## CHAPTER 2 MACROFINANCIAL STABILITY AMId hIGh GLOBAL ECONOMIC uNCERTAINTY

## Source details

**Canonical URL:** [CHAPTER 2 MACROFINANCIAL STABILITY AMId hIGh GLOBAL ECONOMIC uNCERTAINTY](https://www.imf.org/-/media/files/publications/gfsr/2024/october/english/ch2.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/gfsr/2024/october/english/ch2.pdf.md)
- [Structured JSON version](/-/media/files/publications/gfsr/2024/october/english/ch2.pdf.json)

---

### High-level summary
- Uncertainty regarding global economic outcomes and policies has been higher since the COVID-19 pandemic amid inflation shocks and rising geopolitical tensions.
- Macroeconomic uncertainty has remained elevated since the pandemic, with some measures (for example, global economy policy uncertainty of Baker, Bloom, and Davis (2016)) declining in the first quarter of 2024 but rising again in the second quarter amid electoral uncertainty in some major economies.
- Different measures of macroeconomic uncertainty show volatility but on average elevated levels since the pandemic.

### Channels and mechanisms linking uncertainty to macrofinancial stability
- Market channel:
  - Amplifies downside market tail risks in the event of an adverse shock, raising the risk of large negative realized future asset returns and transmitting to the broader economy via balance sheet and financial acceleration effects.
- Real channel:
  - Delays private sector consumption and investment decisions (wait-and-see behavior), slowing economic activity and increasing credit risks for financial institutions, potentially triggering adverse macrofinancial feedback loops.
- Credit channel:
  - Reduces the supply of domestic credit as financial institutions face greater challenges in determining the creditworthiness of new borrowers when the macroeconomic outlook is more uncertain.
- Interaction:
  - These three channels can interact and mutually reinforce each other, magnifying effects on macrofinancial stability.

### Data, empirical scope, and modeling approach
- Sample and outcomes:
  - Panel data from a sample of 43 advanced and emerging market economies since 1990 (or the earliest year for which data are available).
  - Outcomes evaluated: downside risks to future real GDP growth (downside tail risks typically captured by the 5th or 10th percentile of the distribution).
- Modeling approach:
  - Growth-at-Risk (GaR) framework extended to include measures of macroeconomic uncertainty and machine learning tools alongside standard panel quantile regressions.
  - Machine learning GaR models used: panel quantile random forest and panel quantile neural network.

### Key empirical findings and quantitative results
- Effects on GDP downside risk:
  - An increase in macroeconomic uncertainty equivalent to its rise during the global financial crisis reduces the downside outcome (the 10th percentile) of one-year-ahead real GDP growth by, on average, 1.2 percentage points in advanced and emerging market economies.
  - A one-standard-deviation increase in measures of macroeconomic uncertainty reduces one-quarter-ahead real GDP growth (annualized) by 0.5 to 2.0 percentage points.
  - The impact of macroeconomic uncertainty persists up to about seven quarters after the shock.
  - The annual output decline in the bottom 10th percentile of the historical GDP growth distribution for the full sample (advanced and emerging market economies) is 1.2 percent.
  - Macroeconomic uncertainty has a negligible effect on the median of future real GDP growth but large and statistically significant effects on lower and upper quantiles; effects are stronger on downside tail risks (5th or 10th percentiles) than on upside tail risks (90th or 95th percentiles).
- Financial variables and uncertainty:
  - Financial variables explain about 80 percent of the variation in commonly used measures of macroeconomic uncertainty for advanced economies like the United States, and 40 to 50 percent of the variation in those for major emerging markets such as Brazil.
  - Financial market volatility and macroeconomic uncertainty can be disconnected during “macro-market disconnects” (periods of high macroeconomic uncertainty and low financial market volatility), leaving financial indicators insufficient to capture macroeconomic uncertainty—particularly in countries with less-developed financial markets.
- Market, stock, and credit channel quantitative magnitudes (Box 2.1):
  - Sovereign bond spreads:
    - A one-standard-deviation increase in the REU is, on average, associated with an increase of 150 basis points in upside tail risks to sovereign bond spreads (90th percentile) in emerging market economies at a six-month horizon.
    - For the average advanced economy, a one-standard-deviation shock to the REU increases upside tail risks by about 25 basis points.
  - Stock market returns:
    - A one-standard-deviation increase in the REU can raise downside tail risks to stock market returns (10th percentile) by about 30 percentage points, one year after the shock in advanced and emerging market economies.
  - Bank lending:
    - A one-standard-deviation increase in the REU is associated with a decline of about 1 percentage point (annualized) in the 10th percentile of the distribution of one-quarter-ahead real credit growth; this effect persists through about seven quarters.
    - When domestic banks’ exposure to sovereign risk is high (one standard deviation above the mean) the one-year-ahead downside risk to lending (10th percentile) increases by about 1 percentage point compared to the mean level.
- Foreign uncertainty and spillovers:
  - A one-standard-deviation increase in trade-weighted foreign macroeconomic uncertainty (based on the REU) reduces the 10th percentile of the one-quarter-ahead distribution of domestic GDP growth by 1.7 percentage points; this effect peters out in about three quarters.
  - Foreign shocks transmitted via portfolio exposures are notably more persistent; foreign economic policy uncertainty effects can last up to six quarters.

