## EXECUTIVE SUMMARY

## Source details

**Canonical URL:** [EXECUTIVE SUMMARY](https://www.imf.org/-/media/files/publications/pp/2019/ppea2019026.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/pp/2019/ppea2019026.pdf.md)
- [Structured JSON version](/-/media/files/publications/pp/2019/ppea2019026.pdf.json)

---

### Background and scope
- The EBA-Lite methodology was introduced into the Fund’s surveillance in 2015 to provide quantitative inputs for external assessments for members not covered by the EBA (the EBA covers only 49 countries).
- After several rounds of use and feedback from country teams, stakeholders, and the Executive Board, staff conducted the first review of the EBA-Lite in 2018; revisions summarized in this paper supersede the previous EBA-Lite methodology.
- The revised EBA-Lite incorporates changes consistent with the 2018 update of the EBA regression models where data permit and remains underpinned by the EBA conceptual framework in Phillips et. al. (2013).
- Date noted in the document: May 10, 2019.

### Three identified areas for improvement (high-level)
- Expanding fundamentals and policy determinants in the CA and REER regressions to better capture the external balance of EBA-Lite countries.
- Identifying alternatives to regression models for external assessments of large exporters of exhaustible commodities.
- Revising the approach for the assessment of external sustainability in highly indebted economies.

### Structure of the revised methodology (three modules)
- Regression Module
  - Clarifies the role of remittances and aid in the external balance.
  - Incorporates shocks (natural disasters, armed conflicts) to better explain the external balance in EBA-Lite countries.
  - Expands policy determinants by introducing social insurance policies and revising financial policy variables.
- Module for External Assessments of Exporters of Exhaustible Commodities
  - Presents two non-regression complementary approaches for external assessment in large exporters of exhaustible resources.
- Module for the Assessment of External Sustainability
  - Proposes a revised approach that addresses simplifying assumptions in previous assessments for highly externally indebted economies and presents a probabilistic measure.

### Module I — Revisions to the EBA-Lite regressions (key findings and methodological changes)
- Conceptual framework
  - Retains the Phillips et. al. (2013) framework: expand policies influencing CA and REER, and distinguish positive analysis from normative evaluation relative to policy norms.
- Remittances and aid
  - Econometric insight: Regressing the CA on its sub-components (including remittances and aid) is no longer a saving–investment regression and thus does not identify or estimate a CA gap.
  - Conceptual insight: Marginal propensities to save and invest out of current transfers (remittances, aid) differ from those out of other income.
  - Revisions implemented:
    - Aid and remittances are dropped from the CA regression model.
    - Migrant shares are introduced as a covariate to capture variation in CA balances linked to current transfers.
- Migrant share: empirical finding
  - Migrant shares are negatively associated with the current account.
  - Estimated effect: an 8.6 percentage points (or one standard deviation) increase in outward migration lowers the current account balance by about 1% of GDP.
  - Reported regression snippet: CA — Migrant Shares coefficient: -.001**  (.00015); Distinct countries: 126; No. observations: 2,097; Pseudo R-squared: .5681.
- Incorporating shocks: natural disasters
  - Natural disasters are included as indicators; impact allowed to vary non-linearly with financial account openness (Chinn-Ito index).
  - Preferred specification (dummy = 1 if any economic damage reported):
    - When a natural disaster hits a country with a completely open financial account, the CA balance is estimated to decline by about 1 percentage points of GDP.
    - For a country with a completely closed financial account (Chinn-Ito index equal to 1), the marginal effect is an increase in the current account balance of about 0.9 percentage points of GDP.
  - Selected Table 2 coefficients (reported):
    - Column (1): Dummy=1 if $ damage ratio > 0%: -0.013***  (0.00262); Dummy×Chinn-Ito: 0.029***  (0.00515); Observations: 2,097; R-squared: 0.568.
    - Column (2): Dummy=1 if $ damage ratio > 0.05%: -0.00589**  (0.00272); Dummy×Chinn-Ito: 0.0201***  (0.00577); Observations: 2,097; R-squared: 0.561.
    - Column (3): 5YMA disaster dummy: -0.00543***  (0.000802); 5YMA×Chinn-Ito: 0.0130***  (0.00160); Observations: 2,095; R-squared: 0.589.
- Militarized conflicts
  - Conflicts are included as a contemporaneous binary indicator (UCDP/PRIO Dataset Version 17.22).
  - Main estimated impact on current account (controlling for other determinants): the marginal impact of an armed conflict is to raise the current account by 1 percent of GDP (text summary).
  - Reported Table 3 excerpt: conflict indicator coefficient .008*** (.002) in one specification; other specifications report .002 (.003) and .006* (.003). Observations: 2,097, 2,097, 2,094; Distinct countries: 126, 126, 125; Pseudo R-squared: .569, .568, .576.
  - Conflict types covered: internal armed conflicts; internationalized internal armed conflicts; inter-state armed conflicts; extra-systemic armed conflict. Intensity labels: “minor” (25–999 fatalities) and “war” (fatalities > 1,000).
- Social insurance (public health expenditures)
  - Public health expenditures are included in the CA model to capture precautionary saving implications.
  - L.Public Health Spending/GDP reported coefficient: -0.81*** in Annex regressions.
  - Indicative policy norms for public health expenditures modeled as a function of demographics, income inequality, income, and fiscal revenues.
- Financial policy variables
  - Adopted financial variables:
    - DeMeaned private credit-to-GDP ratio (to reflect structural deviations).
    - Credit growth (to capture cyclical considerations).
  - Data limitations preclude BIS-type credit gaps for many EBA-Lite countries.
  - Indicative benchmarks:
    - Auxiliary regression to estimate desirable credit level uses financial development index, GDP per capita, public debt, inflation, inflation-targeting indicator, Euro accession dummy, and financial structure; regressions estimated by country groups.
  - Selected indicative estimated credit gaps and staff views (Table 9 examples):
    - Bhutan, Year 2016: Indicative Estimated Credit Gaps (P-P*) +1.7%; Staff view — “Credit and liquidity conditions should be closely monitored.”
    - Bangladesh, Year 2015: Indicative Estimated Credit Gaps (P-P*) -0.4%; Staff view — “Macroeconomic performance has been strong ... There is room for a counter-cyclical response ...”
    - Cambodia, Year 2016: Indicative Estimated Credit Gaps (P-P*) +25%; Staff view — “Rising financial sector vulnerabilities amid a credit boom ...”
  - Rule of thumb for desirable credit growth:
    - Baseline: medium-term private credit growth = rate of growth of nominal GDP so credit-to-GDP ratio grows at a constant rate.
    - If current credit-to-GDP is above (below) desirable level, desirable credit growth rates should be non-positive (non-negative).
- REER model revisions
  - REER regression incorporates public health and credit growth variables, disaster and conflict indicators, and uses indicative policy norms for credit-to-GDP, credit growth, and public health policy.
  - Annex III selected REER coefficients (reported):
    - L.NFA/GDP: 0.0175***, 0.057*** across columns.
    - DeMeaned private credit/GDP: 0.125***, 0.13***.
    - Real interest rate × K openness: 0.377***, 0.52***.
    - Log Terms of Trade Goods - WEO: 0.0982***.
    - Output Gap: 0.225**.

### Module II — Non-regression-based approaches for exporters of exhaustible commodities
- Limitations of regression approaches for exhaustible resource exporters
  - Regression resource temporariness measures do not capture normative considerations of fiscal policy, do not link resource wealth explicitly to above-ground wealth, and typically focus on oil and gas only.
  - These shortcomings are salient in the EBA-Lite sample, which includes many large exporters of exhaustible commodities and where CA model fit is relatively poor for these countries.
- Two complementary non-regression approaches presented
  - Approach 1: Consumption allocation rules (Bems and de Carvalho Filho, 2009)
    - Distribute resource wealth across periods by consuming an annuity out of resource wealth (below-ground wealth = present value of exports; plus above-ground wealth = net foreign assets).
    - Derive a norm for consumption and from that a saving norm; extension yields a fiscal annuity for government resource wealth.
  - Approach 2: Investment-needs model (Araujo et al., 2015)
    - Dynamic small open-economy model with explicit investment, capital scarcity, and credit constraints.
    - In low-income countries with capital scarcity, allocating part of resource wealth to finance investment may be desirable, lowering CA norms relative to consumption-only rules.
    - Introducing investment frictions can lead to current account norms over 5 percentage points higher than without such frictions (Araujo et al. results referenced).
- Case studies: Ecuador and Nigeria (illustrative)
  - Average 2005–2015 exhaustible commodities’ exports: Ecuador 57 percent of total exports; Nigeria 90 percent.
  - EBA-Lite CA regression selected results (Table 11):
    - Ecuador:
      - CA-Actual -0.2%
      - Cyclical Contributions 0.2%
      - Cyclically adjusted CA -0.4%
      - CA-Norm -3.6%
      - Cyclically adjusted CA Norm -3.7%
      - CA-Gap 3.3%
      - o/w Policy gap -0.6%
      - Elasticity -9.8%
      - REER Gap -33.8%
      - CA-Fitted -4.2%
      - Residual 3.9%
    - Nigeria:
      - CA-Actual 1.9%
      - Cyclical Contributions -0.1%
      - Cyclically adjusted CA 2.0%
      - CA-Norm 1.8%
      - Cyclically adjusted CA Norm 1.9%
      - CA-Gap 0.1%
      - o/w Policy gap -1.4%
      - Elasticity -8.7%
      - REER Gap -1.7%
      - CA-Fitted 0.4%
      - Residual 1.6%
  - Comparative findings:
    - Consumption-based rules imply higher CA and fiscal balance norms than the EBA-Lite regression for both countries.
    - Investment-needs approach yields CA norms lower than consumption-based rules but may be appropriate for countries with large investment needs and limited market access.
    - Example calibrated outcomes:
      - Ecuador: Investment model medium-term CA norm -0.5%; Consumption rules CA norm 0.2%.
      - Nigeria: Investment model CA norm by 2023 about ¼ percent; Consumption-based rule CA norm 4.4 percent.
  - Policy implication: choice of non-regression framework affects inferred fiscal norms and the implied size and composition of adjustment.

