## _wp1112

## Source details

**Canonical URL:** [_wp1112](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1112.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1112.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1112.pdf.json)

---

### I. Introduction — key questions and headline result
- Political instability proxied by Cabinet Changes (number of times in a year in which a new premier is named and/or 50 percent or more of the cabinet posts are occupied by new ministers) is widespread and varies regionally.
- Primary research questions:
  - What are the main transmission channels from political instability to economic growth?
  - How quantitatively important are effects on total factor productivity and on physical and human capital accumulation?
- Core empirical headline:
  - An additional cabinet change (a new premier is named and/or 50 percent of cabinet posts are occupied by new ministers) reduces the annual real GDP per capita growth rate by 2.39 percentage points (Cabinet Changes coefficient: -0.0239***, t = -3.698).
  - More than half of this effect operates through lower total factor productivity growth; physical capital accumulation has a slightly larger effect than human capital accumulation among the remaining channels.
- Policy implication emphasized:
  - Countries need to address political instability, tackling root causes and mitigating effects on the quality and sustainability of economic policies that engender economic growth.

### II. Data, variables, and empirical model
- Sample and sources:
  - Annual data, 1960 to 2004, gathered for 209 countries with up to 169 countries in estimations.
  - Economic data: Penn World Table Version 6.2 – PWT (Heston et al., 2006), World Development Indicators (WDI), Global Development Network Growth Database (GDN), International Financial Statistics (IFS).
  - Political and institutional data: Cross National Time Series Data Archive – CNTS (Databanks International, 2007), Polity IV Database (Marshall and Jaggers, 2005), State Failure Task Force database (SFTF), Gwartney and Lawson (2007).
- Dependent variable and periodicity:
  - GDP per capita growth for consecutive, nonoverlapping, five-year periods: 1960–64, 1965–69, 1970–74, 1975–79, 1980–84, 1985–89, 1990–94, 1995–99, and 2000–04.
- Baseline explanatory variables (averaged over each five-year period except Initial GDP per capita):
  - Initial GDP per capita (log) (PWT)
  - Investment (percent of GDP) (PWT)
  - Primary school enrollment (WDI)
  - Population growth (PWT)
  - Trade openness (PWT)
  - Cabinet Changes (CNTS) — main proxy for political (regime) instability
- Additional variables for macroeconomic stability and institutions:
  - Inflation rate (IFS) defined as ln(1+Inf/100)
  - Government (percent of GDP) (PWT)
  - Index of Economic Freedom (Gwartney and Lawson, 2007)
  - Ethnic Homogeneity Index (SFTF): ranges from 0 to 1
  - Polity Scale (Polity IV): from -10 to 10
- Estimation approach:
  - Dynamic panel data models estimated with system-GMM; all right-hand-side variables treated as endogenous; instruments: lagged levels and lagged differences as described.
  - Two-step GMM with Windmeijer (2005) finite-sample correction for robust standard errors.
  - Significance notation: ***, 1 percent; **, 5 percent; *, 10 percent. T-statistics reported in parentheses.

### III. Descriptive statistics (selected exact values from Table 1)
- Growth of GDP per capita: Obs. 1098; Mean 0.016; St. Dev. 0.037; Min. -0.344; Max. 0.347; Source PWT.
- GDP per capita (log): Obs. 1197; Mean 8.315; St. Dev. 1.158; Min. 5.144; Max. 11.346; Source PWT.
- Growth of TFP: Obs. 703; Mean 0.000; St. Dev. 0.048; Min. -0.509; Max. 0.292; Sources PWT, BL.
- Investment (percent of GDP): Obs. 1287; Mean 14.474; St. Dev. 8.948; Min. 1.024; Max. 91.964; Source PWT.
- Primary School Enrollment: Obs. 1286; Mean 88.509; St. Dev. 27.794; Min. 3.000; Max. 149.240; Source WDI-WB.
- Population Growth: Obs. 1521; Mean 0.097; St. Dev. 0.071; Min. -0.281; Max. 0.732; Source PWT.
- Cabinet Changes: Obs. 1322; Mean 0.044; St. Dev. 0.358; Min. 0.000; Max. 2.750; Source CNTS.
- Index of Economic Freedom: Obs. 679; Mean 5.682; St. Dev. 1.208; Min. 2.004; Max. 8.714; Source EFW.
- Polity Scale: Obs. 1194; Mean 0.239; St. Dev. 7.391; Min. -10.000; Max. 10.000; Source Polity IV.
- Ethnic Homogeneity Index: Obs. 1129; Mean 0.583; St. Dev. 0.277; Min. 0.150; Max. 1.000; Source SFTF.