### Machine learning enhancements and forecast performance
- Predictive gains:
  - ML-GaR models improve out-of-sample prediction accuracy relative to the standard linear quantile GaR benchmark by up to 7 percent at different horizons.
  - Adding measures of macroeconomic uncertainty (for example, the REU) as predictors further improves ML-GaR out-of-sample forecast performance by 5 to 13 percent relative to standard GaR models that exclude uncertainty.
  - ML-GaR models show that macroeconomic uncertainty contributes at least as much as the financial conditions index to predicting downside risk to real GDP growth; on average, the REU contributes more to predictions than the financial conditions index.
- Practical challenges and caveats:
  - Significant data requirements and technological know-how, weak signal-to-noise ratios of financial variables, and poor transparency/interpretability of ML methods.
  - Addressed in the chapter via cross-validation, overfit mitigation, numerical simulations, and analysis of variable importance.

### Interaction with vulnerabilities, macro-market disconnects, and policy trade-offs
- Amplification by vulnerabilities:
  - High real economic uncertainty combined with excessive domestic credit reduces one-quarter-ahead downside tail risk to GDP growth by 0.6 percentage points (interaction effect reported).
  - High public debt levels significantly increase downside risks to GDP growth, particularly when real economic uncertainty is high.
- Macro-market disconnect and intertemporal trade-off:
  - In the short term, easing financial conditions reduces downside tail risks to GDP growth.
  - In the medium term, easy financial conditions encourage a buildup of debt vulnerabilities that exacerbate downside tail risks.
  - Under high macroeconomic uncertainty, looser financial conditions exacerbate medium-term downside tail risks, especially when there is a large macro-market disconnect (defined as the ratio of real economic uncertainty to realized market volatility that is above its mean).
- Role of macroprudential policy:
  - A net tightening of macroprudential policies can help offset the rise in medium-term downside risks associated with easy financial conditions.
  - When the macro-market disconnect is large, a loosening of financial conditions coupled with a net tightening of macroprudential policies is associated with a reduction in downside risks to GDP growth of 0.3 to 0.6 percentage points in the medium to long terms, compared with a scenario with no macroprudential measures.

### Monetary and fiscal framework interactions
- Monetary policy credibility:
  - Countries where inflation expectations deviate more from policy (inflation) targets experience higher levels of economic policy uncertainty.
  - In GaR analysis, the effect of increased macroeconomic uncertainty on downside risk to one-quarter-ahead GDP growth is larger when policy targets were missed by wider margins over the preceding three years (proxy for weaker monetary policy frameworks).
- Fiscal frameworks:
  - Fiscal rules and more stringent fiscal frameworks are associated in the literature with reductions in fiscal policy uncertainty, fiscal procyclicality, market volatility, and overall macroeconomic volatility.
  - Fiscal policies should prioritize debt sustainability to contain adverse effects of elevated public debt levels on borrowing costs.

### Role of technology, information, and text-based measures
- Information amplification:
  - Recent technological innovations and social media can aggravate uncertainty and amplify its effect on market tail risks by making investors and depositors more attentive to surprises in data and news.
  - Fintech developments affect transmission of uncertainty to financial institutions and markets.
- Text-based measures:
  - A bank-level text-based uncertainty measure (February 2024 update of the Loughran-McDonald Master Dictionary) exhibits a low degree of correlation with other measures but remains statistically significant when included.
  - Panel analyses using bank-level text measures show larger negative effects on the 10th percentile of future real credit growth at specific horizons.

