## 4.1 Calibrated parameters and prior distributions (_wp1674)

## Source details

**Canonical URL:** [4.1 Calibrated parameters and prior distributions (_wp1674)](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1674.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1674.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1674.pdf.json)

---

### Calibration — calibrated parameter values
- Patient households’ discount factor βP = 0.9963
- Impatient households’ discount factor βI = 0.975
- Entrepreneurs’ discount factor βE = 0.975
- Inverse of the Frish’s elasticity φ = 2.98
- Share of unconstrained households μ = 1/3
- Weight of housing in the households’ utility function εh = 0.2
- Capital share in the production function α = 0.35
- Depreciation rate of physical capital δ = 0.025
- Markup in the goods market εyεy−1 = 6 (delivers a markup of 20 percent in steady state)
- Markup in the labor market εlεl−1 = 5
- LTV for new mortgages mI = 0.85
- LTV for firm loans mE = 0.25
- Risk weighted/loan ratio in steady state (requirement) νb = 0.12
- Markdown on deposit rate εdεd−1 = -1.1
- Markup on loans to households εbHεbHεbH−1 = 3.2
- Markup on loans to entrepreneurs εbEεbEεbE−1 = 2.4
- Amortization period for existing mortgages calibrated to 50 years
- Mortgage interest deductability calibrated to 30 percent
- Elasticity of substitution of deposits εdt = -1.0738 (matches steady-state spread between deposit rate and interbank rate in the range of 50-100 bps)
- Minimum bank (risk weighted) capital ratio = 0.12
- Bank risk weight for corporate loans = 50 percent (floor)
- Bank risk weight for mortgages = 25 percent (floor)
- Bank capital “depreciation” rate = 0.0658 (ensures bank capital to risk weighted loans ratio = 0.19)

Notes on calibration choices:
- LTV for mortgages set at 0.85 in line with the current LTV cap for new mortgages set by the Swedish FSA.
- Entrepreneur LTV mE calibrated to 0.25 so that the ratio between mortgages and loans to entrepreneurs is about 1.3 as indicated by lending data from Swedish monetary financial institutions.
- Capital share and depreciation chosen as standard in the literature; markup values similar to Gerali et al. (2010) with εt set at 5.

### Prior distributions (specification)
- Priors guided by Adolfson et al. (2008), Lindé et al. (2009), Gerali et al. (2010), Walentin (2014); where lacking, priors kept relatively uninformative.
- Persistence parameters: beta-distribution with prior mean = 0.75 and standard deviation = 0.05.
- Habit persistence in consumption: ah = aP = aI = aE with prior mean = 0.65 and standard deviation = 0.1.
- Monetary policy rule priors: ρR prior mean = 0.8; Φπ prior mean = 1.7; Φy prior mean = 0.13.
- LTV implementation persistence ρmi prior mean = 0.75 (implying implementation takes time).

### Estimation details (posterior sampling)
- Posterior draws obtained via random walk Metropolis algorithm.
- Run 2 parallel chains each of length 12,000,000.
- Scale factor set to deliver acceptance rates near 0.3.
- Convergence assessed using Brooks and Gelman (1998) convergence statistics (individual and multivariate).
- Tables A1 and A2 report posterior means and 90 per cent probability intervals; Figure 7 reports prior and posterior marginal densities.

### Posterior findings (high-level)
- Monetary policy: relatively high degree of interest rate inertia; higher estimated responsiveness to inflation and output (higher Φπ and Φy) than Adolfson et al. (2008).
- Nominal rigidities: wage stickiness is much more significant than price stickiness (wages renegotiated about every three years; collective bargaining objectives).
- Bank interest rate stickiness: deposit rates adjust more rapidly than loan rates to changes in policy rates.
- Mortgage versus firm lending rates: mortgage rates adjust faster than lending rates to firms (reflecting high share of variable rate mortgage contracts in Sweden).

