## _wp0684 - References

## Source details

**Canonical URL:** [_wp0684 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0684.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0684.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0684.pdf.json)

---

### Introduction and research objective
- Research objective:
  - Study changes in Canada’s monetary policy transmission associated with financial-structure changes in the 1990s using:
    - Vector autoregression (VAR) models extended with financial variables emphasized in the credit channel literature, tested for structural breaks or parameter instability around dates of major regulatory changes.
    - Structural econometric (IS/aggregate demand) models augmented with measures of disintermediation (ratio of direct to indirect finance) to test sensitivity of interest-rate transmission.
- Data and sample:
  - VAR models estimated in levels using quarterly data from 1971 to 2005.

### Financial deregulation and disintermediation: documented trends
- Shift toward market-based (direct) finance away from intermediated (bank) finance:
  - Bank lending as a source of funding for the corporate sector declined from approximately 60 percent in 1980 to under 40 percent in the 2000s.
  - Increase in issuance of corporate bonds and commercial paper.
  - Ratio of direct-to-indirect lending increases beginning in the late 1980s; increase in equity issuance in the 1990s.
- Growth in household credit:
  - Credit to the household sector rose from just under 50 percent of GDP in the first half of the 1980s to almost 70 percent of GDP in 2001.
  - Consumer credit share of total (business and consumer credit) increased from 42 percent in 1982 to 55 percent in 2005.
- Institutional drivers:
  - Bank Act amendments in 1980, 1987, 1992, and 1997 contributed to the shift in corporates’ funding mix.
  - Mortgage securitization advanced more slowly in Canada than in the United States.

### Evidence from VAR models (monetary shock impulse responses)
- Benchmark VAR specification:
  - Endogenous vector Yt = [y, p, R, S] where y is quarterly GDP, p is the GDP deflator, R is the 3 month T-bill rate, and S is the exchange rate.
  - Exogenous vector Xt = [USy, USR, cp] with US GDP, Federal Funds rate, and an index of world commodity prices.
  - Extended VARs include block of financial variables F: quantity variables (e.g., total loans to businesses and households, securities) and asset price variables (stock and housing prices, spreads).
- Key empirical findings:
  - Full-sample impulse responses to a contractionary monetary shock show a decline in output and a sluggish decline in prices with an initial price spike (“price puzzle”), broadly similar to other industrialized countries.
  - Structural breaks and parameter instability:
    - Likelihood-ratio tests reject parameter stability at the 5 percent level for all VAR models.
    - Break-test dates examined include 1988:1 and 1993:2 (and intermediate date 1991:1 with qualitatively similar results).
  - Subsample differences:
    - First subsample (1971:1–1991:1):
      - Impulse responses are more tightly estimated; statistically significant decline in output between the 5th and 10th quarters.
      - Prices fall gradually and persistently; exchange rate appreciates sharply on impact and returns to baseline.
    - Second subsample (1991:1–2005:2):
      - Impulse responses show a very different shape and are largely statistically insignificant.
  - Impact of including financial variables (business loans and asset prices):
    - Produces a tighter and improved characterization of transmission for the first subsample.
    - Makes the appreciation of the exchange rate after an interest rate increase statistically significant in the first subsample.
    - Increases the contribution of monetary shocks to output variance:
      - In the first subsample monetary shocks explain close to 35-40 percent of the variance of output in the financial models, while they explain around 27 percent in the baseline model.
  - Volatility of monetary shocks:
    - Standard deviation of the monetary shocks in the second subsample is systematically smaller than in the first subsample, consistent with an increased role for the systematic component of monetary policy (inflation targeting, greater transparency).

### Evidence from structural IS / aggregate-demand models
- Motivation:
  - Test how financial disintermediation affects parameters of aggregate demand — specifically the interest-rate elasticity of the output gap.
- Model specifications:
  - Closed-economy IS (Rudebusch and Svensson, 1999) with coefficient on real interest rate allowed to vary with DIF (ratio of direct to indirect finance).
  - Two DIF measures:
    - DIF1 = ratio of securities to business loans.
    - DIF2 = ratio of securities to total loans (captures rising household lending).
  - Forward-looking/New Keynesian IS variants and open-economy IS variants including real exchange rate gap and U.S. GDP gap.
- Main estimation results:
  - Closed-economy IS:
    - Interest-rate elasticity of the output gap is significantly different from zero in the full sample, but significance concentrated in the more recent subsample.
    - No clear evidence that disintermediation (DIF1 or DIF2) affected the interest-rate elasticity in the closed-economy specification; coefficients on DIF not statistically significant.
    - Contextual numeric values reported in text:
      - “For the average values of the variables DIF1 and DIF2, the interest elasticity coefficients become -0.33 and -0.46, respectively.”
      - Elsewhere: “When calculated using the average sample values of DIF1 and DIF2 both elasticities are respectively -0.15 and -0.09, respectively.”
  - Forward-looking/New Keynesian variants:
    - Replacing expected future gap with actual future gap yields unsatisfactory results: future gap significant but real interest rate loses significance; coefficients on DIF not significant.
  - Open-economy IS:
    - Provides evidence that disintermediation has contributed to changes in monetary transmission.
    - Interest-rate elasticity of aggregate demand is not statistically different from zero for the full sample (1971–2005) when specified linearly, but becomes significant when modeled as a function of DIF.
    - Table 9 results highlighted:
      - Coefficients on DIF1 and DIF2 are significant in the second half of the sample and yield relatively larger interest elasticity estimates (respectively -0.46 and -0.48 for the average sample values of DIF1 and DIF2).

