and yearly averages with axis ranges:
  - Residuals (Yearly Averages) from -.4 to .6
  - Food price shock (Yearly Averages) from -.2 to .6
  - 95% CI and fitted values plotted

### Appendix — Key empirical results (selected table extracts)
- Table 7. Food Price Shocks and Expropriations: Selected coefficients (t statistics in parentheses):
  - Shock (Expropriation dependent): 1.012 (0.97); 1.580 (1.35); 1.688 (1.40); 1.947* (1.93); 2.395* (1.93); 2.254** (2.04)
  - Fiscal Stimuli: 0.268* (1.87); 0.173 (0.97); 0.162 (0.98)
  - Remittances: 0.0105 (1.58); 0.00768 (0.93); 0.00707 (0.85); 0.0101 (1.41); 0.00883 (1.08); 0.00832 (1.20)
  - Shock*Gov Expenditure: 0.107** (2.14); 0.0998 (1.59)
  - Shock*Fiscal Stimuli: 0.807 (0.92); 0.638 (0.75)
  - Shock*Remittances: -0.140** (-2.48); -0.145** (-2.25); -0.133** (-2.30); -0.117** (-2.21)
  - Number of obs: 991, 991, 979, 1674, 1674, 1615
  - Number of countries: 80, 80, 117, 117
- Table 8. Robustness checks: Selected coefficients (t statistics in parentheses):
  - Shock: 3.321*** (3.99); 3.095*** (3.58); 1.346** (2.48); 1.967*** (3.24); 0.457** (2.25); 0.348** (2.44); 1.029** (2.18); 1.535*** (3.54)
  - Gov Expenditure: 0.00312 (0.96); 0.000137 (0.08); 0.000599* (1.75); 0.0000528 (0.05)
  - Fiscal Stimuli: 0.227* (1.81); 0.126* (1.84); 0.0229 (1.11); 0.0930 (1.34)
  - Remittances: 0.00385 (0.50); 0.0000142 (0.00); -0.0317*** (-2.93); -0.0209** (-2.15); 0.000409 (0.44); 0.000150 (0.17); -0.0242* (-1.76); -0.0164 (-1.35)
  - Shock*Fiscal Stimuli: 0.350 (0.51); -0.863* (-1.88); 0.0600 (0.70); -0.684** (-2.25)
  - Shock*Remittances: 0.0468 (0.98); 0.0379 (0.81); -0.160** (-2.49); -0.140** (-2.24); 0.00595 (0.72); 0.00477 (0.71); -0.113** (-2.21); -0.0983** (-2.08)
  - GDPgrowth: -10.26* (-1.78); -8.406* (-1.80); -1.262 (-0.56); -2.779 (-1.12); -1.096* (-1.67); -0.746 (-1.53); -0.442 (-0.24); -1.092 (-0.60)
  - IMF-World Bank Program: -0.540* (-1.75); -0.458* (-1.65); -0.0133 (-0.09); -0.0540 (-0.35); -0.0563* (-1.80); -0.0455* (-1.88); -0.0169 (-0.16); -0.0566 (-0.53)
  - Number of obs: 1793, 2634, 354, 398, 1793, 2634, 354, 398
  - Rsquare (reported for some specs): 0.0540, 0.0429, 0.126, 0.0848

