## _wp1561 - 1. Variables Definitions and Sources

## Source details

**Canonical URL:** [_wp1561 - 1. Variables Definitions and Sources](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1561.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1561.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1561.pdf.json)

---

### Introduction and objectives
- Macroprudential policies considered include: caps on loan to value (LTV) and debt to income (DTI) ratios, limits on credit growth and other balance sheet restrictions, (countercyclical) capital and reserve requirements and surcharges, and Pigouvian levies.
- Two main aims of the paper:
  - Describe usage of a large number of macroprudential policies (12) for a diverse sample of 119 countries over the 2000-13 period.
  - Study relationships between the use of these policies and developments in credit and housing markets to analyze effectiveness in managing credit and financial cycles.
- The dataset is based on a recent IMF survey (Global Macroprudential Policy Instruments, GMPI) carried out by the IMF’s Monetary and Capital Department during 2013-2014.

### Key descriptive findings on policy usage
- Macroprudential policies are used more frequently in emerging economies.
- Foreign exchange related policies are especially used more intensively in emerging economies.
- Borrower-based policies (such as caps on LTV and DTI ratios) are used relatively more in advanced countries, especially recently.
- Almost all countries use some policies to reduce systemic risks arising from intra-financial system vulnerabilities, including from dominant banks and interconnections among banks.
- Usage of macroprudential policies is associated with relatively greater cross-border borrowing.

### Data, survey, sample and coding
- Survey conducted by IMF staff with responses received directly from country authorities; cross-checked with the earlier 2011 survey and other surveys.
- Sample coverage:
  - 119 countries analyzed over 2000-2013: 31 advanced, 64 emerging, and 24 developing.
  - GMPI survey covers 131 countries, but 119 provided sufficiently comprehensive submissions; regressions use 106 countries due to control-variable availability.
- Instruments covered (12 of 18 GMPI instruments, binary coded):
  - General Countercyclical Capital Buffer/Requirement (CTC)
  - Leverage Ratio for banks (LEV)
  - Time-Varying/Dynamic Loan-Loss Provisioning (DP)
  - Loan-to-Value Ratio (LTV)
  - Debt-to-Income Ratio (DTI)
  - Limits on Domestic Currency Loans (CG)
  - Limits on Foreign Currency Loans (FC)
  - Reserve Requirement Ratios (RR)
  - Levy/Tax on Financial Institutions (TAX)
  - Capital Surcharges on SIFIs (SIFI)
  - Limits on Interbank Exposures (INTER)
  - Concentration Limits (CONC)
- Derived variables:
  - LTV_CAP: subset of LTV measures that are strict caps on new loans.
  - RR_REV: subset of RR measures that impose a wedge on foreign currency deposits or are adjusted countercyclically.
  - Borrower union index: 1 if LTV_CAP or DTI is used, 0 otherwise.
  - Borrower intersection index: 1 if both LTV_CAP and DTI are used, 0 otherwise.
  - Macroprudential Index (MPI): simple sum of scores on all 12 policies (MPI, 0-12).
- Coding choices and limitations:
  - Instruments are coded as binary (in place = 1) for the actual period they were operative; intensity and bindingness not quantified.
  - Agency in charge known only for 2013; constructed an index equal to the fraction of instruments decided by the central bank in 2013.

### Descriptive usage statistics and prevalence
- Time profile:
  - Average overall MPI rose from just above 1 in 2000 to almost 2½ in 2013.
- Instrument usage shares (percent of country-year observations across the 119 countries and 14 years):
  - CONC: 75 percent
  - INTER: 29 percent
  - RR_REV: 21 percent
  - LTV_CAP: 21 percent
  - DTI: 15 percent
  - LEV: 15 percent
  - TAX: 14 percent
  - FC: 14 percent
  - CG: 12 percent
  - DP: 9 percent
  - CTC: 2 percent
  - SIFI: 1 percent
- Other descriptive statistics:
  - Overall real credit growth: range -7.9 to 42.6 percentage points; standard deviation 13.1; mean 10 percentage points.
  - MPI: range 0 to 7; mean 1.8; standard deviation 1.5.
  - Central bank decided on use of macroprudential tools in 2013 in 71% of cases.
  - Policy interest rate: range 0.25% to 20%.
  - Credit/GDP (proxy for financial development): range 8% to 175%.

### Empirical approach and identification
- Base regression model components:
  - Dependent variable Y(i,t): aggregate or sectoral real credit growth or real house price growth.
  - One-period lags: lagged dependent variable, Macropru(i,t-1) (MPI or group/individual instruments), GDP(i,t-1) (real GDP growth), Bank Crisis(i,t-1) (Laeven and Valencia 2013 definition), Policy(i,t-1) (central bank policy rate).
  - Country fixed effects μi; White-Huber robust standard errors clustered by country.
- Estimation:
  - Use Arellano-Bond (1991) GMM estimator (STATA “xtdpd”) to address bias from lagged dependent variable with fixed effects; one lag needed to maximize sample size.
  - OLS reported for base regression but GMM used for main results.
  - Lagging macroprudential policy variables and using GMM mitigates, but does not fully eliminate, endogeneity concerns.
- Data treatment and diagnostics:
  - Regressions performed over 2001-2013 (summary statistics over 2002-2013).
  - All variables except categorical ones are winsorized at the 5 percent level.
  - AB AR(1) tests often reject; AB AR(2) tests generally not rejected.
  - Sargan test p-values reported as 1.00 in presented tables.
  - Significance notation preserved: ***, **, and * indicate significance at the 1, 5, and 10 percent levels, respectively.

### Main empirical findings — overall and by country groups
- Baseline and magnitudes:
  - Lagged MPI is negatively and statistically significantly associated with real credit growth (mitigating effect).
  - Lagged credit growth coefficient example: 0.245 (persistence in credit).
  - Banking crisis reduces credit by about 14 percentage points.
  - Lagged policy (interest) rate has a negative coefficient but economically smaller than MPI.
- Key coefficient estimates (selected, preserved as reported):
  - MPI coefficient (All, GMM): -7.637*** [1.876]
  - MPI coefficient (Advanced, OLS): -2.112*** [0.651]
  - MPI coefficient (Emerging, GMM): -1.376* [0.781]
  - MPI coefficient (Open, GMM): -6.743** [3.076]
  - MPI coefficient (Closed, GMM): -6.605*** [2.073]
- Economic magnitudes by income group:
  - Advanced economies: one standard deviation change in MPI reduces credit growth by about 2.2 percentage points (≈ 1/4 the standard deviation in credit growth, 9.04).
  - Emerging markets: one standard deviation change in MPI reduces credit growth by about 8.3 percentage points (≈ 2/3 the standard deviation in credit growth).
- By capital account openness:
  - Macroprudential policies more effective in relatively closed economies; coefficient more than twice as large in closed economies than in open economies, though still significant in open economies.
- Group-level patterns:
  - Emerging markets rely more on macroprudential policies overall; developing countries second; advanced countries least frequent users historically.
  - CONC, INTER, and LEV used consistently across advanced, emerging, and developing countries.
  - LTV more used in advanced countries; RR_REV and FC more in emerging countries; DP and CG more in developing countries.

