Acceso abierto

Conflict and the composition of economic activity in Afghanistan

Virgilio Galdo
Virgilio Galdo
, 
Gladys Lopez Acevedo
Gladys Lopez Acevedo
 y 
Martin Rama
Martin Rama
   | 07 sept 2021
IZA Journal of Development and Migration's Cover Image
Acerca de este artículo
Artículo anterior
Artículo siguiente
Cite
Publicado en línea: 07 sept 2021
Páginas: -
Aceptado: 08 jul 2021
DOI: https://doi.org/10.2478/izajodm-2021-0010
Palabras clave
© 2021 Virgilio Galdo et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1

Conflict-related casualties and injured in Afghanistan. (A) Casualties and injured over time. (B) Casualties and injured across space.Source: Author's calculations based on UNAMA (2019) and GTD (2016). GTD, global terrorism database.
Conflict-related casualties and injured in Afghanistan. (A) Casualties and injured over time. (B) Casualties and injured across space.Source: Author's calculations based on UNAMA (2019) and GTD (2016). GTD, global terrorism database.

Figure 2

Economic activity in Afghanistan. (A) Firm registration by sector. (B) GDP growth rates by sector.Source: World Bank (2018). GDP, gross domestic product.
Economic activity in Afghanistan. (A) Firm registration by sector. (B) GDP growth rates by sector.Source: World Bank (2018). GDP, gross domestic product.

Figure 3

Response estimates of conflict on Economic Activity. (A) Conflict and nighttime light. (B) Conflict and active formal firms. (C) Conflict and land devoted to poppy cultivation.Note: Coefficients from regressing log (sum of lights per area), log (active formal firms per 1,000 inhabitants) and log (Poppy cultivation per 1,000 inhabitants) on lead and lags of conflict, controlling for district and time fixed effects, troops’ presence and aid. Coefficients for t−4 to t−1 come from regressions (A4–A6), while coefficients for t+1 to t+4 come from regressions (A1–A3). Coefficient t−1 refers to the lead (t+1) estimate. Conversely coefficient t+1 refers to the lag (t−1) estimate. Confidence intervals in whiskers are derived from district-clustered standard errors.
Response estimates of conflict on Economic Activity. (A) Conflict and nighttime light. (B) Conflict and active formal firms. (C) Conflict and land devoted to poppy cultivation.Note: Coefficients from regressing log (sum of lights per area), log (active formal firms per 1,000 inhabitants) and log (Poppy cultivation per 1,000 inhabitants) on lead and lags of conflict, controlling for district and time fixed effects, troops’ presence and aid. Coefficients for t−4 to t−1 come from regressions (A4–A6), while coefficients for t+1 to t+4 come from regressions (A1–A3). Coefficient t−1 refers to the lead (t+1) estimate. Conversely coefficient t+1 refers to the lag (t−1) estimate. Confidence intervals in whiskers are derived from district-clustered standard errors.

Figure 4

The impact of a surge in conflict on economic activity. (A) Temporary surge in conflict. (B) Permanent surge in conflict.
The impact of a surge in conflict on economic activity. (A) Temporary surge in conflict. (B) Permanent surge in conflict.

Conflict and nighttime light (placebo test)

log (sum of lights per area)
(1) (2) (3) (4) (5)
Conflicitt+1 −0.000553 (0.00782) 0.00363 (0.00638) 0.00357 (0.00636) 0.00365 (0.00639) 0.00359 (0.00638)
Conflicitt+2 −0.00941 (0.00774) −0.00527 (0.00738) −0.00538 (0.00739) −0.00514 (0.00739) −0.00525 (0.00740)
Conflicitt+3 −0.00223 (0.00642) 0.00213 (0.00572) 0.00205 (0.00573) 0.00224 (0.00571) 0.00216 (0.00572)
Conflicitt+4 0.00366 (0.00638) 0.00552 (0.00588) 0.00544 (0.00587) 0.00560 (0.00588) 0.00553 (0.00587)
Troopst−1 0.0164*** (0.00392) 0.0164*** (0.00392)
  ISAFt−1 0.0172*** (0.00398) 0.0172*** (0.00397)
  ANAt−1 0.0573* (0.0345) 0.0594* (0.0345)
AIDt−1 1.19e–06 (9.56e–07) 1.02e–06 (9.71e–07)
  Securityt−1 1.27e–05*** (4.85e–06) 1.29e–05*** (4.79e–06)
  Other aidt−1 9.39e–07 (9.81e–07) 7.54e–07 (9.97e–07)
District fixed effects Yes Yes Yes Yes Yes
Time fixed effects Yes Yes Yes Yes Yes
Observations 5,572 5,572 5,572 5,572 5,572
Adj. R2 0.993 0.994 0.994 0.994 0.994

