## 5. Response Time for Credit in Different Institutional Arrangements

## Source details

**Canonical URL:** [5. Response Time for Credit in Different Institutional Arrangements](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13166.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13166.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13166.pdf.json)

---

### Introduction and scope
- Objective: Gauge whether institutional arrangements affect the timely use of macroprudential policy instruments by evaluating policy response time across 39 countries using data based on the 2010 IMF survey on Financial Stability and Macroprudential Policy.
- Key assumption: All observed policy actions are warranted by emerging risks. Instruments with a microprudential focus or unrelated to the credit cycle were excluded.
- Focus:
  - Time dimension of systemic risk (timeliness of responses to credit-cycle related risks).
  - Sample: monthly data from 39 countries that used macroprudential instruments during 2008–2011.
  - Instruments included (8): caps on the LTV, caps on the DTI, limits on foreign currency lending, ceilings on credit/credit growth, reserve requirements, capital requirements, provisioning requirements, and restrictions on profit distribution.
- Principal summary finding:
  - There is a negative correlation between policy response time and the central bank’s involvement in the macroprudential policy framework.

### Measuring institutional arrangements
- Five distinguishing dimensions:
  1. Degree of institutional integration between the central bank and financial regulatory/supervisory functions.
  2. Ownership of the macroprudential mandate.
  3. Role of the government (treasury) in macroprudential policy.
  4. Degree of organizational separation of decision-making and control over instruments.
  5. Existence of a coordinating body for macroprudential policy.
- Three indices constructed (score 1 to 4; higher value = more important role):
  - MaPP (macroprudential index — central bank role):
    - 1 – Mandate shared including central bank, no coordination body
    - 2 – Mandate shared including central bank, central bank is a member of a coordination body
    - 3 – Mandate shared including central bank, central bank chairs the coordination body
    - 4 – Central bank (or a committee of it) is the sole owner of the mandate
  - MiPP (microprudential index — central bank supervisory role):
    - 1 – Central bank has no regulatory/supervisory functions
    - 2 – Central bank supervises the banking sector
    - 3 – Central bank supervises the banking sector and part of the nonbank financial sector
    - 4 – Central bank supervises the entire financial sector
  - MoF (government index — ministry of finance role in coordination):
    - 1 – No coordination body or government not a member
    - 2 – Government is a member/observer of the coordination body
    - 3 – Government co-chairs the coordination body with other agencies
    - 4 – Government chairs the coordination body
- Percent of countries by score (Table 1):
  - MaPP: score 1 = 7.7; score 2 = 61.5; score 3 = 10.3; score 4 = 20.5
  - MiPP: score 1 = 33.3; score 2 = 41.0; score 3 = 17.9; score 4 = 7.7
  - MoF: score 1 = 23.1; score 2 = 46.2; score 3 = 5.1; score 4 = 25.6

### Measuring response time (definition and methods)
- Definition: Elapsed time between the emergence of a risk and the subsequent use of a policy instrument. Start-dates for risk emergence are identified by objective changes in the behavior of a risk variable rather than subjective judgments.
- Risk variable used: credit and credit growth (chosen because most countries target credit/credit growth and because this covers the largest number of instrument instances).
- Three approaches to identify the start point of response time:
  - Approach 1 — Break-in-trend analysis:
    - A structural break in level and slope of a trend function denotes a change in behavior of the risk variable.
    - Methodology: Carrion-i-Silvestre, Kim and Perron (2009) to detect break dates by minimizing sum of squared residuals.
  - Approach 2 — Distance-from-peak/trough analysis:
    - A turning point (peak/trough) in the path of the risk variable represents a change; turning points estimated from a six-month moving average of seasonally adjusted data.
  - Approach 3 — Threshold analysis (experimental):
    - Risk considered to have emerged after the risk variable crosses a threshold equal to 1.5 standard deviations from the historical mean (yoy) during the sample period.
    - Approach 3 involved more judgment and excluded more than half of the countries in the sample because many countries took action before the threshold was reached.
- Calculation details:
  - Response times measured separately for upswing and downswing of credit cycle and then averaged.
  - If an instrument is used multiple times in an upswing or downswing, the first implementation time is used.
- Observations on approaches:
  - Average response times under Approaches 1 and 2 are close; Approach 3 gives a substantially longer average response time and excludes many countries.
  - Interpretation: A common cross-country threshold is arbitrary and unrealistic; policymakers often act before such thresholds, relying on behavior over time and multiple indicators.