### Instrument-level and sectoral results
- Borrower-based vs financial-institution-based:
  - Borrower-based measures generally negatively related to credit growth, strongest in emerging markets.
  - Financial institution–based measures associated with lower credit growth, especially in emerging and closed economies.
- Sectoral impacts (high-level summaries):
  - Overall credit:
    - Borrower union index (LTV_CAP or DTI) significant and similar to borrower-based index; intersection index (both used) not significant.
  - Household credit:
    - Borrower-based measures associated with lower household credit growth, especially in emerging markets and significantly in advanced countries.
    - LTV_CAP associated with less household credit in all countries.
    - DTI limits curtail household credit in advanced and emerging markets.
  - Corporate credit:
    - Negative relationships but weaker and often not statistically significant; borrower-based union index significant in advanced countries.
  - House prices:
    - Borrower-based measures’ coefficients negative but not statistically significant — house prices harder to moderate with macroprudential tools.
- Individual instrument highlights (coefficients and relative patterns preserved):
  - LTV_CAP: strongly associated in developing countries with lower overall credit growth; associated with less household credit in all countries.
  - DTI: important for reducing household credit in advanced and emerging markets and corporate credit in emerging markets.
  - FC (limits on foreign currency loans): negatively related to credit growth in all countries, especially emerging and developing; affects corporate credit in emerging markets and household credit in advanced countries.
  - RR_REV (FX and/or countercyclical reserve requirements): strong effects in emerging markets for any credit type, especially corporate credit; positive association with house price growth in emerging markets (potential residual endogeneity).
  - DP (dynamic provisioning): used almost exclusively in emerging markets; negative relation with overall credit growth.
  - LEV and CTC: negative effects in developing countries.
  - SIFI measures: positive relation with overall credit growth in developing countries (driven largely by Mongolia late in sample); otherwise not significant.
  - INTER and CONC: negatively related to credit growth in all markets; INTER effects driven by emerging and developing countries; INTER also appears to reduce house price growth in emerging markets.
  - TAX measures: dampening effect on overall credit in developing countries and on house prices in emerging markets.
  - Many other individual policies not statistically significant for credit and house price growth.

### Circumvention, cross-border effects, and interactions
- Circumvention evidence:
  - Greater use of macroprudential policies is associated with more reliance on cross-border claims (share of cross-border claims to total claims to the non-financial sector).
  - For open economies (column 2, Table 7): one standard deviation increase in MPI increases the cross-border ratio by 6 percentage points (≈ 1/3 the standard deviation of the cross-border ratio).
  - MPI association with cross-border credit ratio example:
    - MPI (All): 1.277 [0.983]
    - Cross-Border Ratio lag coefficient (All): 0.418*** [0.109]
  - Sample sizes for cross-border regressions: Countries: 108 (All), 47 (Open), 60 (Closed); Observations: 1,094 (All), 508 (Open), 575 (Closed).
- Interactions with country characteristics (selected findings from Table 8):
  - Interaction of MPI with log GDP per capita: not significant for any group.
  - Interaction with institutional quality (ICRG): largely not significant (except for closed economies).
  - Interaction with credit/GDP: for developing and closed countries, positive interaction (deeper financial systems harder to enforce policies).
  - Interaction with exchange rate regime: positive for open economies (more flexible exchange rates associated with greater difficulty to control overall credit).
  - Interaction with de jure financial openness: only statistically significant positive for developing countries.
- Financial cycle intensity and phase (selected findings from Table 9):
  - MPI × Credit Growth: negative coefficients for all groups; statistically significant in advanced countries and open economies — macroprudential policies have additional dampening effects when credit growth is higher.
  - Top 10% (boom) and Bottom 10% (bust) dummy interactions:
    - Top-10% dummy * MPI coefficients negative and often significant (e.g., -1.934*, -2.147***, -2.086** in separate regressions).
    - Bottom-10% dummy * MPI coefficients generally positive but often not statistically significant.
    - Single-regression estimates preserve the pattern: MPI: -5.915*** (All) with Top 10% Credit Growth (dummy) * MPI: -1.423* and Bottom 10% Credit Growth (dummy) * MPI: 0.462 (not significant).
  - Conclusion: effects of macroprudential policies depend on both intensity and phase of the financial cycle (stronger in booms).

### Methodology notes, robustness and sample selection
- Primary estimation method: Arellano-Bond GMM treating instruments and control variables as endogenous; OLS reported in some specifications.
- Diagnostic test outcomes reported (examples preserved): AB AR(1) often rejects; AB AR(2) generally not rejected; Sargan p-values reported as 1.00 in tables.
- Robustness checks and sample selection:
  - Excluding countries that did not use any macroprudential policy in 2013 (reduces sample by 11 countries, 100 observations) does not alter main results: MPI remains statistically significant and negative with similar size.
  - Adding time fixed effects and testing agency in charge (central bank index) — no clear evidence that impacts vary by agency.
- Data caveats:
  - Binary coding captures presence/absence, not intensity or bindingness.
  - Some positive associations (e.g., RR_REV with house price growth in emerging markets) may reflect residual endogeneity.