### Model properties — impulse responses to a contractionary monetary policy shock
- Shock: exogenous 100 basis point hike in the monetary policy (repo) rate.
- Key transmission channels:
  - Balance sheet channel: tighter policy lowers housing prices, restricting household borrowing capacity and amplifying shock.
  - Amortization requirements: mortgage debt is “sticky” (cannot be re-optimized each period); amplifies macro effects of rate increases while dampening impact on household debt stock.
  - Banking sector: stickiness of lending rates dampens pass-through to retail rates; bank capital constraint introduces wedge between lending, deposit, and policy rates.
  - International saving choice: households can save in domestic banks or purchase foreign bonds, affecting transmission.

Impulse responses (qualitative magnitudes reported):
- Following a 100 basis point rise in interest rates:
  - Output falls by about 0.6 percent relative to steady state.
  - Inflation falls by about 0.2 percent relative to steady state.
  - Loans to households and firms decline due to lower asset prices and higher real rates.
  - Bank loan rates increase much less than the policy rate; deposit rate increases by almost the full 100 basis points (reducing banks’ interest rate margins).
  - Bank capital initially falls (lower profitability) then increases as margins recover.
  - Endogenous policy response: policy rate would fall below steady state subsequently to stimulate the economy.

### Policy-relevant comparative scenario (amortization and LTV)
- Baseline: Amortization = 50 years, LTV = 85%
- Alternative: Amortization = 30 years, LTV = 80%
- Finding: With lower household debt (faster amortization and lower LTV cap), consumption response to interest rate shocks is smaller because debt service burden rises less for a given rate shock.

### Exploring macroprudential policies — experiment approach
- Permanent change in one macroprudential instrument alters the economy’s structure; study transition paths from one steady state to another.
- Important implementation note: LTV requirement affects only new mortgages and, given currently low amortization rates, can take years to work through the existing mortgage stock. Analysis therefore distinguishes short-term dynamics and long-term new steady state.

### Loan-to-Value (LTV) cap — scenario: 85 → 80 percent
Short run impacts:
- Borrowers’ consumption falls by about 0.2 percent one year after LTV lowered.
- Savers’ consumption falls by less; aggregate consumption falls by 0.02 percent one year after the shock.
- GDP remains lower by 0.1 percent three years after the shock.
- Mechanisms: only new housing acquisition affected initially; falling house prices reduce collateral and reinforce initial tightening.

New steady state impacts:
- Borrower debt declines by about 10 percent cumulatively.
- House prices fall by about 1.5 percent.
- Borrower consumption of goods permanently higher by about 1.8 percent (lower debt service burden, partially offsetting lower housing services consumption).
- Saver consumption permanently lower by about 0.4 percent (driven by bank profits falling by about 4 percent).
- Bank deleveraging cuts loans to firms by about 1.2 percent; output about 0.5 percent lower in new steady state.

### Amortization requirements — scenario: Amortization 50 → 45 years (5-year reduction)
Short run impacts:
- Borrower consumption falls by about 0.03 percent by 4 years after the shocks.
- Monetary authority responds: repo rate lowered by almost 1 percentage point, cushioning effects.
- Aggregate: GDP reduced by 0.02 percent; inflation reduced by almost 0.05 percent.

New steady state impacts:
- Borrower debt falls by about 10 percent.
- Household DTI ratio falls by similar amount.
- House prices down by 0.5 percent.
- Borrower housing stock about 0.5 percent lower.
- Borrower consumption permanently higher by about 1.7 percent.
- Saver consumption declines by about 0.4 percent (bank profit down by almost 5 percent; credit to firms down by more than 1 percent).
- Output lowered by about 0.4 percent.

Comparative transitional dynamics (LTV vs amortization):
- Tightening amortization: household debt falls almost linearly over time.
- Reducing LTV cap: household debt fall accelerates, reinforced by falling house prices.
- Implication: borrower consumption falls faster in the short term under an LTV cap reduction than under amortization tightening.