### Module III — Revised approach for external sustainability assessment
- Motivation
  - External sustainability (ES) methodology traditionally computes the REER adjustment required to satisfy the inter-temporal budget constraint to restore external sustainability.
  - High incidence of highly-indebted economies in the EBA-Lite sample motivated revisions.
- Identified simplifying assumptions in traditional ES approach (four key items)
  - (1) ES does not consider that an adjustment of exchange rates results in a revaluation of the NIIP.
  - (2) ES does not consider rates of return differentials except in a few countries.
  - (3) ES compares debt at a point in the future (rather than its present value) to its level today.
  - (4) ES takes account of the current account projection in just the last year of the WEO horizon.
- Revised ES approach
  - Proposes a deterministic formulation addressing the four simplifications and calculating required REER adjustment via the present-value condition.
  - Core equations and definitions presented (selected):
    - External debt accumulation: D_{t+1} = (1 + r^L_t) D_t + [NX_t + (r^A_t − r^L_t) A_t]
    - Present-value condition (ratios to GDP): d_t ≤ Σ_{j=0}^n ∏_{i=0}^{j} (1+g_{t+i})/(1+r^L_{t+i}) [ nx_{t+j} + (r^A_{t+j} − r^L_{t+j}) a_{t+j} ]
    - Stabilized-debt formulation leading to REER adjustment e* via equation (3) in the source.
- Deterministic vs. probabilistic assessment
  - Deterministic assessment: follows Blanchard and Das (2017) addressing the four concerns.
  - Probabilistic assessment: introduced as a complement, especially when deterministic approach implies a very large required REER depreciation.
    - Construct country-specific joint distribution of shocks to growth, trade balance, returns, and gross positions (e.g., conditional correlations, VAR).
    - For each draw, compute the REER adjustment required to satisfy the inter-temporal budget constraint.
    - Evaluate where current exchange rate stands in the distribution to estimate the probability that a REER depreciation is required.
  - Data requirements for probabilistic approach:
    - Requires reasonably long historical series; example: Blanchard and Das (2017) Chile illustration required 18 years of BOP and NIIP data.
    - Of 58 EBA-Lite countries with 2017 NIIP weaker than -60% of GDP:
      - 40 have a continuous series of at least 25 years of BOP and NIIP data.
      - 43 have a continuous series of 18 or more years.
- Role of financial factors and return differentials
  - Return differentials (r^A_t − r^L_t) can materially affect sustainability assessments.
  - High returns on assets relative to liabilities mitigate the need for future trade surpluses; the reverse amplifies required real depreciations.
- Practical matters on horizon and measurement
  - External adjustment of highly indebted economies should occur over periods longer than the WEO 5-year horizon; suggested extension method: let growth converge linearly to last WEO year forecast q over horizon T to derive steady-state net exports.
  - Measurement errors in positions and flows can mismeasure rates of return; reconciliation data can partially address this.

### Implementation and usage
- The revisions to the EBA-Lite methodology have informed external sector assessments for the EBA-Lite group of countries since the Fall of 2018.

### Annex I and Annex III — variables, data sources, and selected coefficients (selected highlights)
- Key variable definitions and data sources (selected):
  - Output gap — WEO (percent difference between actual and potential GDP).
  - Private credit to GDP — WDI (credit to non-financial private sector), demeaned by historical average.
  - Migrant share — number of outward migrants as ratio to domestic population (UN Population Division).
  - Natural Disasters — EM-DAT (binary = 1 when reported damage in percent of GDP exceeds zero).
  - Armed Conflicts — UCDP/PRIO Armed Conflict Dataset Version 17.22 (binary indicator).
  - Aid and Remittances — IFS, WDI (retained in REER regression but dropped from CA regression).
- Selected Annex III CA regression coefficients (Revised EBA-Lite methodology, reported):
  - Cyclically adjusted Fiscal Balance, instrumented: 0.441***.
  - L. NFA/Y: 0.023***.
  - L. Output per worker, relative to top 3 economies: 0.104***.
  - Commodity ToT gap * Trade Openness: 0.403***.
  - (∆Reserves)/GDP * K controls: 0.856***.
  - DeMeaned Private Credit/GDP: -0.030***.
  - L.Public Health Spending/GDP: -0.81***.
  - Institutional / Political Environment (ICRG-12): -0.034**.
  - Observations: 2,097; Number of countries: 126; R-squared: 0.568.
- Selected Annex III REER regression coefficients (Revised/Updated):
  - Real interest rate × K openness: 0.52***.
  - L.NFA/GDP: 0.057***.
  - DeMeaned private credit/GDP: 0.13***.
  - L.Output per worker - relative to top 3 economies: 0.28***.
  - Log Terms of Trade Goods - WEO: 0.0982***.
  - Observations and R-squared vary across specifications reported (e.g., R-squared: 0.479, 0.490 in reported columns).

*International Monetary Fund — EXECUTIVE SUMMARY (ppea2019026)*

### EXECUTIVE SUMMARY

### EXECUTIVE SUMMARY

### Background and scope
- The EBA-Lite methodology was introduced into the Fund’s surveillance in 2015 to provide quantitative inputs for external assessments for members not covered by the EBA (the EBA covers only 49 countries).
- After several rounds of use and feedback from country teams, stakeholders, and the Executive Board, staff conducted the first review of the EBA-Lite in 2018; revisions summarized in this paper supersede the previous EBA-Lite methodology.
- The revised EBA-Lite incorporates changes consistent with the 2018 update of the EBA regression models where data permit and remains underpinned by the EBA conceptual framework in Phillips et. al. (2013).
- Date noted in the document: May 10, 2019.

### Three identified areas for improvement (high-level)
- Expanding fundamentals and policy determinants in the CA and REER regressions to better capture the external balance of EBA-Lite countries.
- Identifying alternatives to regression models for external assessments of large exporters of exhaustible commodities.
- Revising the approach for the assessment of external sustainability in highly indebted economies.

### Structure of the revised methodology (three modules)
- Regression Module
  - Clarifies the role of remittances and aid in the external balance.
  - Incorporates shocks (natural disasters, armed conflicts) to better explain the external balance in EBA-Lite countries.
  - Expands policy determinants by introducing social insurance policies and revising financial policy variables.
- Module for External Assessments of Exporters of Exhaustible Commodities
  - Presents two non-regression complementary approaches for external assessment in large exporters of exhaustible resources.
- Module for the Assessment of External Sustainability
  - Proposes a revised approach that addresses simplifying assumptions in previous assessments for highly externally indebted economies and presents a probabilistic measure.

### Module I — Revisions to the EBA-Lite regressions (key findings and methodological changes)
- Conceptual framework
  - Retains the Phillips et. al. (2013) framework: expand policies influencing CA and REER, and distinguish positive analysis from normative evaluation relative to policy norms.
- Remittances and aid
  - Econometric insight: Regressing the CA on its sub-components (including remittances and aid) is no longer a saving–investment regression and thus does not identify or estimate a CA gap.
  - Conceptual insight: Marginal propensities to save and invest out of current transfers (remittances, aid) differ from those out of other income.
  - Revisions implemented:
    - Aid and remittances are dropped from the CA regression model.
    - Migrant shares are introduced as a covariate to capture variation in CA balances linked to current transfers.
- Incorporating shocks: natural disasters and militarized conflicts
  - Both are negative income shocks that can affect CA and REER over short horizons; they can raise the CA if national saving rises (or investment falls relative to saving) or lower the CA if countries smooth consumption by borrowing.
  - Empirical results:
    - Disasters tend to lower the CA in countries with open financial accounts but raise it in countries with closed financial accounts.
    - Militarized conflicts have ambiguous impacts on the CA a priori; empirically, they can disrupt consumption, investment and cross-border capital flows with potential offset by official external financing.
  - Revisions implemented:
    - Indicators for natural disasters and militarized conflicts are included in the CA model.
    - The impact of natural disaster shocks is allowed to vary non-linearly with financial account openness.
    - Including these shocks lowers residuals in estimating CA gaps and provides a quantitative measure for ad hoc adjustments to the underlying CA to capture current shocks.
- Social insurance (public health expenditures)
  - Previous EBA-Lite did not include social insurance policies despite their relevance for precautionary saving.
  - Improved data on public health expenditures permits inclusion of public health expenditures in the CA model (as in the EBA).
  - Positive perspective: Low public health expenditures and high exposure to health risks in EBA-Lite countries imply greater household precautionary saving.
  - Normative perspective: Competing demands on public resources and lower fiscal capacity suggest lower feasible public social expenditures; this guides indicative policy norms.
  - Indicative policy norms for public health expenditures are modeled as a function of demographics, income inequality, income (as in the EBA), and fiscal revenues (to reflect lower fiscal capacity).
- Financial policy variables
  - Previous models included financial variables but lacked tailored guidance for EBA-Lite countries.
  - Data limitations preclude BIS-type credit gaps for the EBA-Lite sample.
  - Revised approach uses two financial policy variables:
    - Credit growth to capture cyclical considerations.
    - Private credit in percent of GDP to reflect structural impediments and long-term deviations from desirable saving–investment levels.
  - Indicative benchmarks:
    - Introduces indicative benchmarks to guide appropriate levels of credit-to-GDP, assessed with respect to fundamentals (income level, financial development) and cyclical influences.
    - Proposes rules-of-thumb to assess medium-term credit growth taking into account the desirable level of credit.
- REER model revisions
  - REER regression model incorporates public health and credit growth variables, disaster and conflict indicators, and uses indicative policy norms for credit-to-GDP, credit growth, and public health policy.