Conflict and nighttime light

log (sum of lights per area)
(1) (2) (3) (4) (5)
Conflicitt−1 −0.0110** (0.00488) −0.00833* (0.00499) −0.00825* (0.00499) −0.00826* (0.00500) −0.00821* (0.00496)
Conflicitt−2 −0.0203*** (0.00659) −0.0173*** (0.00632) −0.0173*** (0.00632) −0.0172*** (0.00631) −0.0172*** (0.00631)
Conflicitt−3 −0.0103 (0.00764) −0.00742 (0.00676) −0.00751 (0.00676) −0.00744 (0.00675) −0.00749 (0.00675)
Conflicitt−4 −0.00835 (0.00792) −0.00836 (0.00713) −0.00837 (0.00712) −0.00838 (0.00712) −0.00839 (0.00710)
Troopst−1 0.0211*** (0.00455) 0.0212*** (0.00453)
  ISAFt−1 0.0217*** (0.00476) 0.0215*** (0.00473)
  ANAt−1 −0.00943 (0.0758) 0.00251 (0.0763)
AIDt−1 4.83e–07*** (1.85e–07) 4.91e–07*** (1.84e–07)
  Securityt−1 1.12e–05** (4.81e–06) 1.10e–05** (4.81e–06)
  Other aidt−1 4.40e–07** (1.85e–07) 4.45e–07** (1.83e–07)
District fixed effects Yes Yes Yes Yes Yes
Time fixed effects Yes Yes Yes Yes Yes
Observations 5,970 5,970 5,970 5,970 5,970
Adj. R2 0.994 0.994 0.994 0.994 0.994

Conflict and active formal firms

log (active formal firms per 1,000 inhabitants)
(1) (2) (3) (4) (5)
Conflicitt−1 −0.0512* (0.0271) −0.0488** (0.0247) −0.0490** (0.0247) −0.0489** (0.0246) −0.0492** (0.0246)
Conflicitt−2 −0.0401 (0.0303) −0.0388 (0.0284) −0.0388 (0.0283) −0.0390 (0.0284) −0.0390 (0.0283)
Conflicitt−3 −0.00931 (0.0250) −0.00819 (0.0243) −0.00792 (0.0242) −0.00852 (0.0243) −0.00819 (0.0242)
Conflicitt−4 −0.0338 (0.0251) −0.0318 (0.0236) −0.0316 (0.0235) −0.0320 (0.0237) −0.0318 (0.0235)
Troopst−1 0.0264 (0.0264) 0.0264 (0.0264)
  ISAFt−1 0.0239 (0.0257) 0.0230 (0.0256)
  ANAt−1 0.172 (0.196) 0.224 (0.209)
AIDt−1 −1.42e–06** (6.70e–07) −1.42e–06** (6.72e–07)
  Securityt−1 1.83e–05*** (6.39e–06) 2.09e–05*** (7.92e–06)
  Other aidt−1 −1.56e–06** (6.89e–07) −1.58e06** (6.90e–07)
District fixed effects Yes Yes Yes Yes Yes
Time fixed effects Yes Yes Yes Yes Yes
Observations 3,159 3,159 3,159 3,159 3,159
Adj. R2 0.835 0.836 0.836 0.836 0.836

Conflict and active formal firms (placebo test)

log (active formal firms per 1,000 inhabitants)
(1) (2) (3) (4) (5)
Conflicitt+1 −0.00843 (0.0189) −0.00643 (0.0182) −0.00645 (0.0181) −0.00650 (0.0182) −0.00654 (0.0181)
Conflicitt+2 −0.00230 (0.0200) −0.000640 (0.0215) −0.000756 (0.0215) −0.000623 (0.0216) −0.000808 (0.0215)
Conflicitt+3 −0.0147 (0.0123) −0.0122 (0.0130) −0.0126 (0.0129) −0.0123 (0.0129) −0.0128 (0.0128)
Conflicitt+4 −0.00947 (0.00771) −0.00736 (0.00869) −0.00757 (0.00861) −0.00736 (0.00874) −0.00769 (0.00864)
Troopst−1 0.0198 (0.0193) 0.0196 (0.0193)
  ISAFt−1 0.0181 (0.0189) 0.0169 (0.0189)
  ANAt−1 0.135 (0.168) 0.207 (0.192)
AIDt−1 −2.55e–07 (9.50e–07) −7.21e–08 (1.03e–06)
  Securityt−1 1.51e–05** (6.01e–06) 1.92e–05** (8.79e–06)
  Other aidt−1 −8.04e–07 (9.84e–07) −6.43e–07 (1.02e–06)
District fixed effects Yes Yes Yes Yes Yes
Time fixed effects Yes Yes Yes Yes Yes
Observations 2,673 2,673 2,673 2,673 2,673
Adj. R2 0.889 0.889 0.889 0.889 0.889

