## _wp1694

## Source details

**Canonical URL:** [_wp1694](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1694.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1694.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1694.pdf.json)

---

### Key findings on macroprudential policy effects
- In the two years after the activation of MaPs, bank credit growth falls on average by 7.7 percentage points relative to the counterfactual of no measure.
- Total credit growth falls on average by 4.9 percentage points relative to the counterfactual of no measure.
- The smaller decline in total credit relative to bank credit is explained by an increase in nonbank credit growth (substitution from bank to nonbank intermediation).
- Quantity-based measures slow bank credit growth by 8.7 percentage points over two years relative to the counterfactual of no policy change.
- Quantity-based measures have much stronger effects on credit growth than price-based measures, in both advanced and emerging market economies.
- Substitution effects (bank → nonbank) are stronger in Advanced Economies (AEs) than in Emerging Market Economies (EMEs).
- Substitution effects are stronger for quantity restrictions than for price-based measures.
- The effect on bank credit is always substantially higher than the effect on total credit to the private sector.
- The paper finds strong and statistically significant effects on specific forms of nonbank financial intermediation, such as investment fund assets.

### Cross-country and data measurement
- Bank and nonbank credit measures sourced from the BIS database on private non-financial sector credit (Dembiermont et al., 2013); quarterly series for 40 economies covering the last 40 years.
- Bank credit defined as all loans and debt securities held by domestic and foreign banks (subsidiaries and branches).
- Nonbank credit encompasses loans and debt securities held by all other sectors (insurers, pension funds, investment funds, other firms, households, etc.) and, for some countries, direct cross-border lending by foreign banks.
- Direct cross-border lending amounts to less than 5 percent of nonbank credit in the aggregate BIS reporting sample.
- Supplementary balance-sheet data from World Bank's Financial Development Database (Cihak et al., 2012); cross-section coverage ranges from about 80 countries (investment and pension funds) to over 100 countries (banks and insurance companies); covers period 1980–2012 for different series.
- Panel A (Table 1, 1997–2014 / 1997–2012) summary statistics:
  - Banks represent 85 percent of GDP in AEs and 60 percent of GDP in EMEs.
  - Nonbank credit represents 56 percent of GDP in AEs and 9 percent of GDP in EMEs.
  - Investment fund assets in AEs represent close to a third of GDP, about five times the size as in EMEs.
- Nominal credit growth (year-to-year percentage changes in nominal stock of sectoral credit):
  - EMEs: bank and nonbank credit grew by 10.5 percent and 11.6 percent, respectively.
  - AEs: bank and nonbank credit grew by 6.5 percent and 7.4 percent (reported series). Total credit grew at an average annual rate of 6.7 percent in AEs and 9.8 percent in EMEs.

### Macroprudential policy events and classification
- MaP indicators from Cerutti et al. (2015), based on IMF GMPI survey, covering 120 countries over 2000–13; 12 categories produce 12 dummy variables (1 when used in a given year, 0 otherwise).
- Authors classify MaPs into price-based and quantity-based measures and also by lender- versus borrower-based.
  - Examples: price-based = dynamic provisioning requirements, taxes on financial institutions; quantity-based = limits on interbank and foreign currency exposures, leverage ratios (treated as quantity-based).
- On average: AEs have 1.6 MaPs in place in a given year; EMEs have 2 MaPs.
- Dataset: 171 MaP events, 77 percent of which are quantity-based.
- Most events cluster during 2007–13; prior to that MaPs were mostly activated in EMEs.
- GMPI qualitative answers indicate MaPs primarily aimed at depository institutions (banks), including borrower-based measures.
- MaP event definition: adoption of a new measure (policy tightening); events when a dummy goes from 0 to 1.

### Methodology and identification
- Main research question: whether MaPs lead to substitution from bank-based financial intermediation to nonbank intermediation.
- Approaches used:
  - Event study methodology (leads-and-lags) to analyze timing of effects around activation dates and compute cumulative excess growth rates (CEGR) up to three years before and after.
  - Panel regressions using net sectoral credit flow (quarterly, scaled by total credit) as primary outcome: NetFlow_{c,t} = α_c + β_t + θ_1 BankCrisis_{c,t} + θ_2 MonetaryPolicyΔ_{c,t} + θ_3 MaPΔ_{c,t} + ε_{c,t}; country and time fixed effects; standard errors clustered by country.
  - Regressions control for banking crises (Laeven and Valencia (2013) indicator), monetary policy (year-on-year policy rate changes; year-to-year changes in central bank balance sheet relative to GDP), and lagged macro covariates (GDP growth, current account, capital inflows, central bank rates, growth in central bank assets).
- Net sectoral credit flow measure: difference between quarterly change in bank credit and quarterly change in nonbank credit, scaled by total credit; winsorized at 1 percent tails; Argentina excluded.
- Event‑study excess growth computation: E[ŷ] = y - E[y] with E[y] predicted by covariates x, μ_c, α_t; coefficients φ_i estimated for i periods before/after activation; cumulative excess growth rates and Wald tests used to test anticipation and post-activation effects.
- Robustness: alternative price/quantity classifications tested in online appendix; placebo tests constructed by simulating MaP events to match distribution of actual events.

### Panel regression results (net sectoral credit flows)
- MaP coefficients negative across regressions, consistent with boundary hypothesis (MaP activation shifts provision of credit toward nonbanks).
- Overall sample MaP coefficient: -0.26 (quarterly coefficient), implying during the first year following MaP activation net sectoral credit flows move by about 1 percentage point of total credit in favor of nonbanks (coefficient × 4 = annual change).
- Statistical significance: MaP coefficient significant in most regressions except the EME subsample.
- Banking crisis effects:
  - Banking crises reduce net credit flows from banks by about 5pp of total credit per annum on average; in EMEs the effect is about 10pp per annum.
- Central bank balance sheet expansion:
  - Overall sample: a 1 percent increase in central bank assets is associated with a 2pp per annum shift in net credit flows originating from banks (shift toward nonbanks).
  - Association particularly strong in AEs and bank-based financial systems.
- Price vs quantity MaPs:
  - Quantity-based MaPs: negative and statistically significant coefficients.
    - Overall sample: quantity measures suggest a 2pp relative shift in provision of credit toward nonbanks during the first year after adoption.
    - Effect particularly strong in market-based economies: 3pp.
  - Price-based MaPs: estimated coefficients positive but not statistically significant.

### Event-study findings — timing and magnitudes (CEGR, two-year post-event windows unless otherwise noted)
- Bank credit:
  - CEGR[0,8] = -7.70 (p-value = 0.000).
  - During the two years following MaP activation, bank credit growth is about 8pp below baseline (statistically significant with p < 1 percent).
  - Decline in bank credit begins several quarters prior to MaP activation (CEGR[-8,0] = -2.95, p-value = 0.055), suggesting pre-emptive slowing.
- Nonbank credit:
  - CEGR[0,8] = 9.90 (p-value = 0.000).
  - During the two years following a MaP event nonbank credit growth rises on average by about 10pp above baseline (p < 1 percent).
- Total credit:
  - CEGR[0,8] = -4.95 (p-value = 0.000).
  - Total credit growth declines by about 5pp below baseline during the two years following MaP adoption; rise in nonbank credit does not fully compensate decline in bank credit.
- Net sectoral credit flows:
  - CEGR[-8,0] = -0.99 (p-value = 0.310); CEGR[0,8] = -4.22 (p-value = 0.000).
  - Prior to MaP events the series is statistically indistinguishable from baseline; during two years following the event net flows move about 4pp below baseline.
- Nonbank financial intermediation specifics:
  - Investment fund assets:
    - CEGR[-8,0] = 23.18 (p-value = 0.011); CEGR[0,8] = 19.77 (p-value = 0.019).
    - Over two years following a MaP, investment fund assets rise on average by 20pp above baseline.
  - Domestic private debt issuance:
    - CEGR[-8,0] = 13.30 (p-value = 0.290); CEGR[0,8] = 54.94 (p-value = 0.000).
    - Two years after the policy event, domestic private debt issuance is 55pp above baseline.
    - Note: large CEGR magnitudes partly reflect low initial levels in some countries (e.g., investment funds in EMEs average about 6 percent of GDP).
- Subsample and instrument heterogeneity:
  - Bank credit two-year CEGRs:
    - AEs: -3.2pp (two years after MaP adoption).
    - EMEs: close to -10pp two years after adoption.
  - Quantity vs price measures on bank credit (two years after):
    - AEs: quantity measures → -6.6pp; price measures → 2.0 (not significant).
    - EMEs: quantity measures → -10.4pp; price measures → 1.5 (not significant).
  - Total credit:
    - Negative effect in both AEs and EMEs, statistically significant only in EMEs (e.g., EMEs quantity measures CEGR[0,8] = -6.88, p-value = 0.004).
  - Net sectoral credit flows:
    - Impact negative in both AEs and EMEs on average, statistically significant only for quantity-based measures.