### Policy recommendations and operational guidance
- Overarching goal:
  - Reduce macroeconomic uncertainty and mitigate its adverse effects by strengthening resilience and containing macrofinancial vulnerabilities.
- Reduce policy uncertainty:
  - Enhance credibility of monetary and fiscal policy frameworks through adoption of fiscal and monetary policy rules supported by strong institutions.
  - Improve transparency and adopt well-designed policy communication frameworks to steer market expectations.
  - Maintain a stable financial regulatory framework; announce and implement reforms with clear communication, robust calibration, and appropriate phase-in periods.
- Contain financial stability risks amid high uncertainty:
  - Deploy adequate macroprudential and fiscal policies to contain financial stability risks arising from elevated macrofinancial vulnerabilities.
  - Proactively deploy macroprudential policies (for example, countercyclical capital buffers) when financial conditions are loose and disconnected from elevated uncertainty.
  - Activate borrower-based measures if lax financial conditions amid high uncertainty encourage excessive borrowing for real estate investment.
  - Consider a tighter monetary policy stance where aligned with the central bank’s goal of maintaining price stability.
  - Use the level of macroeconomic uncertainty and its disconnect from financial market volatility to inform choices among policy instruments and response magnitudes.
  - Prudential regulators should ensure bank and nonbank institutions assess vulnerabilities to cross-border spillovers of spikes in foreign macroeconomic uncertainty.
  - Maintain adequate international reserve buffers and greater exchange rate flexibility to cushion adverse impacts of foreign uncertainty shocks.
- Geopolitical uncertainty:
  - Build adequate safety nets to mitigate macrofinancial stability risks from rising geopolitical uncertainty.
  - Strive to reduce geopolitical tensions through diplomacy and multilateral cooperation; where cooperation is elusive, devote resources to identifying, quantifying, managing, and mitigating associated financial stability risks.
  - Ensure adequate international reserves and capital and liquidity buffers at financial institutions.

*Source: Chapter 2, “Macrofinancial Stability Amid High Global Economic Uncertainty,” Global Financial Stability Report, October 2024*

### Chapter 2 at a Glance

### Chapter 2 at a Glance

### High-level summary
- Uncertainty regarding global economic outcomes and policies has been higher since the COVID-19 pandemic amid inflation shocks and rising geopolitical tensions.
- Macroeconomic uncertainty has remained elevated since the pandemic, with some measures (for example, global economy policy uncertainty of Baker, Bloom, and Davis (2016)) declining in the first quarter of 2024 but rising again in the second quarter amid electoral uncertainty in some major economies.
- Different measures of macroeconomic uncertainty show volatility but on average elevated levels since the pandemic.

### Channels through which macroeconomic uncertainty affects macrofinancial stability
- Market channel: amplifies downside market tail risks in the event of an adverse shock, raising the risk of large negative realized future asset returns and transmitting to the broader economy via balance sheet and financial acceleration effects.
- Real channel: delays private sector consumption and investment decisions (wait-and-see behavior), slowing economic activity and increasing credit risks for financial institutions, potentially triggering adverse macrofinancial feedback loops.
- Credit channel: reduces the supply of domestic credit as financial institutions face greater challenges in determining the creditworthiness of new borrowers when the macroeconomic outlook is more uncertain.
- These three channels can interact and mutually reinforce each other, magnifying effects on macrofinancial stability.

### Empirical scope and methods
- Sample: panel data from a sample of 43 advanced and emerging market economies since 1990 (or the earliest year for which data are available).
- Outcomes evaluated: downside risks to future real GDP growth (downside tail risks typically captured by the 5th or 10th percentile of the distribution).
- Modelling approach: the growth-at-risk (GaR) framework is extended by augmenting it with measures of macroeconomic uncertainty and by implementing machine learning tools in addition to standard panel quantile regressions to improve forecasting of downside tail risks.

### Key empirical findings and statistics
- An increase in macroeconomic uncertainty equivalent to its rise during the global financial crisis reduces the downside outcome (the 10th percentile) of one-year-ahead real GDP growth by, on average, 1.2 percentage points in advanced and emerging market economies.
- Financial variables explain about 80 percent of the variation in commonly used measures of macroeconomic uncertainty for advanced economies like the United States, and 40 to 50 percent of the variation in those for major emerging markets such as Brazil.
- Financial market volatility and macroeconomic uncertainty can be disconnected during “macro-market disconnects” (periods of high macroeconomic uncertainty and low financial market volatility), which can leave financial indicators insufficient to capture macroeconomic uncertainty—particularly in countries with less-developed financial markets.