### Descriptive findings (response time and institutional indices)
- Aggregate observations:
  - Average response time appears negatively correlated with the MaPP index under Approaches 1 and 2 (higher MaPP score = shorter response time).
  - Correlation with MiPP is unclear: negative under Approach 1 but not consistent across Approaches 2 and 3.
  - Correlation with MoF is unclear and possibly nonlinear; absence of a coordination body (MoF score 1) has the longest average response time under all three approaches.
- Interpretation:
  - Negative correlation with MaPP suggests desirability of a significant central bank role because it can:
    - Facilitate systemic risk monitoring and identification.
    - Foster cross-agency policy coordination.
  - Central bank comparative advantages cited: monetary policy role, payments-system oversight, expertise in systemic risk monitoring, experience communicating risks via financial stability reports.
  - Government participation can help garner political support, but excessive government role relative to the central bank could introduce short-term political considerations.

### Multivariate analysis (cross-country regressions)
- Estimated equation (37 countries for Approaches 1 and 2):
  - Response time_i = α + β1 MaPP_i + β2 MiPP_i + β3 MoF_i + β4 INT_i + β5 CGDP_i + ɛ_i
    - MaPP_i, MiPP_i, MoF_i: institutional indices at time of policy action
    - INT_i: standardized cumulative change in the policy rate during the period corresponding to response time
    - CGDP_i: average credit-to-GDP ratio during 2008-2011
- Caveats:
  - Small sample size; results sensitive to small variations and outliers.
  - Model is simple; causality cannot be inferred. Monetary policy may be endogenous.

- Main regression results (Table 2):
  - Approach 1:
    - Constant = 10.8 (2.4)***
    - MaPP = -2.6 (0.7)***
    - MiPP = 0.2 (0.7)
    - MoF = -0.6 (0.5)
    - INT = 2.5 (0.8)***
    - CGDP = 0.0 (0.0)***
    - Adj. R2 = 0.47
  - Approach 2:
    - Constant = 15.1 (4.0)***
    - MaPP = -3.1 (1.2)**
    - MiPP = 1.1 (1.1)
    - MoF = -0.9 (0.9)
    - INT = 2.0 (1.2)
    - CGDP = 0.0 (0.0)
    - Adj. R2 = 0.09
  - Note: standard deviations in parentheses; ***, **, * denote significance at 1, 5 and 10 percent respectively.

- Interpretation of coefficients:
  - MaPP: Negative and statistically significant under both Approaches 1 and 2. Magnitude interpreted as roughly a reduction of about three months in response time for each one-point increase in central bank involvement (coefficients: -2.6 and -3.1).
  - MiPP and MoF: Statistically insignificant under both approaches — inconclusive evidence on supervisory integration and government role.
  - INT (policy rate change): Positive and statistically significant under Approach 1 but not Approach 2. Possible interpretations:
    - Larger policy-rate adjustments may reduce the need to use macroprudential instruments early.
    - Or when monetary policy is constrained, macroprudential instruments may be used earlier.
  - CGDP (credit-to-GDP): Statistically significant under Approach 1 but not Approach 2; possible explanations include less-developed financial markets prompting faster macroprudential use, or risks becoming larger the longer policymakers delay.