### Interpretation and policy implications
- Overall conclusions:
  - Macroprudential policies are associated with reductions in credit growth; effectiveness varies by instrument and country circumstances.
  - Policies tend to be more effective in emerging markets and relatively closed economies; weaker associations in advanced and financially open economies.
  - Borrower-based measures (LTV_CAP, DTI) can be effective, especially on household credit; foreign currency–related measures more effective in emerging markets.
  - Macroprudential policy use can be associated with increased cross-border borrowing, indicating circumvention and the importance of integrated policy design (macroprudential + capital flow management).
  - Policies exhibit asymmetric impacts across financial cycle phases: more effective in dampening booms than in preventing busts.
- Practical implications highlighted:
  - Design macroprudential policy packages considering country-specific circumstances (income level, financial openness, financial depth, exchange rate regime).
  - Targeted borrower-based measures suitable for advanced economies (housing-related vulnerabilities); foreign currency measures targeted to emerging markets exposed to volatile capital flows.
  - Address circumvention channels (cross-border banking, nonbank finance) via complementary bank regulation adaptations and capital flow management tools.
  - Central bank involvement: central banks decided on macroprudential tool use in 71% of cases in 2013, but regressions do not show systematic variation in impacts by agency.

### Annex and dataset notes
- Annex documents MPI, borrower-targeted and financial institution–targeted binary indicators annually for 2000–2013 (Tables A1–A3).
- GMPI survey primary source; survey covers about 17 key tools with detailed questions; analysis focuses on 12 instruments due to data coverage constraints.
- Footnote excerpts preserved: the full details of the survey are not open to the public; country officials that participated have access to the entire database.
- Methodological implication reiterated: analyses reflect adoption/use presence rather than calibrated strength or enforcement; cross-country comparability prioritized.

*Source: _wp1561 - 1. Variables Definitions and Sources (IMF working paper content provided).*

### 1. Variables Definitions and Sources ...................................................................................

### _wp1561 - 1. Variables Definitions and Sources

### Introduction and objectives
- Macroprudential policies considered include: caps on loan to value (LTV) and debt to income (DTI) ratios, limits on credit growth and other balance sheet restrictions, (countercyclical) capital and reserve requirements and surcharges, and Pigouvian levies.
- Two main aims of the paper:
  - Describe usage of a large number of macroprudential policies (12) for a diverse sample of 119 countries over the 2000-13 period.
  - Study relationships between the use of these policies and developments in credit and housing markets to analyze effectiveness in managing credit and financial cycles.
- The dataset is based on a recent IMF survey (Global Macroprudential Policy Instruments, GMPI) carried out by the IMF’s Monetary and Capital Department during 2013-2014.

### Key descriptive findings on policy usage
- Macroprudential policies are used more frequently in emerging economies.
- Foreign exchange related policies are especially used more intensively in emerging economies.
- Borrower-based policies (such as caps on LTV and DTI ratios) are used relatively more in advanced countries, especially recently.
- Almost all countries use some policies to reduce systemic risks arising from intra-financial system vulnerabilities, including from dominant banks and interconnections among banks.
- Usage of macroprudential policies is associated with relatively greater cross-border borrowing.

### Main empirical findings on effectiveness
- Some macroprudential policies are associated with reductions in growth rates in (real) credit and house prices.
- Policies appearing especially effective:
  - Borrower-based policies, such as limits on LTVs and DTIs.
  - Financial institutions–based policies, such as limits on leverage and dynamic provisioning.
- Policies seem more effective when growth rates of credit are very high, but provide less supportive impact in busts.
- Evidence of weaker associations between macroprudential policies and credit developments in:
  - Financially more open economies.
  - Economies that have deeper and presumably more sophisticated financial systems.
  - These patterns are suggestive of evasion.
- The association between policy usage and greater cross-border borrowing suggests avoidance channels; countries may limit these through adapting financial sector regulations and adopting capital flow management tools.

### Relation to prior literature and complementary evidence
- The paper contributes to two strands of literature:
  - Cross-country studies linking macroprudential policies with credit growth and financial indicators (examples and findings preserved as stated):
    - Lim et al. (2011): presence of policies such as LTV and DTI limits, ceilings on credit growth, reserve requirements (RR), and dynamic provisioning rules associated with reductions in procyclicality of credit and leverage.
    - IMF (2013b): time-varying capital requirements and RRs negatively associated with credit growth; LTV limits and capital requirements strongly associated with lower house price appreciation rates; reserve requirements associated with a reduction in portfolio inflows in emerging markets with floating exchange rates; LTVs appear to impact overall output growth.
    - Dell'Ariccia et al. (2012): macroprudential policies can reduce incidence of general credit booms and decrease probability booms end badly; reduce risk of a bust and attenuate spillovers to the rest of the economy.
    - Claessens et al. (2013): in 48 countries over 2000-2010, borrower-targeted measures (LTV, DTI, credit growth and foreign currency lending limits) effective in reducing bank leverage growth, asset growth and noncore to core liabilities growth; countercyclical buffers help mitigate increases but few policies stop declines in adverse times.
    - Zhang and Zoli (2014): review use in 13 Asian economies and 33 other economies since 2000; measures helped curb housing price growth, equity flows, credit growth, and bank leverage; loan-to-value ratio caps, housing tax measures, and foreign currency-related measures most effective.
    - Bruno, Shim and Shin (2014): for 12 Asia-Pacific countries, banking sector and bond market capital flow management policies effective in slowing bank and bond inflows respectively; some evidence macroprudential policies more successful when complementing monetary tightening.
    - Crowe et al. (2011) and Cerutti et al. (2015): maximum LTVs have best chance to curb real estate booms.
    - IMF (2011a): LTV tools effective in reducing price shocks and containing feedback between asset prices and credit.
    - Kuttner and Shim (2013): using 57 countries spanning more than three decades, housing credit growth significantly affected by changes in maximum DSTI ratio, maximum LTVs, limits on exposure to housing sector, and housing-related taxes; DSTI limit only significant in some methods; only housing-related tax changes impact house price appreciation.
  - Micro-level studies of specific instruments:
    - Jiménez et al. (2012): Spain—dynamic provisioning useful in taming credit supply cycles though did not stop the boom; during bad times dynamic provisioning helps smooth downturns and uphold firm credit availability.
    - Igan and Kang (2012): Korea—LTV and DTI limits moderate mortgage credit growth.
    - Evidence from Hong Kong (Wong, Fong, Li and Choi, 2011): macroprudential policies targeted at real estate borrowing reduce real estate cycles.
    - Camors and Peydró (2014): Uruguay—large unexpected RR increase in 2008 led to aggregate credit decline but riskier firms got more credit; larger/systemic banks less affected.
    - Aiyar, Calomiris and Wieladek (2013): UK bank-level data 1998-2007—higher bank-specific capital requirements dampened lending with strong aggregate effects.
    - IMF case studies (2014b): Israel and Sweden—evidence that macroprudential measures have effects; for Israel LTVs more effective than DP and CTC over the six-month period following adoption; limited evidence on house price inflation reduction.