### Interaction with vulnerabilities and spillovers
- Macroeconomic uncertainty tends to amplify the effect of prevailing macrofinancial vulnerabilities (for example, excessive leverage in the private and public sectors) on downside risks to future output growth.
- A significant easing of financial conditions amid high macroeconomic uncertainty can exacerbate downside risks to future output growth, particularly during periods of low financial market volatility (that is, during “macro-market disconnects”).
- The effects of macroeconomic uncertainty can spill over across borders through trade and financial interlinkages, increasing the risk of contagion: losses in one region can force investors to sell assets in other countries causing large asset price declines and international financial contagion; weaker domestic demand can reduce imports and raise downside risks for trading partners.

### Role of technology and information
- Recent technological innovations and social media can aggravate uncertainty and amplify its effect on market tail risks by making investors and depositors more attentive to surprises in data and news, intensifying stress episodes.
- Fintech developments also affect the transmission of uncertainty to financial institutions and markets.

### Policy recommendations
- Reduce domestic macroeconomic uncertainty by strengthening the credibility and transparency of frameworks for monetary, fiscal, and financial sector policies and through effective communication strategies.
- Implement adequate fiscal and macroprudential policies to contain macrofinancial vulnerabilities and build resilience against adverse shocks, particularly when macroeconomic uncertainty is high.
- Build adequate international reserve buffers and allow exchange rate flexibility to help cushion the adverse spillover effects of an increase in foreign macroeconomic uncertainty.
- Devote resources to quantifying, managing, and mitigating the risks from rising geopolitical uncertainty on macrofinancial stability.

*Source: Chapter 2, “Macrofinancial Stability Amid High Global Economic Uncertainty,” Global Financial Stability Report, October 2024*

### CHAPTER 2 MACROFINANCIAL STABILITY AMId hIGh GLOBAL ECONOMIC uNCERTAINTY

### CHAPTER 2 MACROFINANCIAL STABILITY AMId hIGh GLOBAL ECONOMIC uNCERTAINTY

### Global landscape of macroeconomic uncertainty
- Macroeconomic uncertainty arises from: (1) innovations affecting the real sector (output, product prices, factor costs, firms’ profitability); (2) monetary, fiscal, trade, and regulatory policies; and (3) geopolitical tensions (conflicts or policy-driven barriers on cross-border trade and capital flows).
- Measures used to quantify these sources:
  - Real sector uncertainty: real economic uncertainty index (REU) of Jurado, Ludvigson, and Ng (2015) and Ludvigson, Ma, and Ng (2021); dispersion in real GDP forecasts from Consensus Economics.
  - Domestic macroeconomic and regulatory policy uncertainty: economic policy uncertainty index of Baker, Bloom, and Davis (2016); world uncertainty index of Ahir, Bloom, and Furceri (2022).
  - Geopolitical uncertainty: geopolitical risk index of Caldara and Iacoviello (2022).
- Correlations across measures:
  - Different measures tend to be positively but not strongly correlated; correlations vary considerably and are generally modest (panel 1, Figure 2.3; 1990–2023).
  - The REU exhibits the strongest correlation with other measures of macroeconomic uncertainty.
- Coverage and horizon differences:
  - Financial uncertainty measures (for example, VIX, option-implied market volatility) often capture short-horizon stock market risk.
  - Macroeconomic policy uncertainty may convey information relevant over much longer horizons (for example, geopolitical shocks, electoral cycles).

### Macro-market disconnect
- Measures of macroeconomic and financial uncertainty may contain complementary information but do not always fluctuate in tandem.
- Evidence of disconnects:
  - In some crises (global financial crisis, COVID-19 pandemic) all measures spiked in tandem (Figure 2.3, panel 2).
  - In other episodes, only some measures respond (for example, the late 1990s US dot-com bubble is captured mainly by financial uncertainty measures; US–China trade tensions starting around 2018 are largely captured by the economic policy uncertainty index).
- Macro-market disconnects can persist: realized and implied stock market volatility may be low while macroeconomic uncertainty is high (Figure 2.3, panels 3 and 4).