### 6.3 Tax deductibility of mortgage interest

### Mechanisms and overview
- A reduction in the tax deductibility of mortgage interest payments increases the cost of servicing a mortgage, tightening households’ budget constraint (negative income effect).
- A reduction also makes debt-financed housing purchases more costly relative to the price of consumption goods (substitution effect).
- The negative income and substitution effects have opposing implications for household consumption:
  - Negative income effect → lower household consumption.
  - Substitution effect → households consume more as consumption becomes relatively cheaper than housing.
- The exercise illustrated is a permanent 5 percentage point reduction in tax deductibility from 30 to 25 percent.

### Short-run impacts (following a 5 percentage point reduction to 25 percent)
- House prices fall by about 0.6 percent in the near term.
- Household debt falls in the short run.
- Borrower consumption increases in the short run because:
  - Consumption becomes cheaper relative to housing (substitution effect).
  - The simulation assumes the government fully redistributes the savings from the reduction in tax deductions, partly offsetting negative income effects.
- Savers:
  - Increase housing investment as prices decline.
  - Benefit from higher transfers, leading to higher consumption.
- Aggregate outcomes in the short run:
  - Slightly higher aggregate consumption.
  - Slightly higher output.
  - Slightly higher inflation.

### New steady state impacts
- Borrower household debt is lowered by about 2.2 percent.
- Borrower debt-to-income (DTI) ratio falls by a similar magnitude.
- Borrower consumption is higher by about 0.1 percent.
- Borrower housing stock is lower by 2 percent.
- Savers’ consumption increases by about 0.1 percent, driven by higher transfers; transfers are sufficient to offset the decline in bank profit so that savers’ consumption does not decline.
- Overall impact on total consumption depends on government use of savings from reduced tax deductibility.

### Redistribution vs no redistribution (initial 20 quarters)
- Baseline: full redistribution of savings from the reduction in tax deductibility via lump-sum transfers to households.
- No redistribution: savings are not redistributed.
- Key qualitative finding:
  - Full redistribution of savings mitigates the negative impact on consumption from reducing tax deductibility.

### 8.2 An alternative specification of welfare

### Methodology and scope
- Alternative welfare measure: sum of households’ utilities (savers and borrowers).
  - Factors included: consumption, leisure, housing stock of savers and borrowers in different steady states.
  - Difference from baseline (Section 7): does not consider stochastic shocks to the economy in the same way as the baseline welfare definition.
- Used to evaluate how saver and borrower utilities change with different intensities of macroprudential measures (combinations of amortization requirements and loan-to-value (LTV) caps).

### Key findings from the alternative specification
- Welfare response to measures:
  - Welfare would improve with less stringent LTV caps.
  - Welfare reaches a peak with an amortization requirement of 40 years.
- Interpretation offered:
  - Taken at face value, result would imply that in Sweden it could be optimal to calibrate the amortization requirement to 40 years, while allowing households the option to refinance by leaving the LTV cap on new mortgage loans at high values.
  - However, authors prefer the baseline welfare function because it fully considers all the future shocks to the economy.

### Figures (descriptions)
- Figure 23: Aggregate of saver’s and borrower’s utilities — amortisation grid across 10 20 30 40 50 60 70 80 90 100 and LTV grid 0.6–1; vertical axis values (utility) from -2.65 to -2.45. Caption: “The figure depicts the sum of saver’s and borrower’s utilities in different steady states that are characterized by a combination of amortization requirements and loan-to-value caps. The light yellow color corresponds to the highest level of utility.”
- Figure 24: Household debt over amortization requirements and LTV — same grids; vertical axis values shown from 0 to 3 with ticks at 0, 0.5, 1, 1.5, 2, 2.5, 3. Caption: “The figure compares borrower’s (mortgage) debt level across the same, as in Figure 23, distribution of macroprudential measures.”