### Contribution, scope, and organization of the paper
- Contribution:
  - Studies impact of a broad set of macroprudential policies (12 instruments) in a large set of 119 countries.
  - Classifies policies between borrower- and lender-based policies.
  - Distinguishes effects on different segments of credit markets (household versus corporate credit) as well as house prices.
- Scope and data:
  - Database spans the 2000-13 period.
  - Data collected via the GMPI IMF survey (2013-2014).
- Paper organization:
  - Section 2 documents data collection and relative use of policies over time, by income levels and degree of de-facto capital account openness.
  - Section 3 presents empirical analysis, methodology, data sources for dependent and control variables, and robustness tests.
  - Section 4 concludes.

*Source: _wp1561 - 1. Variables Definitions and Sources (IMF PDF content provided).*

### Appendix for further details on the data and corresponding questionnaire). The survey was

### _wp1561 - Appendix for further details on the data and corresponding questionnaire). The survey was

### Data, survey and coding
- Survey conducted by IMF staff with responses received directly from country authorities; cross-checked with the earlier 2011 survey and other surveys (e.g., Kuttner and Shim, 2013 and Crowe et al 2011).
- Sample coverage:
  - 119 countries analyzed over 2000-2013: 31 advanced, 64 emerging, and 24 developing.
  - GMPI survey covers 131 countries, but 119 provided sufficiently comprehensive submissions; regressions use 106 countries due to control-variable availability.
- Instruments covered (focus on 12 of 18 detailed GMPI instruments):
  - General Countercyclical Capital Buffer/Requirement (CTC)
  - Leverage Ratio for banks (LEV)
  - Time-Varying/Dynamic Loan-Loss Provisioning (DP)
  - Loan-to-Value Ratio (LTV)
  - Debt-to-Income Ratio (DTI)
  - Limits on Domestic Currency Loans (CG)
  - Limits on Foreign Currency Loans (FC)
  - Reserve Requirement Ratios (RR)
  - Levy/Tax on Financial Institutions (TAX)
  - Capital Surcharges on SIFIs (SIFI)
  - Limits on Interbank Exposures (INTER)
  - Concentration Limits (CONC)
- Additional derived variables:
  - LTV_CAP: subset of LTV measures that are strict caps on new loans.
  - RR_REV: subset of RR measures that impose a wedge on foreign currency deposits or are adjusted countercyclically.
  - Borrower union index: 1 if LTV_CAP or DTI is used, 0 otherwise.
  - Borrower intersection index: 1 if both LTV_CAP and DTI are used, 0 otherwise.
  - Overall macroprudential index (MPI): simple sum of scores on all 12 policies.
- Coding choices and limitations:
  - Instruments are coded as binary (in place = 1) for the actual period they were operative; intensity and bindingness not quantified to avoid subjectivity.
  - Agency in charge known only for 2013; constructed an index equal to the fraction of instruments decided by the central bank in 2013 (also computed separately for borrower- and bank-based instruments).

### Descriptive usage statistics
- Time profile: average overall MPI rose from just above 1 in 2000 to almost 2½ in 2013.
- Instrument usage shares (percent of country-year observations across the 119 countries and 14 years):
  - CONC: 75 percent
  - INTER: 29 percent
  - RR_REV: 21 percent
  - LTV_CAP: 21 percent
  - DTI: 15 percent
  - LEV: 15 percent
  - TAX: 14 percent
  - FC: 14 percent
  - CG: 12 percent
  - DP: 9 percent
  - CTC: 2 percent
  - SIFI: 1 percent
- Other key descriptive statistics:
  - Overall real credit growth: range -7.9 to 42.6 percentage points; standard deviation 13.1; mean 10 percentage points.
  - MPI: range 0 to 7; mean 1.8; standard deviation 1.5.
  - Central bank decided on use of macroprudential tools in 2013 in 71% of cases.
  - Policy interest rate: range 0.25% to 20%.
  - Credit/GDP (proxy for financial development): range 8% to 175%.

### Empirical approach and identification
- Base regression model includes:
  - Dependent variable Y(i,t): aggregate or sectoral real credit growth or real house price growth.
  - One-period lags: lagged dependent variable, Macropru(i,t-1) (MPI or group/individual instruments), GDP(i,t-1) (real GDP growth), Bank Crisis(i,t-1) (Laeven and Valencia 2013 definition), Policy(i,t-1) (central bank policy rate).
  - Country fixed effects μi; White-Huber robust standard errors clustered by country.
- Estimation:
  - Use Arellano-Bond (1991) GMM estimator (STATA “xtdpd”) to address bias from lagged dependent variable with fixed effects; one lag needed to maximize sample size.
  - OLS reported for base regression but GMM used for main results.
  - Lagging macroprudential policy variables and using GMM mitigates, but does not fully eliminate, endogeneity concerns.

### Main empirical findings — overall and by country groups
- Baseline:
  - Lagged MPI is negatively and statistically significantly associated with real credit growth (mitigating effect).
  - Lagged credit growth coefficient is 0.245 (persistence in credit).
  - Banking crisis reduces credit by about 14 percentage points.
  - Lagged policy (interest) rate has a negative coefficient (dampening effect) but economically smaller than MPI.
- Economic magnitudes by income group:
  - Advanced economies (column 3): one standard deviation change in MPI reduces credit growth by about 2.2 percentage points (≈ 1/4 the standard deviation in credit growth, 9.04).
  - Emerging markets (column 4): one standard deviation change in MPI reduces credit growth by about 8.3 percentage points (≈ 2/3 the standard deviation in credit growth).
- By capital account openness:
  - Macroprudential policies more effective in relatively closed economies; coefficient more than twice as large in closed economies than in open economies, though still significant in open economies.
- Group-level patterns:
  - Emerging markets rely more on macroprudential policies overall; developing countries second; advanced countries least frequent users historically.
  - CONC, INTER, and LEV used consistently across advanced, emerging, and developing countries.
  - LTV more used in advanced countries; RR_REV and FC more in emerging countries; DP and CG more in developing countries.