### Macroeconomic uncertainty and downside risk to output
- Modeling approach:
  - An augmented GaR model (Growth-at-Risk) is estimated using panel quantile regressions to examine the full distribution of future GDP growth, focusing on the left tail (10th percentile) as a measure of downside tail risk.
  - The analysis extends Adrian, Boyarchenko, and Giannone (2019) to include measures of macroeconomic and financial uncertainty, controlling for current GDP growth, financial conditions, and country fixed effects.
- Key quantitative findings:
  - A one-standard-deviation increase in measures of macroeconomic uncertainty reduces one-quarter-ahead real GDP growth (annualized) by 0.5 to 2.0 percentage points (Figure 2.4, panel 1).
  - Measures based on real outcomes (REU, GDP forecast dispersion) have the quantitatively largest effect.
  - The impact of macroeconomic uncertainty persists up to about seven quarters after the shock (Figure 2.4, panel 2).
  - In cumulative terms, an increase in the REU equivalent to that observed on average across countries during the global financial crisis translates into a decline in one-year-ahead GaR of about 1.2 percentage points.
  - The annual output decline in the bottom 10th percentile of the historical GDP growth distribution for the full sample (advanced and emerging market economies) is 1.2 percent.
- Asymmetric effects on the distribution of GDP growth:
  - Macroeconomic uncertainty has a negligible effect on the median of future real GDP growth, but large and statistically significant effects on lower and upper quantiles (Figure 2.4, panel 3).
  - Overall, increases in uncertainty exert a stronger effect on downside tail risks (5th or 10th percentiles) than on upside tail risks (90th or 95th percentiles).
  - Some episodes exemplify “good” uncertainty (raising upside tail risks): 1990s US dot-com bubble, mobile phone revolution in Finland, postcrisis reforms in Korea, German reunification in the late 1980s. Episodes of “bad” uncertainty include the onset of the global financial crisis and the COVID-19 pandemic.

### Machine learning enhancements and transmission channels
- Predictive performance improvements:
  - Machine learning GaR models (panel quantile random forest and panel quantile neural network) improve out-of-sample prediction accuracy relative to the standard linear quantile GaR benchmark.
  - Out-of-sample prediction accuracy for advanced and emerging market economies improves by up to 7 percent at different horizons (Figure 2.5, panels 1 and 2, green bars).
  - Adding measures of macroeconomic uncertainty (for example, the REU) as predictors further improves ML-GaR out-of-sample forecast performance by 5 to 13 percent relative to standard GaR models that exclude uncertainty (Figure 2.5, panels 1 and 2, red bars).
  - ML-GaR models show that macroeconomic uncertainty contributes at least as much as the financial conditions index to predicting downside risk to real GDP growth; on average, the REU contributes more to predictions than the financial conditions index (Figure 2.5, panels 3 and 4).
- Transmission channels:
  - Increased macroeconomic uncertainty is associated with a greater likelihood of large negative realizations of stock market returns and spikes in sovereign bond spreads (Box 2.1).
  - Macroeconomic uncertainty influences tail risks to future bank lending, particularly in countries with high banking exposure to sovereign debt.
- Robustness and caveats:
  - The focus is on predicting downside risks to future output, not identifying causal effects; robustness exercises address potential endogeneity concerns.
  - Results are qualitatively robust to alternative uncertainty measures and to excluding major crises from training samples.

*Source: CHAPTER 2 MACROFINANCIAL STABILITY AMId hIGh GLOBAL ECONOMIC uNCERTAINTY, International Monetary Fund, October 2024.*

### CHAPTER 2 MACROFINANCIAL STABILITY AMId hIGh GLOBAL ECONOMIC uNCERTAINTY

### CHAPTER 2 MACROFINANCIAL STABILITY AMId hIGh GLOBAL ECONOMIC uNCERTAINTY

### Macroeconomic uncertainty amplifies macrofinancial vulnerabilities
- High real economic uncertainty combined with excessive domestic credit (measured as the deviation of the credit to the private-sector-to-GDP ratio from its long-term trend) reduces one-quarter-ahead downside tail risk to GDP growth (the 10th percentile of the distribution of future GDP growth) by 0.6 percentage points.
- High public debt levels (captured by the deviation of the public-debt-to-GDP ratio from its long-term trend) significantly increase downside risks to GDP growth, particularly when real economic uncertainty is high.
- Machine learning GaR (growth-at-risk) models increase predictive power relative to a benchmark GaR model, especially when the real economic uncertainty index is included; macroeconomic uncertainty is an important predictor for one- and four-quarters-ahead GaR, particularly in emerging markets.
- A large macro-market disconnect—defined as the ratio of real economic uncertainty to realized market volatility that is above its mean—is associated with an increase in downside tail risks to future output growth.

