## 2. External Financing to Mali

## Source details

**Canonical URL:** [2. External Financing to Mali](https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025130.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025130.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025130.pdf.json)

---

### A. Evidence of Heightened Uncertainty in Mali
- Empirical indicator based on references to “uncertainty” in economic reports was "over 50 percent higher on average" between 2012 and 2022 than the previous decade.
- Key contextual facts:
  - Agriculture is the main source of employment and income for "60 percent" of the active population.
  - Mali ranks fourth globally in a list of countries most affected by terrorism and has the lowest score of all countries for perceptions of safety (Institute for Economics and Peace, 2023 and 2024).
- Business survey evidence (Conseil National du Patronat du Mali, CNPM):
  - The share of businesses reporting a negative effect from "the lack of security, the general economic climate and the social and political situation" was "over 70 percent" in the latest survey.
- External financing and relations with partners:
  - "Budgetary loans to Mali have fallen to zero since 2022."
  - Project loans are "much lower than in the past," while amortization costs continue, increasing fiscal pressure and perceived risks.
- Recent policy announcements that increased uncertainty:
  - Withdrawal request of UN peacekeeping forces (MINUSMA) in mid-2023; exit from G5 Sahel Alliance and Algiers Peace Accord; January 2024 announcement that Mali, Burkina Faso and Niger would leave ECOWAS; cancellation of presidential elections originally scheduled for February 2024.
  - Mentions of uncertainty in published economic reports "more than doubled in the second half of 2023" and rose further at the start of 2024; series remained elevated through 2024 and at the start of 2025 was comparable to 2020 levels.
- Mali-specific increase in uncertainty:
  - Recent increases in measured uncertainty were not mirrored to the same extent in Burkina Faso and Niger during 2024.
  - Ongoing IMF-supported programs in Burkina Faso and Niger may have reduced uncertainty in those countries.

### B. Financial Market Indicators and Interest Premia
- Regional interest premia observations:
  - Mali, Burkina Faso and Niger have paid higher interest rates on regional debt issuances than other WAEMU countries in recent years.
  - During the second half of 2023, Mali paid the highest interest premia among these countries; three-year spreads in Mali rose by more than in Burkina Faso in that period.
- By May 2025:
  - The interest yield on one-year T-bills in Mali was "2.2 percentage points higher than in the rest of the WAEMU."
  - The interest rate premium on three-year bonds was "1.2 percentage points higher."

### C. Impact of Heightened Uncertainty on Economic Activity
- Transmission channels:
  - Business investment:
    - Heightened uncertainty increases the “option value of waiting,” reducing capital spending and depressing the capital stock over time.
    - Public and private investment in Mali have been "extremely low for a number of years"; further reductions would weigh on future productivity and growth.
  - Trade:
    - Reduced investment can lower cross-border trade and associated productivity gains.
    - Uncertainty around tariffs and regulatory alignment (notably the ECOWAS exit) would affect trade more than general economic uncertainty; the share of Mali’s trade affected by the ECOWAS exit is "estimated to be relatively small," but lack of clarity on tariffs/regulations could deter future partners.
    - Uncertainty around the implementation of the revised mining code could discourage investment in a sector that makes up "a large share of Mali’s GDP."
  - Households and consumption:
    - Households undertake precautionary saving under uncertainty (e.g., Deaton, 1989), reducing consumption—especially of durables.
    - Many households are near subsistence levels, limiting the scope for further consumption cuts.
  - Credit and fiscal financing:
    - If uncertainty increases interest premia or lowers credit availability, businesses, households and the government face tighter financing conditions; underdeveloped financial markets can amplify the impact.
- Quantified impacts on growth and risks:
  - A two-standard deviation increase in uncertainty is associated with a "0.9 percentage point" fall in GDP growth in the following quarter (quarterly result).
  - Growth-at-risk analysis (annual model since 1968, quantile regression with controls including gold and cotton prices, money supply, inflation, and coups d’état) implies that a two-standard deviation increase in measured uncertainty:
    - Widens the distribution of predicted annual growth outcomes.
    - Increases the probability of a recession (defined as negative annual growth) by "just under 20 percentage points."