### IV. Empirical results — political instability and growth
- Baseline system-GMM results (Table 2, column 1):
  - Cabinet Changes coefficient: -0.0239*** (t = -3.698) — interpreted as an additional cabinet change per year decreases annual growth by 2.39 percentage points.
  - Initial GDP per capita (log): -0.0087** (t = -2.513) — conditional convergence.
  - Investment (percent of GDP): 0.0009** (t = 2.185).
  - Primary School Enrollment: 0.0003*** (t = 3.097).
  - Population Growth: -0.184*** (t = -3.412).
  - Trade (percent of GDP): 6.70e-05 (t = 0.957), not statistically significant in baseline.
- Extended specifications:
  - Inflation coefficient: -0.0091*** (t = -2.837) when included.
  - Index of Economic Freedom: 0.0109*** (t = 2.824) — reported interpretation: a one-point increase in the index raises annual economic growth by one percentage point (as reported).
  - Area 2 (Legal structure and security of property rights) coefficient: 0.00360* (t = 1.681).
- Robustness:
  - Cabinet Changes remains statistically significant across specifications and when controlling for institutions.
  - Alternative political-instability indices via principal components show expected negative signs; Violence Index is not statistically significant, implying regime instability (not violence alone) drives adverse growth effects.

### V. Index construction and dimensions of political instability
- Five principal-component indices from CNTS variables:
  - Regime Instability Index 1: Cabinet Changes and Executive Changes.
  - Regime Instability Index 2: Cabinet Changes, Constitutional Changes, Coups, Executive Changes, Government Crises.
  - Regime Instability Index 3: Cabinet Changes, Constitutional Changes, Coups, Executive Changes, Government Crises, Number of Legislative Elections, Fragmentation Index.
  - Violence Index: Assassinations, Coups, Revolutions.
  - Political Instability Index: Assassinations, Cabinet Changes, Constitutional Changes, Coups, Revolutions.
- Selected Table 4 coefficients (exact values):
  - Regime Instability Index 1: -0.0198*** (t = -4.851).
  - Regime Instability Index 2: -0.0133*** (t = -3.381).
  - Regime Instability Index 3: -0.0142*** (t = -4.246).
  - Violence Index: -0.0046 (t = -1.197) — not statistically significant.
  - Political Instability Index: -0.0087** (t = -2.255).
- Institutional correlates:
  - Index of Economic Freedom positive and significant (e.g., 0.0084** in Table 4 column 1).
  - Ethnic Homogeneity Index positive and significant (e.g., 0.0497*** in Table 4 column 1).

### VI. Robustness checks — samples and period lengths
- Restricted-sample examples (Table 5) — Cabinet Changes coefficients remain negative and significant:
  - Excluding Industrial Countries: -0.0282*** (t = -3.814)
  - Excluding Africa: -0.0285*** (t = -4.588)
  - Excluding Developing Asia: -0.0342*** (t = -3.583)
  - Excluding the 1960s and 1970s: -0.0309*** (t = -3.108)
  - Excluding the 2000s: -0.0326*** (t = -3.693)
  - Regime Instability Index 1 excluding Industrial Countries: -0.0191*** (t = -3.795)
- Alternative period lengths (Table 6) — consistent negative coefficients:
  - Cabinet Changes: 4-Year Periods -0.0298* (t = -1.683); 6-Year Periods -0.0229** (t = -2.470); 8-Year Periods -0.0121* (t = -1.752); 10-Year Periods -0.0231** (t = -2.004).
  - Regime Instability Index 1: 6-Year Periods -0.0121*** (t = -2.842); 10-Year Periods -0.0213** (t = -2.553).

### VII. Channels of transmission — hypotheses and measurement
- Hypothesized channels:
  - Physical capital accumulation: instability raises uncertainty, reducing investment and slowing accumulation of the physical capital stock.
  - Total factor productivity (TFP): instability may reduce efficient resource allocation, lower R&D, and disrupt firm operations; violence and unrest may damage installed capacity.
  - Human capital accumulation: uncertainty may reduce incentives to invest in education.
- Growth-accounting framework:
  - Production function: Y = A K^(α) (H L)^(1-α) with α = 1/3 (Hall and Jones, 1999 approach).
  - Growth-rate decomposition (equation (10)): contributions computed as growth rates of physical capital accumulation, human capital accumulation, and TFP growth.
- Data construction specifics:
  - Depreciation rate δ assumed to be 6 percent.
  - Factor share α assumed equal to 1/3.
  - Physical capital constructed using perpetual inventory method: K_t = (1 - δ) K_{t-1} + I_t; K_0 = I_0 / (δ + g), where I_0 is investment in 1950 (or first available year) and g is average geometric growth rate of investment in first 10 years available.
  - Human capital-augmented labor H_i = e^{φ(s_i)} L_i with s_i average years of schooling (Barro and Lee, 2001); parameter s reported as 0.134 for s (documented as "0.134 for s").