### Module II — Non-regression-based approaches for exporters of exhaustible commodities
- Limitations of regression approaches for exhaustible resource exporters
  - Resource temporariness measures in regressions do not capture normative considerations (how fiscal policy should affect absorption), do not explicitly link resource wealth and above-ground wealth, and typically focus on oil and gas (excluding large exporters of non-fuel exhaustible commodities).
  - These shortcomings are salient in the EBA-Lite sample, which includes many large exporters of exhaustible commodities.
- Two complementary non-regression approaches presented
  - Approach 1: Allocation of resource wealth for consumption across periods (consumption-based smoothing of resource wealth).
  - Approach 2: Dynamic small open-economy model that explicitly incorporates investment to account for acute investment needs in many commodity exporters.
- Illustrative evidence provided to show how these models can complement external assessments of large exporters of exhaustible commodities.

### Module III — Revised approach for external sustainability assessment
- Motivation
  - External assessments for some highly indebted economies have used an external sustainability (ES) methodology that calculates the REER adjustment required to satisfy the inter-temporal budget constraint to restore external sustainability.
  - The incidence of highly-indebted economies is very high in the EBA-Lite sample.
- Revision objectives
  - Address simplifying assumptions in traditional ES approaches that can affect assessments of sustainability.
  - Present a probabilistic measure to supplement assessments of external sustainability.

### Implementation and usage
- The revisions to the EBA-Lite methodology have informed external sector assessments for the EBA-Lite group of countries since the Fall of 2018.

*International Monetary Fund — EXECUTIVE SUMMARY (ppea2019026)*

### 20.      The review identified several simplifying assumptions in the ES methodology that can

### ppea2019026 - 20.      The review identified several simplifying assumptions in the ES methodology that can

### Simplifying assumptions in the ES methodology and proposed revision
- Identified simplifying assumptions that can significantly affect the assessment of external sustainability:
  - (1) the ES does not consider that an adjustment of exchange rates results in a revaluation of the NIIP;
  - (2) the ES does not consider rates of return differentials except in a few countries;
  - (3) the ES approach compares debt at a point in the future (rather than its present value) to its level today;
  - (4) the ES approach takes account of the current account projection in just the last year of the WEO horizon.
- A revised ES approach that takes account of each of these considerations is proposed.

### Deterministic vs. probabilistic assessment
- The workhorse ES model is deterministic.
- A probabilistic assessment of external sustainability may be more suitable in some cases, as recommended in the IMF’s debt sustainability guidance.
- The revised EBA-Lite introduces a probability approach as a complementary tool, particularly useful when the deterministic approach implies a very large REER adjustment.

### Module I — Revisions to the EBA-Lite regressions: Role of remittances and aid
- Context and conceptual point:
  - EBA-Lite regression models previously included workers’ remittances and foreign aid as explanatory variables.
  - Both remittances and aid are recorded as current transfers in the current account (CA) (secondary income account).
  - Current transfers include personal transfers (workers’ remittances) and government-related transfers (foreign aid, e.g., budgetary grants).
  - In percent of GDP, current transfers are negligible in advanced economies, but significant in low-income and emerging economies.
- Behavioral evidence:
  - Remittances: evidence finds higher marginal propensity to consume from remittance receipts than from other income; higher marginal propensity to invest out of remittances (used for physical capital investment, particularly residential construction).
  - Foreign aid: representative-agent optimal consumption models suggest much moderate aid inflows “are likely to be consumed, not saved”; empirical estimates find marginal propensities to consume out of aid close to unity. Evidence on marginal propensity to invest out of aid is mixed.
- Econometric/accounting implications of including current transfers in CA regression:
  - Including aid and remittances in the CA model means the regression is no longer a pure saving-investment regression and does not estimate a current account gap.
  - Decomposing CA into Aid & Remittances (A&R) and CA_Other with covariates X yields:
    - CA ≡ [CA_other + A&R] = X′β + (A&R)′α + u  (equation (1))
    - Equivalent to CA_other = X′β + (A&R)′[α−1] + u  (equation (2))
  - Regression of equation (2) can estimate a norm and gap for CA_other, but not for the current account.

### EBA-Lite treatment of transfers in CA and REER models
- Accounting/econometric concerns apply to the CA model but not to the REER model.
  - Empirical evidence suggests both foreign aid inflows and remittance inflows result in real exchange rate appreciation (possible Dutch-disease effects raising demand for nontradables).
- Revision adopted:
  - Drop aid and remittances from the CA model but retain them in the REER model.
  - Introduce migrant shares as a driver of saving-investment balances to capture differences in propensities to save and invest where current transfers are large relative to GDP.
- Definition and rationale for migrant share:
  - Migrant share = number of outward migrants in ratio to the domestic population (gross outward migration; migrants included irrespective of legal status; typically excludes seasonal migrants).
  - Migrant share is directly proportional to current transfers in expectation; higher migrant share → larger remittance receipts → higher marginal propensity to consume/invest → lower current account.
  - Migration data source noted as UN Population Division (link given in source).

### Empirical estimates: Migrant shares on the CA balance
- Finding:
  - Migrant shares are negatively empirically associated with the current account.
  - Estimated effect: an 8.6 percentage points (or one standard deviation) increase in outward migration lowers the current account balance by about 1% of GDP.
- Table 1 empirical estimates (as reported):
  - CA — Migrant Shares coefficient: -.001**  (.00015)
  - Distinct countries: 126
  - No. observations: 2,097
  - Pseudo R-squared: .5681

### Country-specific considerations for migrant shares
- The overall assessment should consider country-specific details about migrant populations.
  - If transfers are perceived as temporary, or remittances go to relatively high-income households, a higher migrated share may lead to relatively higher saving from secondary transfers and the saving-investment balance may increase (example: Barai (2012) for Bangladesh).

### Module I.B — Natural disasters and saving-investment balances
- Conceptual points:
  - Natural disasters can affect saving and investment decisions through destruction of wealth (reducing consumption), lower present value of income (deterring investment), and damage to physical capital (raising reconstruction investment).
  - The net qualitative and quantitative effect on CA is an empirical question.
- Data and sample:
  - Analysis uses EM-DAT (Emergency Events Database) with occurrence and damage value for natural disasters; about half of recorded events include U.S. dollar damage values.
  - Focus on events with non-missing damage value.
- Incidence:
  - Natural disasters are more frequent and more intense in non-EBA countries (per square kilometer).
  - Omitting disasters from EBA-Lite regressions could bias coefficients on other regressors (e.g., GDP per capita).
- Empirical results and financial openness interaction:
  - Specifications use disaster dummies equal to 1 when reported damage exceeds thresholds; interact disaster dummy with Chinn-Ito financial openness index.
  - Preferred specification (dummy = 1 if any economic damage reported):
    - When a natural disaster hits a country with a completely open financial account, the CA balance is estimated to decline by about 1 percentage points of GDP.
    - For a country with a completely closed financial account (Chinn-Ito index equal to 1), the marginal effect is an increase in the current account balance of about 0.9 percentage points of GDP.
  - Results robust to threshold of 0.05 percent of GDP to construct dummy.
- Precautionary savings and recurring disasters:
  - Recurring disasters may induce precautionary savings; specification including a lagged five-year moving average (5YMA) of disaster incidence suggests countries hit more frequently tend to save more, except for the most financially-open countries.
  - Sample limitations: cannot identify all effects separately; impact effect estimates lose precision.
- Table 2 regression results (selected reported coefficients and statistics):
  - Column (1):
    - Dummy=1 if $ damage ratio > 0%, relative to world average: -0.013***  (0.00262)
    - Dummy×Chinn-Ito: 0.029***  (0.00515)
    - Observations: 2,097
    - R-squared: 0.568
  - Column (2):
    - Dummy=1 if $ damage ratio > 0.05%, relative to world average: -0.00589**  (0.00272)
    - Dummy×Chinn-Ito: 0.0201***  (0.00577)
    - Observations: 2,097
    - R-squared: 0.561
  - Column (3):
    - 5YMA - disaster dummy if > 0%, relative to world average: -0.00543***  (0.000802)
    - 5YMA×Chinn-Ito: 0.0130***  (0.00160)
    - Observations: 2,095
    - R-squared: 0.589
  - Robust standard errors reported in parentheses. Significance notation: *** p<0.01, ** p<0.05, * p<0.1.