### Summary of empirical magnitudes (event study, two years post-event)
- Bank credit: approximately -8pp (overall); -3.2pp (AEs); ~-10pp (EMEs).
- Nonbank credit: approximately +10pp.
- Total credit: approximately -5pp.
- Net sectoral flows: approximately -4pp.
- Investment fund assets: +20pp.
- Domestic private debt issuance: +55pp.
- Panel regressions: MaP quarterly coefficient -0.26 (implying ~1pp annual shift toward nonbanks); quantity-based MaP quantity index in Table 4: YtY Change in MaP Quantity-Based Index coefficient -0.49*** (ALL).

### Robustness, placebo tests, and GFC split
- Placebo tests:
  - Simulated MaP events drawn via Bernoulli indicators matching relative frequencies; event studies repeated with simulated MaP indicators.
  - Placebo impact window effects mostly statistically indistinguishable from zero, suggesting actual results unlikely driven by spurious trends.
  - Appendix Panel A (placebo) example: All Instruments / All: Bank Credit 0.9, Total Credit 1.1 (post-implementation average cumulative credit growth rates over 2-year window).
- Pre- versus post-GFC (split at 2007Q3):
  - Pre-2007Q3 quantity-based tools reduce bank credit growth in both AEs and EMEs.
  - Cross-sector substitution toward nonbanks statistically and economically significant in both AEs and EMEs, both before and after the GFC.
  - In AEs, cross-sector substitution associated with activation of quantity-based measures is larger during the pre-GFC era.
- Appendix Table A.2 highlights differences:
  - Effects prior to 2007Q3 (two-year windows): All: Bank Credit -8.8***; Total Credit -4.7**.
  - Effects after 2007Q3 (All Instruments): All: Bank Credit -3.7**; Total Credit -1.7.
  - Quantity Measures after 2007Q3: All: Bank Credit -4.8***; Total Credit -2.8**.
  - Price Measures after 2007Q3: All: Bank Credit 4.1*; Total Credit 5.8**.

### Policy implications and recommendations
- Macroprudential monitoring and policy should adopt a broad approach that goes beyond narrow national or sectoral perspectives due to cross-border and cross-sector substitution.
  - Cross-border substitution effects imply macroprudential policy should not take a narrow national perspective.
  - Cross-sector substitution effects imply macroprudential policy should not take a narrow sectoral perspective.
- Support for integrated approaches to monitor and address systemic risks across highly leveraged entities and activities, including maturity and liquidity mismatches, interconnectedness, and misaligned incentives related to too-big-to-fail.
- Uncertainty remains on whether substitution reduces or increases systemic risk:
  - Substitution could lower systemic risk if activities move to less-leveraged, less maturity-mismatched institutions.
  - Substitution could increase systemic risk if market-based intermediation contributes to procyclical leverage, amplifies price changes and flows, or features misaligned incentives.
- Recommendations:
  - Extend macroprudential policy scope beyond banking to focus on systemic risks rather than substitution per se.
  - Use activity-based instruments to target risks of an activity regardless of where it is conducted.
  - Apply instruments similar to those for banks to nonbank institutions when they perform bank-like functions (examples in text):
    - Margin requirements for securities financing transactions analogous to leverage requirements for banks and LTV limits for mortgages.
    - Limits on leverage and liquidity transformation for investment funds engaging in bank-like activities.

### Selected numerical summary statistics (Table 1 highlights, means with standard deviations in parentheses)
- Bank Credit to Private Sector (% of GDP): Advanced Economies 84.79 (36.86); Emerging Market Economies 60.41 (42.48); Whole Sample 77.83 (40.08).
- Non-Bank Credit to Private Sector (% of GDP): Advanced Economies 55.78 (41.25); Emerging Market Economies 9.32 (14.18); Whole Sample 42.49 (41.39).
- Investment fund assets to GDP (%): Advanced Economies 31.11 (68.26); Emerging Market Economies 6.38 (9.68); Whole Sample 21.74 (55.43).
- Bank Credit, YtY % Change: Advanced Economies 6.47 (11.18); Emerging Market Economies 10.53 (16.35); Whole Sample 7.65 (13.03).
- Non-Bank Credit, YtY % Change: Advanced Economies 7.33 (13.95); Emerging Market Economies 11.59 (31.87); Whole Sample 8.57 (20.90).
- Total Credit, YtY % Change: Advanced Economies 6.68 (9.98); Emerging Market Economies 9.78 (15.10); Whole Sample 7.59 (11.79).
- Net Sectoral Credit Flow, % of Total Credit: Advanced Economies 1.29 (6.36); Emerging Market Economies 5.76 (11.10); Whole Sample 2.59 (8.29).
- Overall MaP Index (Cerutti et al., 2015): Advanced Economies 1.56 (1.41); Emerging Market Economies 2.00 (1.63); Whole Sample 1.83 (1.56).
- Price-Based Regulatory Index: Advanced Economies 0.16 (0.38); Emerging Market Economies 0.25 (0.51); Whole Sample 0.22 (0.46).

*Source: Excerpt from the IMF working paper content unit titled "_wp1694 - References ____________________________________________________________ 23".*

### References ____________________________________________________________ 23

### _wp1694 - References ____________________________________________________________ 23

### Key findings on macroprudential policy effects
- In the two years after the activation of MaPs, bank credit growth falls on average by 7.7 percentage points relative to the counterfactual of no measure.
- Total credit growth falls on average by 4.9 percentage points relative to the counterfactual of no measure.
- The smaller decline in total credit relative to bank credit is explained by an increase in nonbank credit growth (substitution from bank to nonbank intermediation).
- Quantity-based measures slow bank credit growth by 8.7 percentage points over two years relative to the counterfactual of no policy change.
- Quantity-based measures have much stronger effects on credit growth than price-based measures, in both advanced and emerging market economies.
- Substitution effects (bank → nonbank) are:
  - Stronger in Advanced Economies (AEs) than in Emerging Market Economies (EMEs).
  - Stronger for quantity restrictions than for price-based measures.
- In the sample, the effect on bank credit is always substantially higher than the effect on total credit to the private sector.
- The paper finds strong and statistically significant effects on specific forms of nonbank financial intermediation, such as investment fund assets.

### Cross-country and institutional context, data coverage, and measurement
- Bank and nonbank credit measures are taken from the BIS database on private non-financial sector credit (Dembiermont et al., 2013).
- The BIS database contains quarterly series of private credit data for 40 economies for a period covering the last 40 years.
- Bank credit is defined as all loans and debt securities held by domestic and foreign banks (subsidiaries and branches).
- Nonbank credit encompasses loans and debt securities held by all other sectors of the economy (e.g., insurers, pension funds, investment funds, other firms, households, etc.) and, for some countries, direct cross-border lending by foreign banks.
- Direct cross-border lending can appear in the nonbank credit measure; for the aggregate sample of BIS reporting countries, direct cross-border lending amounts to less than 5 percent of nonbank credit.
- To supplement BIS coverage, balance-sheet data for banks and nonbank financial institutions come from the World Bank's Financial Development Database (Cihak et al., 2012). Cross-section coverage ranges from about 80 countries (investment and pension funds) to over 100 countries (banks and insurance companies). The database covers the period 1980–2012 for different series.
- Panel A of Table 1 (summary statistics for 1997–2014 / 1997–2012) reports:
  - Banks represent 85 percent of GDP in AEs and 60 percent of GDP in EMEs.
  - Nonbank credit represents 56 percent of GDP in AEs and 9 percent of GDP in EMEs.
  - Investment fund (IF) assets in AEs represent close to a third of GDP, about five times the size as in EMEs.
- Nominal credit growth (year-to-year percentage changes in nominal stock of sectoral credit) is on average higher in EMEs than in AEs:
  - In EMEs, bank and nonbank credit grew by 10.5 percent and 11.6 percent, respectively.
  - In AEs, bank and nonbank credit grew by (value truncated in source content provided).