### VIII. Empirical evidence on channels
- Political instability and physical capital accumulation (Table 7):
  - Cabinet Changes and the three regime instability indexes always statistically significant and negative for growth of physical capital per capita.
  - Population Growth coefficients (selected): -0.299***, -0.272***, -0.212**, -0.216***, -0.192** across columns (t-statistics reported).
  - Trade (percent of GDP): positive effects in some columns, e.g., 0.0001** (t = 2.427) in column 1.
  - Index of Economic Freedom: 0.0070** (t = 2.473) in column 1.
  - Estimation diagnostics examples: Observations 899 (col.1); Countries 155 (col.1); Hansen p-value 0.0535 (col.1).
- Political instability and TFP growth (Table 8):
  - Cabinet Changes negative for TFP growth: -0.0860*** (col.1) and -0.0243* (col.2) reported in different specifications.
  - Regime Instability Index 1: -0.0129** (t = -1.995).
  - Index of Economic Freedom positively associated with TFP growth, e.g., 0.0190*** (col.2).
  - Initial TFP (log) shows convergence: coefficients -0.0338***, -0.0344***, -0.0299***, -0.0308**, -0.0301** across columns.
  - Estimation diagnostics examples: Observations 700 (col.1); Countries 105 (col.1); Hansen p-value 0.501 (col.1).
- Political instability and human capital growth (Table 9):
  - Cabinet Changes negative for human capital growth: -0.0113** (col.1) and -0.00911** (other columns).
  - Regime Instability Index 1: -0.00379**; Index 2: -0.00311**; Index 3: -0.00292*.
  - Polity Scale positively affects human capital growth: 0.0004***, 0.0004***, 0.0005*** in columns 3–5.
  - Trade (percent of GDP) positive association with human capital growth: examples include 0.00009**, 0.00006*, 0.0000721**, 0.0000697**, 0.00006* across columns.
  - Estimation diagnostics examples: Observations 704 (col.1); Countries 105 (col.1); Hansen p-value 0.406 (col.1).

### IX. Decomposition of transmission channels (Table 10)
- Method:
  - Effects on GDP per capita growth obtained by multiplying coefficients from channel regressions by factor shares: physical capital by α = 1/3; TFP by (1-α) = 2/3; human capital by (1-α) = 2/3.
- Main quantitative results (exact values reported):
  - Cabinet Changes:
    - Coefficients: Physical Capital pc = -0.0195***; TFP = -0.0243*; Human Capital pc = -0.00911**.
    - Effect on GDP: Physical = -0.0065; TFP = -0.0162; Human = -0.0061; Total = -0.0288.
    - Percent of Total Effect: 22.59% (physical), 56.30% (TFP), 21.11% (human).
  - Regime Instability Index 1:
    - Coefficients: Physical = -0.0108**; TFP = -0.0129**; Human = -0.00379**.
    - Effect on GDP: -0.0036, -0.0086, -0.0025; Total = -0.0147.
    - Percent of Total Effect: 24.44% (physical), 58.40% (TFP), 17.16% (human).
  - Regime Instability Index 2:
    - Coefficients: Physical = -0.00932**; TFP = -0.00846*; Human = -0.00311**.
    - Effect on GDP: -0.0031, -0.0056, -0.0021; Total = -0.0108.
    - Percent of Total Effect: 28.71% (physical), 52.13% (TFP), 19.16% (human).
  - Regime Instability Index 3:
    - Coefficients: Physical = -0.00906**; TFP = -0.00964**; Human = -0.00292*.
    - Effect on GDP: -0.0030, -0.0064, -0.0019; Total = -0.0114.
    - Percent of Total Effect: 26.51% (physical), 56.41% (TFP), 17.08% (human).
- Interpretation:
  - More than half of the total negative effects of political instability on real GDP per capita growth operate through adverse effects on TFP growth (between 52.13 percent and 58.40 percent depending on the proxy).
  - Physical capital accumulation accounts for 22.59 percent to 28.71 percent of the total effect.
  - Human capital growth accounts for 17.08 percent to 21.11 percent of the total effect.
  - These shares align with growth-accounting literature benchmarks where human capital accounts for 10–30 percent of income differences, physical capital about 20 percent, and TFP 60–70 percent.

### X. Robustness of decomposition and interactions
- Wald tests comparing decomposed total effects to coefficients estimated in GDP per capita growth regressions do not reject equality (chi2 and Prob>chi2 reported in the source).
- Interaction among channels:
  - Growth rates of one channel do not appear to be significantly affected by growth rates of the other channels (authors report no significant interactions; additional estimations available upon request).
- Mechanical correlation avoidance:
  - Variables used to construct series were excluded from corresponding regressions (e.g., Investment percent of GDP excluded when constructing physical capital; Primary School Enrollment excluded when constructing human capital); robustness checks confirm main results when included.

### XI. Conclusions and policy implications
- Political instability significantly and economically reduces growth.
- Main transmission channels and quantitative importance:
  - TFP growth is the primary channel (accounts for 52.13 percent to 58.40 percent of the total negative effect).
  - Physical capital accumulation (22.59 percent to 28.71 percent) and human capital growth (17.08 percent to 21.11 percent) are secondary channels.
- Policy implication emphasized:
  - Governments in politically fragmented countries with high degrees of political instability need to address root causes and mitigate effects on the design and implementation of economic policies to achieve durable economic policies and higher economic growth.

*Source: _wp1112 - 0.134 for s (PDF chapter/section).*

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

### _wp1112 - References .............................................................................................................