### Policy implications and conclusions
- Monetary policy versus demand-side macroprudential instruments:
  - A monetary policy shock initially results in a rise in households’ debt-to-income ratio because household debt is very “sticky” and responds more slowly than household income.
  - Over time, the net effect of an increase in the policy rate is a reduction of the household debt-to-income ratio.
  - Demand-side macroprudential instruments reduce the household debt-to income ratio more effectively than monetary policy, with more limited adverse effects on consumption over the short term.
  - Conclusion: macroprudential policies are better suited to address financial stability risks; monetary policy has a higher cost in terms of foregone consumption.
- Effects of tightening macroprudential measures:
  - Tighter demand-side macroprudential policies in Sweden could significantly reduce households’ DTI ratios, while decreasing consumption and output by a small amount.
  - The housing stock for borrowing households would be slightly lower.
  - A tighter supply-side measure (mortgage risk weights) could lead to an increase in households’ DTI ratios because its impact on output, thus households’ income, could be much larger than its impact on household debt.
  - Macroprudential policies affect consumption, distribution of the housing stock, and other sectors such as the banking sector.
- Welfare analysis main takeaways (policymaker guidance):
  - First: tighter LTV requirements on new mortgages and higher mortgage risk weights improve welfare, although with diminishing returns.
  - Second: the sequence with which macroprudential measures are introduced matters; it is optimal to tighten amortization and reduce tax deductibility only when LTV on new mortgages falls below 80 percent.
  - Third: a mix of the macroprudential measures studied is needed to deliver the maximum level of welfare.
  - Tighter macroprudential policies lead to a more muted response of the economy to banking system shocks (including shocks to bank capital and monopoly power in setting deposit rates), indicating benefits for safeguarding bank intermediation and support to real-economy financing needs.

### Model extensions noted by authors
- Potential extensions for future work:
  - Introduce distribution of LTV ratios, household default, and a housing construction section.
  - Introduce an additional DTI requirement and study its interaction with the existing LTV constraint.
  - Improve model estimation by utilizing external variables (foreign GDP, consumption, trade, financial flow data).

*Source: 4.1 Calibrated parameters and prior distributions; 6.3 Tax deductibility of mortgage interest; 8.2 An alternative specification of welfare — _wp1674*

### 4.1    Calibrated parameters and prior distributions

### 4.1    Calibrated parameters and prior distributions

### Calibration — calibrated parameter values
- Patient households’ discount factor βP = 0.9963
- Impatient households’ discount factor βI = 0.975
- Entrepreneurs’ discount factor βE = 0.975
- Inverse of the Frish’s elasticity φ = 2.98
- Share of unconstrained households μ = 1/3
- Weight of housing in the households’ utility function εh = 0.2
- Capital share in the production function α = 0.35
- Depreciation rate of physical capital δ = 0.025
- Markup in the goods market εyεy−1 = 6 (delivers a markup of 20 percent in steady state)
- Markup in the labor market εlεl−1 = 5
- LTV for new mortgages mI = 0.85
- LTV for firm loans mE = 0.25
- Risk weighted/loan ratio in steady state (requirement) νb = 0.12
- Markdown on deposit rate εdεd−1 = -1.1
- Markup on loans to households εbHεbHεbH−1 = 3.2
- Markup on loans to entrepreneurs εbEεbEεbE−1 = 2.4
- Amortization period for existing mortgages calibrated to 50 years
- Mortgage interest deductability calibrated to 30 percent
- Elasticity of substitution of deposits εdt = -1.0738 (matches steady-state spread between deposit rate and interbank rate in the range of 50-100 bps)
- Minimum bank (risk weighted) capital ratio = 0.12
- Bank risk weight for corporate loans = 50 percent (floor)
- Bank risk weight for mortgages = 25 percent (floor)
- Bank capital “depreciation” rate = 0.0658 (ensures bank capital to risk weighted loans ratio = 0.19)

Notes on calibration choices:
- LTV for mortgages set at 0.85 in line with the current LTV cap for new mortgages set by the Swedish FSA.
- Entrepreneur LTV mE calibrated to 0.25 so that the ratio between mortgages and loans to entrepreneurs is about 1.3 as indicated by lending data from Swedish monetary financial institutions.
- Capital share and depreciation chosen as standard in the literature; markup values similar to Gerali et al. (2010) with εt set at 5.