### D. Policy Implications and Recommendations
- Overarching goals:
  - Reduce uncertainty and its economic costs by addressing sources of fragility and constraints to effective policy-making.
- Specific recommendations:
  - Increase transparency and ensure accurate statistics are available.
  - Remove constraints to growth, including raising education levels and improving governance.
  - Mitigate stressors such as climate risks and protect the most vulnerable to help Mali exit the fragility trap.
- Structural and resilience-building measures:
  - Address both structural fragility (inability to capitalize on positive developments) and fragility to stress (reduced resilience to negative shocks).
  - Many recommendations align with the Country Engagement Strategy in the 2023 Staff Report for Mali.

### Reducing policy uncertainty: recommendations and rationale (from section 19)
- Set out a transparent and detailed medium-term policy framework to provide the basis for strong and sustainable economic growth in Mali and to help households and businesses plan ahead.
- Communicate expected future policy parameters (example given: expected future corporate tax rates) so businesses can prepare budgets.
- Ensure effective and timely communication of policy changes to avoid unwanted responses and speculation around other possible announcements.
- If the ECOWAS exit leads to changes in tariff rates or product regulations, communicate these changes well in advance to avoid costly disruption.

### Sustainable fiscal policy and buffers (from section 19)
- Build fiscal buffers to ensure sufficient resources to respond to negative shocks.
- Achieve fiscal consolidation in a sustainable way with emphasis on revenue mobilization rather than cutting growth-enhancing capital spending.
- Rebuild relationships with international development partners, particularly to facilitate resumption of external budget support.

### Measuring uncertainty: approaches, findings, and caveats
- Economic reports indicator (Annex I):
  - Prominent indicator: number of references to ‘uncertainty’ in economic reports (Ahir and others, 2018).
  - Series available on a quarterly basis; producers recommend using the three-quarter rolling average due to volatility of the underlying data.
  - Finding: uncertainty has risen by much more in Mali than in Burkina Faso or Niger over recent quarters; spikes generally coincide with most high-profile periods of instability.
- Global news coverage indicator (Annex I):
  - Country-specific news-based indicator counts times ‘uncertainty’ is mentioned within ten words of the country name; divided by total number of news articles across all countries to control for increasing article counts.
  - Searches adapted to include related terms in English and French; for Niger, references related to ‘Nigeria’ are removed; exclude news articles with fewer than ninety-nine words.
  - Formula: U_{t,i} = R_{t,i} / A_{t}, results standardized to give a z-score Z_{t,i}.
  - Advantages: wider source coverage than economic reports alone; monthly frequency can attribute rises to individual events; large spikes correspond reasonably well with coups d’état and episodes of wider political instability.
  - Limitations: high volatility; many observations of zero for Burkina Faso and Niger; potential inconsistency over time because similar events may elicit different coverage depending on prior spotlight on a country; number of references influenced by other prominent global events.

### Growth-at-Risk model for Mali: methodology (Annex II)
- Modeling approach:
  - Single-equation quantile regression methodology based on Koenker and Machado, 1999 and Koenker, 2005.
  - Let Δ_h Y_{t+h} denote the log change in real GDP h periods ahead; let X_t denote a vector of key determinants (including past log changes in real GDP). The quantile regression specification:
    - ∆_h ∆Y_{t+h,q} = α_{h,q} + b_{x,h,q} X_t + e_{t,h,q}
- Estimation details:
  - Linear programming algorithm minimizes deviations for the q-th conditional quantile; estimator β̂(q) solves the minimization problem using the check function ρ_q(u) which weighs positive and negative values asymmetrically.
  - Under mild regularity conditions, quantile regression coefficients are asymptotically normally distributed; a bootstrap resampling technique is used to generate a distribution of sample statistics.
- Tail-fitting and distributional choices:
  - Estimation of extreme quantiles (5 percent and 95 percent) can be inaccurate with large confidence error bands; following Adrian and others, 2018, a parametric t-skew distribution is used to fit conditional quantiles.
  - The skewed Student t-distribution (Azzalini, 2003) is used to model tail events more accurately.
- Interpretation:
  - For a given determinant x and quantile q, the sequence of b_{x,h,q} coefficients across horizons h shows how an increase in x changes the q_{t+h} quantile of future GDP growth, providing a “term structure” of growth-at-risk.