### Module I.C — Militarized conflicts and the saving-investment balance
- Conceptual points:
  - Conflicts disrupt consumption, investment and cross-border capital flows: revise down growth expectations, discourage investment, trigger capital flight and raise borrowing costs.
  - Inter-state conflicts may bring sanctions and trade embargoes; disruptions can dislocate labor and depress income and consumption.
- Empirical ambiguity:
  - Direction of impact on trade balance is ambiguous: imports may shrink from lower domestic demand and access to borrowing; exports may fall if production is destroyed.
  - Lower private capital inflows can be offset by non-official flows (aid, official financing).
  - Existing evidence on conflicts’ impact on external accounts is mixed; impacts on saving-investment and external accounts are an empirical question.
- Illustration:
  - Figure 7 documents Ukraine: Net non-official inflows and trade surrounding the Annexation of Crimea (time series labeled by quarters, and annotations for key events in 2013–2015). Source data: IFS, WEO and author’s calculations.

*International Monetary Fund — The Revised EBA-Lite Methodology (excerpts).*

### 40.      The revised EBA-Lite considers the impact of armed conflicts on external positions.

### 40.      The revised EBA-Lite considers the impact of armed conflicts on external positions.

### Definitions and data
- Conflicts data source: UCDP/PRIO Dataset (Allanson and others, 2017); current database is the UCDP/PRIO Armed Conflict Dataset Version 17.22.
- Conflict types included:
  - (a) internal armed conflicts (between the government of a state and one or more internal opposition groups);
  - (b) internationalized internal armed conflicts (between the government of a state and an internal opposition group, with intervention from other states on one or both sides);
  - (c) inter-state armed conflicts;
  - (d) extra-systemic armed conflict (between a state and a non-state group outside its territory).
- Intensity coding in PRIO: “minor” when annual conflict-related human fatalities are between 25 and 999, and “war” when fatalities exceed 1,000 on an annual basis.
- Indicator construction in EBA-Lite: a binary indicator that assumes the value one in every year where a conflict took place and zero otherwise; the indicator covers all conflicts due to interval-censoring of the UCDP data.

### Incidence and patterns (1995–2016)
- Regional shares of conflicts worldwide (1995–2016):
  - African continent: 39 percent
  - Asia: 29 percent
  - Americas: 4 percent
- Temporal patterns:
  - The total number of conflicts has been steady since 1995.
  - Large armed conflicts have been noticeably fewer in the 2000s relative to the 1990s, although they have edged higher recently.
- Sample comparison:
  - Incidence of conflicts during 1995–2016 is higher in the EBA-Lite sample than in the EBA sample, both in absolute terms and in share of their respective samples.
- Visualization: Figure 8 in the source shows time series of “All Armed Conflicts” and “Large Armed Conflicts” for World, EBA, and EBA-Lite (1995–2016).

### Empirical approach
- Purpose: account empirically for the impact of conflicts on current accounts and the real exchange rate within the EBA-Lite regressions.
- Indicator specification: contemporaneous binary indicator for occurrence of a conflict in a given year; lagged indicators were tested but produced jointly noisy estimates and lowered the contribution of the indicator itself, so the final models use only the contemporaneous indicator.
- Rationale for including all conflicts: interval-censoring prevents identifying conflicts with substantial human losses that nonetheless fall below the UCDP/PRIO thresholds for “wars”.

### Key empirical findings
- Main estimated impact on current account (controlling for other CA determinants):
  - Text summary: the marginal impact of an armed conflict is to raise the current account by 1 percent of GDP (Table 3).
- Table 3 regression detail (as reported):
  - Indicator =1 if a conflict begins or continues in any year, 0 otherwise: coefficient reported as .008*** (.002) in one specification and .002 (.003) in another; a lagged value appears in one column reported as .006* (.003).
  - Institutions (ICRG Index) coefficients reported as -.038** (.014), -.034* (.014), and -.035* (.015) across specifications.
  - No. observations: 2097, 2097, 2094.
  - No. of distinct countries: 126, 126, 125.
  - Pseudo R-squared: .569, .568, .576.
- Interpretation notes:
  - The conflict indicator captures a shock distinct, though likely correlated with, institutional quality (ICRG index).
  - Background tests for persistence (lagged conflict indicator) produced expected signs but noisy joint estimates; thus only contemporaneous effects are used in the revised models.
  - Literature citations: mixed evidence—some studies (Bronson 1997; Christiansen et. al. 2009) find conflicts raise the current account balance; others (Morrow et. al. 1999; Penubarti and Ward 2000) find no statistically significant impact on trade.

### Implications for EBA-Lite modeling and policy analysis
- Inclusion of the armed-conflict indicator improves the EBA-Lite’s accounting for shocks to external accounts stemming from episodes of armed conflict.
- Practical modeling choice: use of a simple contemporaneous binary indicator accommodates data interval-censoring and avoids overstating persistence given noisy lagged estimates.
- Caution for interpretation: the indicator reflects occurrence of conflict years regardless of intensity labeling limitations (e.g., major conflicts that straddle years may not be coded as “war” in every year they persist).

*Source: IMF staff calculations.*

### 57.      The conceptual considerations, the country composition of the EBA-Lite sample, and

### 57.      The conceptual considerations, the country composition of the EBA-Lite sample, and

### Choice of financial variables in EBA-Lite regressions
- More than three-fourths of the EBA-Lite sample are low-income, developing or emerging economies where financial development needs tend to dominate cyclical considerations.
- To reflect both financial development and cyclical considerations, EBA-Lite models include:
  - the private credit-to-GDP ratio, demeaned to account for heterogeneity in starting levels of credit-to-GDP;
  - the growth rate of credit, in ratio to GDP, to capture the impact of cyclical changes (e.g. credit booms) on the current account.
- The BIS credit-gap measure, used in the EBA models, cannot be implemented in EBA-Lite because of significant data limitations (e.g., 31 countries in the sample do not have even 10 years of data at annual frequency; nearly no data at quarterly frequency; available series often subject to structural breaks).

### Direct measures of financial policy and financial benchmarks
- EBA-Lite includes two direct measures of financial policy as proxies for both financial excesses and financial depth.
- EBA-Lite retains the demeaned credit-to-GDP ratio as a benchmark measure capturing financial deepening and accumulation of financial excesses.
- To better capture cyclical developments, the revised EBA-Lite includes credit growth as an additional financial policy variable.

### Determinants of Private Credit/GDP (Table 8 results by country group)
- AM (Advanced and Middle?): Observations 637; R-squared 0.475; Number of countries 35.
  - Financial development 0.362** (0.16)
  - GDP per capita 0.459*** (0.12)
  - Public debt -0.001*** (0.0005)
  - Inflation targetter 0.17*** (0.04)
  - Inflation 0.001*** (0.000484)
  - Output gap -0.82** (0.41)
  - WEO growth forecast 5.27*** (1.62)
  - US Federal Funds Rate -0.03** 
  - Euro accession 0.19*** (0.05)
  - Constant -4.03*** (1.40)
- EM (Emerging Markets): Observations 1,078; R-squared 0.391; Number of countries 57.
  - Financial development 0.40*** (0.06)
  - GDP per capita 0.095*** (0.03)
  - Public debt 3e-05 (0.000257)
  - Financial structure -0.0349*** (0.00779)
  - Inflation targetter -0.0360** (0.0162)
  - Inflation 4e-05*** (1e-05)
  - Output gap 0.046 (0.12)
  - WEO growth forecast 0.590* (0.32)
  - Capital control -0.10*** (0.01)
  - US Federal Funds Rate -0.04*** 
  - Constant -0.35 (0.30)
- LIC (Low-Income Countries): Observations 1,506; R-squared 0.447; Number of countries 60.
  - Financial development 0.50*** (0.07)
  - GDP per capita 0.24*** (0.01)
  - Public debt 9.e-05** (4e-05)
  - Inflation targetter -0.0493* (0.0295)
  - Inflation -3e-05*** (8e-06)
  - Output gap -0.31*** (0.04)
  - WEO growth forecast -0.0263 (0.0887)
  - Capital control -0.07*** (0.01)
  - US Federal Funds Rate -0.003
  - Constant -1.70*** (0.10)

### Auxiliary regression to estimate desirable credit level
- An auxiliary regression is estimated to suggest an indicative norm for credit-to-GDP ratios (a “fundamentals-consistent credit gap” approach).
- Explanatory variables include: financial development index, GDP per capita, public debt, inflation, an inflation-targeting indicator, a dummy for Euro accession, and a measure of financial structure.
- Regressions are estimated separately by country groups.
- Statistical models are likely inadequate to accurately define financial policy norms; judgment in applying them is encouraged.
- The estimated desirable credit levels are broadly consistent with staff views for selected examples.