### Methodology and scope
- The paper investigates whether macroprudential policies lead to substitution from bank-based financial intermediation to nonbank intermediation.
- Uses an event study methodology to analyze timing of effects on bank and nonbank intermediation around activation dates.
- Distinguishes between:
  - Quantity versus price-based instruments.
  - Lender (bank)- versus borrower-based instruments.
  - Advanced Economies (AEs) versus Emerging Market Economies (EMEs).
  - Bank- versus market-based financial systems.
- Empirical framework controls for macroeconomic fundamentals to filter out policy effects in a cross-country panel setting.
- The study uses both net flow measures and event study methodology to assess size and timing of cross-sector substitution effects.

### Comparisons with related empirical findings
- Results are of the same order of magnitude as Morgan et al. (2015): economies with LTV policies (classified here as a quantity constraint) experienced residential mortgage loan growth of 6.7 percent per year, while non-LTV economies experienced 14.6 percent per year.
- Results are of the same order of magnitude as Cerutti et al. (2015) for bank credit, including stronger effects in EMEs than in AEs.
- Prior literature documents cross-border leakages and substitution:
  - Aiyar et al. (2014) and Reinhardt and Sowerbutts (2015) find foreign borrowing increases after domestic macroprudential measures affecting domestic banks’ capital.
  - Cerutti et al. (2015) find evidence of greater cross-border borrowing after macroprudential measures.
  - An IMF (2014a) study finds more stringent capital requirements are associated with stronger growth of shadow banking.

### Policy implications and interpretation
- Findings underline the relevance of a broad macroprudential monitoring and policy approach that goes beyond narrow national or sectoral perspectives:
  - Cross-border substitution effects imply macroprudential policy should not take a narrow national perspective.
  - Cross-sector substitution effects imply macroprudential policy should not take a narrow sectoral perspective.
- Results support integrated approaches to monitoring and addressing systemic risks across highly leveraged entities and activities, including maturity and liquidity mismatches, interconnectedness, and misaligned incentives related to too-big-to-fail.
- The paper notes uncertainty about whether substitution reduces or increases systemic risk:
  - Substitution could lower systemic risk if activities move to less-leveraged, less maturity-mismatched institutions.
  - Alternatively, substitution could increase systemic risk if market-based intermediation contributes to procyclical leverage, amplifies price changes and flows, or features misaligned incentives.

### Paper organization (as provided)
- Section II describes the data.
- Section III investigates substitution between bank and nonbank credit by estimating whether MaPs affect flows of bank and nonbank credit as a percentage of total credit.
- Section IV presents event study estimates of MaP effects on sectoral variables and balance-sheet data of nonbank entities, distinguishing instrument types and country groups.
- Section V presents robustness checks.
- Section VI concludes.

*Source: Excerpt from the IMF working paper content unit titled "_wp1694 - References ____________________________________________________________ 23".*

### 6.5 percent and 7.4 percent. Total credit grew at an average annual rate of 6.7 percent in AEs

### _wp1694 - 6.5 percent and 7.4 percent. Total credit grew at an average annual rate of 6.7 percent in AEs

### Measurement of cross‑sector substitution
- Total credit growth:
  - 6.5 percent and 7.4 percent (reported series).
  - Total credit grew at an average annual rate of 6.7 percent in AEs and 9.8 percent in EMEs.
- Net sectoral credit flow (quarterly, scaled by total credit):
  - Defined as the difference between the quarterly change in bank credit and the quarterly change in nonbank credit, scaled by total credit.
  - Positive values indicate bank credit growth outpacing nonbank credit; negative values indicate faster nonbank credit growth.
- Data processing:
  - All credit-related variables are winsorized at the 1 percent level for each tail.
  - Observations for Argentina are excluded.
- Distributional properties:
  - The histogram of the net credit flow measure is positively skewed, with the mean and the mode slightly above 0.
- Caveat:
  - The net sectoral credit flow measure does not reveal whether shifts are driven by banks, nonbanks, or both; direct effects on bank and nonbank credit are estimated elsewhere.

### Macroprudential policy (MaP) events and classification
- Source and construction:
  - MaP indicators from Cerutti et al. (2015), based on the IMF GMPI survey, covering 120 countries over 2000–13.
  - 12 categories of MaPs yield 12 dummy variables (1 when used in a given year, 0 otherwise).
- Classification and prevalence:
  - Cerutti et al. (2015) classify MaPs as lender‑based or borrower‑based; authors further classify MaPs into price‑based and quantity‑based measures.
  - Examples:
    - Price-based: dynamic provisioning requirements, taxes on financial institutions.
    - Quantity-based: limits on interbank and foreign currency exposures, leverage ratios treated as quantity-based in main specification.
  - On average, AEs have 1.6 MaPs in place in a given year; EMEs have 2 MaPs.
  - Total MaP events in the dataset: 171 MaPs, 77 percent of which are quantity-based.
  - Most events cluster during 2007–13; prior to that MaPs were mostly activated in EMEs.
- Targeting:
  - Inspection of GMPI qualitative answers indicates MaPs are primarily aimed at depository institutions (banks), including borrower-based measures.
- MaP events definition:
  - Defined as the adoption of a new measure (policy tightening); events when a dummy goes from 0 to 1.

### Identification strategy and controls (methodological framework)
- Main outcome: net sectoral credit flow (quarterly net difference bank minus nonbank credit scaled by total credit).
- Primary regression specification:
  - NetFlow_{c,t} = α_c + β_t + θ_1 BankCrisis_{c,t} + θ_2 MonetaryPolicyΔ_{c,t} + θ_3 MaPΔ_{c,t} + ε_{c,t}
  - Country and time fixed effects included; standard errors clustered by country.
- Controls to address identification concerns:
  - Banking crises: include Laeven and Valencia (2013) systemic banking crisis indicator.
  - Monetary policy: include year-on-year changes in policy rate and year-to-year changes in central bank balance sheet size relative to GDP.
  - Lagged macroeconomic covariates (in event‑study excess growth regressions): GDP growth, current account balance, gross capital inflows, central bank interest rates, growth in central bank assets.
- Event‑study (leads-and-lags) approach:
  - Compute excessive growth rates E[ŷ] = y - E[y] where E[y] is predicted by covariates x, country fixed effects μ_c, and time fixed effects α_t.
  - Estimate coefficients φ_i for i periods before/after MaP activation; compute cumulative excess growth rates (CEGR) for windows up to three years before and after.
  - Wald tests performed on CEGRs (two years prior and two years after activation) to test anticipation and post-activation effects.
- Robustness:
  - Alternative price/quantity classifications tested in online appendix (results not reproduced here).

