## Measuring the Fiscal Revenue Cost of Conflict in Afghanistan

## Source details

**Canonical URL:** [Measuring the Fiscal Revenue Cost of Conflict in Afghanistan](https://www.imf.org/-/media/files/publications/cr/2017/cr17378.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2017/cr17378.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2017/cr17378.pdf.json)

---

### Overview
- Purpose: Estimate the fiscal revenue cost of conflict in Afghanistan — the loss of government domestic revenue due to conflict — and interpret this as an approximate measure of the potential fiscal gains from peace.
- Main conclusion:
  - Conflict-related violence reduced annual national revenues in 2016 by around 50 percent — around Af 70 billion, or roughly $1.0 billion — relative to 2005 levels of violence.
- Contextual points:
  - Almost all security spending is funded by foreign grants; hence fiscal gains from peace are likely to come principally from increased revenues rather than reduced security spending.
  - The estimated revenue loss matches output loss estimates from Mueller & Tobias 2016; the equivalent reduction in annual revenue growth implied by the study’s estimates is around 6 percentage points.

### Data on Conflict and Revenue
- Conflict data source: Uppsala Conflict Data Program (UCDP).
  - UCDP records over 100,000 deaths from conflict in Afghanistan during 2005–16 in some 20,000 different incidents.
  - UCDP records for Afghanistan are available from 1989; this study restricts to post-2005 to match revenue data.
- Temporal dynamics (2001–16):
  - 2001–05: some 4,500 deaths nationwide.
  - 2006–10: deaths increased above 30,000, over half in the Southern region (Helmand, Kandahar, Nimroz, Uruzgan, Zabul; ~11 percent of national population).
  - 2011–14: conflict deaths declined as foreign troop numbers peaked.
  - 2015–16: troops left; conflict intensified and deaths increased again.
- International comparisons:
  - Recent violence intensity in Afghanistan is at least as severe as that experienced in Iraq in 2014 (UCDP deaths per 1000 population).
  - Syria in 2016: UCDP per capita deaths from conflict were over five times those recorded in Afghanistan.
- Revenue data:
  - Source: Ministry of Finance monthly financial statements; focus on domestic revenue shown annually as a share of GDP.
  - Revenues have risen considerably since 2005 due to efficiency improvements and policy changes (including a tax on mobile phone top-ups).
  - Revenue decline in 2014 was linked to economic slowdown after drawdown of international forces.
- Data caveats:
  - Recording deaths from conflict is challenging and likely subject to under-reporting, but analysis is conducted in relative terms so constant under-reporting is less consequential.
  - Afghanistan has not had a census since 1979; provincial population numbers are rough estimates.

### Estimation Method
- Model: Dynamic panel regression (ordinary least squares) estimating log monthly provincial revenues (y_{i,t}) on log conflict deaths (s_{i,t}), controls, province fixed effects μ_i and period fixed effects δ_t.
- Controls included:
  - Time fixed effects (national macro conditions, seasonality).
  - Province fixed effects (permanent province-specific factors).
  - Provincial average precipitation (Centre for Environmental Data Analysis).
  - Regional opium price.
  - Lags of conflict deaths (to capture delayed effects).
  - Lagged revenues (to control for reverse causality).
- Interpretation: Coefficients are elasticities (e.g., coefficient –0.01 means revenues decrease by 1 percent when conflict increases by 100 percent).
- Specification progression:
  - Column (1) (no fixed effects): positive coefficient on conflict due to common upward trends in revenue and conflict.
  - Columns (2)–(6): progressively add fixed effects, rainfall, opium price, lagged conflict, lagged revenues; contemporaneous or lagged conflict impacts are negative, stable, and statistically significant at the 5 percent level in richer specifications.
  - Column (7): specification with up to six months of past conflict — almost all conflict coefficients negative and jointly significant at the 1 percent level.
  - Column (8): preferred specification replaces lagged conflict variables by their 6-month average.
- Key regression statistics (from Table 1, column (8)):
  - Log conflict deaths (column (8)): 0.017 (standard error (0.010))
  - Log conflict deaths: 6 month lag average (column (8)): -0.042*** (standard error (0.016))
  - Revenue lags in column (8): 12
  - Precipitation lags in column (8): 12
  - Regional opium price lags in column (8): 12
  - Observations (column (8)): 2,405
  - R2 (column (8)): 0.214
  - Adjusted R2 (column (8)): 0.142
  - Note: significance markers are as reported: * p<0.1, ** p<0.05, *** p<0.01. Heteroskedastic and autocorrelation consistent standard errors reported in parentheses.