### I. Introduction — key questions and headline result
- Political instability, proxied by Cabinet Changes (number of times in a year in which a new premier is named and/or 50 percent or more of the cabinet posts are occupied by new ministers), is widespread and varies regionally.
- Primary research questions:
  - What are the main transmission channels from political instability to economic growth?
  - How quantitatively important are effects on total factor productivity and on physical and human capital accumulation?
- Core empirical headline:
  - An additional cabinet change (a new premier is named and/or 50 percent of cabinet posts are occupied by new ministers) reduces the annual real GDP per capita growth rate by 2.39 percentage points.
  - More than half of this effect operates through lower total factor productivity growth; physical capital accumulation has a slightly larger effect than human capital accumulation among the remaining channels.
- Policy implication emphasized:
  - Countries need to address political instability, tackling root causes and mitigating effects on the quality and sustainability of economic policies that engender economic growth.

### II. Data and empirical model — coverage, variables, and estimation approach
- Sample and sources:
  - Annual data, 1960 to 2004, gathered for 209 countries with up to 169 countries in estimations.
  - Economic data: Penn World Table Version 6.2 – PWT (Heston et al., 2006), World Development Indicators (WDI), Global Development Network Growth Database (GDN), International Financial Statistics (IFS).
  - Political and institutional data: Cross National Time Series Data Archive – CNTS (Databanks International, 2007), Polity IV Database (Marshall and Jaggers, 2005), State Failure Task Force database (SFTF), Gwartney and Lawson (2007).
- Dependent variable and periodicity:
  - GDP per capita growth for consecutive, nonoverlapping, five-year periods: 1960–64, 1965–69, 1970–74, 1975–79, 1980–84, 1985–89, 1990–94, 1995–99, and 2000–04.
- Baseline explanatory variables (averaged over each five-year period except Initial GDP per capita):
  - Initial GDP per capita (log) (PWT)
  - Investment (percent of GDP) (PWT)
  - Primary school enrollment (WDI)
  - Population growth (PWT)
  - Trade openness (PWT)
  - Cabinet Changes (CNTS) — main proxy for political (regime) instability
- Additional variables for macroeconomic stability:
  - Inflation rate (IFS) defined as ln(1+Inf/100)
  - Government (percent of GDP) (PWT)
- Institutional controls in extended model:
  - Index of Economic Freedom (Gwartney and Lawson, 2007)
  - Ethnic Homogeneity Index (SFTF): ranges from 0 to 1
  - Polity Scale (Polity IV): from -10 to 10
- Estimation approach:
  - Dynamic panel data models estimated with system-GMM to address endogeneity and dynamic panel bias.
  - All right-hand-side variables treated as endogenous; instruments: lagged levels and lagged differences as described.
  - Two-step GMM with Windmeijer (2005) finite-sample correction for robust standard errors.

### III. Descriptive statistics (selected exact values from Table 1)
- Growth of GDP per capita: Obs. 1098; Mean 0.016; St. Dev. 0.037; Min. -0.344; Max. 0.347; Source PWT.
- GDP per capita (log): Obs. 1197; Mean 8.315; St. Dev. 1.158; Min. 5.144; Max. 11.346; Source PWT.
- Growth of TFP: Obs. 703; Mean 0.000; St. Dev. 0.048; Min. -0.509; Max. 0.292; Sources PWT, BL.
- Investment (percent of GDP): Obs. 1287; Mean 14.474; St. Dev. 8.948; Min. 1.024; Max. 91.964; Source PWT.
- Primary School Enrollment: Obs. 1286; Mean 88.509; St. Dev. 27.794; Min. 3.000; Max. 149.240; Source WDI-WB.
- Population Growth: Obs. 1521; Mean 0.097; St. Dev. 0.071; Min. -0.281; Max. 0.732; Source PWT.
- Cabinet Changes: Obs. 1322; Mean 0.044; St. Dev. 0.358; Min. 0.000; Max. 2.750; Source CNTS.
- Index of Economic Freedom: Obs. 679; Mean 5.682; St. Dev. 1.208; Min. 2.004; Max. 8.714; Source EFW.
- Polity Scale: Obs. 1194; Mean 0.239; St. Dev. 7.391; Min. -10.000; Max. 10.000; Source Polity IV.
- Ethnic Homogeneity Index: Obs. 1129; Mean 0.583; St. Dev. 0.277; Min. 0.150; Max. 1.000; Source SFTF.

### IV. Empirical results — political instability and growth (system-GMM)
- Baseline results (Table 2, column 1):
  - Cabinet Changes coefficient: -0.0239*** (t-statistic -3.698) → interpreted as an additional cabinet change per year decreases annual growth by 2.39 percentage points.
  - Initial GDP per capita (log): -0.0087** (t = -2.513) — evidence of conditional convergence.
  - Investment (percent of GDP): 0.0009** (t = 2.185).
  - Primary School Enrollment: 0.0003*** (t = 3.097).
  - Population Growth: -0.184*** (t = -3.412).
  - Trade (percent of GDP): 6.70e-05 (t = 0.957), not statistically significant in baseline.
- Extended models and institutional controls:
  - Including Inflation and Government (column 2): Inflation coefficient -0.0091*** (t = -2.837).
  - Index of Economic Freedom (column 3): coefficient 0.0109*** (t = 2.824). A one-point increase in the index raises annual economic growth by one percentage point (as reported).
  - Area 2 (Legal structure and security of property rights) (column 5): coefficient 0.00360* (t = 1.681).
- Robustness of political instability effect:
  - Cabinet Changes remains statistically significant across specifications and when controlling for institutions (Tables 2 and 3).
  - Alternative political-instability indices constructed via principal components all show expected negative signs; Violence Index is not statistically significant, implying regime instability (not violence alone) drives adverse growth effects (Table 4).