### Cross‑sector substitution: panel regression results
- MaP effect on net sectoral credit flows:
  - MaP coefficients are negative across regressions, consistent with the boundary hypothesis: MaP activation shifts provision of credit toward nonbanks.
  - Overall sample coefficient: -0.26 (quarterly coefficient), implying during the first year following MaP activation net sectoral credit flows move by about 1 percentage point (pp) of total credit in favor of nonbanks (coefficient × 4 = annual change).
  - Statistical significance: coefficient significant in most regressions except for the EME subsample.
- Banking crisis effect:
  - Banking crisis coefficients negative and statistically significant.
  - Banking crises reduce net credit flows from banks by about 5pp of total credit per annum on average; in EMEs the effect is about 10pp per annum.
- Monetary policy and central bank balance sheets:
  - Relationship with policy rates: ambiguous (positive in AEs, negative in EMEs, insignificant in the pooled sample).
  - Central bank balance sheet expansion: negatively related to net credit flows in most specifications.
    - In the overall sample, a 1 percent increase in central bank assets is associated with a 2pp per annum shift in net credit flows originating from banks (i.e., shift toward nonbanks).
    - Association particularly strong in AEs and bank‑based financial systems.
- Price vs quantity MaPs:
  - Quantity-based MaPs: estimated coefficients negative and statistically significant.
    - In the overall sample, quantity measures suggest a 2pp relative shift in provision of credit toward nonbanks during the first year after adoption.
    - Effect particularly strong in market-based economies: 3pp.
  - Price-based MaPs: estimated coefficients positive but not statistically significant.
- Summary conclusion from regressions:
  - Evidence broadly consistent with the boundary hypothesis: MaP adoption directed at banks induces a relative shift from bank to nonbank credit, especially for quantity-based measures.

### Event‑study findings (timing and magnitudes)
- Bank credit (CEGR):
  - During the two years following MaP activation, bank credit growth is about 8pp below the baseline (statistically significant with p < 1 percent).
  - Decline in bank credit begins several quarters prior to MaP activation, suggesting pre-emptive slowing.
- Nonbank credit (CEGR):
  - During the two years following a MaP event nonbank credit growth rises on average by about 10pp above baseline (statistically significant at 1 percent).
  - Pattern is almost the mirror image of bank credit CEGR.
- Total credit (CEGR):
  - Total credit growth declines by about 5pp below baseline during the two years following MaP adoption.
  - Interpretation: rise in nonbank credit does not fully compensate the decline in bank credit.
- Net sectoral credit flows (CEGR):
  - Prior to MaP events the series is statistically indistinguishable from baseline (p = 0.31).
  - During the two years following the event, net sectoral credit flows move about 4pp below baseline (denominated in terms of total credit).
  - Absence of pre-existing trend supports causal interpretation of MaP impact.
- Effects on specific nonbank finance sources:
  - Investment fund assets:
    - CEGR begins to pick up 6 quarters prior to activation, accelerates around activation, decelerates thereafter.
    - Over two years following a MaP, investment fund assets rise on average by 20pp above baseline (p-value = 0.019).
  - Domestic private debt issuance:
    - Exhibits strong positive growth up to 6 quarters following MaP activation (p = 0.00).
    - Two years after the policy event, domestic private debt issuance is 55pp above baseline.
    - Note: large CEGR magnitudes partly reflect low initial levels of these series in some countries (e.g., investment funds in EMEs average about 6 percent of GDP).
- Subsample and instrument heterogeneity (two‑year post‑event windows):
  - AEs vs EMEs — bank credit:
    - AEs: bank credit slows by 3.2pp below baseline two years after MaP adoption.
    - EMEs: slowdown close to 10pp below baseline two years after adoption.
  - Quantity vs price measures:
    - Quantity-based measures drive most of the decline in bank credit in both AEs and EMEs.
      - AEs: quantity measures lead to a 6.6pp contraction below baseline two years after event.
      - EMEs: quantity measures lead to a 10.4pp contraction below baseline two years after event.
    - Price-based measures show no statistically distinguishable impact on banks in AEs or EMEs.
  - Total credit:
    - Negative effect in both AEs and EMEs, statistically significant only in EMEs.
    - Difference between bank credit and total credit effects is larger in AEs (especially for quantity constraints), possibly due to greater substitution opportunities in more developed financial systems.
  - Net sectoral credit flows:
    - Impact of MaPs is statistically negative in both AEs and EMEs on average, but statistically significant only for quantity-based measures.

### Summary of empirical conclusions
- Activation of MaPs directed at banks is associated with:
  - A relative shift in credit provision from banks to nonbanks (consistent with the boundary hypothesis).
  - The substitution effect is especially pronounced for quantity‑based MaP measures.
  - Price‑based MaP measures do not show detectable substitution effects in the main findings.
- Magnitudes from event study (two years post‑event):
  - Bank credit: approximately -8pp (overall), -3.2pp (AEs), ~-10pp (EMEs).
  - Nonbank credit: approximately +10pp.
  - Total credit: approximately -5pp.
  - Net sectoral flows: approximately -4pp.
  - Investment fund assets: +20pp.
  - Domestic private debt issuance: +55pp.
- Policy implication:
  - Quantity-based constraints on banks are more likely to induce cross‑sector leakages toward nonbank finance than price-based measures; regulators should consider substitution channels when designing MaPs.

*Source: _wp1694 - 6.5 percent and 7.4 percent. Total credit grew at an average annual rate of 6.7 percent in AEs (IMF working paper content provided).*

### 1. MaP measures tend to slow the growth rate of bank credit.

### 1. MaP measures tend to slow the growth rate of bank credit.

### Key empirical findings
- 1. MaP measures tend to slow the growth rate of bank credit.
- 2. MaP measures tend to increase the growth rate of nonbank credit.
- 3. MaP measures tend to reduce the net sectoral credit flow (i.e., stimulate cross-sector substitution to nonbank credit).
- 4. MaP measures tend to reduce the growth rate of total credit (i.e., substitution effects do not fully compensate the impact on bank credit).
- 5. Substitution effects are stronger in AEs than in EMEs.
- 6. The effects of MaPs are stronger when the measures directly constrain credit.
- On average, bank credit falls by almost 8 percentage points in the two years following the adoption of macroprudential policy measures.
- Cross-sector substitution is particularly pronounced following the adoption of quantity-based MaPs, in advanced economies and bank-based financial systems.
- The growth of investment funds and of capital market debt issuance following macroprudential measures illustrates how these measures are offset by new forms of credit growth outside the banking sector.

### Robustness checks — placebo tests
- Placebo testing procedure:
  - For each country-year observation in the original Cerutti et al. (2015) dataset draw a Bernoulli distributed indicator to simulate placebo dates for each of the 12 MaP tools.
  - Match the distribution of simulated and actual MaP events by setting the Bernoulli probability parameter to the relative frequency of the corresponding measure in the original dataset.
  - Compute aggregate MaP indices from simulated MaP and derive MaP event indicators; repeat the event study from Section IV using simulated MaP indicators.
- Results:
  - Impact window effects in most cases are statistically indistinguishable from zero when using simulated MaP events.
  - Interpretation: results reported for actual MaP activations are unlikely to be driven by spurious trends in the data.

### Robustness checks — pre- and post-global financial crisis (GFC)
- Method:
  - Repeat event studies from Section IV separately for periods before and after the onset of the GFC, taken to be the third quarter of 2007.
  - Because of lack of price-based MaP events prior to the GFC, report only quantity-based measures for the pre-GFC period.
- Results:
  - Quantity-based tools during the pre-GFC years are found to reduce bank credit growth in both AEs and EMEs.
  - Cross-sector substitution towards nonbanks is statistically and economically significant in both AEs and EMEs, both before and after the GFC.
  - In AEs, the cross-sector substitution associated with activation of quantity-based measures is larger during the pre-GFC era.

### Interpretation of substitution effects and systemic implications
- Factors that temper concern about cross-sector substitution:
  - Nonbank financial institutions are generally less leveraged and have less liquidity risks than the banking sector.
  - Nonbank institutions are separated from systemic functions related to the payments infrastructure.
  - The nonbank financial sector generally does not have access to public sector safety nets, such as deposit insurance and central bank liquidity support; a shift to market-based financing can act as a “spare tire” in the supply of credit in times of systemic banking crises.
- Potential new systemic risks from substitution:
  - Credit bubbles shifting from banks to financial markets may leave macroeconomic vulnerabilities if households or corporates continue to accumulate debt owed to nonbanks.
  - Investment funds purchasing illiquid debt securities while promising liquidity to end investors create refinancing and sudden price shock risks.
  - Interconnections between nonbank financial sector and formal banking sector (credit lines, participation in banks’ debt issuances, ownership links) can transmit shocks between sectors.