### Estimated Effects and Scenarios
- Short- and long-run effects (preferred specification using average lagged conflict):
  - Long-run elasticity computed directly: (0.017 - 0.042)/(1 - 0.735) = –0.102 (an elasticity of 10.2 percent).
- Scenario responses (illustrated in Figures 5 and 6):
  - Temporary doubling of conflict deaths in first period (then return): overall loss cumulates to around 10 percent of one month’s revenues.
  - Permanent doubling of violence: long-run impact is a revenue loss of over 10 percent per month.
  - Initial short-term positive response in revenues observed (may reflect local demand or pre-collection of debts) — effect only just significant at the 10 percent level in specification (7).
- Aggregated national impact:
  - National revenue loss computed by holding provincial deaths at their 2005 levels as counterfactual and aggregating provincial effects.
  - Estimated monthly revenue loss series plotted 2005–2016 shows increasing loss as conflict grows.
  - 2016 estimated revenue loss:
    - Around 50 percent of revenues.
    - About Af 70 billion.
    - About 5 percent of GDP.
    - Roughly $1 billion.
- Distributional insights:
  - Cross-provincial patterns account for about half the estimated lost revenue.
  - If all provinces had experienced the same relative increase in conflict, the national revenue loss would be about half the size.
  - Provinces that generate more revenue experienced larger increases in violence, producing outsized impacts on national totals.
- Additional illustrative calculation:
  - Conflict deaths increased tenfold; model predicts a long-run impact of around log(10) * –0.102 = –0.24, i.e., a 24 percent loss.

### Key Findings and Implications
- Quantitative findings:
  - UCDP: over 100,000 conflict deaths during 2005–16 across ~20,000 incidents.
  - 2016 revenue loss due to conflict: around 50 percent of revenues ≈ Af 70 billion ≈ $1.0 billion ≈ 5 percent of GDP.
  - Long-run elasticity of revenue to conflict (preferred estimate): –0.102.
  - Temporary doubling of violence → cumulated loss ~10 percent of one month’s revenues; permanent doubling → >10 percent monthly revenue loss.
- Policy-relevant implications:
  - Lost domestic revenue is a major component of the humanitarian cost of conflict because it reduces government capacity to provide health, education, security, and infrastructure.
  - A sizeable potential fiscal dividend exists should peace or a significant reduction in violence materialize, primarily via revenue gains rather than reduced security spending (as security spending is largely grant-funded).
  - Provincial targeting matters: provinces with higher revenue bases experiencing larger increases in violence drive national revenue losses; policy to protect revenue-generating provinces may yield outsized national benefits.

### Conclusion
- The fiscal revenue cost of conflict in Afghanistan is substantial and robust in order of magnitude, even accounting for uncertainty in estimates; the study’s central estimate is around $1 billion of annual revenue loss in 2016.
- These results imply a sizeable potential fiscal dividend if significant reductions in violence occur, improving the government’s ability to finance basic services.