### Intertemporal trade-off from easing financial conditions and role of macroprudential policy
- Easing financial conditions:
  - In the short term, an easing of financial conditions (rising asset valuations and compression of credit spreads and stock market volatility) reduces downside tail risks to GDP growth.
  - In the medium term, easy financial conditions encourage a buildup of debt vulnerabilities that exacerbate downside tail risks to GDP growth.
- Amplification by uncertainty:
  - Under high macroeconomic uncertainty, looser financial conditions exacerbate downside tail risks to GDP growth in the medium term.
  - The impact of looser financial conditions is more pronounced when there is a large macro-market disconnect, since compressed market volatility can reverse quickly in the face of shocks.
- Mitigation by macroprudential policies:
  - A net tightening of macroprudential policies can help offset the rise in medium-term downside risks associated with easy financial conditions, especially when there is a macro-market disconnect.
  - When the macro-market disconnect is large, a loosening of financial conditions coupled with a net tightening of macroprudential policies is associated with a reduction in downside risks to GDP growth of 0.3 to 0.6 percentage points in the medium to long terms, compared with a scenario with no macroprudential measures.
- Policy implication:
  - Policymakers may need to be more proactive in deploying policies aimed at preserving financial stability in periods when macroeconomic uncertainty is high relative to market volatility.
  - Credible policy frameworks may help reduce macroeconomic uncertainty and its impact on downside risks to output.

### Downside risks from foreign macroeconomic uncertainty and cross-border spillovers
- Construction of foreign uncertainty measures:
  - Foreign uncertainty measures are weighted averages of macroeconomic uncertainty in a country’s major trading and financial partners, using weights based on trade intensity (exports and imports) or banking and portfolio investment exposures.
- Spillover magnitudes and persistence:
  - A one-standard-deviation increase in trade-weighted foreign macroeconomic uncertainty (based on the REU) reduces the 10th percentile of the one-quarter-ahead distribution of domestic GDP growth by 1.7 percentage points; this effect is less persistent than a similar increase in domestic macroeconomic uncertainty and peters out in about three quarters.
  - Increases in REU among partners with strong banking relationships or cross-border portfolio investment exposures produce similar declines in domestic GaR; the effect from portfolio exposures is notably more persistent, suggesting nonbank financial intermediaries can play an important role in transmitting macroeconomic uncertainty across borders.
  - For uncertainty related to foreign economic policy, the impact on domestic downside risks to output is found to be more persistent, lasting up to six quarters.
- Mitigants to foreign uncertainty spillovers:
  - Building adequate international reserve buffers and greater exchange rate flexibility can help mitigate the adverse implications of foreign macroeconomic uncertainty for domestic downside risks to output.

*Sources: IMF staff calculations and the chapter text.*

### Conclusion and Policy Recommendations

### Conclusion and Policy Recommendations

### Macroeconomic uncertainty: effects and transmission
- Macroeconomic uncertainty remains elevated globally since the COVID-19 pandemic and increases downside risks to future real GDP growth, stock and bond market returns, and bank lending.
- Macrofinancial vulnerabilities (for example, high ratios of public and private sector debt to GDP) can interact with high macroeconomic uncertainty to amplify the effects of adverse shocks on future output growth.
- A macro-market disconnect (an easing of financial conditions accompanied by low financial market volatility) worsens the intertemporal trade-off posed by easing financial conditions for downside risk to medium-term output growth.
- Spillovers:
  - The impact of macroeconomic uncertainty tends to spill over across borders through trade and financial linkages, raising the risk of international contagion in the face of large adverse shocks.

### Quantitative findings from market and credit channels (Box 2.1)
- Sovereign bond spreads (market channel):
  - A one-standard-deviation increase in the real economic uncertainty index (REU) is, on average, associated with an increase of 150 basis points in upside tail risks to sovereign bond spreads (defined as the 90th percentile of the distribution of sovereign bond spreads) in emerging market economies at a six-month horizon.
  - For the average advanced economy, a one-standard-deviation shock to the REU increases upside tail risks by about 25 basis points.
  - The impact on sovereign bond spreads is more pronounced when fiscal vulnerabilities such as public debt service and banks’ exposure to public debt are high rather than low in emerging market economies.
- Stock market returns:
  - A one-standard-deviation increase in the REU can raise downside tail risks to stock market returns (the 10th percentile of the distribution of stock market returns) by about 30 percentage points, one year after the shock in advanced and emerging market economies.
- Bank lending (credit channel):
  - A one-standard-deviation increase in the REU is associated with a decline of about 1 percentage point (annualized) in the 10th percentile of the distribution of one-quarter-ahead real credit growth.
  - This effect persists through about seven quarters, although it becomes smaller over time.
  - Results are qualitatively robust across various measures of uncertainty, including a bank-level text-based measure constructed from earnings call transcripts.
  - Existing financial vulnerabilities amplify effects: for instance, a one-standard-deviation increase in the REU is associated with an increase in the one-year-ahead downside risk to lending (10th percentile) of about 1 percentage point when domestic banks’ exposure to sovereign risk is high (one standard deviation above the mean) compared to at the mean level.
- Data and methods notes:
  - Panel quantile regressions use samples such as 18 advanced and 13 emerging market economies (credit lending estimates) with data from 2001 to 2023; other panel samples include monthly data for 20 advanced and 9 emerging market economies (sovereign spreads) and 21 advanced and 19 emerging markets (stock returns) covering 1990:M1 to 2023:M12.
  - A bank-level text-based uncertainty measure follows the approach of Soto (2021) and uses the February 2024 update of the Loughran-McDonald Master Dictionary; this measure generally exhibits a low degree of correlation with other measures of uncertainty but remains statistically significant when included.