### Structural break tests, variance decompositions and shock statistics (selected results)
- Structural break tests in VAR models:
  - Table 1 (break in 1988:1): F(24, 80) value for 5 percent significance is 1.65. LR test is Chi-square with degrees of freedom equal to number of restrictions (100 for benchmark, 120 for financial models). Chi-square values for 120 degrees of freedom: 146.6 (5%) and 159.0 (1%).
  - Selected single-equation F-statistics reported (examples preserved exactly as in source):
    - GDP: 1.777 *, 2.008 *, 2.217 *, 2.221 *, 2.233 *
    - Prices: 1.118 1.296 1.004 2.094 * 1.892 *
    - Ints. Rate: 1.755 * 1.953 * 1.436 1.565 1.836 *
    - Exch.Rate: 1.894 * 2.024 * 2.582 * 3.135 * 3.116 *
    - Loans: 3.751 * ...... 4.069 *
    - Asset Prices: ......... 1.972 * 1.636 *
  - Tables 2 and 3 provide analogous F-statistics and LR-Test values for breaks in 1993:2 and 1991:1.
- Variance decompositions of GDP to monetary shock (in percent) — selected horizon entries preserved:
  - Baseline Model (Quarter):
    - 4: 0.81.6 1.3
    - 8: 9.22 0.41 1.1
    - 12: 15.6 27.1 1.2
    - 16: 15.9 26.9 1.2
  - Model with Loans:
    - 4: 1.6 0.6 9.1
    - 8: 9.8 22.7 8.7
    - 12: 14.5 33.7 9.2
    - 16: 15.5 32.2 19.2
  - Model with Asset Prices:
    - 4: 1.2 2.1 2.4
    - 8: 8.7 22.8 2.0
    - 12: 16.2 35.3 3.7
    - 16: 18.7 34.6 4.2
  - Model with Loans and Asset Prices:
    - 4: 1.4 0.5 4.5
    - 8: 9.8 25.0 3.3
    - 12: 16.7 39.6 3.2
    - 16: 19.2 35.1 4.0
- Changes in the standard deviations of monetary shocks (Interest Rate Equation) — Table 5:
  - Baseline VAR: 1971–1991 = 0.49; 1991–2005 = 0.41; SD2/SD1 = 0.84
  - VAR with Loans: 1971–1991 = 0.49; 1991–2005 = 0.46; SD2/SD1 = 0.94
  - VAR with Asset Prices: 1971–1991 = 0.52; 1991–2005 = 0.44; SD2/SD1 = 0.85
  - VAR with Loans and Asset Prices: 1971–1991 = 0.49; 1991–2005 = 0.44; SD2/SD1 = 0.90

### IS equation estimation results (selected coefficients and diagnostics)
- Closed Economy IS Equation (1971:1–2005:2) — Table 6 (models: Closed Economy IS Equation / Model with DIF1 / Model with DIF2):
  - α0: 0.473 0.483 0.476 (p-values: (0.300)(0.325)(0.309))
  - α1: 1.141 1.141 1.141 (p-values: (0.000)(0.000)(0.000))
  - α2: -0.298 -0.298 -0.298 (p-values: (0.000)(0.000)(0.000))
  - α30: -0.156 -0.137 -0.147 (p-values: (0.029)(0.684)(0.592))
  - α31: ... -0.010 -0.015 (...(0.954)(0.974))
  - R2: 0.818 0.818 0.818
  - DW: 2.00 2.00 2.00
- Open Economy IS Equation (1971:1–2005:2) — Table 8 (Model with DIF1 / Model with DIF2):
  - α0: -0.096 0.137 -0.054 (p-values (0.837)(0.768)(0.906))
  - α1: 1.021 0.972 0.979 (p-values (0.000)(0.000)(0.000))
  - α2: -0.252 -0.229 -0.229 (p-values (0.001)(0.003)(0.003))
  - α30: -0.072 0.874 0.658 (p-values (0.312)(0.022)(0.031))
  - α31: ... -0.505 -1.191 ...(0.012)(0.014)
  - α4: -0.027 -0.018 -0.019 (p-values (0.628)(0.747)(0.737))
  - α5: 0.440 0.628 0.602 (p-values (0.000)(0.000)(0.000))
  - R2: 0.838 0.845 0.845
  - DW: 1.79 1.76 1.78
- Open Economy IS Equation with sample break in 1988:1 — Table 9 reports parameters for (1971:1–1988:1) and (1988:1–2005:2) with R2 values ranging from 0.771 to 0.948 and DW statistics as reported in the source.