### Box 1 — Key Facts About Afghanistan’s Financial Sector (selected entries)
- Financial sector composition (selected counts by year):
  - 2014: 16 16 35 34 62 30 91 1,772
  - 2015: 15 17 35 35 83 33 41 1,415
  - 2016: 15 17 45 36 69 34 82 2,192
  - 2017: 15 17 44 36 06 29 11 1,705
- Key financial indicators (December, 2016):
  - Credit to GDP: 4 percent
  - Deposit dollarization: 66.2 percent
  - Loan dollarization: 61.4 percent
  - Currency to deposit ratio: 93.4 percent
  - Loans to deposit ratio: 19.9 percent
- Microfinance sector (Q1 2017 and banking sector comparison):
  - Microfinance loan portfolio: Af 10.5 billion (0.8 percent of GDP)
  - Microfinance savings deposits: Af 2.5 billion (0.2 percent of GDP)
  - Banking sector loan portfolio: Af 47.4 billion (3.6 percent of GDP)
  - Banking sector total deposits: Af 234 billion (17.7 percent of GDP)
- Financial access for firms (selected indicators):
  - Percent of firms with a checking or savings account: Afghanistan 43.7; South Asia 77.6; All countries 86.8
  - Percent of firms with a bank loan/line of credit: Afghanistan 5.1; South Asia 27.0; All countries 33.8
  - Percent of firms identifying access to finance as a major constraint: Afghanistan 47.6; South Asia 26.5; All countries 26.5
- Policy recommendations to boost financial inclusion:
  - Strengthen the legal framework and remove barriers to financial development while preserving financial stability.
  - Strengthen and extend the perimeter of financial supervision by including microfinance institutions.
  - Develop mobile banking to improve access in remote areas; draw on international experience for regulations and policies.
  - Develop Sharia compliant instruments to attract customers who do not use existing financial institutions for religious reasons.
  - Reduce costs of operation of financial services providers; introduce unique biometric identification for all Afghanistan citizens to resolve documentation problems.
  - Develop the insurance sector to reduce risks for banks and likely reduce costs associated with lending.
  - Promote financial literacy (international experience summarized in Box 2).

*Prepared by Philip Barrett (FAD) with Mariusz Sumlinski (MCD) and research support from Tetyana Sydorenko.*

### References _____________________________________________________________________________ 11

### Measuring the Fiscal Revenue Cost of Conflict in Afghanistan

### Overview
- Purpose: Estimate the fiscal revenue cost of conflict in Afghanistan — the loss of government domestic revenue due to conflict — and interpret this as an approximate measure of the potential fiscal gains from peace.
- Main conclusion: Conflict-related violence reduced annual national revenues in 2016 by around 50 percent — around Af 70 billion, or roughly $1.0 billion — relative to 2005 levels of violence.
- Contextual points:
  - Almost all security spending is funded by foreign grants; hence fiscal gains from peace are likely to come principally from increased revenues rather than reduced security spending.
  - The estimated revenue loss matches output loss estimates from Mueller & Tobias 2016; the equivalent reduction in annual revenue growth implied by the study’s estimates is around 6 percentage points.

### Data on Conflict and Revenue
- Conflict data source: Uppsala Conflict Data Program (UCDP).
  - UCDP records over 100,000 deaths from conflict in Afghanistan during 2005–16 in some 20,000 different incidents.
  - UCDP records for Afghanistan are available from 1989; this study restricts to post-2005 to match revenue data.
- Temporal dynamics (2001–16):
  - 2001–05: some 4,500 deaths nationwide.
  - 2006–10: deaths increased above 30,000, over half in the Southern region (Helmand, Kandahar, Nimroz, Uruzgan, Zabul; ~11 percent of national population).
  - 2011–14: conflict deaths declined as foreign troop numbers peaked.
  - 2015–16: troops left; conflict intensified and deaths increased again.
- International comparisons:
  - Recent violence intensity in Afghanistan is at least as severe as that experienced in Iraq in 2014 (UCDP deaths per 1000 population).
  - Syria in 2016: UCDP per capita deaths from conflict were over five times those recorded in Afghanistan.
- Revenue data: Ministry of Finance monthly financial statements; focus on domestic revenue shown annually as a share of GDP.
  - Revenues have risen considerably since 2005 due to efficiency improvements and policy changes (including a tax on mobile phone top-ups).
  - Revenue decline in 2014 was linked to economic slowdown after drawdown of international forces.
- Data caveats:
  - Recording deaths from conflict is challenging and likely subject to under-reporting, but analysis is conducted in relative terms so constant under-reporting is less consequential.
  - Afghanistan has not had a census since 1979; provincial population numbers are rough estimates.