### Robustness check: dummy-variable specification
- Dummy variables:
  - D1 = 1 if central bank chairs coordination body or is sole agency with mandate
  - D2 = 1 if banking supervision is inside the central bank
  - D3 = 1 if ministry of finance chairs the coordination body
- Results (Table 3):
  - Approach 1:
    - Constant = 5.2 (1.7)***
    - D1 = -4.3 (1.5)***
    - D2 = 0.7 (1.5)
    - D3 = -0.4 (1.7)
    - INT = 2.6 (0.9)***
    - CGDP = 0.0 (0.0)**
    - Adj. R2 = 0.37
  - Approach 2:
    - Constant = 8.8 (2.6)***
    - D1 = -5.1 (2.4)**
    - D2 = 2.6 (2.2)
    - D3 = -0.5 (2.5)
    - INT = 1.7 (1.3)
    - CGDP = -0.0 (0.0)
    - Adj. R2 = 0.03
- Interpretation:
  - D1 (central bank decision role): Negative and statistically significant under both approaches — corroborates negative correlation between response time and central bank role.
  - D2 (supervisory integration) and D3 (government chairs coordination): Statistically insignificant — inconclusive evidence.

### Conclusions and caveats
- Main conclusion:
  - Negative correlation between policy response time and the involvement of the central bank in the macroprudential framework — giving the central bank an important role is conducive to reduced policy response time.
- Additional insights:
  - Monetary policy may affect timing of macroprudential use; evidence is inconclusive but suggests coordination between macroprudential and monetary policies is important.
  - A single common cross-country risk threshold triggering policy action likely does not exist; many countries take action before arbitrary thresholds are breached and rely on multiple indicators and judgment.
- Limitations:
  - Policy response time is only one aspect of an effective macroprudential framework. Institutional arrangements must also address the cross-sectional dimension of systemic risk (common exposures, linkages, interdependencies).
  - Cross-country results hold “on average”; country-specific factors (quality of supervision, policy coordination mechanisms, approach to economic management) are important and not fully captured in this cross-country study.

*Source: IMF staff calculations and analysis based on the 2010 IMF survey on Financial Stability and Macroprudential Policy (content unit: 5. Response Time for Credit in Different Institutional Arrangements).*

### 1. Institutional Arrangement Indices ...................................................................................

### 1. Institutional Arrangement Indices

### Main sections (as listed)
- 1. Institutional Arrangement Indices ..........................................................................................5
- 2. Response Time and Institutional Arrangements ..................................................................12
- 3. Response Time and Dummy Variables ................................................................................14
- 4. Institutional Arrangement Indices ........................................................................................16

### Figures (as listed)
- 1. Break-in-Trend Analysis ........................................................................................................6
- 2. Distance-from Peak/Trough Analysis ....................................................................................7
- 3. Threshold Analysis ................................................................................................................7
- 4. Calculated Response Time .....................................................................................................9

*Source: _wp13166 - 1. Institutional Arrangement Indices (PDF)._*

### 5. Response Time for Credit in Different Institutional Arrangements ....................................10

### 5. Response Time for Credit in Different Institutional Arrangements

### Introduction
- Objective: Gauge whether institutional arrangements affect the timely use of macroprudential policy instruments by evaluating policy response time across 39 countries using data based on the 2010 IMF survey on Financial Stability and Macroprudential Policy.
- Key assumption: All observed policy actions are warranted by emerging risks. Instruments with a microprudential focus or unrelated to the credit cycle were excluded from the sample.
- Principal summary finding:
  - There is a negative correlation between policy response time and the central bank’s involvement in the macroprudential policy framework, consistent with the Fund’s position that “the central bank needs to play an important role.”
- Scope and focus:
  - Time dimension of systemic risk (timeliness of responses to credit-cycle related risks).
  - Sample: monthly data from 39 countries that used macroprudential instruments during 2008–2011.
  - Eight macroprudential instruments used in calculations: caps on the LTV, caps on the DTI, limits on foreign currency lending, ceilings on credit/credit growth, reserve requirements, capital requirements, provisioning requirements, and restrictions on profit distribution.