### VAR impulse-response visuals and data series (figures and panels)
- Financial structure and time series visuals:
  - Figure 1. Nonfinancial Corporate Financing Sources, 1971–2005 (stocks outstanding, in percent of total). Source: Bank of Canada.
  - Figure 2. Credit Flows, 1971–2005 (billion Canadian dollars). Source: Bank of Canada.
  - Figure 3. Ratio of Direct to Indirect Private Lending, 1971–2005. Source: Bank of Canada.
  - Figure 4. Outstanding Credit, 1971–2005 (billion Canadian dollars): household mortgages, business loans, consumer credit; axes range 0 to 700. Source: Bank of Canada.
- VAR impulse-response results (Figures 5–8):
  - Panels show impulse responses of:
    - GDP_R to TBILL3MO
    - GDP_P to TBILL3MO
    - TBILL3MO to TBILL3MO
    - XR_INDEX_LEV to TBILL3MO
  - Responses presented for:
    - Full sample (1971-Q1 to 2005-Q2)
    - 1971-Q1 to 1991-Q1
    - 1991-Q1 to 2005-Q2
  - Scales include small magnitude responses (axes approximately -.004 to .004 for GDP_R / GDP_P) and larger scales for exchange rate responses (e.g., -2 to 3).
  - Source: Fund staff calculations.

### Conclusions and policy implications
- Timing and drivers of structural change:
  - Monetary transmission in Canada changed markedly since the late 1980s; VAR models show a clear break beginning in 1988 after regulatory changes that initiated disintermediation.
  - Both changes in financial structure (disintermediation, increased household credit) and institutional changes in monetary policy (inflation targeting, greater systematic policy) contributed to the evolving transmission mechanism.
- Role of financial variables:
  - Inclusion of financial variables in VARs improves characterization of transmission in the pre-break period and increases the share of output variance explained by monetary shocks in financial VARs.
  - Structural IS estimates indicate disintermediation contributed to changes in the interest-rate elasticity of aggregate demand in the 1990s — often implying increased responsiveness (larger absolute value of the elasticity) associated with greater use of market-based finance.
  - Mechanism suggested: relationship-based bank lending is less interest-rate sensitive than price-sensitive market funding; disintermediation (more direct finance) can increase aggregate-demand responsiveness to interest-rate changes.
- Current and future considerations:
  - Monetary policy appears to have become more effective in the 1990s when measured as the average impact of interest rate changes on the output gap or aggregate demand.
  - This increased effectiveness may be undermined by the more recent increase in household borrowing and the relative decline in corporate securities issuance (declining ratio of direct to indirect finance since 2002 for the corporate sector).
  - Policy recommendation:
    - Monetary policy analysis would benefit from more explicit consideration of the evolving roles of financial markets and intermediaries, given their measurable influence on transmission parameters.

*Source: _wp0684 - 8. Impulse-Response Functions from VAR including Asset Prices and Business Loans (PDF chapter).*

### References..............................................................................................................

### _wp0684 - References

### Tables
- 1. Tests for Break in VAR Model in 1988:1....................................................................17
- 2. Tests for Break in VAR Model in 1993:2....................................................................18
- 3. Tests for Break in VAR Model in 1991:1....................................................................19
- 4. Variance Decompositions of GDP to Monetary Shock ...............................................20
- 5. Changes in the Standard Deviations of Shocks ...........................................................21
- 6. Estimates of Closed Economy IS Equation (1971:1–200:2) .......................................22
- 7. Estimates of Closed Economy IS Equation (Sample break in 1988:1) .......................23
- 8. Estimates of Open Economy IS Equation (1971:1–200:2)..........................................24
- 9. Estimates of Open Economy IS Equation (Sample break in 1988:1) ..........................25