### Estimation Method
- Model: Dynamic panel regression (ordinary least squares) estimating log monthly provincial revenues (y_{i,t}) on log conflict deaths (s_{i,t}), controls, province fixed effects μ_i and period fixed effects δ_t.
- Controls included:
  - Time fixed effects (national macro conditions, seasonality).
  - Province fixed effects (permanent province-specific factors).
  - Provincial average precipitation (Centre for Environmental Data Analysis).
  - Regional opium price.
  - Lags of conflict deaths (to capture delayed effects).
  - Lagged revenues (to control for reverse causality).
- Interpretation: Coefficients are elasticities (e.g., coefficient –0.01 means revenues decrease by 1 percent when conflict increases by 100 percent).
- Specification progression:
  - Column (1) (no fixed effects): positive coefficient on conflict due to common upward trends in revenue and conflict.
  - Columns (2)–(6): progressively add fixed effects, rainfall, opium price, lagged conflict, lagged revenues; contemporaneous or lagged conflict impacts are negative, stable, and statistically significant at the 5 percent level in richer specifications.
  - Column (7): specification with up to six months of past conflict — almost all conflict coefficients negative and jointly significant at the 1 percent level.
  - Column (8): preferred specification replaces lagged conflict variables by their 6-month average.
- Key regression statistics (from Table 1, exact figures preserved):
  - Log conflict deaths (column (8)): 0.017 (standard error (0.010))
  - Log conflict deaths: 6 month lag average (column (8)): -0.042*** (standard error (0.016))
  - Revenue lags in column (8): 12
  - Precipitation lags in column (8): 12
  - Regional opium price lags in column (8): 12
  - Observations (column (8)): 2,405
  - R2 (column (8)): 0.214
  - Adjusted R2 (column (8)): 0.142
  - Note: significance markers are as reported: * p<0.1, ** p<0.05, *** p<0.01. Heteroskedastic and autocorrelation consistent standard errors reported in parentheses.

### Estimated Effects and Scenarios
- Short- and long-run effects (preferred specification using average lagged conflict):
  - Long-run elasticity computed directly: (0.017 - 0.042)/(1 - 0.735) = –0.102 (an elasticity of 10.2 percent).
- Scenario responses (illustrated in Figures 5 and 6):
  - Temporary doubling of conflict deaths in first period (then return): overall loss cumulates to around 10 percent of one month’s revenues.
  - Permanent doubling of violence: long-run impact is a revenue loss of over 10 percent per month.
  - Initial short-term positive response in revenues observed (may reflect local demand or pre-collection of debts) — effect only just significant at the 10 percent level in specification (7).
- Aggregated national impact:
  - National revenue loss computed by holding provincial deaths at their 2005 levels as counterfactual and aggregating provincial effects.
  - Estimated monthly revenue loss series plotted 2005–2016 shows increasing loss as conflict grows.
  - 2016 estimated revenue loss:
    - Around 50 percent of revenues.
    - About Af 70 billion.
    - About 5 percent of GDP.
    - Roughly $1 billion.
- Distributional insights:
  - Cross-provincial patterns account for about half the estimated lost revenue.
  - If all provinces had experienced the same relative increase in conflict, the national revenue loss would be about half the size.
  - Provinces that generate more revenue experienced larger increases in violence, producing outsized impacts on national totals.
- Additional illustrative calculation:
  - Conflict deaths increased tenfold; model predicts a long-run impact of around log(10) * –0.102 = –0.24, i.e., a 24 percent loss.