### Measuring Institutional Arrangements
- Five distinguishing dimensions identified from Fund’s previous work:
  1. Degree of institutional integration between the central bank and financial regulatory/supervisory functions.
  2. Ownership of the macroprudential mandate.
  3. Role of the government (treasury) in macroprudential policy.
  4. Degree of organizational separation of decision-making and control over instruments.
  5. Existence of a coordinating body for macroprudential policy.
- Three indices constructed (score 1 to 4; higher value = more important role). Indices measure de facto arrangements and are not mutually exclusive:
  - MaPP (macroprudential index — central bank role):
    - 1 – Mandate shared including central bank, no coordination body
    - 2 – Mandate shared including central bank, central bank is a member of a coordination body
    - 3 – Mandate shared including central bank, central bank chairs the coordination body
    - 4 – Central bank (or a committee of it) is the sole owner of the mandate
  - MiPP (microprudential index — central bank supervisory role):
    - 1 – Central bank has no regulatory/supervisory functions
    - 2 – Central bank supervises the banking sector
    - 3 – Central bank supervises the banking sector and part of the nonbank financial sector
    - 4 – Central bank supervises the entire financial sector
  - MoF (government index — ministry of finance role in coordination):
    - 1 – No coordination body or government not a member
    - 2 – Government is a member/observer of the coordination body
    - 3 – Government co-chairs the coordination body with other agencies
    - 4 – Government chairs the coordination body
- Table 1 (percent of countries by score):
  - MaPP: score 1 = 7.7; score 2 = 61.5; score 3 = 10.3; score 4 = 20.5
  - MiPP: score 1 = 33.3; score 2 = 41.0; score 3 = 17.9; score 4 = 7.7
  - MoF: score 1 = 23.1; score 2 = 46.2; score 3 = 5.1; score 4 = 25.6

### Measuring Response Time
- Definition: Elapsed time between the emergence of a risk and the subsequent use of a policy instrument. Start-dates for risk emergence are identified by objective changes in the behavior of a risk variable rather than subjective judgments.
- Risk variable used: credit and credit growth (chosen because most countries target credit/credit growth and because this covers the largest number of instrument instances).
- Three approaches to identify the start point of response time:
  - Approach 1 — Break-in-trend analysis:
    - A structural break in level and slope of a trend function denotes a change in behavior of the risk variable.
    - Methodology: Carrion-i-Silvestre, Kim and Perron (2009) to detect break dates by minimizing sum of squared residuals.
  - Approach 2 — Distance-from-peak/trough analysis:
    - A turning point (peak/trough) in the path of the risk variable represents a change; turning points estimated from a six-month moving average of seasonally adjusted data.
  - Approach 3 — Threshold analysis (experimental):
    - Risk considered to have emerged after the risk variable crosses a threshold equal to 1.5 standard deviations from the historical mean (yoy) during the sample period.
    - Approach 3 involved more judgment and excluded more than half of the countries in the sample because many countries took action before the threshold was reached.
- Calculation details:
  - Response times measured separately for upswing and downswing of credit cycle and then averaged.
  - If an instrument is used multiple times in an upswing or downswing, the first implementation time is used.
- Observations on approaches:
  - Average response times under Approaches 1 and 2 are close; Approach 3 gives a substantially longer average response time and excludes many countries.
  - Interpretation: A common cross-country threshold is arbitrary and unrealistic; policymakers often act before such thresholds, relying on behavior over time and multiple indicators.

### Response Time and Institutional Indices (Descriptive)
- Visual/aggregate observation:
  - Average response time appears negatively correlated with the MaPP index under Approaches 1 and 2 (higher MaPP score = shorter response time).
  - Correlation with MiPP is unclear: negative under Approach 1 but not consistent across Approaches 2 and 3.
  - Correlation with MoF is unclear and possibly nonlinear; absence of a coordination body (MoF score 1) has the longest average response time under all three approaches.
- Interpretation:
  - Negative correlation with MaPP suggests it is desirable for the central bank to play an important role because it can (i) facilitate systemic risk monitoring and identification, and (ii) foster cross-agency policy coordination.
  - The central bank’s unique positioning (monetary policymaker, payments system supervisor, expertise in systemic risk monitoring, experience communicating risks via financial stability reports) supports a role that can reduce response time.
  - Government participation can help garner political support, but excessive government role relative to the central bank could introduce short-term political considerations.