### Appendix — Variables, Definitions and Sources (selected)
- Price shocks: see section (3). Data: Authors, using IMF WEO (2017) and FAO STATISTIQUES (2016).
- Civil conflict: Dummy = 1 if intrastate conflict with more than 25 battle deaths; source PRIO Uppsala Conflict Data Program (UCDP) (2017).
- Conflict Intensity: yearly number of conflicts with at least 25 battle deaths; source PRIO UCDP (2017).
- Government Crisis: Dummy = 1 if any manifestation of government crisis; source CNTS.
- Expropriation: Dummy = 1 if any manifestation with a risk of expropriation; source CNTS.
- Gov Expenditure: Government consumption expenditure annual growth; source Authors, using WDI (2015).
- Fiscal Stimuli: Dummy = 1 if there are fiscal stimuli; source Authors, using WDI (2015).
- Remittances: Net personal remittances received as % of GDP; source WDI (2015).
- GDPgrowth: Annual rate of GDP growth; computed using WEO (2016) data.
- IMF-World Bank Program: Dummy = 1 if country is on program; source Dreher, Sturm, and Vreeland (2009) and latest updated version.
- Private Credit: Bank and financial credit held by private sector (as a share of GDP); source WDI (2015).
- Financial Crisis: Number of Systemic Banking Crisis, Currency Crisis, Sovereign Debt Crisis and Sovereign Debt Restructuring; source Laeven & Valencia (2018).

### Appendix — Summary statistics (Table 10, selected)
- Civil Conflict: Observations 677; Mean 0,69; Standard. Deviation. 0,46; Minimum 0,00; Maximum 1,00
- Conflict Intensity: Observations 741; Mean 1,80; Standard. Deviation. 1,33; Minimum 1,00; Maximum 9,00
- Government Crisis: Observations 4432; Mean 0,10; Standard. Deviation. 0,29; Minimum 0,00; Maximum 1,00
- Expropriations: Observations 4432; Mean 0,03; Standard. Deviation. 0,16; Minimum 0,00; Maximum 1,00
- Shock: Observations 3779; Mean 0,07; Standard. Deviation. 0,16; Minimum -0,33; Maximum 0,84
- Positive Shock: Mean 2,30; Standard. Deviation. 1,39; Minimum 0,00; Maximum 4,00
- Gov expenditure: Observations 2429; Mean 5,14; Standard. Deviation. 28,36; Minimum -82,33; Maximum 1004,60
- Fiscal Stimuli: Observations 4431; Mean 0,37; Standard. Deviation. 0,48; Minimum 0,00; Maximum 1,00
- Remittances: Observations 3145; Mean 4,67; Standard. Deviation. 8,56; Minimum 0,00; Maximum 106,48
- GDPgrowth: Observations 3929; Mean 0,02; Standard. Deviation. 0,07; Minimum -0,61; Maximum 1,40
- IMF-World Bank Program: Observations 4399; Mean 0,04; Standard. Deviation. 0,21; Minimum 0,00; Maximum 1,00
- Private Credit: Observations 2702; Mean 0.08; Standard. Deviation. 0.30; Minimum 0; Maximum 2

### Appendix — Country coverage (Table 11)
- Lists of countries considered with at least one observation in regressions of government crises and regressions of civil conflict, including (among others): Algeria; Bangladesh; Belize; Benin; Bhutan; Bolivia; Botswana; Brazil; Burkina Faso; Burundi; Cambodia; Cameroon; Cape Verde; Chad; Colombia; Comoros; Congo; Costa Rica; Cote d'Ivoire; Djibouti; Dominican Republic; Ecuador; Egypt; El Salvador; Equatorial Guinea; Fiji; Gabon; Gambia; Ghana; Guatemala; Guinea; Guinea‑Bissau; India; Indonesia; Iran; Jamaica; Jordan; Kenya; Laos; Lebanon; Lesotho; Madagascar; Malawi; Mali; Mauritania; Mauritius; Mexico; Morocco; Mozambique; Nepal; Niger; Nigeria; Pakistan; Panama; Papua New Guinea; Paraguay; Peru; Philippines; Rwanda; Senegal; Sierra Leone; South Africa; Sri Lanka; Sudan; Suriname; Swaziland; Syria; Tanzania; Thailand; Togo; Tunisia; Turkey; Uganda; Vanuatu; Venezuela; Vietnam; etc.