### Key Findings and Implications
- Quantitative findings:
  - UCDP: over 100,000 conflict deaths during 2005–16 across ~20,000 incidents.
  - 2016 revenue loss due to conflict: around 50 percent of revenues ≈ Af 70 billion ≈ $1.0 billion ≈ 5 percent of GDP.
  - Long-run elasticity of revenue to conflict (preferred estimate): –0.102.
  - Temporary doubling of violence → cumulated loss ~10 percent of one month’s revenues; permanent doubling → >10 percent monthly revenue loss.
- Policy-relevant implications:
  - Lost domestic revenue is a major component of the humanitarian cost of conflict because it reduces government capacity to provide health, education, security, and infrastructure.
  - A sizeable potential fiscal dividend exists should peace or a significant reduction in violence materialize, primarily via revenue gains rather than reduced security spending (as security spending is largely grant-funded).
  - Provincial targeting matters: provinces with higher revenue bases experiencing larger increases in violence drive national revenue losses; policy to protect revenue-generating provinces may yield outsized national benefits.

### Conclusion
- The fiscal revenue cost of conflict in Afghanistan is substantial and robust in order of magnitude, even accounting for uncertainty in estimates; the study’s central estimate is around $1 billion of annual revenue loss in 2016.
- These results imply a sizeable potential fiscal dividend if significant reductions in violence occur, improving the government’s ability to finance basic services.

*Prepared by Philip Barrett (FAD) with Mariusz Sumlinski (MCD) and research support from Tetyana Sydorenko.*

### References

### References

### Academic studies on conflict, violence, and economic costs
- Abadie, Alberto, and Javier Gardeazabal, 2003, "The Economic Costs of Conflict: A case Study of the Basque Country," The American Economic Review, Vol. 93 No. 1, pp. 113–32.
- Besley, Timothy, Thiemo Fetzer, and Hannes Mueller, 2015, "The Welfare cost of Lawlessness: Evidence from Somali Piracy, "Journal of the European Economic Association,” Vol. 13 No. 2, pp. 203–39.
- Besley, Timothy, and Hannes Mueller, 2012, "Estimating the Peace Dividend: The impact of Violence on House Prices in Northern Ireland," The American Economic Review, Vol. 102, No. 2, pp. 810–33.
- Collier, Paul, 1999, "On the Economic Consequences of Civil War," Oxford Economic Papers, Vol. 51 No. 1, pp. 168–83.
- Dube, Oeindrila, and Juan F. Vargas, 2013, "Commodity Price Shocks and Civil Conflict: Evidence from Colombia," The Review of Economic Studies, Vol. 80 No. 4, pp. 1384–1421.
- Gupta, Sanjeev, Benedict Clements, Rina Bhattacharya, and Shamit Chakravarti, 2004, "Fiscal Consequences of Armed Conflict and Terrorism in Low-and Middle-Income Countries,” European Journal of Political Economy, Vol. 20 No. 2, pp. 403–21.
- Knight, Malcolm, Norman Loayza, and Delano Villanueva, 1996, "The Peace Dividend: Military Spending Cuts and Economic Growth," IMF Staff Papers,” Vol. 43 No. 1, pp. 1-37 (Washington: International Monetary Fund).
- Mueller, Hannes, 2016, "Growth and Violence: Argument for a Per Capita Measure of Civil War," Economica, Vol. 83 No. 331, pp. 473–97.
- Mueller, H. and Tobias, J., (2016), “The Cost of Violence: Estimating the Economic Impact of Conflict,” IGC Growth Brief Series 007, (London: International Growth Centre).
- Besley, Timothy, and Hannes Mueller (listed above) — studies emphasize measurable welfare and price impacts of violence.

### IMF and multilateral macroeconomic and trade analyses
- International Monetary Fund (IMF), 2017, “Gaining Momentum?” World Economic Outlook, Chapter 1, Box 1.1, (Washington: International Monetary Fund).
- International Monetary Fund, Middle East and Central Asia Department, Regional Economic Outlook, October 2016 (Washington: International Monetary Fund).
- World Bank publications cited in the report (see separate reference list entries below).