*Source: IMF staff, The Economic Impact of Uncertainty in Mali, July 16, 2025.*

### 2. External Financing to Mali ____________________________________________________________ 3

### 2. External Financing to Mali

### A. Evidence of Heightened Uncertainty in Mali
- Multiple challenges since 2012 have increased economic uncertainty; an empirical indicator based on references to “uncertainty” in economic reports was "over 50 percent higher on average" between 2012 and 2022 than the previous decade.
- Key contextual facts:
  - Agriculture is the main source of employment and income for "60 percent" of the active population.
  - Mali ranks fourth globally in a list of countries most affected by terrorism and has the lowest score of all countries for perceptions of safety (Institute for Economics and Peace, 2023 and 2024).
- Business survey evidence (Conseil National du Patronat du Mali, CNPM):
  - The share of businesses reporting a negative effect from "the lack of security, the general economic climate and the social and political situation" was "over 70 percent" in the latest survey.
- External financing and relations with partners:
  - "Budgetary loans to Mali have fallen to zero since 2022."
  - Project loans are "much lower than in the past," while amortization costs continue, increasing fiscal pressure and perceived risks.
- Recent policy announcements that increased uncertainty:
  - Withdrawal request of UN peacekeeping forces (MINUSMA) in mid-2023; exit from G5 Sahel Alliance and Algiers Peace Accord; January 2024 announcement that Mali, Burkina Faso and Niger would leave ECOWAS; cancellation of presidential elections originally scheduled for February 2024.
  - Mentions of uncertainty in published economic reports "more than doubled in the second half of 2023" and rose further at the start of 2024; series remained elevated through 2024 and at the start of 2025 was comparable to 2020 levels.
- Mali-specific increase in uncertainty:
  - Recent increases in measured uncertainty were not mirrored to the same extent in Burkina Faso and Niger during 2024.
  - Ongoing IMF-supported programs in Burkina Faso and Niger may have reduced uncertainty in those countries.

### B. Financial Market Indicators and Interest Premia
- Interest premia observations:
  - Mali, Burkina Faso and Niger have paid higher interest rates on regional debt issuances than other WAEMU countries in recent years.
  - During the second half of 2023, Mali paid the highest interest premia among these countries; three-year spreads in Mali rose by more than in Burkina Faso in that period.
  - By May 2025:
    - The interest yield on one-year T-bills in Mali was "2.2 percentage points higher than in the rest of the WAEMU."
    - The interest rate premium on three-year bonds was "1.2 percentage points higher."

### C. Impact of Heightened Uncertainty on Economic Activity
- Channels through which uncertainty affects the economy:
  - Business investment:
    - Heightened uncertainty increases the “option value of waiting,” reducing capital spending and depressing the capital stock over time.
    - Public and private investment in Mali have been "extremely low for a number of years"; further reductions would weigh on future productivity and growth.
  - Trade:
    - Reduced investment can lower cross-border trade and associated productivity gains.
    - Uncertainty around tariffs and regulatory alignment (notably the ECOWAS exit) would affect trade more than general economic uncertainty; the share of Mali’s trade affected by the ECOWAS exit is "estimated to be relatively small," but lack of clarity on tariffs/regulations could deter future partners.
    - Uncertainty around the implementation of the revised mining code could discourage investment in a sector that makes up "a large share of Mali’s GDP."
  - Households and consumption:
    - Households undertake precautionary saving under uncertainty (e.g., Deaton, 1989), reducing consumption—especially of durables.
    - Many households are near subsistence levels, limiting the scope for further consumption cuts.
  - Credit and fiscal financing:
    - If uncertainty increases interest premia or lowers credit availability, businesses, households and the government face tighter financing conditions; underdeveloped financial markets can amplify the impact.

- Quantified impacts on growth and risks:
  - A two-standard deviation increase in uncertainty is associated with a "0.9 percentage point" fall in GDP growth in the following quarter (quarterly result).
  - Growth-at-risk analysis (annual model since 1968, quantile regression with controls including gold and cotton prices, money supply, inflation, and coups d’état) implies that a two-standard deviation increase in measured uncertainty:
    - Widens the distribution of predicted annual growth outcomes.
    - Increases the probability of a recession (defined as negative annual growth) by "just under 20 percentage points."

### D. Policy Implications and Recommendations
- To reduce uncertainty and its economic costs, the authorities should address sources of fragility and constraints to effective policy-making:
  - Increase transparency and ensure accurate statistics are available.
  - Remove constraints to growth, including raising education levels and improving governance.
  - Mitigate stressors such as climate risks and protect the most vulnerable to help Mali exit the fragility trap.
- Structural and resilience-building measures are emphasized:
  - Address both structural fragility (inability to capitalize on positive developments) and fragility to stress (reduced resilience to negative shocks).
  - Many recommendations align with the Country Engagement Strategy in the 2023 Staff Report for Mali.