### Prior distributions (specification)
- Priors guided by Adolfson et al. (2008), Lindé et al. (2009), Gerali et al. (2010), Walentin (2014); where lacking, priors kept relatively uninformative.
- Persistence parameters: beta-distribution with prior mean = 0.75 and standard deviation = 0.05.
- Habit persistence in consumption: ah = aP = aI = aE with prior mean = 0.65 and standard deviation = 0.1.
- Monetary policy rule priors: ρR prior mean = 0.8; Φπ prior mean = 1.7; Φy prior mean = 0.13.
- LTV implementation persistence ρmi prior mean = 0.75 (implying implementation takes time).

### Estimation details (posterior sampling)
- Posterior draws obtained via random walk Metropolis algorithm.
- Run 2 parallel chains each of length 12,000,000.
- Scale factor set to deliver acceptance rates near 0.3.
- Convergence assessed using Brooks and Gelman (1998) convergence statistics (individual and multivariate).
- Tables A1 and A2 report posterior means and 90 per cent probability intervals; Figure 7 reports prior and posterior marginal densities.

### Posterior findings (high-level)
- Monetary policy: relatively high degree of interest rate inertia; higher estimated responsiveness to inflation and output (higher Φπ and Φy) than Adolfson et al. (2008).
- Nominal rigidities: wage stickiness is much more significant than price stickiness (wages renegotiated about every three years; collective bargaining objectives).
- Bank interest rate stickiness: deposit rates adjust more rapidly than loan rates to changes in policy rates.
- Mortgage versus firm lending rates: mortgage rates adjust faster than lending rates to firms (reflecting high share of variable rate mortgage contracts in Sweden).

*Source: 4.1 Calibrated parameters and prior distributions — _wp1674*

### Model properties — impulse responses to a contractionary monetary policy shock
- Shock: exogenous 100 basis point hike in the monetary policy (repo) rate.
- Key transmission channels highlighted:
  - Balance sheet channel: tighter policy lowers housing prices, restricting household borrowing capacity and amplifying shock.
  - Amortization requirements: mortgage debt is “sticky” (cannot be re-optimized each period); amplifies macro effects of rate increases while dampening impact on household debt stock.
  - Banking sector: stickiness of lending rates dampens pass-through to retail rates; bank capital constraint introduces wedge between lending, deposit, and policy rates.
  - International saving choice: households can save in domestic banks or purchase foreign bonds, affecting transmission.

Impulse responses (qualitative magnitudes reported):
- Following a 100 basis point rise in interest rates:
  - Output falls by about 0.6 percent relative to steady state.
  - Inflation falls by about 0.2 percent relative to steady state.
  - Loans to households and firms decline due to lower asset prices and higher real rates.
  - Bank loan rates increase much less than the policy rate; deposit rate increases by almost the full 100 basis points (reducing banks’ interest rate margins).
  - Bank capital initially falls (lower profitability) then increases as margins recover.
  - Endogenous policy response: policy rate would fall below steady state subsequently to stimulate the economy.

Policy-relevant comparative scenario:
- Baseline: Amortization = 50 years, LTV = 85%
- Alternative: Amortization = 30 years, LTV = 80%
- Finding: With lower household debt (faster amortization and lower LTV cap), consumption response to interest rate shocks is smaller because debt service burden rises less for a given rate shock.

### Exploring macroprudential policies — experiment approach
- Permanent change in one macroprudential instrument alters the economy’s structure; study transition paths from one steady state to another.
- Important implementation note: LTV requirement affects only new mortgages and, given currently low amortization rates, can take years to work through the existing mortgage stock. Analysis therefore distinguishes short-term dynamics and long-term new steady state.

### Loan-to-Value (LTV) cap — scenario: 85 → 80 percent
Short run impacts:
- Borrowers’ consumption falls by about 0.2 percent one year after LTV lowered.
- Savers’ consumption falls by less; aggregate consumption falls by 0.02 percent one year after the shock.
- GDP remains lower by 0.1 percent three years after the shock.
- Mechanisms: only new housing acquisition affected initially; falling house prices reduce collateral and reinforce initial tightening.