### World Bank and development organization studies on landlocked countries, logistics, and corridors
- Arvis, Jean-Franҫois, Gaël Raballand, and Jean-Franҫois Marteau, 2010, “The Cost of Being Landlocked; Logistics Costs and Supply Chain Reliability” (Washington: The World Bank).
- Arvis, Jean-Franҫois, Robin Carruthers, Graham Smith, and Christopher Willoughby, 2011, “Connecting Landlocked Developing Countries to Market—Trade Corridors in the 21st Century” (Washington: The World Bank).
- Cali, Massimiliano, 2015, “Trading Away from Conflict: Using Trade to Increase Resilience in Fragile States” (Washington: The World Bank).
- Rastogi, Cordula, and Jean-Franҫois Arvis, 2014, “The Eurasian Connection—Supply Chain Efficiency along the Modern Silk Route through Central Asia” (Washington: The World Bank).
- Rocha, Nadia, 2017, “Trade as a Vehicle for Growth in Afghanistan: Challenges and Opportunities” (Washington: The World Bank).
- World Bank, 2012, “Afghanistan Diagnostics Trade Integration Study (DTIS)” (Washington: The World Bank).
- World Bank, 2017, “Afghanistan Development Update”, May 2017 (Washington: The World Bank).

### Country and regional publications related to Afghanistan connectivity and trade
- Islamic Republic of Afghanistan, Ministry of Foreign Affairs, Regional Economic Cooperation Conference on Afghanistan (RECCA), 2015, “Towards Regional Economic Growth and Stability: The Silk Road Through Afghanistan” (Kabul).
- Islamic Republic of Afghanistan, 2016, Annual Review, “From Negotiation to Investment, Construction, and Trade: A New Decade of Progress” (Kabul).
- United Nations Conference on Trade and Development, 2013, “The Way to The Ocean—Transit Corridors Servicing the Trade of Landlocked Developing Countries” (New York and Geneva: United Nations).

*Italic: References as listed in the content unit cr17378 - References*

### Box 1. Key Facts About Afghanistan’s Financial

### Box 1. Key Facts About Afghanistan’s Financial Sector

### Financial sector composition (selected counts by year)
- Columns: Year; # of commercial banks; # of specialized microfinance banks; # of Islamic banking windows; # of Insurance companies; # of Microfinance institutions; # of Electronic money institutions; # of Money service providers; # of MSPs' representative offices; # of Foreign exchange dealers
- 2014: 16 16 35 34 62 30 91 1,772
- 2015: 15 17 35 35 83 33 41 1,415
- 2016: 15 17 45 36 69 34 82 2,192
- 2017: 15 17 44 36 06 29 11 1,705

### Key financial indicators (December, 2016)
- Credit to GDP: 4 percent
- Deposit dollarization: 66.2 percent
- Loan dollarization: 61.4 percent
- Currency to deposit ratio: 93.4 percent
- Loans to deposit ratio: 19.9 percent

### Institutional and sectoral features
- Afghanistan is a cash economy with high level of dollarization.
- Afghanistan's financial sector is dominated by commercial banks.
- DAB regulates and supervises all financial institutions except insurance companies (regulated by the MoF).
- Non-depository microfinance institutions are not supervised by any institution.
- The other financial service providers mostly handle the current transactions of their clients and do not take deposits.

### Microfinance sector (Q1 2017 and banking sector comparison)
- Microfinance loan portfolio: Af 10.5 billion (0.8 percent of GDP)
- Microfinance savings deposits: Af 2.5 billion (0.2 percent of GDP)
- Banking sector loan portfolio: Af 47.4 billion (3.6 percent of GDP)
- Banking sector total deposits: Af 234 billion (17.7 percent of GDP)
- Microfinance is the second biggest component of Afghanistan's financial sector.