Conflict and land devoted to poppy cultivation

log (Poppy cultivation per 1,000 inhabitants)
(1) (2) (3) (4) (5)
Conflicitt−1 0.00054 (0.0395) 0.000385 (0.0395) 0.0107 (0.0397) −0.00158 (0.0395) 0.00817 (0.0398)
Conflicitt−2 0.0640* (0.0365) 0.0613* (0.0365) 0.0640* (0.0369) 0.0603* (0.0365) 0.0630* (0.0369)
Conflicitt−3 0.045 (0.0530) 0.0429 (0.0528) 0.053 (0.0525) 0.045 (0.0526) 0.0539 (0.0524)
Conflicitt−4 0.0674 (0.0614) 0.0667 (0.0610) 0.0626 (0.0593) 0.0676 (0.0610) 0.0637 (0.0595)
Troopst−1 −0.00704 (0.00910) −0.00729 (0.00903)
  ISAFt−1 0.0105 (0.0124) 0.00877
(0.0122)
  ANAt−1 −0.0478** (0.0186) −0.0444** (0.0187)
AIDt−1 −5.10e–06** (2.41e–06) −4.79e–06** (2.42e–06)
  Securityt−1 −0.000274*** (9.41e–05) −0.000233** (9.14e–05)
  Other aidt−1 −5.05e–06** (2.41e–06) −4.77e–06** (2.42e–06)
District fixed effects Yes Yes Yes Yes Yes
Time fixed effects Yes Yes Yes Yes Yes
Observations 3,582 3,582 3,582 3,582 3,582
Adj. R2 0.835 0.836 0.836 0.836 0.836

Summary statistics

Variable N Mean Min Max Std. Dev.
Nighttime light (nanowatts) 5,970 0.68 0.00 89.43 5.14
Active formal firms (per 1,000 inhabitants) 3,159 1.12 0.01 18.71 1.68
Poppy cultivation (hectares per 1,000 inhabitants) 3,582 6.73 0.00 416.5 27.20
Killings (per 1,000 inhabitants) 5,970 0.06 0.00 6.81 0.22
ISAF troops (per 1,000 inhabitants) 646 2.66 0.00 38.73 5.46
ANA troops (per 1,000 inhabitants) 646 4.64 0.00 19.32 5.50
Security-related aid (USD per quarter per 1,000 inhabitants) 646 119 0.00 16,555 878
Other aid (USD per quarter per 1,000 inhabitants) 646 15,492 0.00 242,573 27,538

Conflict and land devoted to poppy cultivation (placebo test)

log (Poppy cultivation per 1,000 inhabitants)
(1) (2) (3) (4) (5)
Conflicitt+1 −0.0248 (0.0514) −0.0120 (0.0519) −0.0123 (0.0521) −0.0143 (0.0522) −0.0151 (0.0524)
Conflicitt+2 0.0140 (0.0305) 0.0212 (0.0319) 0.0212 (0.0319) 0.0207 (0.0319) 0.0207 (0.0319)
Conflicitt+3 −0.0537 (0.0775) −0.0699 (0.0664) −0.0705 (0.0669) −0.0702 (0.0666) −0.0718 (0.0669)
Conflicitt+4 0.0668 (0.0551) 0.0689 (0.0595) 0.0687 (0.0595) 0.0698 (0.0597) 0.0695 (0.0599)
Troopst−1 −0.0250*** (0.00897) −0.0248*** (0.00893)
  ISAFt−1 −0.0264* (0.0136) −0.0287** (0.0132)
  ANAt−1 −0.0223 (0.0193) −0.0176 (0.0193)
AIDt−1 −1.58e–05** (6.61e–06) −1.58e–05** (6.63e–06)
  Securityt−1 −0.000682** (0.000275) −0.000704*** (0.000266)
  Other aidt−1 −1.56e–05** (6.59e–06) −1.58e–05** (6.59e–06)
District fixed effects Yes Yes Yes Yes Yes
Time fixed effects Yes Yes Yes Yes Yes
Observations 3,184 3,184 3,184 3,184 3,184
Adj. R2 0.778 0.813 0.813 0.813 0.813
eISSN:
2520-1786
Idioma:
Inglés
Calendario de la edición:
Volume Open
Temas de la revista:
Business and Economics, Political Economics, Microeconomics, Macroecomics, Economic Policy, Mathematics and Statistics for Economists, Econometrics
RSS Feed de revista