### Response Time in a Multivariate Analysis
- Cross-country regression equation estimated (37 countries for Approaches 1 and 2):
  - Response time_i = α + β1 MaPP_i + β2 MiPP_i + β3 MoF_i + β4 INT_i + β5 CGDP_i + ɛ_i
    - MaPP_i, MiPP_i, MoF_i: institutional indices at time of policy action
    - INT_i: standardized cumulative change in the policy rate during the period corresponding to response time
    - CGDP_i: average credit-to-GDP ratio during 2008-2011 (depth of financial sector)
- Caveats:
  - Small sample size; results sensitive to small variations and outliers.
  - Model is simple; causality cannot be inferred. Monetary policy may be endogenous.
- Main regression results (Table 2):
  - Approach 1:
    - Constant = 10.8 (2.4)***
    - MaPP = -2.6 (0.7)***
    - MiPP = 0.2 (0.7)
    - MoF = -0.6 (0.5)
    - INT = 2.5 (0.8)***
    - CGDP = 0.0 (0.0)***
    - Adj. R2 = 0.47
  - Approach 2:
    - Constant = 15.1 (4.0)***
    - MaPP = -3.1 (1.2)**
    - MiPP = 1.1 (1.1)
    - MoF = -0.9 (0.9)
    - INT = 2.0 (1.2)
    - CGDP = 0.0 (0.0)
    - Adj. R2 = 0.09
  - Note: standard deviations in parentheses; ***, **, * denote significance at 1, 5 and 10 percent respectively.
- Interpretation of coefficients:
  - MaPP: Negative and statistically significant under both Approaches 1 and 2. Magnitude: -2.6 (Approach 1) and -3.1 (Approach 2) — interpreted as roughly a reduction of about three months in response time for each one-point increase in central bank involvement.
  - MiPP and MoF: Coefficients statistically insignificant under both approaches — inconclusive evidence on supervisory integration and government role.
  - INT (policy rate change): Positive and statistically significant under Approach 1 but not Approach 2 — possible interpretations:
    - Larger policy-rate adjustments may reduce the need to use macroprudential instruments early.
    - Or when monetary policy is constrained, macroprudential instruments may be used earlier (coordination between macroprudential and monetary policies matters).
  - CGDP (credit-to-GDP): Statistically significant under Approach 1 but not Approach 2; possible explanations include less-developed financial markets prompting faster macroprudential use, or risks becoming larger the longer policymakers delay.
- Robustness check with dummy-variable specification (Equation b):
  - D1 = 1 if central bank chairs coordination body or is sole agency with mandate
  - D2 = 1 if banking supervision is inside the central bank
  - D3 = 1 if ministry of finance chairs the coordination body
- Results with dummy variables (Table 3):
  - Approach 1:
    - Constant = 5.2 (1.7)***
    - D1 = -4.3 (1.5)***
    - D2 = 0.7 (1.5)
    - D3 = -0.4 (1.7)
    - INT = 2.6 (0.9)***
    - CGDP = 0.0 (0.0)**
    - Adj. R2 = 0.37
  - Approach 2:
    - Constant = 8.8 (2.6)***
    - D1 = -5.1 (2.4)**
    - D2 = 2.6 (2.2)
    - D3 = -0.5 (2.5)
    - INT = 1.7 (1.3)
    - CGDP = -0.0 (0.0)
    - Adj. R2 = 0.03
  - Interpretation:
    - D1 (central bank decision role): Negative and statistically significant under both approaches — corroborates negative correlation between response time and central bank role.
    - D2 (supervisory integration) and D3 (government chairs coordination): Statistically insignificant — again inconclusive evidence.