New steady state impacts:
- Borrower debt declines by about 10 percent cumulatively.
- House prices fall by about 1.5 percent.
- Borrower consumption of goods permanently higher by about 1.8 percent (lower debt service burden, partially offsetting lower housing services consumption).
- Saver consumption permanently lower by about 0.4 percent (driven by bank profits falling by about 4 percent).
- Bank deleveraging cuts loans to firms by about 1.2 percent; output about 0.5 percent lower in new steady state.

### Amortization requirements — scenario: Amortization 50 → 45 years (5-year reduction)
Short run impacts:
- Borrower consumption falls by about 0.03 percent by 4 years after the shocks.
- Monetary authority responds: repo rate lowered by almost 1 percentage point, cushioning effects.
- Aggregate: GDP reduced by 0.02 percent; inflation reduced by almost 0.05 percent.

New steady state impacts:
- Borrower debt falls by about 10 percent.
- Household DTI ratio falls by similar amount.
- House prices down by 0.5 percent.
- Borrower housing stock about 0.5 percent lower.
- Borrower consumption permanently higher by about 1.7 percent.
- Saver consumption declines by about 0.4 percent (bank profit down by almost 5 percent; credit to firms down by more than 1 percent).
- Output lowered by about 0.4 percent.

Comparative transitional dynamics (LTV vs amortization):
- Tightening amortization: household debt falls almost linearly over time.
- Reducing LTV cap: household debt fall accelerates, reinforced by falling house prices.
- Implication: borrower consumption falls faster in the short term under an LTV cap reduction than under amortization tightening.

### 6.3    Tax deductibility of mortgage interest

### 6.3    Tax deductibility of mortgage interest

### Mechanisms and overview
- A reduction in the tax deductibility of mortgage interest payments increases the cost of servicing a mortgage, tightening households’ budget constraint (negative income effect).
- A reduction also makes debt-financed housing purchases more costly relative to the price of consumption goods (substitution effect).
- The negative income and substitution effects have opposing implications for household consumption:
  - Negative income effect → lower household consumption.
  - Substitution effect → households consume more as consumption becomes relatively cheaper than housing.
- The exercise illustrated is a permanent 5 percentage point reduction in tax deductibility from 30 to 25 percent.

### Short-run impacts (following a 5 percentage point reduction to 25 percent)
- House prices fall by about 0.6 percent in the near term.
- Household debt falls in the short run.
- Borrower consumption increases in the short run because:
  - Consumption becomes cheaper relative to housing (substitution effect).
  - The simulation assumes the government fully redistributes the savings from the reduction in tax deductions, partly offsetting negative income effects.
- Savers:
  - Increase housing investment as prices decline.
  - Benefit from higher transfers, leading to higher consumption.
- Aggregate outcomes in the short run:
  - Slightly higher aggregate consumption.
  - Slightly higher output.
  - Slightly higher inflation.

### New steady state impacts
- Borrower household debt is lowered by about 2.2 percent.
- Borrower debt-to-income (DTI) ratio falls by a similar magnitude (text: "a similar change in the DTI ratio").
- Borrower consumption is higher by about 0.1 percent.
- Borrower housing stock is lower by 2 percent.
- Savers’ consumption increases by about 0.1 percent, driven by higher transfers; transfers are sufficient to offset the decline in bank profit so that savers’ consumption does not decline.
- Overall impact on total consumption depends on government use of savings from reduced tax deductibility.

### Redistribution vs no redistribution
- The simulation compares two scenarios over the initial 20 quarters:
  - Baseline: full redistribution of savings from the reduction in tax deductibility via lump-sum transfers to households.
  - No redistribution: savings are not redistributed.
- Key qualitative finding:
  - Full redistribution of savings mitigates the negative impact on consumption from reducing tax deductibility.
- Illustrated dynamic responses (as described):
  - Saver consumption: baseline (full re-distribution) vs no re-distribution shows deviations from steady state up to roughly -0.06 percent (figure scale).
  - Borrower consumption: baseline shows positive deviations up to about 0.15 percent in the early quarters (figure scale).