### Policy implications and recommendations
- Extend macroprudential policy scope beyond banking to focus on systemic risks rather than substitution per se.
- Use activity-based instruments to target risks of an activity regardless of where it is conducted.
- Apply instruments similar to those for banks to nonbank institutions when they perform bank-like functions:
  - Margin requirements for securities financing transactions may perform a similar function as leverage requirements for banks and LTV limits for mortgages.
  - Limits on leverage and liquidity transformation can ensure that investment funds engaging in bank-like activities and taking on bank-like risks face comparable requirements.

*Source: _wp1694 - 1. MaP measures tend to slow the growth rate of bank credit.*

### REFERENCES

### REFERENCES

### Bibliographic entries
- Adrian, Tobias. 2014. “Financial Stability Policies for Shadow Banking.” Federal Reserve Bank of New York Staff Report No. 664.
- Adrian, Tobias. and Hyun Song Shin. 2009. “Money, Liquidity and Monetary Policy.” American Economic Review: Papers & Proceedings 99 (2): 600–605.
- Adrian, Tobias. and Hyun Song Shin. 2010. “Liquidity and Leverage.” Journal of Financial Intermediation 19: 418–43.
- Aiyar, Shekhar, Charles W. Calomiris and Tomasz Wieladek. 2014. “Does Macro-Prudential Regulation Leak? Evidence from a U.K. Policy Experiment.” Journal of Money, Credit and Banking 46(s1): 181–214.
- Akinci, Ozge and Jane Olmstead-Rumsey. 2015. “How Effective Are Macroprudential Policies? An Empirical Investigation.” Board of Governors of the Federal Reserve System, International Finance Discussion Paper No. 1136.
- Arregui, Nicolas, Jaromír Beneš, Ivo Krznar, Srobona Mitra and Andre Oliveira Santos. 2013. “Evaluating the Net Benefits of Macroprudential Policy: A Cookbook.” IMF Working Paper No. 13/167, International Monetary Fund, Washington.
- Atanasov, Vladimir A. and Black, Bernard S. 2016. “Shock-Based Causal Inference in Corporate Finance and Accounting Research.” Critical Finance Review 6.
- Azis, Iwan J. and Hyun Song Shin. 2015. Managing Elevated Risk Global Liquidity, Capital Flows, and Macroprudential Policy—An Asian Perspective. Singapore: Springer-Verlag.
- Bruno, Valentina and Hyun Song Shin. 2014. “Assessing Macroprudential Policies: Case of South Korea.” The Scandinavian Journal of Economics, 116(1): 128–157.
- Buch, Claudia M., Sandra Eickmeier and Esteban Prieto. 2014. “In search for yield? Survey-based evidence on bank risk taking.” Journal of Economic Dynamics and Control, 43: 12–30.
- Berger, Allen and Gregory F. Udell. 1992. “Some Evidence on the Empirical Significance of Credit Rationing.” Journal of Political Economy, 100(5): 1047–107.
- Cerutti, Eugenio, Stijn Claessens, and Luc Laeven. 2015. “The Use and Effectiveness of Macroprudential Policies.” IMF Working Paper No 15/61, International Monetary Fund, Washington.
- Cihak, Martin, Asli Demirgüç-Kunt, Erik Feyen, and Ross Levine. 2012. “Benchmarking Financial Systems Around the World.” World Bank Policy Research Working Paper No. 6175, World Bank, Washington.
- Claessens, Stijn, Swati R. Ghosh, and Roxana Mihet. 2013. “Macro-Prudential Policies to Mitigate Financial System Vulnerabilities.” Journal of International Money and Finance 39: 153–85.
- Clement, Piet. 2010. “The Term “Macroprudential: Origins and Evolution.” BIS Quarterly Review, March, Bank for International Settlements, Basel.
- Dembiermont, Christian, Mathias Drehmann, and Siriporn Muksakunratana. 2013. “How Much Does the Private Sector Really Borrow? A New Database for Total Credit to the Private Non-Financial Sector.” BIS Quarterly Review, March, Bank for International Settlements, Basel.
- European Commission (EC). 2015. “Action Plan on Building a Capital Markets Union.” Com 468 Final, Brussels.
- European Systemic Risk Board (ESRB). 2013. “Recommendation on Intermediate Objectives and Instruments of Macroprudential Policy.” April, Frankfurt.
- European Systemic Risk Board (ESRB). 2014. “The ESRB Handbook on Operationalising Macroprudential Policy in the Banking Sector.” March, Frankfurt.
- European Systemic Risk Board (ESRB). 2015. “A Review of Macroprudential Policy in the EU One Year After the Introduction of the CRD/CRR.” June, Frankfurt.
- Feroli, Michael, Anil K. Kashyap, Kermit L. Schoenholtz and Hyun Song Shin. 2014. “Market Tantrums and Monetary Policy.” Chicago Booth Research Paper No. 14–09.
- Financial Stability Board (FSB). 2014. “Global Shadow Banking Monitoring Report 2014.” Basel.
- Frost, Jon and Ruben van Tilburg. 2014. “Financial Globalization or Great Financial Expansion? The Impact of Capital Flows on Credit and Banking Crises.” DNB Working Paper No. 441, De Nederlandsche Bank, Amsterdam.
- Galati, Gabriele, and Richhild Moessner. 2013. “Macroprudential Policy—A Literature Review.” Journal of Economic Surveys 27 (5): 846–78.
- Galati, Gabriele, and Richhild Moessner. 2014. “What Do We Know About the Effects of Macroprudential Policy?” DNB Working Paper No. 440, De Nederlandsche Bank, Amsterdam.
- Gambacorta, Leonardo, Jing Yang and Kostas Tsatsaronis. 2014. “Financial Structure and Growth.” BIS Quarterly Review, March, Bank for International Settlements, Basel.
- Gertler, Mark and Simon Gilchrist. 1991. “Monetary Policy, Business Cycles and the Behavior of Small Manufacturing Firms.” NBER Working Paper No. 3892, National Bureau of Economic Research, Cambridge.
- Goodhart, Charles. 2008. “The Boundary Problem in Financial Regulation.” National Institute Economic Review, 206(1): 48–55.
- Igan, Deniz Igan and Heedon Kang. 2011. “Do Loan-to-Value and Debt-to-Income Limits Work? Evidence from Korea.” IMF Working Paper No. 11/297, International Monetary Fund, Washington.
- International Monetary Fund (IMF). 2013. Global Macroprudential Policy Instruments Database, 2013, Washington.
- International Monetary Fund (IMF). 2014a. “Shadow Banking Around the Globe: How Large and How Risky?” In Global Financial Stability Report, International Monetary Fund, Washington.
- International Monetary Fund (IMF). 2014b. “Staff Guidance Note on Macroprudential Policy.” IMF Policy Paper, International Monetary Fund, Washington.
- International Monetary Fund (IMF). 2015. “The Asset Management Industry and Financial Stability.” In Global Financial Stability Report, International Monetary Fund, Washington.
- Jiménez, Gabriel, Steven Ongena, José-Luis Peydró and Jesús Saurina. 2012. “Macroprudential Policy, Countercyclical Bank Capital Buffers and Credit Supply: Evidence from the Spanish Dynamic Provisioning Experiments.” Economics Working Paper No. 1315, Universitat Pompeu Fabra.
- Kashyap, Anil K., Jeremy C. Stein and David W. Wilcox. 1993. “Monetary Policy and Credit Conditions: Evidence from the Composition of External Finance.” American Economic Review, 83(1): 78–93.
- Krznar, Ivo and James Morsink. 2014. “With Great Power Comes Great Responsibility: Macroprudential Tools at Work in Canada.” IMF Working Paper No. 14/83, International Monetary Fund: Washington.
- Laeven, Luc, and Fabián Valencia. 2013. “Systemic Banking Crises Database.” IMF Economic Review 61 (2): 225–70.
- Lim, C., F. Columba, A. Costa, P. Kongsamut, A. Otani, M. Saiyid, T. Wezel, and X. Wu. 2011. “Macroprudential Policy: What Instruments and How to Use Them? Lessons from Country Experiences.” IMF Working Paper No. 11/238, International Monetary Fund, Washington.
- McDonald, Chris. 2015. “When Is Macroprudential Policy Effective?” BIS Working Paper No. 496, Bank for International Settlements, Basel.
- Morgan, Peter J., Paulo José Regis, and Nimesh Salike. 2015. “Loan-to-Value Policy as a Macroprudential Tool: The Case of Residential Mortgage Loans in Asia.” ADBI Working Paper, No. 528, Asian Development Bank Institute, Tokyo.
- Pattipeilohy, Christiaan, Jan Willem van den End, Mostafa Tabbae, Jon Frost and Jakob De Haan. 2013. “Unconventional Monetary Policy of the ECB during the Financial Crisis: an Assessment and New Evidence.” in: Morten Balling and Ernest Gnan (eds.). 50 Years of Money and Finance: Lessons and Challenges. Vienna: SUERF.
- Rajan, Raghuram. 2006. “Has Finance Made the World Riskier?” European Financial Management, 12(4): 499–533.
- Reinhardt, Dennis and Rhiannon Sowerbutts. 2015. “Regulatory Arbitrage in Action: Evidence from Banking Flows and Macroprudential Policy.” Bank of England Working Paper No. 546, London.
- Schoenmaker, Dirk and Peter Wierts. 2015. “Regulating the Financial Cycle: An Integrated Approach with a Leverage Ratio.” Economics Letters, 136: 70–72.