### Figures
- 1.         Nonfinancial         Corporate         Financing Sources, 1971–2005 .............................................26
- 2. Credit Flows, 1971–2005.............................................................................................27
- 3. Ratio of Direct to Indirect Private Lending, 1971–2005 .............................................28
- 4. Outstanding Credit, 1971–-2005..................................................................................29
- 5. Impulse-Response Functions from Benchmark VAR..................................................30
- 6. Impulse-Response Functions from VAR including Business Loans ...........................31
- 7. Impulse-Response Functions from VAR including Asset Prices ................................32

*Source: _wp0684 - References*

### 8. Impulse-Response Functions from VAR including Asset Prices

### 8. Impulse-Response Functions from VAR including Asset Prices and Business Loans

### Introduction
- Research objective: Study changes in Canada’s monetary policy transmission associated with financial-structure changes in the 1990s using two methodologies:
  - Vector autoregression (VAR) models extended with financial variables emphasized in the credit channel literature, tested for structural breaks or parameter instability around dates of major regulatory changes.
  - Structural econometric (IS/aggregate demand) models augmented with measures of disintermediation (ratio of direct to indirect finance) to test sensitivity of interest-rate transmission.
- Data and sample: VAR models estimated in levels using quarterly data from 1971 to 2005.

### Financial deregulation and disintermediation: documented trends
- Two principal trends in the Canadian financial system over the last two decades:
  - Shift toward market-based (direct) finance away from intermediated (bank) finance.
    - Bank lending as a source of funding for the corporate sector declined from approximately 60 percent in 1980 to under 40 percent in the 2000s.
    - Increase in issuance of corporate bonds and commercial paper.
    - Ratio of direct-to-indirect lending increases beginning in the late 1980s; increase in equity issuance in the 1990s.
  - Growth in household credit:
    - Credit to the household sector rose from just under 50 percent of GDP in the first half of the 1980s to almost 70 percent of GDP in 2001.
    - Consumer credit share of total (business and consumer credit) increased from 42 percent in 1982 to 55 percent in 2005.
- Institutional drivers: Bank Act amendments in 1980, 1987, 1992, and 1997 contributed to the shift in corporates’ funding mix; mortgage securitization advanced more slowly in Canada than in the United States.

### Evidence from VAR models (monetary shock impulse responses)
- Benchmark VAR specification:
  - Endogenous vector Yt = [y, p, R, S] where y is quarterly GDP, p is the GDP deflator, R is the 3 month T-bill rate, and S is the exchange rate.
  - Exogenous vector Xt = [USy, USR, cp] with US GDP, Federal Funds rate, and an index of world commodity prices.
  - Extended VARs include block of financial variables F: quantity variables (e.g., total loans to businesses and households, securities) and asset price variables (stock and housing prices, spreads).
- Key empirical findings:
  - Full-sample impulse responses to a contractionary monetary shock show a decline in output and a sluggish decline in prices with an initial price spike (“price puzzle”), broadly similar to other industrialized countries.
  - Statistical tests indicate structural breaks in VAR coefficients around regulatory change dates; break-test dates examined include 1988:1 and 1993:2 (and intermediate date 1991:1 with qualitatively similar results). Likelihood-ratio tests reject parameter stability at the 5 percent level for all VAR models.
  - Subsample differences:
    - First subsample (1971:1–1991:1): Impulse responses are more tightly estimated; statistically significant decline in output between the 5th and 10th quarters; prices fall gradually and persistently; exchange rate appreciates sharply on impact and returns to baseline.
    - Second subsample (1991:1–2005:2): Impulse responses show a very different shape and are largely statistically insignificant.
  - Including financial variables (notably business loans and asset prices) in the VAR:
    - Produces a tighter and improved characterization of transmission for the first subsample.
    - Makes the appreciation of the exchange rate after an interest rate increase statistically significant in the first subsample.
    - Increases the contribution of monetary shocks to output variance: in the first subsample monetary shocks explain close to 35-40 percent of the variance of output in the financial models, while they explain around 27 percent in the baseline model.
  - Volatility of monetary shocks:
    - Standard deviation of the monetary shocks in the second subsample is systematically smaller than in the first subsample, consistent with an increased role for the systematic component of monetary policy (inflation targeting, greater transparency).