*Source: wpiea2020248-print-pdf — Authors’ computations and analysis from content in wpiea2020248-print-pdf - 11. List of countries (excerpt).*

### References  ______________________________________________________________ 25

### References — Figures and Tables index

### Figures
- 1. Food Price Shocks and Food Price Growth  ___________________________________  11
- 2. Socio-political Instability: Some Heterogeneities _______________________________  12
- 3. Patterns of Food Commodity Prices from 1980 to 2012 __________________________  29
- 4. Socio-political instability with Fiscal stimuli versus no Fiscal Stimuli  ______________  29
- 5. Remittances and Food Price Shocks _________________________________________  30

### Tables
- 1. Correlation Matrix _______________________________________________________  11
- 2. Baseline Results: Food Price Shocks and Conflicts: The Roles of Fiscal Policy and        Remittances ______________________________________________________________  16
- 3. Food price shocks and socio-political instability: geographical heterogeneities  _______  18
- 4. Food Price Shocks and Political Instability: The Role of Fiscal Policy  
  by Level of Income ________________________________________________________  19
- 5. Food Price Shocks and Political Instability: by Level of Private Credit Access  _______  21
- 6. Positive Food Price Shocks and Civil Conflicts: The Roles of Fiscal  
  Policies and Remittances ____________________________________________________  22
- 7. Food Price Shocks and Expropriations: The Roles of Remittances and Fiscal Policy ....... 31
- 8. Food price shocks and instability: Additional Robustness Checks ..................................... 32
- 9. Variables, Definitions and Sources ..................................................................................... 33
- 10. Summary statistics ............................................................................................................. 34

*Source: wpiea2020248-print-pdf - References  ______________________________________________________________ 25*

### 11. List of countries ..................................................................................................

### 11. List of countries

### Introduction and objective
- Political stability fosters sustainable growth and social well-being; import food price shocks can trigger socio-political instability.
- Paper objective: assess whether fiscal policy and remittances mitigate the impact of import food price shocks on socio-political instability.
- Sample: 101 low- and middle-income countries over 1980-2012.

### Data, variables, and empirical approach
- Food price shocks:
  - Constructed using IMF WEO prices and FAO trade quantities; country-specific weighted price index across wheat, sugar, soybeans, soybean oil, maize, and rice.
  - Main shock variable: residual from country-by-country regression of ln(price index) on trend and first and second lags (equation (3)).
  - Alternative shock: frequency of positive residuals within non-overlapping four-year periods.
- Fiscal policy:
  - Two measures: (i) annual growth rate of government consumption expenditure as share of GDP; (ii) fiscal stimuli proxy based on residuals from country regressions of government consumption on cyclical GDP growth (Cycle_GDPgrowth, Hodrick-Prescott λ = 6.25), inflation, and inflation squared, then a 4-year rolling standard deviation dummy.
- Socio-political instability (dependent variables):
  - Civil conflict: dummy = 1 if intrastate conflict with more than 25 battle deaths in a year.
  - Conflict intensity: number of conflicts with more than 25 battle-related deaths in a year.
  - Expropriations: dummy = 1 if systemic elimination of political opposition (jailing/execution) in a year.
  - Government crisis: dummy = 1 for rapidly evolving contexts threatening regime downfall (excluding revolt aimed at overthrow).
- Controls: GDP per capita growth (first lag instrumented), IMF-World Bank poverty reduction program dummy, macro-financial crises (Systemic Banking Crisis, Currency Crisis, Sovereign Debt Crisis, Sovereign Debt Restructuring).
- Estimators:
  - Probit for binary outcomes; maximum likelihood Tobit for conflict counts.
  - IV Probit and IV Tobit (instruments: lagged remittances and lagged GDP per capita growth) and two-stage least squares for robustness.
  - Standard errors clustered at country level; Probit standard errors bootstrapped (250 replications) for some checks.