### V. Index construction and dimensions of political instability
- Five principal-component indices constructed from CNTS variables:
  - Regime Instability Index 1: Cabinet Changes and Executive Changes.
  - Regime Instability Index 2: Cabinet Changes, Constitutional Changes, Coups, Executive Changes, Government Crises.
  - Regime Instability Index 3: Cabinet Changes, Constitutional Changes, Coups, Executive Changes, Government Crises, Number of Legislative Elections, Fragmentation Index.
  - Violence Index: Assassinations, Coups, Revolutions.
  - Political Instability Index: Assassinations, Cabinet Changes, Constitutional Changes, Coups, Revolutions.
- Table 4 coefficients (selected exact values):
  - Regime Instability Index 1: -0.0198*** (t = -4.851).
  - Regime Instability Index 2: -0.0133*** (t = -3.381).
  - Regime Instability Index 3: -0.0142*** (t = -4.246).
  - Violence Index: -0.0046 (t = -1.197) — not statistically significant.
  - Political Instability Index: -0.0087** (t = -2.255).
- Institutional correlates:
  - Index of Economic Freedom: positive and significant across specifications (e.g., 0.0084** in Table 4 column 1).
  - Ethnic Homogeneity Index: positive and significant (e.g., 0.0497*** in Table 4 column 1).

### VI. Robustness checks — samples and period lengths
- Restricted-sample robustness (Table 5):
  - Cabinet Changes coefficients across seven restricted samples remain negative and statistically significant; examples:
    - Excluding Industrial Countries: -0.0282*** (t = -3.814)
    - Excluding Africa: -0.0285*** (t = -4.588)
    - Excluding Developing Asia: -0.0342*** (t = -3.583)
    - Excluding the 1960s and 1970s: -0.0309*** (t = -3.108)
    - Excluding the 2000s: -0.0326*** (t = -3.693)
  - Regime Instability Index 1 example: -0.0191*** (t = -3.795) when excluding Industrial Countries.
- Alternative period lengths (Table 6):
  - Estimations using 4-, 6-, 8-, and 10-year non-overlapping periods yield consistently negative and generally significant coefficients for Cabinet Changes and the regime instability indexes. Examples:
    - Cabinet Changes: 4-Year Periods -0.0298* (t = -1.683); 6-Year Periods -0.0229** (t = -2.470); 8-Year Periods -0.0121* (t = -1.752); 10-Year Periods -0.0231** (t = -2.004).
    - Regime Instability Index 1: 6-Year Periods -0.0121*** (t = -2.842); 10-Year Periods -0.0213** (t = -2.553).

### VII. Channels of transmission — approach and hypotheses
- Hypothesized channels:
  - Physical capital accumulation: political instability raises uncertainty, reducing investment and slowing accumulation of the physical capital stock.
  - Total factor productivity (TFP): instability may reduce efficient resource allocation, lower R&D, and disrupt firm operations; violence and unrest may damage installed capacity.
  - Human capital accumulation: uncertainty may reduce incentives to invest in education.
- Measurement approach:
  - Production function assumed: Y = A K^(α) (H L)^(1-α) with α = 1/3 (Hall and Jones, 1999 approach).
  - Physical capital stock K constructed using perpetual inventory method:
    - K_t = (1 - δ) K_{t-1} + I_t, with δ assumed to be 6 percent.
    - Initial capital stock K_0 = I_0 / (δ + g), where I_0 is investment in 1950 (or first available year) and g is average geometric growth rate of investment in first 10 years available.
  - Human capital-augmented labor H_i = e^{φ(s_i)} L_i, where s_i is average years of schooling (Barro and Lee, 2001), and φ(s_i) is piecewise linear (construction continues beyond excerpt).

*Source: _wp1112 - References.*

### 0.134 for s

### _wp1112 - 0.134 for s

### Growth accounting framework and data construction
- Production function expression used (after dividing by population N):
  - α α N H A N K N Y 1   = (equation (8) in source)
  - Alternative expression: (α α Ahky  = 1) (equation (9) in source) where:
    - y is real GDP per capita
    - k is physical capital per capita
    - A is total factor productivity (TFP)
    - h is human capital per capita
- Growth-rate decomposition (equation (10)):
  - (hAky11)
  - Contributions computed as growth rates: physical capital accumulation, human capital accumulation, and TFP growth.
- Data construction notes:
  - Depreciation rate of physical capital assumed to be 6 percent.
  - Factor share α assumed equal to 1/3.
  - Series constructed using PWT 6.2 variables: Y = rgdpch*(pop*1000), I = (ki/100)*rgdpl*(pop*1000), L = rgdpch*(pop*1000)/rgdpwok.
  - Population variable pop of PWT 6.2 is scaled in thousands (hence multiplication by 1000).