### Figures and tables referenced (selected numeric results, charts and notes)
- Figures 1A–1C and 1B: Growth series plotted with LOESS smoothing parameter set to 0.1; 95 percent confidence interval shown. Axes labeled "Year-on-year Growth in Bank Credit (in %, quarterly)" and "Year-on-year Growth in Non-Bank Credit (in %, quarterly)".
- Figure 2: Histogram of substitution measure defined as: 11 [Bank Credit ][Non-Bank Credit ] 44 [Quarterly Net Sectoral Credit Flow] 100* [Total Credit ] ,4 YtYYtY ctct ct ct     . (Actual axis ticks shown between 0.00 and 0.35 and substitution bins between -10 and 6).
- Figure 3: Activation of Macroprudential Policy Measures (2000–2013). Source: Cerutti et al. (2015). Blank spaces refer to no new MaPs in a country in a given year.
- Figure 4 (Cumulative Excess Credit Growth Rates Around Macroprudential Policy Measures):
  - 4A. Bank Credit: CEGR[-8,0] = -2.95 (p-value = 0.055); CEGR[0,8] = -7.70 (p-value = 0.000).
  - 4B. Nonbank Credit: CEGR[-8,0] = 2.03 (p-value = 0.436); CEGR[0,8] = 9.90 (p-value = 0.000).
  - 4C. Total Credit: CEGR[-8,0] = -2.43 (p-value = 0.060); CEGR[0,8] = -4.95 (p-value = 0.000).
  - 4D. Net Sectoral Credit Flows: CEGR[-8,0] = -0.99 (p-value = 0.310); CEGR[0,8] = -4.22 (p-value = 0.000).
- Figure 5 (Cumulative Excess Growth of Nonbank FIs' Assets Around Macroprudential Policy Measures):
  - A. Investment Fund Asset Growth: CEGR[-8,0] = 23.18 (p-value = 0.011); CEGR[0,8] = 19.77 (p-value = 0.019).
  - B. Domestic Private Debt Issuance: CEGR[-8,0] = 13.30 (p-value = 0.290); CEGR[0,8] = 54.94 (p-value = 0.000).
- Figure 6 (Impact of Macroprudential Policy Measures on Bank Credit):
  - Advanced Economies, Quantitative Measures: CEGR[-8,0] = 1.57 (p-value = 0.344); CEGR[0,8] = -6.65 (p-value = 0.000).
  - Advanced Economies, Price Measures: CEGR[-8,0] = 3.15 (p-value = 0.212); CEGR[0,8] = 2.08 (p-value = 0.414).
  - Emerging Market Economies, Quantitative Measures: CEGR[-8,0] = 0.36 (p-value = 0.891); CEGR[0,8] = -10.36 (p-value = 0.000).
  - Emerging Market Economies, Price Measures: CEGR[-8,0] = -0.06 (p-value = 0.978); CEGR[0,8] = 1.47 (p-value = 0.508).
- Figure 7 (Impact on Nonbank Credit):
  - Advanced Economies, Quantitative Measures: CEGR[-8,0] = 4.79 (p-value = 0.263); CEGR[0,8] = 7.81 (p-value = 0.056).
  - Advanced Economies, Price Measures: CEGR[-8,0] = 0.87 (p-value = 0.801); CEGR[0,8] = 6.97 (p-value = 0.046).
  - Emerging Market Economies, Quantitative Measures: CEGR[-8,0] = 8.54 (p-value = 0.292); CEGR[0,8] = 3.72 (p-value = 0.661).
  - Emerging Market Economies, Price Measures: CEGR[-8,0] = 22.55 (p-value = 0.086); CEGR[0,8] = 23.09 (p-value = 0.079).
- Figure 8 (Impact on Total Credit):
  - Advanced Economies, Quantitative Measures: CEGR[-8,0] = -0.91 (p-value = 0.577); CEGR[0,8] = -1.51 (p-value = 0.358).
  - Advanced Economies, Price Measures: CEGR[-8,0] = 0.71 (p-value = 0.756); CEGR[0,8] = 2.12 (p-value = 0.358).
  - Emerging Market Economies, Quantitative Measures: CEGR[-8,0] = 0.25 (p-value = 0.915); CEGR[0,8] = -6.88 (p-value = 0.004).
  - Emerging Market Economies, Price Measures: CEGR[-8,0] = -1.22 (p-value = 0.843); CEGR[0,8] = -2.77 (p-value = 0.652).
- Figure 9 (Impact on Net Sectoral Credit Flows):
  - Advanced Economies, Quantity Measures: CEGR[-8,0] = 0.03 (p-value = 0.978); CEGR[0,8] = -4.59 (p-value = 0.000).
  - Advanced Economies, Price Measures: CEGR[-8,0] = 2.91 (p-value = 0.116); CEGR[0,8] = -1.15 (p-value = 0.539).
  - Emerging Market Economies, Quantity Measures: CEGR[-8,0] = -1.43 (p-value = 0.417); CEGR[0,8] = -6.48 (p-value = 0.000).
  - Emerging Market Economies, Price Measures: CEGR[-8,0] = -1.05 (p-value = 0.730); CEGR[0,8] = 1.11 (p-value = 0.715).