### Key empirical findings
- Main effects of food price shocks:
  - Positive food price shocks significantly increase the probability of civil conflict, government crises, and expropriations; increase conflict intensity.
  - Quantitative example: "a deviation of the food price of one unit from its long run trend leads to an increase of 0.8 to 4 % in the likelihood of civil conflict" (column 3 discussion).
  - Baseline coefficient estimates (selected from Table 2):
    - Shock coefficients reported across models include: 0.604**, -1.288, 4.425***, 0.696, 5.152***, 8.035*** (t-statistics reported in table).
- Role of remittances:
  - Remittances dampen the adverse effect of food price shocks on the likelihood of civil conflict and expropriations.
  - Example magnitude: "an additional 1% of received remittances (in share GDP) leads to a decrease in likelihood of civil conflict of 0.09."
  - Interaction Shock*Remittances coefficients (Table 2): -0.483**, -0.509***, -0.487***, -0.324**, -0.123 (with t-statistics shown).
- Role of fiscal policy:
  - Evidence is mixed across specifications and fiscal measures.
  - Government consumption growth alone generally not significant in reducing conflict likelihood.
  - Fiscal stimuli interaction with shocks sometimes significantly negative: Shock*Fiscal Stimuli = -2.721** (Table 2) in one specification, indicating fiscal stimuli episodes can mitigate shock effects in some models.
  - Overall finding: mitigating role of fiscal policy is "barely significant" across full sample; stronger mitigation in countries with low private credit and low income.
- Robustness and additional results:
  - Using frequency of positive shocks (Positive_Shock): positive and significant association with civil conflict. Positive_Shock coefficients: 0.235***, 0.342***, 0.311***, 0.202***, 0.337***, 0.205** (Table 6).
  - Positive_Shock*Remittances interactions robustly negative in multiple specifications (e.g., -0.0479***, -0.0388***, -0.0384**).
  - Results robust to IV methods, Tobit, and when controlling for regional dummies and additional controls.
  - Structural reforms (IMF-World Bank program) reduce conflict intensity in some models (e.g., coefficient -0.866** in Table 2, column 5).

### Heterogeneity results
- By income level:
  - Median GDP per capita in sample = $ 35,599.5 (constant base 2011).
  - Food price shocks increase civil conflict likelihood in both lower- and higher-income subsamples, but coefficient in lower-income sample is more than double that in higher-income sample (Table 4).
  - Government crises increase significantly with shocks mainly in lower-income subsample.
- By access to private credit:
  - Median private credit in sample = 22% of GDP.
  - Shock effects on conflict and government crisis stronger and more statistically significant in countries with lower access to private credit.
  - Remittances reduce civil conflict likelihood more clearly in higher credit access countries in some specifications; results on remittances and government crisis are mixed and sometimes counterintuitive (Table 5).
- By region:
  - Continental dummies included; some findings unchanged. African countries in sample found less likely to experience government crises than other continents in one specification, but caution advised due to limited representation of highly unstable cases.

### Robustness checks and sensitivity
- Alternative shock measure (frequency of positive shocks) confirms main conclusions on remittances damping effect.
- Using expropriations as dependent variable: food price shocks generally increase expropriation likelihood; remittances generally mitigate shock effects; fiscal policy mitigation not consistently observed.
- Lagged dependent variables and additional lags of shock variables tested; cumulative shocks increase expropriation likelihood.
- Additional IV regressions treating both GDP growth and remittances as endogenous yield similar conclusions; two-stage least squares show GDP growth and IMF-World Bank program often significantly and negatively associated with instability.

### Policy implications and recommendations
- Findings underline adverse effects of import food price shocks on political instability, particularly in lower-income countries and those with limited private credit access.
- Remittances play a vital damping role on the probability of civil conflict and expropriations during positive import food price shocks.
  - Policy recommendation: measures to favor remittance inflows, including reduction of transfer costs.
- Fiscal policy recommendation: fiscal authorities should adopt more countercyclical measures targeting low-income households and vulnerable groups during import food price shocks.
- General recommendation: policies that reduce developing countries’ vulnerability to import food price shocks are highly recommended.