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1674.pdf*

### 8.2    An alternative specification of welfare

### 8.2    An alternative specification of welfare

### Methodology and scope
- Alternative welfare measure: sum of households’ utilities (savers and borrowers).
  - Factors included: consumption, leisure, housing stock of savers and borrowers in different steady states.
  - Difference from baseline (Section 7): does not consider stochastic shocks to the economy in the same way as the baseline welfare definition.
- The alternative specification is used to evaluate how saver and borrower utilities change with different intensities of macroprudential measures (combinations of amortization requirements and loan-to-value (LTV) caps).

### Key findings from the alternative specification
- Welfare response to measures:
  - Welfare would improve with less stringent LTV caps.
  - Welfare reaches a peak with an amortization requirement of 40 years (see Figures 23 and 24).
- Interpretation offered in the text:
  - Taken at face value, result would imply that in Sweden it could be optimal to calibrate the amortization requirement to 40 years, while allowing households the option to refinance by leaving the LTV cap on new mortgage loans at high values.
  - However, authors prefer the baseline welfare function because it fully considers all the future shocks to the economy.

### Figures (as described in the source)
- Figure 23: Aggregate of saver’s and borrower’s utilities — interaction between amortization requirements and LTV.
  - Amortisation (years) shown across 10 20 30 40 50 60 70 80 90 100.
  - LTV grid: 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95, 1.
  - Vertical axis values (utility) shown from -2.65 to -2.45 in increments shown on the figure.
  - Caption: “The figure depicts the sum of saver’s and borrower’s utilities in different steady states that are characterized by a combination of amortization requirements and loan-to-value caps. The light yellow color corresponds to the highest level of utility.”
- Figure 24: Household debt over amortization requirements and LTV.
  - Same amortisation and LTV grid as Figure 23.
  - Vertical axis values shown from 0 to 3 with tick marks at 0, 0.5, 1, 1.5, 2, 2.5, 3.
  - Caption: “The figure compares borrower’s (mortgage) debt level across the same, as in Figure 23, distribution of macroprudential measures.”

### Related findings and context from the Conclusion (implications for policy)
- Monetary policy versus demand-side macroprudential instruments:
  - A monetary policy shock initially results in a rise in households’ debt-to-income ratio because household debt is very “sticky” and responds more slowly than household income.
  - Over time, the net effect of an increase in the policy rate is a reduction of the household debt-to-income ratio.
  - Demand-side macroprudential instruments reduce the household debt-to income ratio more effectively than monetary policy, with more limited adverse effects on consumption over the short term.
  - Conclusion: macroprudential policies are better suited to address financial stability risks; monetary policy has a higher cost in terms of foregone consumption.
- Effects of tightening macroprudential measures:
  - Tighter demand-side macroprudential policies in Sweden could significantly reduce households’ DTI ratios, while decreasing consumption and output by a small amount.
  - The housing stock for borrowing households would be slightly lower.
  - A tighter supply-side measure (mortgage risk weights) could lead to an increase in households’ DTI ratios because its impact on output, thus households’ income, could be much larger than its impact on household debt.
  - Macroprudential policies affect consumption, distribution of the housing stock, and other sectors such as the banking sector.
- Welfare analysis main takeaways (policymaker guidance):
  - First: tighter LTV requirements on new mortgages and higher mortgage risk weights improve welfare, although with diminishing returns.
  - Second: the sequence with which macroprudential measures are introduced matters; it is optimal to tighten amortization and reduce tax deductibility only when LTV on new mortgages falls below 80 percent.
  - Third: a mix of the macroprudential measures studied is needed to deliver the maximum level of welfare.
  - Tighter macroprudential policies lead to a more muted response of the economy to banking system shocks (including shocks to bank capital and monopoly power in setting deposit rates), indicating benefits for safeguarding bank intermediation and support to real-economy financing needs.