### Evidence from structural IS / aggregate-demand models
- Motivation: Test how financial disintermediation affects parameters of aggregate demand — specifically the interest-rate elasticity of the output gap.
- Model specifications estimated:
  - Closed-economy IS (Rudebusch and Svensson, 1999) with coefficient on real interest rate allowed to vary with DIF (ratio of direct to indirect finance). Two DIF measures used:
    - DIF1 = ratio of securities to business loans.
    - DIF2 = ratio of securities to total loans (captures rising household lending).
  - Forward-looking/New Keynesian IS variants and open-economy IS variants including real exchange rate gap and U.S. GDP gap.
- Main estimation results:
  - Closed-economy IS:
    - Interest-rate elasticity of the output gap is significantly different from zero in the full sample, but significance concentrated in the more recent subsample (contrasting with the VAR result that shows weaker responses in the recent sample for unexpected shocks).
    - No clear evidence that disintermediation (DIF1 or DIF2) affected the interest-rate elasticity in the closed-economy specification; coefficients on DIF not statistically significant in those specifications.
    - Contextual numeric values from text: “For the average values of the variables DIF1 and DIF2, the interest elasticity coefficients become -0.33 and -0.46, respectively.” Also elsewhere for a particular specification: “When calculated using the average sample values of DIF1 and DIF2 both elasticities are respectively -0.15 and -0.09, respectively.” (These appear in different model contexts described in the text.)
  - Forward-looking/New Keynesian variants:
    - Replacing expected future gap with actual future gap yields unsatisfactory results: future gap significant but real interest rate loses significance; coefficients on DIF not significant.
  - Open-economy IS (includes real exchange rate gap and U.S. GDP gap):
    - Provides evidence that disintermediation has contributed to changes in monetary transmission.
    - Interest-rate elasticity of aggregate demand is not statistically different from zero for the full sample (1971–2005) when specified linearly, but becomes significant when modeled as a function of DIF.
    - Table 9 results noted in the text: coefficients on DIF1 and DIF2 are significant in the second half of the sample and yield relatively larger interest elasticity estimates (respectively -0.46 and -0.48 for the average sample values of DIF1 and DIF2).

### Conclusions and policy implications
- Timing and drivers of structural change:
  - Monetary transmission in Canada changed markedly since the late 1980s; VAR models show a clear break beginning in 1988 after regulatory changes that initiated disintermediation.
  - Both changes in financial structure (disintermediation, increased household credit) and institutional changes in monetary policy (inflation targeting, greater systematic policy) contributed to the evolving transmission mechanism.
- Role of financial variables:
  - Inclusion of financial variables in VARs improves characterization of transmission in the pre-break period and increases the share of output variance explained by monetary shocks in financial VARs.
  - Structural IS estimates indicate the process of disintermediation contributed to changes in the interest-rate elasticity of aggregate demand in the 1990s — often implying increased responsiveness (larger absolute value of the elasticity) associated with greater use of market-based finance.
  - Possible mechanism: relationship-based bank lending is less interest-rate sensitive than price-sensitive market funding; disintermediation (more direct finance) can increase aggregate-demand responsiveness to interest-rate changes.
- Current and future considerations:
  - Monetary policy appears to have become more effective in the 1990s when measured as the average impact of interest rate changes on the output gap or aggregate demand.
  - This increased effectiveness may be undermined by the more recent increase in household borrowing and the relative decline in corporate securities issuance (declining ratio of direct to indirect finance since 2002 for the corporate sector).
  - Policy recommendation: Monetary policy analysis would benefit from more explicit consideration of the evolving roles of financial markets and intermediaries, given their measurable influence on transmission parameters.

*Source: _wp0684 - 8. Impulse-Response Functions from VAR including Asset Prices and Business Loans (PDF chapter).*