### Instrument-level and sectoral results
- Borrower-based vs financial-institution-based:
  - Borrower-based measures generally negatively related to credit growth, strongest in emerging markets.
  - Financial institution–based measures associated with lower credit growth, especially in emerging and closed economies.
- Sectoral credit and house prices (summary of Table 6):
  - Overall credit: borrower union index (LTV_CAP or DTI) significant and similar to borrower-based index; intersection index (both used) not significant — no clear complementarities.
  - Household credit:
    - Borrower-based measures associated with lower household credit growth, especially in emerging markets and significantly in advanced countries.
    - LTV_CAP associated with less household credit in all countries.
    - DTI limits curtail household credit in advanced and emerging markets.
  - Corporate credit:
    - Negative relationships but weaker and often not statistically significant; borrower-based union index significant in advanced countries (possible owner-financed firms).
  - House prices:
    - Borrower-based measures’ coefficients negative but not statistically significant — house prices harder to moderate with macroprudential tools.
- Individual instrument highlights:
  - LTV_CAP: strongly associated in developing countries with lower overall credit growth; associated with less household credit in all countries.
  - DTI: important for reducing household credit in advanced and emerging markets and corporate credit in emerging markets.
  - FC (foreign currency limits): negatively related to credit growth in all countries, especially emerging and developing; affects corporate credit in emerging markets and household credit in advanced countries.
  - RR_REV (reserve requirement revisions): strong effects in emerging markets for any credit type, especially corporate credit; positive association with house price growth in emerging markets (potential residual endogeneity).
  - DP (dynamic provisioning): used almost exclusively in emerging markets; negative relation with overall credit growth.
  - LEV and CTC: negative effects in developing countries.
  - SIFI measures: positive relation with overall credit growth in developing countries (driven largely by Mongolia late in sample); otherwise not significant.
  - INTER and CONC: negatively related to credit growth in all markets; INTER effects driven by emerging and developing countries; INTER also appears to reduce house price growth in emerging markets.
  - TAX measures: dampening effect on overall credit in developing countries and on house prices in emerging markets.
  - Many other individual policies not statistically significant for credit and house price growth.

### Circumvention, cross-border effects, and interactions
- Circumvention evidence:
  - Greater use of macroprudential policies is associated with more reliance on cross-border claims (share of cross-border claims to total claims to the non-financial sector).
  - For open economies (column 2, Table 7): one standard deviation increase in MPI increases the cross-border ratio by 6 percentage points (≈ 1/3 the standard deviation of the cross-border ratio).
  - Implies need to consider macroprudential policies together with capital flow management.
- Interactions with country characteristics (Table 8; summary):
  - Interaction of MPI with log GDP per capita: not significant for any group.
  - Interaction with institutional quality (ICRG): largely not significant (except for closed economies).
  - Interaction with credit/GDP: for developing and closed countries, positive interaction (deeper financial systems harder to enforce policies).
  - Interaction with exchange rate regime: positive for open economies (more flexible exchange rates associated with greater difficulty to control overall credit).
  - Interaction with de jure financial openness: only statistically significant positive for developing countries.
- Financial cycle intensity and phase (Table 9; summary):
  - Interaction MPI × credit growth: negative coefficients for all groups; statistically significant in advanced countries and open economies — macroprudential policies have additional dampening effects when credit growth is higher (intense cycles).
  - Top 10% (boom) and bottom 10% (bust) dummy interactions:
    - Dummies have predicted opposite signs: negative if growth in top 10%, positive if growth in bottom 10% (consistent across groups with some exceptions).
    - When MPI and both dummies included, macroprudential policies show additional dampening during exceptionally high positive credit growth; interaction with bottom-10% dummy mostly positive but not statistically significant.
  - Conclusion: effects of macroprudential policies depend on both intensity and phase of the financial cycle (stronger in booms).
- Robustness and sample selection:
  - Excluding countries that did not use any macroprudential policy in 2013 (reduces sample by 11 countries, 100 observations) does not alter main results: MPI remains statistically significant and negative with similar size.
  - Additional checks (not reported): adding time fixed effects, testing whether agency in charge (central bank index) changes impact — no clear evidence that impacts vary by agency.

### Interpretation and policy implications
- Overall conclusions:
  - Macroprudential policies are associated with reductions in credit growth; effectiveness varies by instrument and country circumstances.
  - Macroprudential policies tend to be more effective in emerging markets and relatively closed economies; weaker associations in advanced and financially open economies.
  - Borrower-based measures (LTV_CAP, DTI) can be effective, especially on household credit; foreign currency–related measures more effective in emerging markets.
  - Macroprudential policy use can be associated with increased cross-border borrowing, indicating circumvention and the importance of integrated policy design (macroprudential + capital flow management).
  - Policies exhibit asymmetric impacts across financial cycle phases: more effective in dampening booms than in preventing busts.
- Practical implications highlighted:
  - Consider country-specific circumstances (income level, financial openness, financial depth, exchange rate regime) when designing macroprudential policy packages.
  - Targeted borrower-based measures suitable for advanced economies (housing-related vulnerabilities); foreign currency measures targeted to emerging markets exposed to volatile capital flows.
  - Need to address circumvention channels (cross-border banking, nonbank finance) possibly via complementary bank regulation adaptations and capital flow management tools.
  - Central bank involvement: 71% of cases in 2013, but regressions do not show clear evidence that impacts systematically vary by the agency in charge.

*Source: Appendix for further details on the data and corresponding questionnaire). The survey was (IMF working paper content provided).*

### References

### _wp1561 - References

### Key empirical findings from regressions
- Macroprudential Index (MPI) is associated with lower real credit growth:
  - MPI coefficient (All, GMM): -7.637*** [1.876]
  - MPI coefficient (Advanced, OLS): -2.112*** [0.651]
  - MPI coefficient (Emerging, GMM): -1.376* [0.781]
  - MPI coefficient (Open, GMM): -6.743** [3.076]
  - MPI coefficient (Closed, GMM): -6.605*** [2.073]
- Credit Growth persistence:
  - Lagged Credit Growth coefficient ranges and examples:
    - All: 0.245*** [0.0715]
    - Advanced (OLS): 0.324*** [0.0512]
    - Emerging: 0.485*** [0.134]
- Crisis dummy large negative association with credit growth:
  - Crisis coefficient (All): -14.24** [6.669]
  - Crisis coefficient (Emerging, OLS): -5.967*** [1.706]
- Monetary policy (Policy Rate) shows a negative association with credit growth:
  - Policy Rate coefficient (All): -1.071*** [0.340]
  - Policy Rate coefficient (Advanced, OLS): -0.697*** [0.196]
  - Policy Rate coefficient (Emerging): -0.952** [0.417]

### Effects by instrument group (selected coefficients)
- Borrower-targeted instruments:
  - BORROWER (All): -11.06** [4.496]
  - BORROWER (Emerging): -8.389** [3.637]
- Financial institution-targeted instruments:
  - FINANCIAL (All): -8.838*** [2.523]
  - FINANCIAL (Emerging): -6.625*** [2.213]
- Individual instrument examples (All):
  - LTV_CAP: -12.35* (coefficient presented)
  - DTI: -24.16** (coefficient presented)
  - DP (Dynamic provisioning): -16.39*** (coefficient presented)
  - INTER (Limits on interbank exposures): -35.46** (coefficient presented)
  - CONC (Concentration limits): -29.84* (coefficient presented)
  - RR_REV (FX and/or countercyclical reserve requirements): -42.84* (coefficient presented)
  - Note: coefficients are reported in Table 6 with significance markers; columns combine many outcomes for compactness.