_Italic: Source: Authors’ computations and analysis from content in wpiea2020248-print-pdf - 11. List of countries (excerpt). _

### REFERENCES

### REFERENCES

### Key literature cited
- Literature on links between food prices, remittances, fiscal policy, and sociopolitical instability, including studies by Agnello et al.; Aguirre; Aisen and Veiga; Annett; Arezki and Bruckner; Barnett; Barro and Sala-i-Martin; Baser and Swain; Bellemare; Besley and Persson; Blanchard; Brückner and Ciccone; Chami et al.; Collier and Dehn; Collier and Hoeffler; Combes and Ebeke; Craigwell et al.; Deaton and Miller; Devereux and Wen; Draman et al.; Dreher et al.; Ebeke and Combes; FAO/IFAD/OECD/IMF/WFP/UNCTAD/World Bank/WTO/IFPRI/UN HLTF; Fatas and Mihov; Finlay and Magnusson; Greene; Gries & Meierrieks; Hendrix & Brinkman; Gupta; Laeven & Valencia; Maddala & Lahiri; Miranda & Rabe‑Hesketh; Ortiz & Cummins; Pinstrup‑Andersen & Shimokawa; Raleigh et al.; Ravn & Uhlig; Smit & Skinner; Smith (1997); Smith (2014); Sobek & Boehmer; Wooldridge; Ziesemer.

### Appendix — Figures
- Figure 3: Patterns of Food Commodity Prices from 1980 to 2012.
  - Note: World prices of the six staple foods considered. Data from IMF WEO database.
- Figure 4: Socio-political instability with Fiscal stimuli versus no Fiscal Stimuli.
  - Source: Authors, using data from Prio/Upsala, Cross-National Time-Series Data Archive (CNTS) and World Bank Indicators.
  - On the x-axis, Fiscal stimuli refers to situations where there was at least one fiscal stimuli.
  - Visual values shown (yearly averages): No Fiscal Stimuli vs Fiscal Stimuli
    - Civil Conflict: 0.70 (No Fiscal Stimuli), 0.17 (Fiscal Stimuli)
    - Conflict Intensity, Likelihood of Civ conflict or Gov Crisis: 1.32 (No Fiscal Stimuli), 0.76 (Fiscal Stimuli)
    - Govt Crisis: 0.20 (No Fiscal Stimuli), 1.25 (Fiscal Stimuli)
    - (Chart labels shown: 0, .5, 1, 1.5 on x-axis)
- Figure 5: Remittances and Food Price Shocks.
  - Source: Authors, using IMF WEO, FAO Statistics and World Bank Indicators.
  - Residuals extracted from regression of food price shock variables on net personal received remittances. Figures represent yearly averages.
  - Visual axis ticks and ranges shown: Residuals (Yearly Averages) from -.4 to .6; Food price shock (Yearly Averages) from -.2 to .6; 95% CI and fitted values plotted.

### Appendix — Key empirical results (Tables)
- Table 7. Food Price Shocks and Expropriations: The Roles of Remittances and Fiscal Policy
  - Models: Probit, IV Probit, Tobit variants. Standard errors clustered at country level. Year fixed effects included.
  - Selected coefficient estimates (t statistics in parentheses). Significance: * p < 0.10, ** p < 0.05, *** p < 0.01.
  - Shock (Expropriation dependent):
    - 1.012 (0.97)
    - 1.580 (1.35)
    - 1.688 (1.40)
    - 1.947* (1.93)
    - 2.395* (1.93)
    - 2.254** (2.04)
  - Fiscal Stimuli:
    - 0.268* (1.87)
    - 0.173 (0.97)
    - 0.162 (0.98)
  - Remittances:
    - 0.0105 (1.58)
    - 0.00768 (0.93)
    - 0.00707 (0.85)
    - 0.0101 (1.41)
    - 0.00883 (1.08)
    - 0.00832 (1.20)
  - Shock*Gov Expenditure:
    - 0.107** (2.14)
    - 0.0998 (1.59)
  - Shock*Fiscal Stimuli:
    - 0.807 (0.92)
    - 0.638 (0.75)
  - Shock*Remittances:
    - -0.140** (-2.48)
    - -0.145** (-2.25)
    - -0.133** (-2.30)
    - -0.117** (-2.21)
  - _Model summary counts:_
    - Number of obs: 991, 991, 979, 1674, 1674, 1615
    - Number of countries: 80, 80, 117, 117
  - Notes: Results remain unchanged when dependent variable is number of conflicts during the year. Europe and Oceania dummies dropped.