### Selected indicative estimated credit gaps and staff views (Table 9)
- Bhutan, Year 2016: Staff view — “Credit and liquidity conditions should be closely monitored.” Indicative Estimated Credit Gaps (P-P*) +1.7%
- Bangladesh, Year 2015: Staff view — “Macroeconomic performance has been strong over the past two years, underpinned by prudent policies. There is room for a counter-cyclical response to temporary growth shocks.” Indicative Estimated Credit Gaps (P-P*) -0.4%
- Cambodia, Year 2016: Staff view — “Rising financial sector vulnerabilities amid a credit boom pose significant risks to macro-financial stability.” Indicative Estimated Credit Gaps (P-P*) +25%

### Rule of thumb for desirable credit growth
- Baseline: set medium-term private credit growth equal to the rate of growth of nominal GDP so the credit-to-GDP ratio grows at a constant rate (steady-state assumption).
- Alternatively: if current credit-to-GDP ratio is above (below) its desirable level, desirable credit growth rates should be non-positive (non-negative).
- Example operationalization: choose the annual rate of credit growth that would achieve the desirable credit-to-GDP ratio over a specified medium-term horizon.
- If the implied growth rate is undesirably high because the desirable ratio is significantly larger than the current ratio, staff may consider a longer horizon (beyond the five-year WEO horizon).
- Country teams may consider other toolkits for assessment of systemic risk to inform judgment.

### Module II — Non-regression-based approaches for exporters of exhaustible resources
- Important considerations for exporters of exhaustible resources are not captured by current account regression approaches: exhaustible resource income streams are temporary and may warrant smoothing of domestic absorption for inter-generational equity and precautionary motives.
- Limitations of treating resource temporariness as a regression fundamental:
  - Does not capture normative considerations (how fiscal policy should affect absorption of resource wealth).
  - Does not explicitly link aspects of countries’ balance sheets (e.g., above-ground wealth differences).
  - EBA and EBA-Lite specifications focus on oil and gas exporters only; many countries export other non-fuel exhaustible commodities.
- Incidence of large exporters in EBA-Lite is significant; EBA-Lite CA model fit is relatively poor for these countries.
- Tight connection between external and fiscal balances for exporters of exhaustible resources implies understanding fiscal policy's role in current account gaps is important.

### Two alternative non-regression approaches discussed
- Consumption allocation rules (Bems and de Carvalho Filho, 2009):
  - Distribute resource wealth across periods by consuming an annuity out of resource wealth (below-ground wealth = present value of exports of exhaustible commodities, plus above-ground wealth = net foreign assets).
  - Derive a norm for consumption and from that a saving norm.
  - Extension: define an annuity for fiscal expenditures drawing from government resource wealth (present value of resource-related revenues plus net government assets) to derive fiscal savings norms.
- Investment needs model (Araujo et al., 2015):
  - Small open economy model that explicitly incorporates the role of investment, capital scarcity, and credit constraints.
  - In low-income countries where capital is scarce, allocating part of resource wealth to finance investment may be desirable, which lowers CA norms relative to pure consumption-allocation rules.
  - CA gaps depend on calibration of investment inefficiencies; introducing investment frictions can lead to current account norms over 5 percentage points higher than without such frictions (Araujo et al. results referenced).

### Considerations on precautionary motives and investment inefficiencies
- Precautionary motives (to buffer volatile revenues) can increase current account and fiscal policy norms, but simulations suggest they play a secondary role relative to consumption smoothing.
- Investment frictions:
  - Investment inefficiencies: one dollar invested translates into less than one dollar of productive capital.
  - Absorptive capacity constraints: adjusting capital stocks is costly.
  - Larger inefficiencies lower optimal investment and therefore lead to higher CA norms.
  - Calibrations of inefficiencies can be large in many resource-rich developing countries (Pritchett, 2000; IMF 2012).

### Case studies: Ecuador and Nigeria (illustrative comparisons of EBA-Lite and non-regression approaches)
- Average 2005–2015 exhaustible commodities’ exports: 57 percent of total exports in Ecuador; 90 percent in Nigeria.
- Table 11 — EBA-Lite CA regression results (selected):
  - Ecuador:
    - CA-Actual -0.2%
    - Cyclical Contributions (from model) 0.2%
    - Cyclically adjusted CA -0.4%
    - CA-Norm -3.6%
    - Cyclically adjusted CA Norm -3.7%
    - CA-Gap 3.3%
    - o/w Policy gap -0.6%
    - Elasticity -9.8%
    - REER Gap -33.8%
    - CA-Fitted -4.2%
    - Residual 3.9%
  - Nigeria:
    - CA-Actual 1.9%
    - Cyclical Contributions (from model) -0.1%
    - Cyclically adjusted CA 2.0%
    - CA-Norm 1.8%
    - Cyclically adjusted CA Norm 1.9%
    - CA-Gap 0.1%
    - o/w Policy gap -1.4%
    - Elasticity -8.7%
    - REER Gap -1.7%
    - CA-Fitted 0.4%
    - Residual 1.6%
- Under EBA-Lite:
  - Ecuador current account norm: -3.6 percent (fitted regression value -4.2 percent; policy contributions 0.6 percent, with fiscal policy gap contributing 1.3 percent to the norm).
  - Nigeria current account norm: 1.8 percent.

*Source: IMF staff estimates and calculations.*

### 70.      The consumption-based rules approach suggests higher current account and fiscal

### 70.      The consumption-based rules approach suggests higher current account and fiscal 

### Consumption-based rules: findings and implications
- Consumption-based rules imply higher current account and fiscal balance norms relative to the EBA-Lite CA regression model for both country cases.
- Ecuador:
  - EBA-Lite model: produces a large current account deficit norm.
  - Consumption-based allocation approach: suggests the norm is a small surplus (Figure 17a).
  - A constant real per capita annuity is presented as the preferred normative measure for external-sector assessments because it ensures each individual in each generation is allocated the same real resources out of the country’s wealth.
  - Under a constant real annuity, the norm accounting for population growth is shown separately (Figure 4b).
  - The consumption-based approach indicates a higher fiscal norm for 2017 compared to the medium-term projection, implying a significant fiscal adjustment over the medium term (Figure 17c).
- Nigeria:
  - EBA-Lite current account model suggests a norm of 1.8 percent.
  - Consumption-based rules suggest surplus current account norms that are significantly higher than the regression approaches when inter-generational and precautionary needs are taken into account (Figures 17b, 17d).

### Investment-needs approach: findings and suitability
- Findings:
  - The investment-needs approach suggests current account (CA) norms that are lower than those implied by consumption-based rules (Figures 18a, 18b).
  - The small-open economy model is calibrated using the latest available country data as initial conditions; medium-term norms are the focus to filter out cyclical components.
  - Ecuador:
    - Investment model medium-term CA norm: -0.5 percent.
    - Consumption rules model CA norm: 0.2 percent.
  - Nigeria:
    - Investment model CA norm by 2023: about ¼ percent.
    - Consumption-based rule (constant real per capita annuity) CA norm: 4.4 percent.
- Suitability and policy use:
  - The investment-needs model is better suited for countries with large investment needs and limited market access but with institutions able to effectively implement investments.
  - The model emphasizes public and private investment efficiency and can illustrate how external misalignments could change if investment efficiency is improved, motivating policies beyond external-sector measures.
- Interaction with fiscal norms:
  - The investment-needs framework does not explicitly include a government sector but can be used to gauge the fiscal norm when public sector investment represents the lion’s share of total investment.
  - In such cases, the fiscal gap could be smaller than in the consumption-based framework by exactly the same amount as the investment-needs CA norm is smaller than the consumption-model CA norm.

### On fiscal norms (non-regression-based methods)
- Two additional approaches to derive fiscal norms are provided beyond standard methods (e.g., debt-stabilizing fiscal balance):
  - The consumption-rules framework yields a fiscal norm consistent with intergenerational equity.
  - The investment-needs framework can be used to infer a fiscal norm when public investment dominates total investment (see above).

### External sustainability: analytical framework (inter-temporal budget constraint)
- Core concept:
  - Net external liabilities are sustainable if they are less than or equal to the present value of net exports, plus the rate of return differential between external assets and external liabilities times the gross asset position (Box 1).
- Equations and definitions (as presented):
  - External debt accumulation:
    - D_{t+1} = (1 + r^L_t) D_t + [NX_t + (r^A_t − r^L_t) A_t]    (1)
    - D_t is net external liabilities at the beginning of t.
    - NX_t is net exports in t.
    - r^L_t and r^A_t are real rates of return on liabilities and assets respectively.
    - A_t denotes gross assets.
  - Present-value condition (ratios to GDP):
    - d_t ≤ Σ_{j=0}^n ∏_{i=0}^{j} (1+g_{t+i})/(1+r^L_{t+i}) [ nx_{t+j} + (r^A_{t+j} − r^L_{t+j}) a_{t+j} ]    (2)
    - d_t denotes the ratio of NIIP to GDP.
    - nx_t denotes the ratio of net exports to GDP (inclusive of current and capital transfers).
    - g_t denotes the rate of growth of real GDP.
    - a_t denotes the ratio of external assets to GDP.
  - Stabilized-debt formulation and REER adjustment:
    - d_t − d* (∏_{i=0}^{n} (1+g_{t+i})/(1+r^L_{t+i})) ≤ Σ_{j=0}^n (∏_{i=0}^{j} (1+g_{t+i})/(1+r^L_{t+i})) [ nx_{t+j} + (r^A_{t+j} − r^L_{t+j}) a_{t+j} ]    (3)
    - Let e* be the REER that satisfies equation (3). The exchange rate adjustment required for debt to be sustainable is (e* − e).