### Effects on different credit aggregates and prices
- Dependent-variable specific highlights:
  - HH Credit Growth mean: 6.74, median: 4.95, Min: -4.34, Max: 26.64, Standard Deviation: 7.82, Observations: 5131, Number of Countries: 131
  - Corp Credit Growth mean: 4.47, median: 2.99, Min: -5.81, Max: 19.32, Standard Deviation: 6.86, Observations: 3513, Number of Countries: 131
  - House Price Growth mean: 2.18, median: 1.43, Min: -10.87, Max: 17.24, Standard Deviation: 7.28, Observations: 4994, Number of Countries: 49
- Table 6 indicates heterogeneous effects of individual instruments on Corporate Credit Growth, HH Credit Growth, Credit Growth, and House Price Growth (coefficients and significance vary by instrument and subgroup).

### Cross-border credit ratio results
- MPI association with cross-border credit ratio:
  - MPI (All): 1.277 [0.983]
  - Cross-Border Ratio lag coefficient (All): 0.418*** [0.109]
- Regression controls (examples):
  - GDP Growth coefficient (All): -0.0383 [0.123]
  - Crisis coefficient (All): 3.873 [5.303]
  - Policy Rate coefficient (All): -0.0579 [0.234]
- Sample sizes:
  - Countries: 108 (All), 47 (Open), 60 (Closed)
  - Observations: 1,094 (All), 508 (Open), 575 (Closed)

### Variable definitions and data sources (selected)
- Key survey instruments (binary 0-1 unless noted):
  - LTV: Loan-to-Value Ratio — constrains highly levered mortgage downpayments.
  - DTI: Debt-to-Income Ratio — constrains household indebtedness.
  - DP: Time-Varying/Dynamic Loan-Loss Provisioning — requires banks to hold more loan-loss provisions during upturns.
  - CTC: General Countercyclical Capital Buffer/Requirement — requires banks to hold more capital during upturns.
  - LEV: Leverage Ratio — limits banks from exceeding a fixed minimum leverage ratio.
  - SIFI: Capital Surcharges on SIFIs — requires Systemically Important Financial Institutions to hold higher capital.
  - INTER: Limits on Interbank Exposures — limits fraction of liabilities held by banking sector.
  - CONC: Concentration Limits — limits fraction of assets held by a limited number of borrowers.
  - FC: Limits on Foreign Currency Loans — reduces vulnerability to foreign-currency risks.
  - RR: Reserve Requirement Ratios — limits credit growth; can target foreign-currency credit growth.
  - CG: Limits on Domestic Currency Loan — limits credit growth directly.
  - TAX: Levy/Tax on Financial Institution — taxes revenues of financial institutions.
- Derived instruments:
  - LTV_CAP: Loan-to-Value Ratio Caps (strictly enforced caps on new loans).
  - RR_REV: FX and/or Countercyclical Reserve Requirements (reserve wedge on foreign currency deposits or countercyclical adjustments).
- Group indices:
  - Macroprudential Index (MPI, 0-12) = LTV_CAP + DTI + DP + CTC + LEV + SIFI + INTER + CONC + FC + RR_REV + CG + TAX
  - BORROWER (0-2) = LTV_CAP + DTI
  - FINANCIAL (0-10) = DP + CTC + LEV + SIFI + INTER + CONC + FC + RR_REV + CG + TAX
- Data sources cited for variables:
  - Credit series: Adjusted BIS Domestic Bank Credit to Private non-financial sector or IMF IFS; deflated by yearly CPI growth from World Bank WDI.
  - GDP Growth: World Bank WDI.
  - Exchange Rate Regime: Updated Ilzetzki, Reinhart, and Rogoff (2004) classification.
  - Crisis indicator: Laeven and Valencia (2013) systemic banking crisis database.
  - Policy Rate: IFS Central Bank Policy Rate when available; otherwise Discount Rate or Repurchase Agreement Rate; ECB deposit facility rate for Eurozone countries.
  - ICRG Institutional Quality Rating: International Country Risk Guide (0-22).
  - De Jure Openness Index: Updated dataset from Quinn, Schindler, and Toyoda (2011).

### Descriptive statistics and country classifications
- Summary for MPI and instrument prevalence (Table 3 excerpts):
  - MPI mean: 1.85, median: 2.00, Min: 0.00, Max: 8.00, Standard Deviation: 1.57, Observations: 1,428, Number of Countries: 119
  - LTV_CAP mean: 0.18, median: 0.00, Std. Dev.: 0.38
  - DTI mean: 0.13, median: 0.00, Std. Dev.: 0.34
- Frequency of instrument use by subgroup (Table 2, percent of country-years using instrument over 2000-2013):
  - LTV_CAP: Total Countries 21%, Advanced 40%, Emerging Markets 20%, Developing 6%, Open 29%, Closed 14%
  - DTI: 15%, 13%, 21%, 0%, 19%, 12%
  - DP: 9%, 5%, 6%, 19%, 5%, 11%
  - CONC: 75%, 69%, 76%, 77%, 72%, 78%
  - INTER: 29%, 33%, 32%, 17%, 34%, 26%
  - RR_REV: 21%, 0%, 24%, 33%, 4%, 32%
  - CG: 12%, 0%, 11%, 26%, 9%, 14%
  - TAX: 14%, 14%, 14%, 11%, 17%, 12%
- Country subgroup classifications provided for IMF WEO (April 2014) income groups and de facto financial openness (open vs closed economies); lists of countries are included in the source.