### Selected table figures and regression coefficients (exact values)
- Table 1 (Summary statistics, mean values; standard deviations in parentheses; 1997–2014, Quarterly):
  - Bank Credit to Private Sector (% of GDP), Source: BIS: Advanced Economies 84.79 (36.86); Emerging Market Economies 60.41 (42.48); Whole Sample 77.83 (40.08).
  - Non-Bank Credit to Private Sector (% of GDP), Source: BIS: Advanced Economies 55.78 (41.25); Emerging Market Economies 9.32 (14.18); Whole Sample 42.49 (41.39).
  - Investment fund assets to GDP (%), Source: WB-GFDD: Advanced Economies 31.11 (68.26); Emerging Market Economies 6.38 (9.68); Whole Sample 21.74 (55.43).
  - Bank Credit, YtY % Change, Source: BIS: Advanced Economies 6.47 (11.18); Emerging Market Economies 10.53 (16.35); Whole Sample 7.65 (13.03).
  - Non-Bank Credit, YtY % Change, Source: BIS: Advanced Economies 7.33 (13.95); Emerging Market Economies 11.59 (31.87); Whole Sample 8.57 (20.90).
  - Total Credit, YtY % Change, Source: BIS: Advanced Economies 6.68 (9.98); Emerging Market Economies 9.78 (15.10); Whole Sample 7.59 (11.79).
  - Net Sectoral Credit Flow, % of Total Credit, Source: BIS: Advanced Economies 1.29 (6.36); Emerging Market Economies 5.76 (11.10); Whole Sample 2.59 (8.29).
  - Overall Index, Source: Cerutti et al. (2015): Advanced Economies 1.56 (1.41); Emerging Market Economies 2.00 (1.63); Whole Sample 1.83 (1.56).
  - Quantity-Based Regulatory Index, Source: Cerutti et al. (2015): Advanced Economies 1.57 (1.43); Emerging Market Economies 1.96 (1.52); Whole Sample 1.81 (1.50).
  - Price-Based Regulatory Index, Source: Cerutti et al. (2015): Advanced Economies 0.16 (0.38); Emerging Market Economies 0.25 (0.51); Whole Sample 0.22 (0.46).
  - Inflation, average consumer prices, Source: IMF-WEO: Advanced Economies 2.91 (2.75); Emerging Market Economies 7.27 (7.45); Whole Sample 6.00 (6.74).
  - YtY Real % Growth in GDP, Source: IMF-IFS: Advanced Economies 3.58 (7.48); Emerging Market Economies 5.59 (7.78); Whole Sample 5.00 (7.75).
  - Current account balance, Source: IMF-WEO: Advanced Economies 1.90 (11.53); Emerging Market Economies -5.03 (10.35); Whole Sample -3.04 (11.15).
  - General government net lending/borrowing, Source: IMF-WEO: Advanced Economies -0.10 (7.28); Emerging Market Economies -2.14 (5.56); Whole Sample -1.54 (6.18).
  - Equity inflows, % of GDP, Source: IMF-IFS: Advanced Economies 6.17 (11.56); Emerging Market Economies 1.43 (4.81); Whole Sample 2.79 (7.71).
  - Debt inflows, % of GDP, Source: IMF-IFS: Advanced Economies 11.78 (33.73); Emerging Market Economies 1.52 (8.11); Whole Sample 4.47 (19.89).
  - CB Lending Rate (in %), Source: IMF-IFS: Advanced Economies 7.26 (4.65); Emerging Market Economies 16.73 (9.93); Whole Sample 13.82 (9.70).
  - YtY % Growth in CB Assets, Source: WB-GFDD: Advanced Economies 8.97 (47.49); Emerging Market Economies 13.00 (43.89); Whole Sample 11.81 (45.02).
  - GDP per capita, current prices, Source: IMF-WEO: Advanced Economies 29,042 (17,691); Emerging Market Economies 2,943 (2,951); Whole Sample 31,043 (15,342).
  - Banking crisis dummy (1=banking crisis, 0=none), Source: WB-GFDD: Advanced Economies 0.19 (0.39); Emerging Market Economies 0.03 (0.16); Whole Sample 0.07 (0.26).

- Table 2 (Classification of Macroprudential Policies; Number of Events; Price/Quantity):
  - LTV Loan-to-Value Ratio: 32 [18] Borrower Quantity.
  - DTI Debt-to-Income Ratio: 23 [13] Borrower Quantity.
  - DP Time-Varying/Dynamic Loan-Loss Provisioning: 10 [5] Lender Price.
  - CTC General Countercyclical Capital Buffer/Requirement: 6 [3] Lender Price.
  - LEV Leverage Ratio: 13 [7] Lender Quantity.
  - SIFI Capital Surcharges on SIFIs: 7 [4] Lender Price.
  - INTER Limits on Interbank Exposures: 16 [9] Lender Quantity.
  - CONC Concentration Limits: 22 [12] Lender Quantity.
  - FC Limits on Foreign Currency Loans: 15 [8] Lender Quantity.
  - RR Reserve Requirement Ratios: 12 [7] Lender Quantity.
  - CG Limits on Domestic Currency Loans: 7 [4] Lender Quantity.
  - TAX Levy/Tax on Financial Institutions: 17 [9] Lender Price.

- Table 3 (Substitution from Bank to Nonbank Credit; selected coefficients):
  - Banking Crisis Dummy: -1.18*** (0.11) in ALL; -0.59*** (0.09) in AE; -2.48*** (0.36) in EM; -0.55*** (0.20) Market-Based; -1.37*** (0.14) Bank-Based.
  - YtY Change in MaP Index: -0.26** (0.13) in ALL; -0.27*** (0.10) in AE; -0.30 (0.29) in EM; -0.34* (0.21) Market-Based; -0.24* (0.16) Bank-Based.
  - YtY Change in CB Lending Rate: 0.00 (0.01) in ALL; 0.04*** (0.01) in AE; -0.09*** (0.02) in EM; 0.03*** (0.01) Market-Based; -0.08*** (0.02) Bank-Based.
  - YtY Change in Log of CB BS Size: -0.46*** (0.11) in ALL; -0.68*** (0.08) in AE; 0.54* (0.35) in EM; -0.08 (0.18) Market-Based; -0.58*** (0.14) Bank-Based.
  - Constant: 0.61*** (0.23) in ALL; 0.63*** (0.16) in AE; -0.15 (0.83) in EM; 0.26 (0.33) Market-Based; 0.78** (0.31) Bank-Based.
  - R-squared: 0.133 (ALL); 0.123 (AE); 0.289 (EM); 0.154 (Market-Based); 0.153 (Bank-Based).
  - Observations: 322 (ALL); 242 (AE); 29 (EM); 193 (Market-Based); 106 (Bank-Based).
  - # of Countries: 312 (ALL); 291 (AE); 10 (EM); 21 (Market-Based); 63 (Bank-Based).
  - Dependent variable: Net Bank/NonBank Credit Flow.

- Table 4 (Quantity vs Price measures; selected coefficients):
  - Banking Crisis Dummy: -1.20*** (0.11) in ALL; -0.61*** (0.09) in AE; -2.51*** (0.36) in EM; -0.59*** (0.20) Market-Based; -1.40*** (0.14) Bank-Based.
  - YtY Change in MaP Quantity-Based Index: -0.49*** (0.15) in ALL; -0.45*** (0.12) in AE; -0.38 (0.34) in EM; -0.69*** (0.25) Market-Based; -0.42** (0.18) Bank-Based.
  - YtY Change in MaP Price-Based Index: 0.56** (0.26) in ALL; 0.12 (0.20) in AE; 0.81 (0.74) in EM; 0.36 (0.42) Market-Based; 0.71** (0.34) Bank-Based.
  - Other coefficients and statistics as in Table 3 with R-squared: 0.137 (ALL); 0.125 (AE); 0.292 (EM); 0.158 (Market-Based); 0.158 (Bank-Based).
  - Observations and # of countries same as Table 3.

- Table 5 (Drivers of Growth in Credit, Investment Fund Assets and Domestic Private Debt Securities; selected coefficients and significance):
  - Dependent variables columns: (1) 1-Year Growth in Bank Credit to Private Sector; (2) 1-Year Growth in NonBank Credit to Private Sector; (3) 1-Year Growth in Total Credit to Private Sector; (4) 1-Year Growth in Investment Fund Assets; (5) Issuance of Domestic Private Debt Securities.
  - YtY % Real GDP growth: 0.85*** (0.14) in (1); 1.05*** (0.22) in (2); 0.76*** (0.12) in (3); 0.63*** (0.23) in (4); 0.31** (0.16) in (5).
  - Banking crisis dummy: -5.73*** (1.73) in (1); -5.68** (2.36) in (2); -6.19*** (1.65) in (3); 1.23 (4.57) in (4); 4.24*** (1.34) in (5).
  - Log GDP per capita: -9.13*** (2.64) in (1); 4.51 (4.15) in (2); -6.71*** (2.11) in (3); -22.50** (9.03) in (4); 0.19 (6.31) in (5).
  - Inflation: 0.25 (0.18) in (1); 0.18 (0.19) in (2); 0.17 (0.15) in (3); 0.05 (0.41) in (4); 0.94* (0.51) in (5).
  - Current account balance (% of GDP): -0.01 (0.18) in (1); -0.06 (0.15) in (2); 0.02 (0.16) in (3); -0.89*** (0.31) in (4); -0.54* (0.29) in (5).
  - Equity inflows (% of GDP): -0.11*** (0.04) in (1); -0.11 (0.10) in (2); -0.06 (0.04) in (3); 0.03 (0.14) in (4); 0.15 (0.09) in (5).
  - YtY % Growth in CB Assets: -0.02** (0.01) in (1); 0.01 (0.01) in (2); -0.01 (0.01) in (3); -0.05*** (0.02) in (4); -0.03 (0.02) in (5).
  - R-squared: 0.427 (1); 0.183 (2); 0.430 (3); 0.331 (4); 0.231 (5).
  - Observations and number of countries vary by regression (details in table).