- Table 8. Food Price shocks and instability: Additional Robustness Checks
  - Models: IV Tobit and Two-Stage least squares (2SLS).
  - Selected coefficient estimates (t statistics in parentheses). Significance: * p < 0.10, ** p < 0.05, *** p < 0.01.
  - Shock:
    - 3.321*** (3.99)
    - 3.095*** (3.58)
    - 1.346** (2.48)
    - 1.967*** (3.24)
    - 0.457** (2.25)
    - 0.348** (2.44)
    - 1.029** (2.18)
    - 1.535*** (3.54)
  - Gov Expenditure:
    - 0.00312 (0.96)
    - 0.000137 (0.08)
    - 0.000599* (1.75)
    - 0.0000528 (0.05)
  - Fiscal Stimuli:
    - 0.227* (1.81)
    - 0.126* (1.84)
    - 0.0229 (1.11)
    - 0.0930 (1.34)
  - Remittances:
    - 0.00385 (0.50)
    - 0.0000142 (0.00)
    - -0.0317*** (-2.93)
    - -0.0209** (-2.15)
    - 0.000409 (0.44)
    - 0.000150 (0.17)
    - -0.0242* (-1.76)
    - -0.0164 (-1.35)
  - Shock*Gov Expenditure:
    - -0.0158 (-0.69)
    - 0.0105 (0.89)
    - -0.00449* (-1.80)
    - 0.00607 (0.77)
  - Shock*Fiscal Stimuli:
    - 0.350 (0.51)
    - -0.863* (-1.88)
    - 0.0600 (0.70)
    - -0.684** (-2.25)
  - Shock*Remittances:
    - 0.0468 (0.98)
    - 0.0379 (0.81)
    - -0.160** (-2.49)
    - -0.140** (-2.24)
    - 0.00595 (0.72)
    - 0.00477 (0.71)
    - -0.113** (-2.21)
    - -0.0983** (-2.08)
  - GDPgrowth:
    - -10.26* (-1.78)
    - -8.406* (-1.80)
    - -1.262 (-0.56)
    - -2.779 (-1.12)
    - -1.096* (-1.67)
    - -0.746 (-1.53)
    - -0.442 (-0.24)
    - -1.092 (-0.60)
  - IMF-World Bank Program:
    - -0.540* (-1.75)
    - -0.458* (-1.65)
    - -0.0133 (-0.09)
    - -0.0540 (-0.35)
    - -0.0563* (-1.80)
    - -0.0455* (-1.88)
    - -0.0169 (-0.16)
    - -0.0566 (-0.53)
  - Model counts and fit:
    - Number of obs: 1793, 2634, 354, 398, 1793, 2634, 354, 398
    - Rsquare: 0.0540, 0.0429, 0.126, 0.0848 (reported for some specifications)
  - Notes: Instruments include first lag of GDP per capita growth and first lag of net personal received remittances. Results unchanged when dependent variables are number of conflicts during the year.