### Empirical strategy
- System-GMM estimations for dynamic panel-data models.
- Sample period: 1960–2004.
- All explanatory variables treated as endogenous; lagged values two periods used as instruments in first-difference equations and once-lagged first-differences used in levels equation.
- Two-step results using robust standard errors corrected for finite samples (Windmeijer’s, 2005, correction).
- Significance notation: ***, 1 percent; **, 5 percent; *, 10 percent.
- T-statistics reported in parentheses in tables.

### Political instability and physical capital accumulation (Table 7)
- Main findings:
  - Cabinet Changes and the three regime instability indexes are always statistically significant and negative for growth of physical capital per capita.
  - Population Growth coefficients: -0.299***, -0.272***, -0.212**, -0.216***, -0.192** across columns (t-statistics in parentheses).
  - Trade (percent of GDP) shows small positive effects: 0.0001** (column 1) with t-statistic (2.427); other columns show smaller or insignificant magnitudes.
  - Log Physical Capital per capita (-1) and (-2) lags: positive and negative significant coefficients respectively (e.g., 0.1000*** and -0.109*** in column 1).
  - Index of Economic Freedom: 0.0070** in column 1 (t-statistic 2.473); weaker in other columns.
  - Ethnic Homogeneity Index: 0.0343* in column 1 (t-statistic 1.825); not significant in other columns.
- Estimation diagnostics (examples):
  - Number of Observations: 899 (col.1), 531 (col.2–4), 529 (col.5)
  - Number of Countries: 155 (col.1), 108 (col.2–4), 107 (col.5)
  - Hansen test (p-value): 0.0535 (col.1), 0.553, 0.195, 0.426, 0.213
  - AR1 test (p-value): 0.0000009, 0.00002, 0.0001, 0.0002, 0.00006
  - AR2 test (p-value): 0.182, 0.905, 0.987, 0.987, 0.928

### Political instability and TFP growth (Table 8)
- Main findings:
  - Cabinet Changes is statistically significant and negative for TFP growth (e.g., -0.0860*** in column 1; -0.0243* in column 2).
  - Regime instability indexes show negative effects on TFP growth:
    - Regime Instability Index 1: -0.0129** (t-statistic -1.995)
    - Regime Instability Index 2: -0.0084* (t-statistic -1.700)
    - Regime Instability Index 3: -0.0096** (t-statistic -1.976)
  - Index of Economic Freedom positively associated with TFP growth:
    - 0.0190*** (col.2), 0.0225** (col.3), 0.0225** (col.4), 0.0197** (col.5) with respective t-statistics reported.
  - Initial TFP (log) shows convergence: coefficients -0.0338***, -0.0344***, -0.0299***, -0.0308**, -0.0301** across columns.
- Estimation diagnostics (examples):
  - Number of Observations: 700 (col.1), 502 (col.2–4), 498 (col.5)
  - Number of Countries: 105 (col.1), 91 (col.2–5)
  - Hansen test (p-value): 0.501, 0.614, 0.472, 0.253, 0.242
  - AR1 test (p-value): 0.0064, 0.00004, 0.00004, 0.00005, 0.00005
  - AR2 test (p-value): 0.677, 0.898, 0.907, 0.823, 0.811

### Political instability and human capital growth (Table 9)
- Main findings:
  - Cabinet Changes and regime instability indexes are statistically significant and negative for human capital growth:
    - Cabinet Changes: -0.0113** (col.1) and -0.00911** (col.3/4) with corresponding t-statistics.
    - Regime Instability Index 1: -0.00379**; Index 2: -0.00311**; Index 3: -0.00292*.
  - Polity Scale (democracy) positively affects human capital growth:
    - Polity Scale coefficients: 0.0004***, 0.0004***, 0.0005*** in columns 3–5 (t-statistics ~3.17–3.22).
  - Trade (percent of GDP) positively associated with human capital growth:
    - Coefficients include 0.00009**, 0.00006*, 0.0000721**, 0.0000697**, 0.00006* across columns with t-statistics reported.
  - Ethnic Homogeneity Index shows weak positive evidence in some specifications (e.g., 0.00998* in one column).
- Estimation diagnostics (examples):
  - Number of Observations: 704 (col.1), 504 (col.2–4), 500 (col.5)
  - Number of Countries: 105 (col.1), 91 (col.2–5)
  - Hansen test (p-value): 0.406, 0.699, 0.672, 0.703, 0.678
  - AR1 test (p-value): 0.0000001, 0.00001, 0.00001, 0.00002, 0.00003
  - AR2 test (p-value): 0.718, 0.581, 0.525, 0.623, 0.675