### Deterministic and probabilistic assessments in the revised EBA-Lite methodology
- Deterministic assessment:
  - Adopts the approach in Blanchard and Das (2017) that addresses four simplifying concerns of the traditional ES approach and calculates the required REER adjustment per equation (3).
- Probabilistic assessment:
  - Complements the deterministic approach, especially when deterministic results indicate a very large required depreciation.
  - Uses country historical data to construct a country-specific joint distribution of shocks to growth, the trade balance, returns, and gross positions (e.g., via conditional correlations or a VAR).
  - For each draw from this distribution, compute the REER adjustment required to satisfy the inter-temporal budget constraint (Box 1, equation 3).
  - Assess where the current exchange rate stands in the distribution of required REER adjustments to yield the probability that a REER depreciation is required to restore sustainability.
  - Recognizes that due to high variability of real and financial shocks, the support of the distribution can be large; deterministic unsustainability does not imply probability one of needing depreciation.

### Role of financial factors and key simplifications in ES assessments
- Financial factors:
  - Gross stocks of foreign assets and liabilities now exceed national GDP in many advanced and emerging economies, elevating the role of return differentials.
  - High returns on assets relative to liabilities can mitigate the need for future trade surpluses; the reverse can amplify required real depreciations.
- Four key simplifications of the traditional ES approach (as identified):
  - It does not consider that exchange rate adjustments affect both net exports and the valuation of the current NIIP (revaluation linked to foreign currency denomination).
  - It generally does not consider return differentials except for a few countries.
  - It requires stabilizing debt at its level today at a point in the future, rather than comparing discounted future debt to today’s level.
  - It takes into account just the medium-term CA projection rather than the entire stream of projected CAs as required by the inter-temporal condition.
- Countries explicitly considered for return differentials in the traditional ES approach include: the United States, Brazil, China, Hong Kong SAR and Switzerland.

### Probabilistic approach: data requirements and practical matters
- Data requirements:
  - Requires a reasonably long historical data series to construct empirical joint distributions.
  - Example: the Chile illustration in Blanchard and Das (2017) required 18 years of BOP and NIIP data at annual frequency.
  - Of the 58 EBA-Lite countries whose 2017 NIIP was weaker than -60% of GDP:
    - 40 have a continuous series of at least 25 years of BOP and NIIP data.
    - 43 have a continuous series of 18 or more years.
- Measurement and horizon issues:
  - Measurement errors in positions and flows can significantly mismeasure rates of return; reconciliation data can partially address this.
  - External adjustment of highly indebted economies should desirably occur over a period longer than the WEO 5-year horizon, requiring longer forecasts for assessment.
  - A suggested approach to extend WEO forecasts: assume steady-state characterized by balanced growth, let q be WEO forecast of growth of potential output in the last WEO year and T be the long horizon; growth rates converge linearly to q from the last WEO year through T (illustrated for the United States: March 2018 WEO q = 1.4%; example T = 15 yields net exports converging to a steady-state ratio of -2.5 percent).

*Source: THE REVISED EBA-LITE METHODOLOGY, INTERNATIONAL MONETARY FUND*

### Annex I. Glossary of Variables and Data Sources

### Annex I. Glossary of Variables and Data Sources

### Overview
- All covariates enter the regressions either in deviations from the contemporaneous GDP-weighted global average (in the CA model) or in deviations from the contemporaneous trade-weighted average (in the REER model) to ensure multilateral consistency of estimated CA and REER norms.
- Where not explicitly available, time series are extended (e.g., using IMF commodity price projections) and HP filtered; many variables are demeaned or interacted with indices such as trade openness or capital account openness.

### Cyclical Variables
- Output gap
  - Percent difference between actual and potential GDP in percent of potential GDP (based on WEO estimates; HP filtered where unavailable).
- Commodity terms of trade gap
  - Constructed from a commodity index: ratio of a geometric weighted-average price of 43 commodity export categories to a geometric weighted-average price of 43 commodity imports, each relative to manufactured goods prices in advanced economies; weights are commodity shares in exports/imports.
  - Time series extended into medium term using IMF commodity price projections and HP filtered separately for each country; zero country-specific mean.
  - In the CA model, interacted with trade openness (average projected trade balance in percent of GDP over next 5 years).

### Policy Variables
- Cyclically-adjusted fiscal balance (instrumented)
  - Based on WEO projections or computed as residual from country-specific regression of overall fiscal balance on output gap.
  - Instrumented with lagged global variables: world real GDP growth, world output gap, the world cyclically-adjusted fiscal balance, and global risk aversion proxied by lagged U.S. corporate credit spreads; also controls for country-specific factors (GDP per capita, exchange rate regime, democracy ranking) and other predetermined variables.
- Public health spending/GDP
  - Total government expenditures on health schemes, in percent of GDP.
- Capital controls index
  - Chinn-Ito index varying between 0 (no controls) and 1 (full controls).
- Change in Foreign Exchange Reserves (instrumented)
  - Change in central bank foreign exchange reserves (including off-balance sheet changes) during the year, in ratio to nominal GDP in U.S. dollars; enters interacted with capital controls index.
  - Instrumented with: difference between M2/GDP and reserves/GDP, global reserves accumulation and an EMDE indicator (all interacted with capital controls index); first stage also controls for predetermined regressors.
- Private credit to GDP
  - Credit to non-financial private sector by domestic non-bank financial and banking institutions, demeaned by historical (1996-2016) average.
- Growth rate of private credit
  - Change in private credit between year t and t-1, normalized by year t nominal GDP.

### Fundamentals
- Net foreign assets (NFA) to GDP ratio
  - Enters in level terms and interacted with indicator equal to one if NFA < -60 percent of GDP.
  - Data are updated EWN dataset (Lane and Milesi-Ferretti (2007) extension).
- Output per worker, relative to top 3 economies
  - Ratio of GDP (PPP) to working age population relative to average of Germany, Japan, and U.S., demeaned; also interacted with capital account openness.
- Oil and gas trade balance, adjusted for resource “temporariness”
  - Exports minus imports of oil and gas in percent of GDP; uses net oil and gas external balance (five-year moving average, percent of GDP) multiplied by temporariness = (current oil extraction / proven reserves) relative to Norway in 2010.
- 5-year growth forecast
  - IMF WEO 5-year ahead real GDP growth as proxy for potential output growth.
- Population growth
  - Annual growth rate of total population.
- Old age dependency ratio
  - Population aged over 65 divided by population between 30 and 64.
- Share of prime-aged savers
  - Share of ages 45-64 as proportion of working-age population (30-64).
- Life expectancy
  - Life expectancy of a current prime-aged saver.
- Future Old Age Dependency Ratio
  - Average of old age dependency ratio 15, 20 and 25 years ahead; enters only interacted with life expectancy.
- Own currency share in global reserves
  - Share of a country’s currency in total world reserves (2017 shares noted: US dollar 67 percent, euro 20 percent, yen 4.7 percent, sterling 4.4 percent); enters alone and interacted with VIX and capital account openness.
- VIX
  - Index of implied U.S. stock market volatility (S&P 500); enters interacted with capital account openness and with capital account openness × own currency share.
- Institutional quality
  - 12 ICRG sub-indicators normalized to range 0–1 (higher = less risk).
- Migrant Share
  - Number of outward migrants as ratio to domestic population.
- Natural Disasters
  - Binary variable: 1 when reported damage in percent of GDP exceeds zero; enters directly and interacted with capital account openness.
- Armed Conflicts
  - Binary variable equal to 1 in years with an armed conflict (PRIO database).

### Additional variables in the EBA-Lite REER Regression
- Aid and Remittances
  - Current transfers in secondary income account, ratio to GDP.
- Real interest rate
  - Nominal short-term interest rate minus annual inflation rate.
- Commodity terms of trade (REER)
  - Ratio of geometric weighted-average price of six commodity categories (food, fuels, agricultural raw materials, metals, gold, beverages) to manufacturing goods prices in advanced economies; weights are time-average export and import shares of each category in total trade.