### Appendix — Variables, Definitions and Sources (Table 9)
- Price shocks: see section (3). Data: Authors, using IMF WEO (2017) and FAO STATISTIQUES (2016).
- Civil conflict: Dummy = 1 if intrastate conflict with more than 25 battle deaths; source PRIO Uppsala Conflict Data Program (UCDP) (2017).
- Conflict Intensity: yearly number of conflicts with at least 25 battle deaths; source PRIO UCDP (2017).
- Government Crisis: Dummy = 1 if any manifestation of government crisis; source CNTS.
- Expropriation: Dummy = 1 if any manifestation with a risk of expropriation; source CNTS.
- Gov Expenditure: Government consumption expenditure annual growth; source Authors, using WDI (2015).
- Fiscal Stimuli: Dummy = 1 if there are fiscal stimuli; source Authors, using WDI (2015).
- Remittances: Net personal remittances received as % of GDP; source WDI (2015).
- GDPgrowth: Annual rate of GDP growth; computed using WEO (2016) data.
- IMF-World Bank Program: Dummy = 1 if country is on program; source Dreher, Sturm, and Vreeland (2009) and latest updated version.
- Private Credit: Bank and financial credit held by private sector (as a share of GDP); source WDI (2015).
- Financial Crisis: Number of Systemic Banking Crisis, Currency Crisis, Sovereign Debt Crisis and Sovereign Debt Restructuring; source Laeven & Valencia (2018).

### Appendix — Summary statistics (Table 10)
- Civil Conflict: Observations 677; Mean 0,69; Standard. Deviation. 0,46; Minimum 0,00; Maximum 1,00
- Conflict Intensity: Observations 741; Mean 1,80; Standard. Deviation. 1,33; Minimum 1,00; Maximum 9,00
- Government Crisis: Observations 4432; Mean 0,10; Standard. Deviation. 0,29; Minimum 0,00; Maximum 1,00
- Expropriations: Observations 4432; Mean 0,03; Standard. Deviation. 0,16; Minimum 0,00; Maximum 1,00
- Shock: Observations 3779; Mean 0,07; Standard. Deviation. 0,16; Minimum -0,33; Maximum 0,84
- Positive Shock: Mean 2,30; Standard. Deviation. 1,39; Minimum 0,00; Maximum 4,00
- Gov expenditure: Observations 2429; Mean 5,14; Standard. Deviation. 28,36; Minimum -82,33; Maximum 1004,60
- Fiscal Stimuli: Observations 4431; Mean 0,37; Standard. Deviation. 0,48; Minimum 0,00; Maximum 1,00
- Remittances: Observations 3145; Mean 4,67; Standard. Deviation. 8,56; Minimum 0,00; Maximum 106,48
- GDPgrowth: Observations 3929; Mean 0,02; Standard. Deviation. 0,07; Minimum -0,61; Maximum 1,40
- IMF-World Bank Program: Observations 4399; Mean 0,04; Standard. Deviation. 0,21; Minimum 0,00; Maximum 1,00
- Private Credit: Observations 2702; Mean 0.08; Standard. Deviation. 0.30; Minimum 0; Maximum 2

### Appendix — Country coverage (Table 11)
- Lists of countries considered with at least one observation in regressions of government crises and regressions of civil conflict. (Countries include, among others: Algeria; Bangladesh; Belize; Benin; Bhutan; Bolivia; Botswana; Brazil; Burkina Faso; Burundi; Cambodia; Cameroon; Cape Verde; Chad; Colombia; Comoros; Congo; Costa Rica; Cote d'Ivoire; Djibouti; Dominican Republic; Ecuador; Egypt; El Salvador; Equatorial Guinea; Fiji; Gabon; Gambia; Ghana; Guatemala; Guinea; Guinea‑Bissau; India; Indonesia; Iran; Jamaica; Jordan; Kenya; Laos; Lebanon; Lesotho; Madagascar; Malawi; Mali; Mauritania; Mauritius; Mexico; Morocco; Mozambique; Nepal; Niger; Nigeria; Pakistan; Panama; Papua New Guinea; Paraguay; Peru; Philippines; Rwanda; Senegal; Sierra Leone; South Africa; Sri Lanka; Sudan; Suriname; Swaziland; Syria; Tanzania; Thailand; Togo; Tunisia; Turkey; Uganda; Vanuatu; Venezuela; Vietnam; etc.)

*Source: wpiea2020248-print-pdf - REFERENCES*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020248-print-pdf.pdf_