### Conclusion
- Main conclusion:
  - The paper finds a negative correlation between policy response time and the involvement of the central bank in the macroprudential framework — giving the central bank an important role is conducive to reduced policy response time.
- Practical rationale:
  - Central bank advantages: capacity to monitor macrofinancial linkages (monetary policymaker and payments-system supervisor), expertise in systemic risk identification and monitoring, and experience communicating risks (financial stability reports).
- Additional insights:
  - Monetary policy may affect timing of macroprudential use; evidence is inconclusive but suggests coordination between macroprudential and monetary policies is important.
  - A single common cross-country risk threshold triggering policy action likely does not exist; many countries take action before arbitrary thresholds are breached and rely on multiple indicators and judgment.
- Limitations and caveats:
  - Policy response time is only one aspect of an effective macroprudential framework. Institutional arrangements must also address the cross-sectional dimension of systemic risk (common exposures, linkages, interdependencies).
  - Cross-country results hold “on average”; country-specific factors (quality of supervision, policy coordination mechanisms, approach to economic management) are important and not fully captured in this cross-country study.

*Source: IMF staff calculations and analysis based on the 2010 IMF survey on Financial Stability and Macroprudential Policy (content unit: 5. Response Time for Credit in Different Institutional Arrangements).*

### REFERENCES

### _wp13166 - REFERENCES

### IMF publications
- International Monetary Fund, 2011a, “Macroprudential Policy: An Organizing Framework” (Washington: International Monetary Fund, March).
- International Monetary Fund, 2011b, “Macroprudential Policy: An Organizing Framework – Background Paper” (Washington: International Monetary Fund, March).
- International Monetary Fund, 2011c, “Toward Operationalizing Macroprudential Policies: When to Act?” Chapter 3, Global Financial Stability Report (Washington: International Monetary Fund, September).
- International Monetary Fund, 2012a, “Policies for Macrofinancial Stability: How to Deal with Credit Booms,” IMF Staff Discussion Note 12/06 (Washington: International Monetary Fund, June).
- International Monetary Fund, 2012b, “The Interaction of Monetary and Macroprudential Policies” (Washington: International Monetary Fund, December).
- International Monetary Fund, 2013, “Key Aspects of Macroprudential Policy” (Washington: International Monetary Fund, May).

### IMF and related working papers, studies, and assessments
- Arregui, Nicolas, Jaromir Benes, Ivo Krznar, Srobona Mitra, and Andre Oliveira Santos, 2013a, “Evaluating the Net Benefits of Macroprudential Policy: A Cookbook,” forthcoming (Washington: International Monetary Fund).
- 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 11/238 (Washington: International Monetary Fund, October).
- Masciandaro, Donato, Rosaria Vega Pansini, and Marc Quintyn, 2011, “The Economic Crisis: Did Financial Supervision Matter?,” IMF Working Paper 11/261 (Washington: International Monetary Fund, November).
- Nier, Erlend W., Jacek Osiński, Luis I. Jácome, and Pamela Madrid, 2011, “Towards Effective Macroprudential Policy Frameworks: An Assessment of Stylized Institutional Models,” IMF Working Paper 11/250 (Washington: International Monetary Fund, November).

### Other research and chapters
- Kaufman, Daniel, Aart Kraay and Massimo Mastruzzi, 2010, “The Worldwide Governance Indicators – Methodology and Analytical Issues,” World Bank Policy Research Working Paper 5430 (Washington: The World Bank, September).
- Repullo, R. and J. Saurina (2012): “The Countercyclical Capital Buffer of Basel III: A Critical Assessment,” Chapter 3 (45-67) of the book The Crisis Aftermath: New Regulatory Paradigms, edited by Mathias Dewatripont and Xavier Freixas, CEPR March.
- Repullo, R.; Trucharte, C. and J. Saurina (2010): “Mitigating the procyclicality of Basel II.” Economic Policy, October, 659-702.

* _wp13166 - REFERENCES *

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13166.pdf_