### Methodology notes and robustness
- Estimation approaches:
  - Primary method: Arellano-Bond GMM treating instrument and controls (credit growth, GDP growth, crisis dummy, policy rate) as endogenous.
  - Column 2 of main regressions estimated via OLS.
- Sample period: regressions performed over 2001-2013 (summary statistics over 2002-2013).
- Data treatment:
  - All variables except categorical ones are winsorized at the 5 percent level.
  - Country fixed effects control for individual trends.
- Diagnostic tests (selected):
  - AB AR(1) tests often reject (e.g., 0.00 in many specifications), AB AR(2) tests generally not rejected (e.g., 0.11, 0.18, 0.13 in various columns).
  - Sargan test p-values reported as 1.00 in presented tables indicating non-rejection of over-identifying restrictions.
- Significance notation preserved: ***, **, and * indicate significance at the 1, 5, and 10 percent levels, respectively.

*Source: _wp1561 - References*

### 2013. Arellano-Bond robust standard errors clustered by countries are in brackets. ***, **, and *

### _wp1561 - 2013. Arellano-Bond robust standard errors clustered by countries are in brackets. ***, **, and *

### Interactions with Country Control Variables (Table 8)
- Table header context: "AllAdvancedEmergingDevelopingOpenClosed (1)(2)(3)(4)(5)(6)"
- GDP/Capita (log) * MPI: 0.3944.569-1.61111.292.1931.699
- ICRG index * MPI: -0.05860.0888-0.1320.130.198-0.512**
- Credit/GDP * MPI: 0.01160.02880.02810.414**0.0240.0548**
- Exchange Rate Regime * MPI: -0.1650.06810.08930.2370.426**-0.363
- De Jure Openness Index * MPI: -0.0267-0.0908-0.01180.140**-0.0956-0.0392
- Notes excerpt (preserved wording):
  - "The estimates are determined using Arellano-Bond GMM treating the instrument and the control variables as endogeneous. The dependent variable is real credit growth. MPI, credit growth, GDP growth, crisis, and the policy rate are included in each regression (omitted in the table). The other regressors are added separately to the baseline regression (except the interaction terms, which always enter with the associated independent variable), but their coefficients are represented in the same column for compactness. All variables except the categorical ones are winsorized at the 5 percent level. Country fixed effects control for individual trends. The regressions are performed over the period 2001-2013. Arellano-Bond robust standard errors are in brackets. ***, **, and * indicate significance at the 1, 5, and 10 percent levels, respectively."

### Interactions with the Financial Cycle (Table 9)
- Table header context: "AllAdvancedEmergingDevelopingOpenClosed (1)(2)(3)(4)(5)(6)"
- Separate Regressions:
  - MPI * Credit Growth: -0.0707-0.157**-0.0537-0.0628-0.100**-0.07
  - Top 10% Credit Growth (dummy) * MPI: -1.934*-2.147***-2.086**-0.0786-2.014***-1.862*
  - Bottom 10% Credit Growth (dummy) * MPI: 0.7621.777*0.137-1.2651.182*-0.0594
- Single Regression:
  - MPI: -5.915***-1.115-3.473***-5.935**-1.518*-5.214***
  - Top 10% Credit Growth (dummy) * MPI: -1.423*-1.562-2.125***-0.479-1.565***-1.669*
  - Bottom 10% Credit Growth (dummy) * MPI: 0.4621.396-0.188-2.5560.6630.0579
- Subsample: MPI>0 in 2013
  - Lag MPI: -6.995***-1.593*-5.223***-5.788*-2.687**-6.525***
- Notes excerpt (preserved wording):
  - "The estimates are determined using Arellano-Bond GMM treating the instrument and the control variables as endogeneous. The dependent variable is real credit growth. MPI, credit growth, GDP growth, crisis, and the policy rate are included in each regression (omitted in the table). In the first section of the table, the other regressors are added separately to the baseline regression (except the interaction terms, which always enter with the associated independent variable), but their coefficients are represented in the same column for compactness. In the second section of the table, all of the coefficients are determined in a single regression. In the third section of the table, countries whose MPI is equal to 0 in 2013 are omitted from the regression. All variables except the categorical ones are winsorized at the 5 percent level. Country fixed effects control for individual trends. The regressions are performed over the period 2001-2013. Arellano-Bond robust standard errors are in brackets. ***, **, and * indicate significance at the 1, 5, and 10 percent levels, respectively."

### Figures (as labeled)
- Figure 1: "The Macroprudential Policy Index by Income Level" — chart axis and legend labels preserved in source (years 2000–2012; series: All Countries, Advanced Countries, Emerging Markets, Developing Countries; title: Mean MPI: By Income Group).
- Figure 2: "The Relative Use of Macroprudential Policies over Time, by Income Group" — figure present in source.