### Financial depth and inclusion (summary findings)
- Domestic private credit to GDP ratio was only 4 percent in 2014 and likely around this level since overall domestic credit to GDP remains below 4 percent in 2017.
- Afghanistan's financial sector is shallow even relative to other low income and fragile countries.
- World Bank: 80 percent of adults in FCA countries remain outside of the formal financial system and only 15 percent of adults have an account at a formal financial institution compared to more than 40 percent in the rest of the developing world and 62 worldwide.
- In FCA countries there is high demand but low supply of financial services.

### Barriers to use of formal accounts (World Bank Global Findex survey observations)
- Most common barrier: Poverty (not having enough money).
- Other barriers: religious reasons; lack of trust; documentation requirements; distance to the nearest financial institution; costs of financial account ownership.
- In FCA countries: 48 percent of adults borrowed money but only six percent used a formal financial institution for borrowing.

### Financial access indicators in Afghanistan
- Number of ATMs per 100,000 adults lags other LICs, FCAs, and neighboring countries.
- Number of commercial bank branches per 100,000 adults is small.
- Some regions have no access to basic financial services due to remoteness, security challenges, or lack of government control.
- Accounts in financial institutions are less than average in other LICs even when controlled for income level.
- Household bank accounts mostly used as repositories for remittances or wages, not for transactions.
- Afghanistan lags the LICs average in number of accounts used for business purposes (reluctance possibly motivated by tax evasion).

### Unique constraints in Afghanistan
- High illiteracy rate: 60 percent.
- Low participation of women in the economy.
- Higher cost of using financial services possibly due to compliance with anti-corruption rules, "Know Your Customer" procedures, and enhanced diligence.
- Until conditions normalize, compliance costs should be borne as benefits likely outweigh costs.

### Borrowing and firm finance behavior
- Afghans prefer borrowing from private lenders, relatives, and friends.
- Global Findex: of the 41 percent of adults that borrowed money only 11 percent borrowed from a formal financial institution.
- Firms relying on bank loans to finance investments: around 3 percent in Afghanistan versus 18 percent in other fragile countries.

### Table 1. Financial Access for Firms (selected indicators)
- Percent of firms with a checking or savings account: Afghanistan 43.7; South Asia 77.6; All countries 86.8
- Percent of firms with a bank loan/line of credit: Afghanistan 5.1; South Asia 27.0; All countries 33.8
- Proportion of loans requiring collateral (%): Afghanistan 70.7; South Asia 81.1; All countries 79.4
- Percent of firms whose recent loan application was rejected: Afghanistan 35.3; South Asia 14.4; All countries 12.2
- Percent of firms using banks to finance investments: Afghanistan 2.0; South Asia 21.8; All countries 25.3
- Proportion of investments financed internally (%): Afghanistan 84.5; South Asia 73.9; All countries 71.2
- Proportion of investments financed by banks (%): Afghanistan 1.5; South Asia 14.4; All countries 14.3
- Percent of firms using banks to finance working capital: Afghanistan 3.9; South Asia 25.0; All countries 30.4
- Percent of firms identifying access to finance as a major constraint: Afghanistan 47.6; South Asia 26.5; All countries 26.5

### Policy measures and recommendations to boost financial inclusion (from Conclusions)
- Strengthen the legal framework and remove barriers to financial development while preserving financial stability.
- Strengthen and extend the perimeter of financial supervision by including microfinance institutions.
- Developing mobile banking to improve access in remote areas; Afghanistan could benefit from experience of other countries when developing relevant regulations and policies.
- Developing Sharia compliant instruments to attract customers who do not use existing financial institutions for religious reasons.
- Reducing costs of operation of financial services providers; introduce unique biometric identification for all Afghanistan citizens to resolve documentation problems.
- Developing the insurance sector to reduce risks for banks and likely reduce costs associated with lending.
- Promoting financial literacy (international experience summarized in Box 2).

*Source: Box 1. Key Facts About Afghanistan’s Financial Sector, IMF country report content provided.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2017/cr17378.pdf_