### Machine learning, AI tools, and systemic risk monitoring
- Machine learning models can improve the forecasting capacity of systemic risk assessment frameworks such as the GaR framework.
- Regulatory and policy institutions can enhance systemic risk monitoring by explicitly considering macroeconomic uncertainty as a key determinant of systemic risk and by exploiting AI tools:
  - Machine learning models for predicting downside tail risks to output and financial markets.
  - Other AI tools (such as natural language models) to extract high-frequency information from text-based sources (firms’ earnings call reports, social media, local and global news) to enhance real-time monitoring of systemic risk.
- Caveats and challenges in applying machine learning methods:
  - Significant data requirements and technological know-how, which may pose challenges for many emerging market and developing economies with data, skill, and technological constraints.
  - Weak signal-to-noise ratios of financial variables can lead large models to perform poorly out of sample.
  - Poor transparency and interpretability of machine learning methods.
  - The chapter addresses these shortcomings using cross-validation methods for model selection and overfit mitigation, numerical simulations, and analysis of variable importance (see Online Annexes 2.3, 2.4, and 2.5 for methodological details).

### Policy recommendations
- Overarching goal:
  - Policy actions should focus on reducing macroeconomic uncertainty and on mitigating its adverse effects by strengthening resilience and containing macrofinancial vulnerabilities.

- Reducing policy uncertainty:
  - Enhance credibility of monetary and fiscal policy frameworks through adoption of fiscal and monetary policy rules supported by strong institutions (Box 2.2).
  - Improve transparency and adopt well-designed policy communication frameworks to steer market expectations and make policy decisions more predictable and less uncertain.
  - Maintain a stable financial regulatory framework; announce and implement reforms with clear communication strategies, robust calibration, phase-in periods as necessary, and clear and practical use of supervisory discretion and enforcement.

- Mitigating financial stability risks amid high macroeconomic uncertainty:
  - Deploy adequate macroprudential and fiscal policies to contain financial stability risks arising from elevated macrofinancial vulnerabilities.
  - Proactively deploy macroprudential policies (for example, countercyclical capital buffers) to limit vulnerabilities, especially when financial conditions are loose and disconnected from elevated uncertainty (a macro-market disconnect).
  - Activate borrower-based measures if lax financial conditions amid high uncertainty encourage excessive borrowing for real estate investment.
  - Consider a tighter monetary policy stance where aligned with the central bank’s goal of maintaining price stability.
  - Use the level of macroeconomic uncertainty and its disconnect from financial market volatility to inform choices among policy instruments and the magnitude of the response, given different implementation and transmission lags for monetary and macroprudential policies.
  - Fiscal policies should prioritize debt sustainability to contain adverse effects of elevated public debt levels on borrowing costs (see the October 2024 Fiscal Monitor).
  - Prudential regulators and supervisors should ensure that bank and nonbank financial institutions assess vulnerabilities to cross-border spillovers of spikes in foreign macroeconomic uncertainty.
  - At the country level, maintain adequate reserve buffers and greater exchange rate flexibility to cushion potential adverse impacts of foreign uncertainty shocks.

- Geopolitical uncertainty:
  - Governments should build adequate safety nets to mitigate macrofinancial stability risks arising from rising geopolitical uncertainty.
  - Policymakers should strive to reduce geopolitical tensions through diplomacy and multilateral cooperation; where cooperation is elusive, devote resources to identifying, quantifying, managing, and mitigating financial stability risks associated with increases in geopolitical tensions and uncertainty.
  - Ensure an adequate level of international reserves and of capital and liquidity buffers at financial institutions to mitigate adverse consequences of increasing geopolitical risks.