### Annex 1: Macro‑Prudential Dataset (survey and instrument coverage)
- Dataset primary source:
  - "The main source of the aggregated dataset put together for our analysis of the use and effectiveness of macroprudential policies is the IMF survey on Global Macroprudential Policy Instruments (GMPI), which was carried out by Luis Jacome, Yitae Kim and Claudia Jadrijevic (all IMF staff) during 2013-2014."
  - "The central banks/national authorities of 125 member countries and the Central Bank of West African States (BCEAO) provided responses to more than 100 detailed questions on about 17 key macroprudential policy tools. In addition to these responses, we also used several of the more than 350 attachment files that countries included in the survey to complement the responses. We also cross-checked GMPI-responses with those in other surveys (e.g., Kuttner and Shim, 2013 and Crowe et al, 2011) as well as our own web-based and other searches, all to further ensure a high quality dataset. We focus on 12 macroprudential instruments included in the GMPI Survey and compute time series dummy indicators on the usage of each instrument for each of the 120 countries included in our sample during the period 2000-2013. The instruments covered and the main questions used from the GMPI Survey are detailed below (following the survey’s original numbering of sections and questions):"
  - Footnote excerpts preserved:
    - "The full details of the survey are currently not open to the public, but country officials that have participated in the survey have access to the entire database."
    - "The IMF GMPI survey covers 18 sections/instruments, due to lack of enough data and cross-sectional coverage, we have not included in the analysis Sector Specific Capital Buffer/Requirement (section 4 of the survey), Liquidity Requirements/Buffers (section 14), Loan-to-Deposit ratio (section 15), Margins/Haircuts on Collateralized Financial Market Transactions (section 16), Limits on Open FX Positions or Currency mismatches (section 17), and Other policies (section 18, the rest category)."
- List of 12 instruments covered (survey question snippets preserved):
  1. General Countercyclical Capital Buffer/Requirement
     - "1.1.9 Please specify the date when this instrument was introduced."
     - "1.1.9.1 Please specify whether any changes have been made to the countercyclical capital buffer/requirement since 2000. Yes or no"
     - "1.1.9.1.1 Please describe the changes (level and design of the instrument) made to the countercyclical capital requirement, together with the dates of such changes, since 2000."
  2. Leverage Ratio
     - "2.1.10 Please specify when this instrument was introduced."
     - "2.1.10.1 Please specify whether any changes have been made to the leverage ratio since 2000. Yes or no"
     - "2.1.10.1.1 Please describe the changes (level and design of the instrument) made to the leverage ratio, together with the dates of such changes, since 2000."
  3. Time‑Varying/Dynamic Loan‑Loss Provisioning
     - "3.1.9 Please specify the date when this instrument was introduced."
     - "3.1.9.1 Please specify whether any changes have been made to the time-varying provisioning scheme since 2000. Yes or No"
     - "3.1.9.1.1 Please describe the changes (level and design of the instrument) made to the provisioning, together with the dates of such changes, since 2000."
  5. Loan‑to‑Value (LTV) Ratio
     - "5.1.8 Please specify the date when this instrument was introduced."
     - "5.1.8.1 Please specify whether changes have been made to the ratios or other elements of this instrument since 2000. Yes or no"
     - "5.1.8.1.1 Please describe the changes (level and design of the instrument) made in the LTV ratio, together with the dates of such changes, since 2000."
  6. Debt‑to‑Income (DTI) Ratio
     - "6.1.7 Please specify the date when this instrument was introduced."
     - "6.1.7.1 Please specify whether any changes have been made to this instrument since 2000. Yes or No"
     - "6.1.7.1.1 Please describe the changes (level and design of the instrument) the DTI ratio, together with the dates of such changes since 2000."
  7. Limits on Domestic Currency Loans
     - "7.1.8 Please specify the date when this instrument was introduced."
     - "7.1.8.1 Please specify whether any changes have been made to the limit on domestic currency loans since 2000. Yes or no"
     - "7.1.8.1.1 Please describe the changes (level and design of the instrument) made to the limits together with the dates of such change since 2000."
  8. Limits on Foreign Currency Loans
     - "8.1.9 Please specify when this instrument was introduced."
     - "8.1.9.1 Please specify whether any changes have been made to the limits since 2000. Yes or No"
     - "8.1.9.1.1 Please describe the changes (level and design of the instrument) made to the limits, together with the dates of such changes, since 2000."
  9. Reserve Requirement Ratios
     - "9.1.9 Please specify the date when this instrument was introduced."
     - "9.1.9.1 Please specify if any changes have been made to the reserve requirements since 2000. Yes or no"
     - "9.1.9.1.1 Please describe the changes (level and design of the instrument) made to the reserve requirements, together with the dates of such changes, since 2000."
  10. Levy/Tax on Financial Institutions
     - "10.1.8 Please specify when this instrument was introduced."
     - "10.1.8.1 Please specify whether any changes have been made to the levy/tax ratios or other elements of this instrument since 2000. Yes or No"
     - "10.1.8.1.1 Please describe the changes (level and design of the instrument) the levy/tax on banks, together with the dates of such changes since 2000."
  11. Capital Surcharges on SIFIs
     - "11.1.7.2 Please specify when this instrument was introduced."
     - "11.1.8 Please specify whether any changes have been made to the surcharges on SIFIs since 2000. Yes or no"
     - "11.1.8.1.1 Please describe the changes (level and design of the instrument) made to the capital surcharges on SIFIs, together with the dates of such changes, since 2000."
  12. Limits on Interbank Exposures
     - "12.1.4 Please specify when this instrument was introduced."
     - "12.1.5 Please specify whether any changes have been made to this instrument since 2000. Yes or No"
     - "12.1.5.1 Please describe the changes (level and design of the instrument) the limits on interbank exposures, together with the dates of such changes since 2000."

*Source: _wp1561 - 2013. Arellano-Bond robust standard errors clustered by countries are in brackets. ***, **, and *.*

### 13. Concentration Limits

### 13. Concentration Limits

### Instrument introduction and changes
- Instruments are coded for the period they were actually in place, i.e., from the date they were introduced until the day they were discontinued (if this occurred during our sample period).
- The survey asks whether instruments were introduced and whether any changes have been made since 2000.
- The dataset does not attempt to capture the intensity of measures or changes in intensity over time because:
  - Constructing an intensity measure involves subjectivity the authors avoid.
  - The survey data do not allow constructing objective measures across countries and over time denoting when instruments are binding.
  - Levels/thresholds may change over time but may not capture how binding instruments actually are.
  - Variations in use are difficult to code objectively as tightening or loosening.
- As a result, the authors construct simple binary measures of whether or not instruments were part of policy choices.

### Coding and measurement approach
- Binary coding: each instrument is recorded as present or absent for the exact period it was in force.
- No numerical intensity or strength is assigned to instruments in the dataset.
- The approach prioritizes cross-country comparability and objectivity over capturing nuanced changes in design or enforcement.

### Macroprudential Index (MPI) and sub-aggregates
- The annex includes aggregated indices:
  - Macroprudential Index (MPI)
  - Two main sub-aggregates: Borrower-Targeted Instruments and Financial Institution-Targeted Instruments
  - Variables capturing Central Banks’ Oversight of Macroprudential Policies
- The tables present MPI and sub-aggregate indicators over the period 2000–2013.

### Tables and data coverage
- Time coverage in the tables spans 2000–2013.
- Country coverage is broad and includes low-, middle-, and high-income economies (as shown in the annex tables).
- Tables A1–A3 present annual binary indicators (and in some table views aggregated fractions) for:
  - Table A1: Macroprudential Index (MPI)
  - Table A2: Borrower-Targeted Macroprudential Instruments
  - Table A3: Financial Institution-Targeted Macroprudential Instruments
- The authors note availability of these tables and the individual instrument sheets in Excel on the IMF website.

### Key methodological implications for analysis
- Analyses using these data reflect the presence or absence of instruments, not their calibrated strength or enforcement.
- Time-series comparisons should interpret changes as changes in the adoption/use of instruments rather than their intensity.
- Cross-country comparisons are facilitated by the binary approach but may understate heterogeneity in actual policy tightness.

*Source: Authors' estimations based on IMF GMPI Survey and other sources.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1561.pdf_