### Decomposition of transmission channels (Table 10)
- Method:
  - Effects of each channel on GDP per capita growth obtained by multiplying:
    - coefficient for growth of Physical Capital per capita by α = 1/3
    - coefficient for growth of TFP by (1-α) = 2/3
    - coefficient for growth of Human Capital per capita by (1-α) = 2/3
  - Applied equation: (hαAαkαy11)
- Main quantitative results (for each proxy of political instability):
  - Cabinet Changes:
    - Coefficients: Physical Capital pc = -0.0195***; TFP = -0.0243*; Human Capital pc = -0.00911**
    - Effect on GDP: -0.0065 (physical capital), -0.0162 (TFP), -0.0061 (human capital); Total = -0.0288
    - Percent of Total Effect: 22.59% (physical capital), 56.30% (TFP), 21.11% (human capital)
  - Regime Instability Index 1:
    - Coefficients: Physical Capital pc = -0.0108**; TFP = -0.0129**; Human Capital pc = -0.00379**
    - Effect on GDP: -0.0036, -0.0086, -0.0025; Total = -0.0147
    - Percent of Total Effect: 24.44%, 58.40%, 17.16%
  - Regime Instability Index 2:
    - Coefficients: Physical Capital pc = -0.00932**; TFP = -0.00846*; Human Capital pc = -0.00311**
    - Effect on GDP: -0.0031, -0.0056, -0.0021; Total = -0.0108
    - Percent of Total Effect: 28.71%, 52.13%, 19.16%
  - Regime Instability Index 3:
    - Coefficients: Physical Capital pc = -0.00906**; TFP = -0.00964**; Human Capital pc = -0.00292*
    - Effect on GDP: -0.0030, -0.0064, -0.0019; Total = -0.0114
    - Percent of Total Effect: 26.51%, 56.41%, 17.08%
- Interpretation:
  - More than half of the total negative effects of political instability on real GDP per capita growth operate through adverse effects on TFP growth (between 52.13 percent and 58.40 percent depending on the proxy).
  - Physical capital accumulation accounts for 22.59 percent to 28.71 percent of the total effect.
  - Human capital growth accounts for 17.08 percent to 21.11 percent of the total effect.
  - These shares are consistent with growth-accounting literature benchmarks where human capital accounts for 10–30 percent of income differences, physical capital about 20 percent, and TFP 60–70 percent.

### Robustness and additional notes
- Wald tests comparing total effects from decomposition to coefficients estimated in GDP per capita growth regressions do not reject equality (examples provided with chi2 and Prob>chi2 values).
- Interaction among transmission channels:
  - Growth rates of one channel do not appear to be significantly affected by growth rates of the other channels (authors report no significant interactions; additional estimations available upon request).
- Variables used to construct series were excluded from corresponding regressions to avoid mechanical correlation (e.g., Investment percent of GDP not included when constructing physical capital; Primary School Enrollment not included when constructing human capital), with robustness checks confirming main results when included.

### Conclusions and policy implications
- Political instability significantly and economically reduces growth.
- Main transmission channels identified and quantified:
  - TFP growth is the primary channel through which political instability reduces GDP per capita growth.
  - Physical capital accumulation and human capital growth are secondary channels.
- Policy implication emphasized:
  - Governments in politically fragmented countries with high degrees of political instability need to address root causes and mitigate effects on the design and implementation of economic policies to achieve durable economic policies and higher economic growth.

*Source: _wp1112 - 0.134 for s (PDF chapter/section, canonical source URL provided in content unit)*

### References

### _wp1112 - References

### Institutions and Growth
- Acemoglu, D., Johnson, S. and Robinson, J. (2001). “The colonial origins of comparative development: An empirical investigation.” American Economic Review 91, 1369–1401.  
- Acemoglu, D., Johnson, S., Robinson, J. and Yared, P. (2008), “Income and Democracy.” American Economic Review 98(3), 808–842.  
- Caselli, F. (2005), “Accounting for cross-country income differences,” in P. Aghion and S. Durlauf, eds, Handbook of Economic Growth, Amsterdam: North Holland, pp. 679–741.  
- Caselli, F., Esquivel, G. and Lefort, F. (1996). “Reopening the Convergence Debate: A New Look at Cross-Country Growth Empirics.” Journal of Economic Growth 1, 363–390.  
- Glaeser, E., La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (2004). “Do institutions cause growth?” Journal of Economic Growth 9, 271–303.  
- Hall, R. and Jones, C. (1999). “Why do some countries produce so much more output per worker than others?” Quarterly Journal of Economics 114, 83–116.  
- Mankiw, N. G., Romer, D. and Weil, D. (1992), “A contribution to the empirics of economic growth.” Quarterly Journal of Economics 107, 407–437.  
- Rodrik, D. (1991). “Policy uncertainty and private investment in developing countries.” Journal of Development Economics 36, 229–242.  
- Tavares, J. and Wacziarg, R. (2001) “How democracy affects growth.” European Economic Review 45, 1341–1378.  
- De Haan, J. (2007). “Political institutions and economic growth reconsidered.” Public Choice 127, 281–292.  
- Durlauf, S., Johnson, P. and Temple, J. (2005). “Growth econometrics.” In: Aghion, P., Durlauf, S. (Eds.), Handbook of Economic Growth. Amsterdam: North Holland, pp. 555–677.  
- Caselli, F., Esquivel, G. and Lefort, F. (1996). “Reopening the Convergence Debate: A New Look at Cross-Country Growth Empirics.” Journal of Economic Growth 1, 363–390.  