### Data Sources (selected mapping)
- Current Account (% of GDP) — WEO
- Output gap (% of potential GDP) — WEO, FAD Fiscal Monitor and Estimation
- Commodity terms of trade gap — WEO and World Bank, WITS
- Cyclically adjusted fiscal balance (% of GDP) — WEO
- Public health expenditures (% of GDP) — WDI
- Capital Controls (index) — Chinn-Ito Database
- Change in foreign exchange reserves (interacted with capital controls) — WEO
- Private Credit (% of GDP) — WDI
- Growth rate of Private Credit (% of GDP) — WDI and WEO
- Net foreign assets (NFA), (% of GDP) — EWN
- Output per worker, relative to top 3 economies — WEO and UN World Population Prospects
- Oil and gas trade balance, adjusted for resource temporariness — WEO, WITS, and British Petroleum Statistical Review of World Energy
- Expected real GDP growth, 5-years ahead (%) — WEO
- Population growth (%) — UN, World Population Prospects, 2017 Revision
- Own currency share in global reserves (%) — COFER Database
- VIX — Haver
- Institutional Quality — PRS Group, International Country Risk Guide
- Natural Disasters — Emergency Events Database (EM-DAT)
- Armed Conflicts — UCDP/PRIO Armed Conflict Dataset
- REER data — IMF, Information Notice System (INS)
- Real interest rate (%) — IFS, WEO and the Haver Database
- Aid and Remittances (% of GDP) — IFS, WDI

### Annex III — Estimated Coefficients of EBA-Lite CA and REER Regression Models (selected coefficients and model statistics)
- EBA-Lite CA Model (columns shown for EBAlite 1.0 and Revised EBA-Lite methodology):
  - Cyclically adjusted Fiscal Balance, instrumented: 0.527*** (EBAlite 1.0), 0.441*** (Revised)
  - L. NFA/Y: 0.009***, 0.023***
  - L. Output per worker, relative to top 3 economies: 0.103***, 0.104***
  - Commodity ToT gap * Trade Openness: 0.178***, 0.403***
  - (∆Reserves)/GDP * K controls: 0.598***, 0.856***
  - DeMeaned Private Credit/GDP: -0.052***, -0.030***
  - L.demeaned VIX * K openness: -0.002*** ; L.demeaned VIX * K openness * share in world reserves: 0.002**
  - Life Expectancy at Prime Age: -0.001*** ; Life Expectancy * Future OADR: 0.002***
  - Prime Savers Share: 0.127***
  - L.Public Health Spending/GDP: -0.81***
  - Own currency’s share in world reserves: -0.072***
  - Institutional / Political Environment (ICRG-12): -0.034**
  - Observations: 2,313 (EBAlite 1.0), 2,097 (Revised)
  - Number of countries: 150, 126
  - R-squared: 0.45, 0.568
  - Years: 1995-2013, 1995-2016

- EBA-Lite REER Model (columns shown for EBAlite 1.0 and Updated EBA-Lite methodology):
  - Change in reserves/GDP × K Controls (instrumented): -2.670***, 0.24
  - Real interest rate × K openness: 0.377***, 0.52***
  - L.NFA/GDP: 0.0175***, 0.057***
  - DeMeaned private credit/GDP: 0.125***, 0.13***
  - L.Output per worker - relative to top 3 economies: 0.548***, 0.28***
  - Life Expectancy at age 45: -0.04***
  - Population growth: -1.92***
  - GDP growth - forecast in 5 years: -0.54*
  - Share of reserve: 0.67***
  - Log Terms of Trade Goods - WEO: 0.0982***
  - L.Trade openness (exp + imp)/2GDP: -0.358***
  - Output Gap: 0.225**
  - Constant: 4.180***, 4.65***
  - Observations: 189, 315, 12 (table shows multiple observation counts across columns)
  - R-squared: 0.479, 0.490

- Significance notation, where shown, uses stars as presented in the tables (e.g., ***, **, *).

*THE REVISED EBA-LITE METHODOLOGY — Annex I. Glossary of Variables and Data Sources*

### References

### ppea2019026 - References

### External balance, current account, and sovereign external positions
- Araujo, J.D., B.G. Li, M. Poplawski-Ribeiro. and L.-F. Zanna. 2016. “Current account norms in natural resource rich and capital scarce economies,” Journal of Development Economics, Vol. 120, pp. 144-156.  
- Beidas-Strom, S. and P. Cashin. 2011. Are Middle Eastern Current Account Imbalances Excessive? IMF Working Paper 11/195.  
- Bems, R., I. de Carvalho Filho. 2009. “Exchange Rate Assessments: Methodologies for Oil Exporting Countries,” IMF Working Paper 09/281.  
- _________, 2011. “The current account and precautionary savings for exporters of exhaustible resources,” Journal of International Economics, Vol. 84, pp. 48-64.  
- Bénétrix, A., Lane. P., and Shambaugh, J. 2015, “International currency exposures, valuation effects, and the global financial crisis.”, Journal of International Economics, 96, 98-109.  
- Chinn, M., Prasad, E., 2003. Medium-term determinants of current accounts in industrial and developing countries: an empirical exploration. Journal of International Economics 59 (1), 47e76.  
- Gourinchas, P.O and Rey H. 2013. “External adjustment, global imbalances, valuation effects”, Handbook of International Economics, Vol. 14.  
- Gruber, J. and S. B. Kamin, 2005, “Explaining the Global Pattern of Current Account Imbalances,” mimeo (Washington, D.C.: Board of Governors of the Federal Reserve System, September).  
- Lane, P. and G. M. Milesi-Ferretti. 2005. “A Global Perspective on External Positions,” IMF Working Paper 05/161.  
- ________, 2006. “The External Wealth of Nations Mark II: Revised and Extended Estimates of External Assets and Liabilities, 1970–2004,” IMF Working Paper 06/69.  
- ________, 2009. “Where did all the borrowing go? A forensic analysis of the U.S. external position”, Journal of the Japanese and International Economies, 2, 177-199.  
- Phillips, S., L. Catao., L. Ricci., R. Bems., M. Das., J. Di Giovanni., D. Filiz Unsal., M. Castillo., J. Lee., J. Rodriguez., and M. Vargas. 2013. The external balance assessment (EBA) methodology, IMF working paper, WP/13/272. https://www.imf.org/external/pubs/ft/wp/2013/wp13272.pdf  
- Prati, A., L.A. Ricci, L. Christiansen, S. Tokarick and T. Tressel. 2011. “External Performance in Low-Income Countries,” IMF Occasional Paper 272.  
- Cubeddu, L., S. Krogstrup, G. Adler, P. Rabanal, M.C. Dao, S. Ahmed Hannan, L. Juvenal, C. Osorio Buitron, C. Rebillard, D. Garcia-Macia, C. Jones, J. Rodriguez, K.S. Chang, D. Gautam, Z. Wang, N. Li. 2019. :The External Balance Assessment Methodology: 2018 Update,” forthcoming, IMF WP.  

### Financial development, credit cycles, leverage, and stability
- Adrian, T., Shin, H. 2010. "Liquidity and leverage," Journal of Financial Intermediation, Elsevier, vol. 19(3), pages 418-437, July.  
- Borio, C. Furfine, C., Lowe, P. 2001. Procyclicality of the financial system and financial stability: issues and policy options. Marrying the macro- and microprudential dimensions of financial stability. BIS Papers 1, 1–57.  
- Cottarelli, Carlo, G. Dell’Ariccia, and I. Vladkova-Hollar (2003). “Early Birds, Late Risers, and Sleeping Beauties: Bank Credit Growth to the Private Sector in Central and Eastern Europe and the Balkans.” IMF Working Paper 03/213.  
- Drehmann, M., Tsatsaronis, K., 2014. The credit-to-GDP gap and countercyclical capital buffers: questions and answers. BIS Quart. Rev. March.  
- Iossifov, P. and M. Y. Khamis 2009. “Credit Growth in Sub-Saharan Africa—Sources, Risks, and Policy Responses” IMF Working Paper 09/180.  
- Jordà, O., M. Schularick, and A. M. Taylor. 2011. “Financial Crises, Credit Booms, and External Imbalances: 140 Years of Lessons.” International Monetary Fund (IMF) Economic Review, 59, 340–78.  
- Mendoza, E. G., V. Quadrini, and J. Ríos-Rull, 2009. “Financial Integration, Financial Deepness and Global Imbalances,” Journal of Political Economy 117(3): 317-416.  
- Rozenov, R. 2017. “Public Debt Sustainability under Uncertainty: An Invariant Set Approach”, IMF WP/17/57. Washington, DC: International Monetary Fund.  
- Sahay, R, M. Cihák, P. N’Diaye, A. Barajas, R. Bi, D. Ayala, Y. Gao, A. Kyobe, L. Nguyen, C. Saborowski, K. Svirydzenka, and S. Yousefi, 2015. “Rethinking Financial Deepening: Stability and Growth in Emerging Markets.” IMF Staff Discussion Note 15/08. Washington, DC: International Monetary Fund (May).  
- Sahay, R. M. Cihák, and P. N’Diaye. 2015. “How Much Finance is Too Much? Stability, Growth, and Emerging Markets.” IMF Blog Post.  
- Schularick, M., A. Taylor. 2012. “Credit booms gone bust: monetary policy, leverage cycles and financial crises, 1870–2008”, American Economic Review, vol 102(2), pp 1029–61.  
- Shin, H. S. 2013, “Procyclicality and the Search for Early Warning Indicators,” in S. Claessens, M. A. Kose, L. Laeven, and F. Valencia, eds., Financial Crises, Consequences, and Policy Responses, forthcoming.  
- Svirydzenka, K. 2016. Introducing a new broad-based index of financial development. IMF Working paper 16/5.  
- Guo, K. and V. Stepanyan. 2011. “Determinants of Bank Credit in Emerging Market Economies,” Working Paper No. 11/51, International Monetary Fund, March.  
- King, R.G., Levine, R., 1993. Finance and growth: Schumpeter might be right. Quarterly Journal of Economics 108, 717e738.  
- Levine, Ross, 2002. “Bank-based or Market-based Financial Systems: Which is Better?” Journal of Financial Intermediation, Vol 11, Issue 4, pp.398-428.  
- Levine, R. N. Loyaza, and T. Beck. 2000. “Financial Intermediation and Growth: Causality and Causes,” Journal of Monetary Economics, Vol 46 pp. 31-77.  
- Wurgler, J. 2000. Financial markets and the allocation of capital. Journal of Financial Economics 58.  
- Geanakoplos, J. A. Fostel. 2008. "Leverage Cycles and the Anxious Economy," American Economic Review, American Economic Association, vol. 98(4), pages 1211-1244, September.  
- Boyd, John, Ross Levine, and Bruce Smith. 1996. “Inflation and Financial Market Performance,” Federal Reserve Bank of Minneapolis, Working Paper 573D.  
- Khan, M. A. Senhadji, and B. Smith. 2001. “Inflation and Financial Depth,” IMF Working Paper 01/44 (Washington, International Monetary Fund).  