### References

### _wp0684 - References

### References cited
- Allen, Franklin, and Douglas Gale, 2000, Comparing Financial Systems (Cambridge, Massachusetts: MIT Press).
- Atoyan, Ruben, 2005, “Econometric Investigation of Policy Preference Evolution: The Bank of Canada Case” (unpublished; Washington: International Monetary Fund).
- Bean, Charles, Jens Larsen, and Kalin Nikolov, 2002, “Financial Frictions and the Monetary Transmission Mechanism: Theory, Evidence, and Policy Implications,” ECB Working Paper No. 113 (Frankfurt: European Central Bank).
- Berg, Andrew, Philippe Karam, and Douglas Laxton, 2006, “A Practical Model-Based Approach to Monetary Policy Analysis—Overview,” IMF Working Paper forthcoming (Washington: International Monetary Fund).
- Bernanke, Ben, and Mark Gertler, 1995, “Inside the Black Box: The Credit Channel of Monetary Policy Transmission,” Journal of Economic Perspectives, Vol. 9, No. 4, pp. 27–48.
- Bernanke, Ben, and Mark Watson, 1997, “Systematic Monetary Policy and the Effect of Oil Price Shocks,” Brookings Papers on Economic Activity, No.1, pp. 91–142.
- Bernanke, Ben, and Simon Gilchrist, 1999, “The Financial Accelerator in a Quantitative Business Cycle Framework,” Handbook of Macroeconomics, ed. by J.B. Taylor and M. Woodford (Amsterdam: Elsevier Science).
- Boivin, Jean, and Marc Giannoni, 2002, “Assessing Changes in the Monetary Transmission Mechanism: A VAR Approach,” Federal Reserve Bank of New York Economic Policy Review, pp. 97–111.
- Calmes, Christian, 2004, “Regulatory Changes and Financial Structure: The Case of Canada,” Bank of Canada Working Paper, 2004–26, Ottawa.
- Cecchetti, Stephen G., “Legal Structure, Financial Structure and the Monetary Policy Transmission Mechanism,” Federal Reserve Bank of New York Economic Policy Review, July 1999, pp. 19–28.
- Clarida, Richard, Jordi Gali, and Mark Gertler, 1999, “The Science of Monetary Policy,” Journal of Economic Literature, Vol. 37, No. 4, pp. 1661–1707.
- Clarida, Richard, Jordi Gali, and Mark Gertler, 2001, “Optimal Monetary Policy in Open versus Closed Economies: An Integrated Approach,” American Economic Review, Vol. 91, No. 2, pp. 248–252.
- Christiano, Lawrence, Martin Eichenbaum, and Charles Evans, 1999, “Monetary Policy Shocks: What Have We learned and to What End?” Handbook of Macroeconomics, ed. by J.B. Taylor and M. Woodford (Amsterdam: Elsevier Science).
- Estrella, Arturo, 2001, “Financial Innovation and the Monetary Transmission Mechanism,” mimeo, Federal Reserve Bank of New York.
- Estrella, Arturo, 2002, “Securitization and the Effectiveness of Monetary Policy,” Federal Reserve Bank of New York Economic Policy Review.
- Favero, Carlo, 2001, Applied Macroeconometrics (London: Oxford University Press).
- Freedman, Charles, and Walter Engert, 2003, “Financial Developments in Canada: Past Trends and Future Challenges,” Bank of Canada Review, Summer, p. 3–16.
- Fung, Ben S.C., and Mingwei Yuan, 1999, “Measuring the Stance of Monetary Policy,” in Money, Monetary Policy and Transmission Mechanisms—Proceedings of a Conference held at the Bank of Canada.
- Kashyap, Anil K., and Jeremy C. Stein, 1994, “Monetary Policy and bank Lending”, Monetary Policy, NBER Studies in Business Cycles, ed. by N. Gregory Mankiw, Vol. 29, pp. 221–61 (Chicago: University of Chicago Press).
- Levin, Andrew T., Fabio Natalucci and Egon Zakrajsek, 2004, “The Magnitude and Cyclical Behavior of Financial Market Frictions,” Finance and Economic Discussion Series, 2004-70, Federal Reserve Board, Washington D.C.
- McCallum, Bennett, and Edward Nelson, 1999, “Nominal Income Targeting in an Open-Economy Optimizing Model,” Journal of Monetary Economics, 43, pp. 553–78.
- Morsink, James, and Tamim Bayoumi, 2001, “A Peek Inside the Black Box: The Monetary Transmission Mechanism in Japan”, IMF Staff Papers, Vol. 48, No. 1, pp. 22–57.
- Peersman, Gert, and Frank Smets, 2001, “The Monetary Transmission Mechanism in the Euro Area: More Evidence from VAR Analysis,” European Central Bank WP No. 91.
- Rudebusch, Glenn D., and Lars E. O. Svensson, 1999, “Policy Rules for Inflation Targeting,” Monetary Policy Rules, ed. by John B. Taylor (Chicago: University of Chicago Press).
- Stock, James H., and Mark W. Watson, 2003, “Has the Business Cycle Changed and Why?,” in NBER Macro Annual 2000, ed. by M. Gertler and K. Rogoff (Cambridge Massachussets: MIT Press).
- Stock, James H., and Mark W. Watson, 2005, “Understanding Changes in International Business Cycle Dynamics,” Journal of the European Economic Association, Vol. 3, No.5, pp. 968–1006.