*Source: IMF staff, The Economic Impact of Uncertainty in Mali, July 16, 2025.*

### 19.      The authorities could reduce policy uncertainty by setting out medium-term economic

### 19.      The authorities could reduce policy uncertainty by setting out medium-term economic

### Reducing policy uncertainty: recommendations and rationale
- Set out a transparent and detailed medium-term policy framework to provide the basis for strong and sustainable economic growth in Mali and to help households and businesses plan ahead.
- Communicate expected future policy parameters (example given: expected future corporate tax rates) so businesses can prepare budgets.
- Ensure effective and timely communication of policy changes to avoid unwanted responses and speculation around other possible announcements.
- If the ECOWAS exit leads to changes in tariff rates or product regulations, communicate these changes well in advance to avoid costly disruption.

### Sustainable fiscal policy and buffers
- Build fiscal buffers to ensure sufficient resources to respond to negative shocks.
- Achieve fiscal consolidation in a sustainable way with emphasis on revenue mobilization rather than cutting growth-enhancing capital spending.
- Rebuild relationships with international development partners, particularly to facilitate resumption of external budget support.

### Measuring uncertainty using economic reports (Annex I): approach and findings
- Prominent indicator: number of references to ‘uncertainty’ in economic reports (Ahir and others, 2018).
- Series are available on a quarterly basis; producers recommend using the three-quarter rolling average due to volatility of the underlying data.
- Finding: uncertainty has risen by much more in Mali than in Burkina Faso or Niger over recent quarters; spikes generally coincide with most high-profile periods of instability.

### Measuring uncertainty using global news coverage (Annex I): methodology and caveats
- Country-specific news-based indicator: record number of times ‘uncertainty’ is mentioned within ten words of the country name; divide by total number of news articles across all countries to control for increasing article counts.
- Searches adapted to include related terms in English and French; for Niger, references related to ‘Nigeria’ are removed; exclude news articles with fewer than ninety-nine words.
- Formula presented: U_{t,i} = R_{t,i} / A_{t}, where R_{t,i} is number of references to uncertainty in country i during period t and A_{t} is number of articles produced in period t; results standardized to give a z-score Z_{t,i}.
- Advantages: wider source coverage than economic reports alone; can be produced at monthly frequency to attribute rises in uncertainty to individual events; large spikes correspond reasonably well with coups d’état and episodes of wider political instability.
- Limitations: high volatility; many observations of zero for Burkina Faso and Niger; potential inconsistency over time because similar events may elicit different coverage depending on prior spotlight on a country; number of references influenced by other prominent global events.

### Growth-at-Risk model for Mali (Annex II): methodology
- Model uses a single-equation quantile regression methodology based on Koenker and Machado, 1999 and Koenker, 2005.
- Let Δ_h Y_{t+h} denote the log change in real GDP h periods ahead; let X_t denote a vector of key determinants (including past log changes in real GDP). The quantile regression specification:
  - ∆_h ∆Y_{t+h,q} = α_{h,q} + b_{x,h,q} X_t + e_{t,h,q}
- Estimation:
  - Linear programming algorithm minimizes deviations for the q-th conditional quantile; estimator β̂(q) solves the minimization problem using the check function ρ_q(u) which weighs positive and negative values asymmetrically.
  - Under mild regularity conditions, quantile regression coefficients are asymptotically normally distributed; a bootstrap resampling technique is used to generate a distribution of sample statistics.
- Tail-fitting and distributional choices:
  - Estimation of extreme quantiles (5 percent and 95 percent) can be inaccurate with large confidence error bands; following Adrian and others, 2018, a parametric t-skew distribution is used to fit conditional quantiles.
  - The skewed Student t-distribution (Azzalini, 2003) is used to model tail events more accurately.
- Interpretation:
  - For a given determinant x and quantile q, the sequence of b_{x,h,q} coefficients across horizons h shows how an increase in x changes the q_t+h quantile of future GDP growth, providing a “term structure” of growth-at-risk.

### Key methodological references and notes
- Quantile regression foundations: Koenker, 2005; Koenker and Machado, 1999.
- Tail modeling: Adrian and others, 2018; Azzalini, 2003.
- Uncertainty indicator source: World Uncertainty Index; methodology adaptations use Economist Intelligence Unit Reports and Factiva for news articles; three-quarter rolling averages excluding observations of zero noted in figure notes.

*Source: Extracted from sipea2025130 - 19.      The authorities could reduce policy uncertainty by setting out medium-term economic (PDF).*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025130.pdf_