*International Monetary Fund | October 2024*

### 1. Real Economic Uncertainty Index

### 1. Real Economic Uncertainty Index

### Empirical effects on downside risk to credit and GDP
- Panels 1 and 2 estimate the effect of a one-standard-deviation increase in uncertainty measures on the 10th percentile of future aggregate real credit growth (average quarterly rate, annualized) at multiple horizons.  
  - Panel 1: Real economic uncertainty index effects shown for horizons labeled Quarters 1 2 3 4 5 6 7 8 9 10 11 12. Numeric axis markers appearing in the figure: −1.5, 0.5, −1.0, −0.5, 0, −1.5, −2.0, 0.5, −1.0, −0.5, 0.  
  - Panel 2: Bank-level text-based uncertainty measure effects shown for horizons labeled Quarters 1 2 3 4 5 6 7 8 9 10 11 12. Numeric axis markers appearing in the figure: −3.5, 0, −3.0, −2.5, −2.0, −1.5, −1.0, −0.5.  
- Graph lines show estimated effects and shaded areas represent 90 percent confidence intervals. Results are obtained from panel quantile regressions using country-level data for both advanced and emerging market economies.
- Finding: A one-standard-deviation increase in real economic uncertainty or bank-level uncertainty shifts the 10th percentile of future real credit growth downward (annualized), with larger negative effects at specific horizons shown in the panels.

### Bank-level uncertainty and sovereign exposure (amplification)
- Panel 3 reports the Amplification Effect from Exposure to Sovereign Risk, One Year Ahead (percentage points, annualized).  
- Measures considered: Real economic uncertainty; Dispersion of GDP growth forecast; Economic policy uncertainty; Bank-level text-based uncertainty measure.  
- Interaction design: The “Base” effect (green bars) shows the coefficient for the uncertainty measure alone; “With interaction” (red bar) shows the effect when banks’ exposure to sovereign debt (share of banks’ holdings of sovereign debt in an economy as a share of banks’ total assets) is one standard deviation above the mean. The measures and sovereign exposure variable are standardized.  
- Estimated coefficients (solid bars) are statistically significant at the 10 percent level. Error bars represent 90 percent confidence intervals for the sum of the coefficients (base + interaction).  
- Key qualitative finding: Uncertainty’s negative impact on the lower tail of one-year-ahead real credit growth is larger when banks’ exposure to sovereign debt is higher (“its impact increases when financial vulnerabilities are high”).

### Monetary policy frameworks, policy credibility, and uncertainty
- Empirical relationship: Countries where inflation expectations deviate more from the policy (inflation) targets experience higher levels of economic policy uncertainty.  
  - Figure labeling related to this result includes regimes “Low deviation of inflation from target” and “High deviation of inflation from target” and the sample periods 2000–09 and 2010–23. Numeric axis markers in related figure: −4.0, −3.5, −3.0, −2.5, −2.0, −1.5, −1.0, −0.5, 0.  
- In a growth-at-risk framework, the effect of increased macroeconomic uncertainty (real or policy related) on downside risk to one-quarter-ahead GDP growth is larger when policy targets were missed by wider margins over the preceding three years (proxy for weaker monetary policy frameworks).  
- Conclusion: Enhanced credibility and stronger monetary policy frameworks tend to reduce economic policy uncertainty and mitigate the adverse implications of increased uncertainty for macrofinancial stability.

### Fiscal policy frameworks
- Fiscal rules and more stringent fiscal frameworks are associated in the literature with reductions in fiscal policy uncertainty, fiscal procyclicality, market volatility, and overall macroeconomic volatility (hence lower real economic uncertainty).  
- Evidence cited: fiscal rules can reduce budget balances or debt volatility, lower sovereign risk premiums, and enhance fiscal sustainability; discretionary fiscal policy is prone to deficit bias that fiscal rules can partly offset.

### Methods, significance, and notes
- Data sources: Bank for International Settlements; Fitch Analytics; IMF, Global Data Source and International Financial Statistics databases; Organisation for Economic Co-operation and Development, Main Economic Indicators database; and IMF staff calculations.  
- Estimation approach: Panel quantile regression for country-level data (advanced and emerging market economies).  
- Statistical notation: Confidence intervals shown are 90 percent; estimated coefficients in interaction analysis are statistically significant at the 10 percent level. See Online Annex 2.5 for further details (as referenced in the source).

*Source: Bank for International Settlements; Fitch Analytics; IMF, Global Data Source and International Financial Statistics databases; Organisation for Economic Co-operation and Development, Main Economic Indicators database; and IMF staff calculations.*

---


_Source: https://www.imf.org/-/media/files/publications/gfsr/2024/october/english/ch2.pdf_