### Structural break tests in VAR models (Canada)
- Table 1. Tests for Break in VAR Model in 1988:1 (F-Statistics; 1971:1–1988:1; 1988:1–2005:2)
  - Selected single-equation F-statistics:
    - GDP: 1.777 *, 2.008 *, 2.217 *, 2.221 *, 2.233 *
    - Prices: 1.118 1.296 1.004 2.094 * 1.892 *
    - Ints. Rate: 1.755 * 1.953 * 1.436 1.565 1.836 *
    - Exch.Rate: 1.894 * 2.024 * 2.582 * 3.135 * 3.116 *
    - Loans: 3.751 * ...... 4.069 *
    - CP-Spread: ...... 1.275 ...... 
    - Asset Prices: ......... 1.972 * 1.636 *
  - LR-Test values: 293.043 3.049 8.342 4.760 6.4
  - Notes:
    - LR test is Chi-square with degrees of freedom equal to number of restrictions (100 for benchmark, 120 for financial models).
    - Chi-square values for 120 degrees of freedom: 146.6 (5%) and 159.0 (1%).
    - F(24, 80) value for 5 percent significance is 1.65.
- Table 2. Tests for Break in VAR Model in 1993:2 (F-Statistics; 1971:1–1993:2; 1993:2–2005:2)
  - Selected single-equation F-statistics:
    - GDP: 1.487 1.859 * 1.355 1.479 2.214 *
    - Prices: 1.853 * 2.418 * 1.825 * 2.197 * 2.193 *
    - Ints. Rate: 1.114 0.929 1.179 0.959 0.637
    - Exch. Rate: 1.567 1.699 * 1.431 1.739 * 2.237 *
    - Loans: ... 3.060 * ...... 2.896 *
    - Asset Prices: ......... 1.406 1.237
  - LR-Test values: 328.544 5.848 9.341 6.860 9.6
  - Same LR and F-test notes as Table 1.
- Table 3. Tests for Break in VAR Model in 1991:1 (F-Statistics; 1971:1–1991:1; 1991:1–2005:2)
  - Selected single-equation F-statistics:
    - GDP: 1.276 1.615 1.239 1.406 1.946
    - Prices: 1.876 2.375 1.990 2.392 2.341
    - Ints. Rate: 1.500 1.186 1.028 0.952 1.038
    - Exch.Rate: 1.580 1.810 1.505 1.944 2.137
    - Loans: ... 4.114 ...... 4.479
    - Asset Prices: ......... 1.482 1.240
  - LR-Test values: 284.243 0.144 7.036 6.156 0.2
  - Same LR and F-test notes as above.

### Variance decompositions and shock statistics
- Table 4. Variance Decompositions of GDP to Monetary Shock (in percent)
  - Baseline Model (Quarter): Full Sample / 1971-1991 / 1991-2005
    - 4: 0.81.6 1.3
    - 8: 9.22 0.41 1.1
    - 12: 15.6 27.1 1.2
    - 16: 15.9 26.9 1.2
  - Model with Loans
    - 4: 1.6 0.6 9.1
    - 8: 9.8 22.7 8.7
    - 12: 14.5 33.7 9.2
    - 16: 15.5 32.2 19.2
  - Model with Asset Prices
    - 4: 1.2 2.1 2.4
    - 8: 8.7 22.8 2.0
    - 12: 16.2 35.3 3.7
    - 16: 18.7 34.6 4.2
  - Model with Loans and Asset Prices
    - 4: 1.4 0.5 4.5
    - 8: 9.8 25.0 3.3
    - 12: 16.7 39.6 3.2
    - 16: 19.2 35.1 4.0
  - Source: Author's calculations.
- Table 5. Changes in the Standard Deviations of Shocks (Interest Rate Equation)
  - Model / 1971–1991 / 1991–2005 / SD2/SD1
    - Baseline VAR: 0.49 0.41 0.84
    - VAR with Loans: 0.49 0.46 0.94
    - VAR with Asset Prices: 0.52 0.44 0.85
    - VAR with Loans and Asset Prices: 0.49 0.44 0.90
  - Source: Author's calculations.
  - Note: Last column shows the ratio of the standard deviation of the monetary shock in the second and first subsamples.

### IS equation estimates (Canada)
- Table 6. Closed Economy IS Equation (1971:1–2005:2)
  - Models: Closed Economy IS Equation / Model with DIF1 / Model with DIF2
  - α0: 0.473 0.483 0.476 (p-values: (0.300)(0.325)(0.309))
  - α1: 1.141 1.141 1.141 (p-values: (0.000)(0.000)(0.000))
  - α2: -0.298 -0.298 -0.298 (p-values: (0.000)(0.000)(0.000))
  - α30: -0.156 -0.137 -0.147 (p-values: (0.029)(0.684)(0.592))
  - α31: ... -0.010 -0.015 (...(0.954)(0.974))
  - R2: 0.818 0.818 0.818
  - DW: 2.00 2.00 2.00
  - Notes:
    - DIF1 is the ratio of direct to indirect finance for the corporate sector; DIF2 is the ratio of direct finance to total lending--including the household sector.
    - p-values in parenthesis.
- Table 7. Closed Economy IS Equation (Sample break in 1988:1)
  - Models report parameters for (1971:1–1988:1) and (1988:1–2005:2) under Model with DIF1 and Model with DIF2
  - Selected parameter examples (α coefficients and p-values preserved as in source):
    - α0 ranges from 0.216 to 0.723 across subsamples and models (p-values reported accordingly).
    - α1 is consistently highly significant: 1.029 1.425 0.971 1.418 0.976 1.411 (p-values (0.000) in all cases).
    - α2 examples: -0.273 -0.511 -0.225 -5.507 -0.244 -0.503 (p-values included).
  - R2 for subsamples: 0.687 0.938 0.708 0.938 0.704 0.938
  - DW: values reported (1.98, 2.04, 2.00, 2.03, 1.96, 2.02)
  - Notes:
    - Benchmark IS Equation.
    - p-values in parenthesis.
    - DIF1 and DIF2 defined as above.