### Model and extensions noted by authors
- Potential extensions that could affect findings and are added to the research agenda:
  - Introduce distribution of LTV ratios, household default, and a housing construction section.
  - Introduce an additional DTI requirement and study its interaction with the existing LTV constraint.
  - Improve model estimation by utilizing external variables (foreign GDP, consumption, trade, financial flow data).

*Source: _wp1674 - 8.2    An alternative specification of welfare*

### References

### _wp1674 - References

### DSGE models, monetary policy, and open-economy frameworks
- Adolfson, M., Laséen, S., Lindé, J., and Villani, M. (2008). Evaluating an estimated new keynesian small open economy model. Journal of Economic Dynamics and Control, 32(8):2690–2721.
- Gerali, A., Neri, S., Sessa, L., and Signoretti, F. M. (2010). Credit and banking in a dsge model of the euro area. Journal of Money, Credit and Banking, 42(s1):107–141.
- Iacoviello, M. (2005). House prices, borrowing constraints, and monetary policy in the business cycle. American economic review, pages 739–764.
- Iacoviello, M. M. and Neri, S. (2008). Housing market spillovers: Evidence from an estimated dsge model. Bank of Italy Temi di Discussione (Working Paper) No, 659.
- Lindé, J., Nessén, M., and Söderström, U. (2009). Monetary policy in an estimated open-economy model with imperfect pass-through. International Journal of Finance & Economics, 14(4):301–333.
- Smets, F. and Wouters, R. (2003). An estimated dynamic stochastic general equilibrium model of the euro area. Journal of the European economic association, 1(5):1123–1175.
- Quinta, D. and Rabanal, P. (2014). Monetary and macroprudential policy in an estimated dsge model of the euro area. International Journal of Central Banking.

### Housing markets, mortgage markets, and household debt (focus on Sweden)
- Emanuelsson, R. (2015). Supply of housing in sweden. Sveriges Riksbank Economic Review, 2.
- Ho, G. (2015). Housing supply constraints in sweden. IMF Selected Issue Papers, Sweden Country Report 15/330.
- Leonhard, A., Hansson, B., and Johansson, S. (2012). The era of cheap mortgages is at an end. Sweden’s National Housing Credit Guarantee Board’s Market Report.
- Skingsley, C. (2007). Household debt under the microscope. Sabos finansdag, Operaterrassen, Stockholm.
- Sveriges Riskbank, . (2014). From a to z: the swedish mortgage market and its role in the financial system. Riksbank Studies.
- Turk, R. (2015). Housing prices and household debt interactions in sweden. IMF Working Paper 15/276.
- Walentin, K. (2014). Housing collateral and the monetary transmission mechanism. Scandinavian Journal of Economics, 116(3):635–668.

### Macroprudential policy, financial stability, and capital requirements
- Angelini, P., Neri, S., and Panetta, F. (2014). The interaction between capital requirements and monetary policy. Journal of Money, Credit and Banking, 46:1073–1112.
- Bernanke, B. (2015). Should monetary policy take into account risks to financial stability? Ben Bernanke’s blog, Brookings Institution.
- Chen, J. (2014). An overview of macroprudential policy tools. IMF Selected Issue Papers, Sweden Country Report 14/262.
- Claessens, S. (2014). An overview of macroprudential policy tools. IMF Working Paper 14-214.
- IMF. (2015). Monetary policy and financial stability. IMF Policy Papers.
- Svensson, L. (2016). Cost-benefit analysis of leaning against the wind: are costs larger also with less effecitive macroprudential policy? IMF Working Paper 16/3.

### Methods, estimation, and computational approaches
- Brooks, S. P. and Gelman, A. (1998). General methods for monitoring convergence of iterative simulations. Journal of computational and graphical statistics, 7(4):434–455.
- Schmitt-Grohé, S. and Uribe, M. (2007). Optimal simple and implementable monetary and fiscal rules. Journal of monetary Economics, 54(6):1702–1725.

*References (pages 57–58)*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1674.pdf_