### Debt sustainability, public investment, and macro fiscal frameworks
- Barroy, H., S. Sparkes. and E. Dale. 2016, “Assessing fiscal space for health expansion in low-and-middle income countries: A review of the evidence”, Health financing working paper No. 3.  
- Berg, A., R. Portillo, S.-C. S. Yang and L.-F. Zanna. 2013. “Public Investment in Resource-Abundant Developing Countries,” IMF Economic Review, Vol. 61, No. 1.  
- Blanchard, O., Das, M. 2017, “A New Index of External Debt Sustainability”, NBER Working Paper No. 24068. https://www.nber.org/papers/w24068  
- Bouabdallah, O., Checherita-Westphal, C., Warmedinger, T. de Stefani, R., Drudi, F., Setzer, R.,Westphal. A. 2017. “Debt sustainability Analysis for Euro Area Sovereigns: A Methodological Framework”. ECB Occasional Paper Series, No. 185/April.  
- Barrett, P. 2018. “Interest-growth differentials and debt limits in advanced economies”, IMF Working Paper, forthcoming.  
- Rozenov, R. 2017. “Public Debt Sustainability under Uncertainty: An Invariant Set Approach”, IMF WP/17/57. Washington, DC: International Monetary Fund.  
- IMF. 2002. “Assessing Sustainability”. IMF Policy Paper.  
- IMF. 2018. “Scoping note for the Review of the Debt Sustainability Framework for Market Access Countries”. Washington, DC: International Monetary Fund.  

### Remittances, migration, and development finance
- Adams, R. 1998, “Remittances, Investment, and Rural Asset Accumulation in Pakistan,” Economic Development and Cultural Change, 47, 155–73.  
- Barai, M. 2012. “Development Dynamics of Remittances in Bangladesh”, Sage Publishing.  
- Chami, R., A. Barajas, T. Cosimano, C. Fullenkamp, M. Gapen and P. Montiel. 2008. “The Macroeconomic Consequences of Remittances”, International Monetary Fund, Occasional Paper 259.  
- Durand, J, W. Kandel, E. A. Parrado, and D. Massey. 1996. “International Migration and Development in Mexican Communities,” Demography, 33, pp. 249–64.  
- Glytsos, N. 1993, “Measuring the Income Effects of Migrant Remittances: A Methodological Approach Applied to Greece,” Economic Development and Cultural Change, 42, pp. 131–68.  
- Ratha, D. 2005. "Workers’ Remittances: An Important and Stable Source of External Development Finance" Economics Seminar Series, World Bank.  
- World Bank, 2006, The Development Impact of Workers’ Remittances in Latin America, Vol. 2: Detailed Findings, Report No. 37026 (Washington).  

### Natural disasters, conflict, violence, and trade impacts
- Blomberg, B., G. Hess. 2006. How Much Does Violence Tax Trade? Review of Economics and Statistics, 88(4): 599-612.  
- Ciccone, Antonio. 2011. "Economic Shocks and Civil Conflict: A Comment." American Economic Journal: Applied Economics, 3 (4): 215-27.  
- Davies, V. 2010. “Capital Flight and Violent Conflict: A Review of the Literature”, World Bank.  
- Felbermayr, G. and J. Gröschl. 2014. “Naturally Negative: The growth effects of natural disasters,” Journal of Development Economics, Vol. 111, pp. 92-106.  
- Laframboise, N. and B. Loko. 2012. “Natural Disasters: Mitigating Impact, Managing Risks,” IMF Working Paper 12/245.  
- Noy, I. 2009. “The macroeconomic consequences of disasters,” Journal of Development Economics, Vol. 88, pp. 221-231.  
- Gleditsch, N., P. Wallensteen, M. Eriksson, M. Sollenberg and H. Strand. 2002. “Armed Conflict 1946–2001: A New Dataset”, Journal of Peace Research, 39, 615–637.  
- Glick, R and A. Taylor. 2007. “Collateral Damage: Trade Disruption and the Economic Impact of War”, The Review of Economics and Statistics, MIT Press, vol. 92(1) ol. 92(1), pages 102-127, 05.  
- Fielding, D. 2003. "How Does Civil War Affect the Magnitude of Capital Flight? Evidence from Israel during the Intifada," Discussion Papers in Economics 03/10, Dept of Economics, University of Leicester.  
- Rasmussen, T.N. 2004. “Macroeconomic Implications of Natural Disasters in the Caribbean,” IMF Working Paper 04/224.  

### Health spending, public spending, and social sectors
- Barroy, H., S. Sparkes. and E. Dale. 2016, “Assessing fiscal space for health expansion in low-and-middle income countries: A review of the evidence”, Health financing working paper No. 3.  
- Moreno-Serra, R. and P. C. Smith. 2015. Broader health coverage is good for the national’s health: evidence from country level panel data, Journal of the Royal Statistical Society A (2015) 178, pp. 101-124.  
- Savedoff, William D. 2007. What should a country spend on health care? Health Affairs 26, 962-970.  
- Xu, K., P. Saksena, and A. Holly, 2011, The determinants of health expenditure: A country-level panel data analysis, Word Health Organization, Working Paper, December 2011.  

### Foreign aid, public spending, and governance
- Alesina, A., B. Weder. 2002, ”Do Corrupt Governments Receive Less Foreign Aid?” American Economic Review, Vol. 92, No. 4, pp. 1126-1137.  
- Boone, P. 1995. “The impact of foreign aid on savings and growth,” mimeo (London School of Economics).  
- _________, 1996. “Politics and the Effectiveness of Foreign Aid”, European Economic Review, 40, 289-329.  
- Dollar, D. and V. Levin. 2006. “The Increasing Selectivity of Foreign Aid, 1984-2003,” World Development, Vol. 34 (12), pp. 2034-2046.  
- Feyzioglu, T., S. Swaroop and M. Zhu. 1996. “Foreign Aid's Impact on Public Spending,” World Bank Policy Research Working Paper No. 1610.  

### Trade, politics, and institutions
- Mansfield, E. and R. Bronson. 1997. “Alliances, Preferential Trading Arrangements, and International Trade”, American Political Science Review 91(1): 94-107.  
- Penubarti, M. and M. Ward. 2000. Commerce and Democracy. Center for Statistics and the Social Sciences Working Paper No. 6, University of Washington.  
- Pollins, B. 1989b. “Does Trade Still Follow the Flag?” American Political Science Review, 83, 465-480.  
- Morrow, J, R. Siverson and T Tabares. 1999. Correction to: ‘The Political Determinants of International Trade. American Political Science Review 93(4): 931-933.  
- Blomberg, B., G. Hess. 2006. How Much Does Violence Tax Trade? Review of Economics and Statistics, 88(4): 599-612.  
- Glick, R and A. Taylor. 2007. “Collateral Damage: Trade Disruption and the Economic Impact of War”, The Review of Economics and Statistics, MIT Press, vol. 92(1) ol. 92(1), pages 102-127, 05.  

### Methodologies, datasets, and indices
- Ev ans, M. 2012. “Beyond CGER: Implementation of a new external assessment procedure”, IMF working paper.  
- Has anov, F. and A. Senhadji. 2008. “Macroeconomic Balance Approach and Assessment of Exchange Rate Misalignments,” in Assessment of Exchange Rates, Middle East and Central Asia Department, International Monetary Fund.  
- International Monetary Fund (IMF). 2006. Methodology for CGER exchange rate assessments. IMF Occasional Paper.  
- Lane, P. and G. M. Milesi-Ferretti. 2006. “The External Wealth of Nations Mark II: Revised and Extended Estimates of External Assets and Liabilities, 1970–2004,” IMF Working Paper 06/69.  
- Solt, F. 2017. The standardized world income inequality database (SWIID), 6.0.  
- Svirydzenka, K. 2016. Introducing a new broad-based index of financial development. IMF Working paper 16/5.  

### Country and regional studies, central bank guidance, and reports
- Garcia Martinez, P. and M. Soto. 2017. Enhancing medium-term fiscal policy framework: key imperative for the UAE, United Arab Emirates Selected Issues, 2017, International Monetary Fund.  
- IMF. 2017. “Banking Challenges in the Western Balkans: Prospects and Challenges”, Regional Economic Outlook: Europe Hitting Its Stride.  
- Reserve Bank of India. 2013. “Report of the internal working group on implementation of countercyclical capital buffer”, draft, December.  
- Reserve Bank of South Africa. 2011. Financial Stability Report, September.  
- World Bank, 2010 Comments on the consultative document countercyclical capital buffer proposal.  

*References as listed in ppea2019026 - References (source PDF).*

---


_Source: https://www.imf.org/-/media/files/publications/pp/2019/ppea2019026.pdf_