- Table 8. Estimates of Open Economy IS Equation (1971:1–2005:2)
  - Model with DIF1 / Model with DIF2
  - α0: -0.096 0.137 -0.054 (p-values (0.837)(0.768)(0.906))
  - α1: 1.021 0.972 0.979 (p-values (0.000)(0.000)(0.000))
  - α2: -0.252 -0.229 -0.229 (p-values (0.001)(0.003)(0.003))
  - α30: -0.072 0.874 0.658 (p-values (0.312)(0.022)(0.031))
  - α31: ... -0.505 -1.191 ...(0.012)(0.014)
  - α4: -0.027 -0.018 -0.019 (p-values (0.628)(0.747)(0.737))
  - α5: 0.440 0.628 0.602 (p-values (0.000)(0.000)(0.000))
  - R2: 0.838 0.845 0.845
  - DW: 1.79 1.76 1.78
  - Notes:
    - DIF1 and DIF2 definitions as above.
    - p-values in parenthesis.

- Table 9. Open Economy IS Equation (Sample break in 1988:1)
  - Open Economy IS Equation parameters reported for (1971:1–1988:1) and (1988:1–2005:2) across Model with DF1 and Model with DF2
  - Selected parameters:
    - α0 examples: -1.567 0.337 -1.495 1.209 -1.292 1.346 (p-values (0.058)(0.428)(0.072)(0.043)(0.142)(0.025))
    - α1 examples: 0.747 1.303 0.740 1.172 0.734 1.143 (p-values (0.000) for all listed)
    - α2 examples: -0.204 -0.427 -0.187 -0.333 -0.192 -0.313 (p-values shown)
    - α5 examples: 1.025 0.375 0.956 0.594 0.999 0.626 (p-values include (0.000), (0.017), (0.000), (0.002), (0.000), (0.001))
  - R2: 0.771 0.943 0.773 0.947 0.773 0.948
  - DW: 1.80 1.93 1.82 1.82 1.80 1.80
  - Notes:
    - p-values in parenthesis.
    - DIF1 and DIF2 definitions as above.

### Financial structure and time series visuals (Canada)
- Figure 1. Nonfinancial Corporate Financing Sources, 1971–2005 (stocks outstanding, in percent of total)
  - Series plotted: Loans, Equities, Other, Bonds, debentures, and short-term paper (percent composition over 1971–2005).
  - Source: Bank of Canada.
- Figure 2. Credit Flows, 1971–2005
  - Panel displays credit flows in billion Canadian dollars for Bonds and debentures, Loans, Commercial paper over 1971–2005 and 1991–2005 subperiod.
  - Source: Bank of Canada.
- Figure 3. Ratio of Direct to Indirect Private Lending, 1971–2005
  - Values range from 0.0 to 1.0 across 1971–2005.
  - Source: Bank of Canada.
- Figure 4. Outstanding Credit, 1971–2005 (billion Canadian dollars)
  - Series: Household mortgages, Business loans, Consumer credit; values range from 0 to 700 (billion Canadian dollars) across 1971–2005.
  - Source: Bank of Canada.

### VAR impulse-response results (Canada)
- Figures 5–8: Impulse-Response Functions for various VAR specifications (Benchmark VAR; VAR including Business Loans; VAR including Asset Prices; VAR including Asset Prices and Business Loans)
  - Panels show responses of:
    - GDP_R to TBILL3MO
    - GDP_P to TBILL3MO
    - TBILL3MO to TBILL3MO
    - XR_INDEX_LEV to TBILL3MO
  - Responses are presented for:
    - Full sample (1971-Q1 to 2005-Q2)
    - 1971-Q1 to 1991-Q1
    - 1991-Q1 to 2005-Q2
  - Scales shown include small magnitude responses (e.g., axes spanning approximately -.004 to .004 for GDP_R / GDP_P responses) and larger scales for exchange rate responses (e.g., -2 to 3).
  - Source: Fund staff calculations.

*Source: _wp0684 - References (PDF content provided).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0684.pdf_