### Political Instability, Social Conflict, and Democracy
- Aisen, A. and Veiga, F.J. (2006). “Does Political Instability Lead to Higher Inflation? A Panel Data Analysis.” Journal of Money, Credit and Banking 38(5), 1379–1389.  
- Alesina, A. and Perotti, R. (1996). “Income distribution, political instability, and investment.” European Economic Review 40, 1203- 1228.  
- Alesina, A., Ozler, S., Roubini, N. and Swagel, P. (1996). “Political instability and economic growth.” Journal of Economic Growth 1, 189–211.  
- Benhabib, J. and Rustichini, A. (1996). “Social conflict and growth.” Journal of Economic Growth 1, 125–142.  
- Campos, N. and Nugent, J. (2002). “Who is afraid of political instability?” Journal of Development Economics 67, 157–172.  
- Easterly, W., Ritzen, J. and Wollcock, M. (2006). “Social cohesion, institutions and growth.” Economics & Politics 18(2), 103–120.  
- Jong-a-Pin, R. (2009). “On the measurement of political instability and its impact on economic growth.” European Journal of Political Economy 25, 15–29.  
- Klomp, J. and de Haan, J. (2009). “Political institutions and economic volatility.” European Journal of Political Economy 25, 311–326.  
- Mauro, P. (1995). “Corruption and growth.” Quarterly Journal of Economics 110(3), 681–712.  
- Perotti, R. (1996). “Growth, income distribution, and democracy: what the data say.” Journal of Economic Growth 1, 149–187  
- Özler, S. and Rodrik, D. (1992). "External shocks, politics and private investment: Some theory and empirical evidence." Journal of Development Economics 39(1), 141–162.  

### Finance, Trade, and International Integration
- Beck, T. (2008), “The econometrics of finance and growth.” Policy Research Working Paper, WPS4608, World Bank.  
- Edison, H. J., Levine, R., Ricci, L. and Sløk, T. (2002). “International financial integration and economic growth.” Journal of International Money and Finance 21, 749–776.  
- La-Porta, R., Lopez-De-Silanes, F., Shleifer, A. and Vishny, R. (1997), “Legal determinants of external finance.” The Journal of Finance 52, 1131–1150.  
- Levine, R., Loayza, N. and Beck, T. (2000), “Financial intermediation and growth: Causality and causes.” Journal of Monetary Economics 46, 31–77.  

### Education, Human Capital, and Productivity
- Barro, R. and Lee, J. (2001). “International data on educational attainment: updates and implications.” Oxford Economic Papers 53, 541–563.  
- Hsieh, C.T. and Klenow, P. (2010). “Development Accounting.” American Economic Journal: Macroeconomics 2(1), 207–223.  

### Econometric Methods, Panels, and Inference
- Arellano, M. and Bond, S. (1991). “Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations.” The Review of Economic Studies 58, 277–297.  
- Arellano, M. and Bover, O. (1995). “Another look at the instrumental variable estimation of error-component models.” Journal of Econometrics 68, 29–51.  
- Baltagi, B. H. (2008). Econometric Analysis of Panel Data. 4th ed. Chichester: John Wiley & Sons.  
- Blundell, R. and Bond, S. (1998). “Initial conditions and moment restrictions in dynamic panel data models.” Journal of Econometrics 87, 115–143.  
- Bond, S., Hoeffler, A.and Temple, J. (2001). “GMM Estimation of Empirical Growth Models” Center for Economic Policy Research, 3048  
- Holtz-Eakin, D., Newey, W. and Rosen, H.S. (1988). “Estimating vector autoregressions with panel data.” Econometrica 56, 1371–1395.  
- Judson, R.A. and Owen, A.L. (1999). “Estimating dynamic panel data models: A practical guide for macroeconomists.” Economics Letters 65, 9–15.  
- Windmeijer, F. (2005). “A finite sample correction for the variance of linear efficient two-step GMM estimators.” Journal of Econometrics 126, 25–51.  
- Elder, J. (2004). “Another perspective on the effects of inflation uncertainty.” Journal of Money, Credit and Banking 36(5), 911–28.  

### Data Sources and Compilations
- Databanks International (2007). Cross National Time Series Data Archive, 1815–2007. Binghampton, NY (http://www.databanksinternational.com/).  
- Heston, A., Summers, R. and Aten, B. (2006). Penn World Table Version 6.2. Center for International Comparisons at the University of Pennsylvania (CICUP). Data set downloadable at: http://pwt.econ.upenn.edu/.  
- Marshall, M. and Jaggers, K. (2005). Polity IV Project: Political Regime Characteristics and Transitions, 1800–2004. Center for Global Policy, George Mason University. Data set downloadable at: http://www.systemicpeace.org/polity/polity4.htm.  
- Gwartney, J. and Lawson, R. (2007). Economic Freedom of the World - 2007 Annual Report. Vancouver, BC: Fraser Institute.  

_References list from _wp1112 - References_

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1112.pdf_