- Table 6 (Summary of Event Study Results):
  - Panel A. Effects on Bank and Total Credit (All Instruments / Quantity Measures / Price Measures):
    - All: Bank Credit -7.7***; Total Credit -4.9***; Quantity Measures: Bank Credit -8.7***; Total Credit -4.1***; Price Measures: Bank Credit 1.7; Total Credit 1.2.
    - AEs: Bank Credit -3.2**; Total Credit -1.6; Quantity Measures: Bank Credit -6.6***; Total Credit -1.5; Price Measures: Bank Credit 2.0; Total Credit 2.1.
    - EMEs: Bank Credit -9.9***; Total Credit -6.5***; Quantity Measures: Bank Credit -10.4***; Total Credit -6.9***; Price Measures: Bank Credit 1.5; Total Credit -2.8.
  - Panel B. Cross-Sector Credit Substitution (All Instruments / Quantity Measures / Price Measures):
    - All: -4.3*** (All Instruments); -5.2*** (Quantity Measures); 2.3 (Price Measures).
    - AEs: -4.1*** (All Instruments); -4.6*** (Quantity Measures); -1.2 (Price Measures).
    - EMEs: -6.2*** (All Instruments); -6.5*** (Quantity Measures); 1.1 (Price Measures).
  - Note: Effects reported are average cumulative credit growth rates during the 2-year period following activation of macroprudential policies, adjusted for baseline rates implied by countries' macroeconomic fundamentals.
  - *, p < .1; **, p < .05; ***, p < .01.
  - Emerging market economy list: Brazil, China, Hungary, Indonesia, India, Mexico, Malaysia, Thailand, Turkey, and South Africa.
  - Advanced economy list: Australia, Austria, Belgium, Canada, Switzerland, Czech Republic, Germany, Denmark, Spain, Finland, France, United Kingdom, Greece, Hong Kong, Ireland, Italy, Japan, Luxembourg, Netherlands, Norway, Poland, Portugal, Russia, Saudi Arabia, Singapore, Sweden, and the United States.

*Italic: Source: _wp1694 - REFERENCES (source PDF content as provided).*

### APPENDIX I. PLACEBO TESTS

### APPENDIX I. PLACEBO TESTS

### Panel A — Effects on Bank and Total Credit (Event Study Using Placebo Event Dataset)
- Post-implementation Effect of Macroprudential Policies (average cumulative credit growth rates during the 2-year period following activation; growth rates adjusted for baseline rates implied by countries' macroeconomic fundamentals)
  - All Instruments
    - All: Bank Credit 0.9, Total Credit 1.1
    - AEs: Bank Credit 2.0, Total Credit 1.8
    - EMEs: Bank Credit -4.5, Total Credit -6.0*
  - Quantity Measures
    - All: Bank Credit -0.8, Total Credit 0.1
    - AEs: Bank Credit -0.2, Total Credit 0.3
    - EMEs: Bank Credit -4.2, Total Credit -6.6
  - Price Measures
    - All: Bank Credit 1.6, Total Credit -0.2
    - AEs: Bank Credit 2.9, Total Credit -0.9
    - EMEs: Bank Credit -7.9, Total Credit -8.4

- Notes on statistical significance
  - * p < .1, ** p < .05, *** p < .01 (Standard errors in parentheses).

- Country group composition (as used in these tables)
  - Emerging market economy group: Brazil, China, Hungary, Indonesia, India, Mexico, Malaysia, Thailand, Turkey, and South Africa.
  - Advanced economy group: Australia, Austria, Belgium, Canada, Switzerland, Czech Republic, Germany, Denmark, Spain, Finland, France, United Kingdom, Greece, Hong Kong, Ireland, Italy, Japan, Luxembourg, Netherlands, Norway, Poland, Portugal, Russia, Saudi Arabia, Singapore, Sweden, and the United States.

### Panel B — Cross-Sector Credit Substitution (Placebo Event Dataset)
- Net sectoral credit flow cumulated over the 2-year period following activation of macroprudential policies (measure defined in source)
  - All Instruments
    - All: -0.1, -1.2, 4.4*
    - AEs: 1.5, 0.2, 5.4*
    - EMEs: -3.8, -6.2, 0.7

- Notes on statistical significance
  - * p < .1, ** p < .05, *** p < .01 (Standard errors in parentheses).

### Table A.2 — Event Study on MaP Effects Prior and During the GFC — Summary of Results

- Panel A: Effects on Bank and Total Credit
  - Effects prior to 2007Q3 (average cumulative credit growth rates during the 2-year period following activation; growth rates adjusted for baseline rates)
    - All: Bank Credit -8.8***, Total Credit -4.7**
    - AEs: Bank Credit -7.3**, Total Credit -6.5*
    - EMEs: Bank Credit -9.2**, Total Credit -3.3
  - Effects after 2007Q3 (post-implementation effect of MaP)
    - All Instruments
      - All: Bank Credit -3.7**, Total Credit -1.7
      - AEs: Bank Credit -1.3*, Total Credit 0.4
      - EMEs: Bank Credit -9.6***, Total Credit -7.5***
    - Quantity Measures
      - All: Bank Credit -4.8***, Total Credit -2.8**
      - AEs: Bank Credit -1.6*, Total Credit -0.4
      - EMEs: Bank Credit -10.5***, Total Credit -8.6***
    - Price Measures
      - All: Bank Credit 4.1*, Total Credit 5.8**
      - AEs: Bank Credit 1.0, Total Credit 2.2
      - EMEs: Bank Credit 3.5, Total Credit 5.8*

- Panel B: Cross-Sector Credit Substitution (effects prior to 2007Q3 and effects after 2007Q3)
  - Post-Implementation Effect of MaP (as reported)
    - All Instruments / Quantity Measures / Price Measures
      - All: Bank Credit -3.7**, Total Credit -1.7; Bank Credit -4.8***, Total Credit -2.8**; Bank Credit 4.1*, Total Credit 5.8**
      - AEs: Bank Credit -1.3*, Total Credit 0.4; Bank Credit -1.6*, Total Credit 0.4; Bank Credit 1.0, Total Credit 2.2
      - EMEs: Bank Credit -9.6***, Total Credit -7.5***; Bank Credit -10.5***, Total Credit -8.6***; Bank Credit 3.5, Total Credit 5.8*
    - Additional reported rows
      - All: -5.7***, -5.7***, NA
      - AEs: -6.0***, -6.0***, NA
      - All Instruments / Quantity Measures / Price Measures: -6.1***, -6.1***, NA

- Notes on statistical significance
  - * p < .1, ** p < .05, *** p < .01 (Standard errors in parentheses).

- Country group composition (repeated)
  - Emerging market economy group: Brazil, China, Hungary, Indonesia, India, Mexico, Malaysia, Thailand, Turkey, and South Africa.
  - Advanced economy group: Australia, Austria, Belgium, Canada, Switzerland, Czech Republic, Germany, Denmark, Spain, Finland, France, United Kingdom, Greece, Hong Kong, Ireland, Italy, Japan, Luxembourg, Netherlands, Norway, Poland, Portugal, Russia, Saudi Arabia, Singapore, Sweden, and the United States.

### Figure A1 — Simulated Macroprudential Policy Measures Used in the Placebo Tests
- Visualization note (from source)
  - Dots correspond to simulated MaP events.
  - Blank spaces refer to no new MaPs in a country in a given year.

*Source: Authors’ calculations.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1694.pdf_