Abstract
Selected Issues
Macroprudential Policy and High Household Debt1
Household debt rose rapidly, posing risks to growth and financial stability. Example of other countries that safely operate with higher levels of household debt suggests that steps to strengthen financial resilience can lessen these risks. Macroprudential policies have been tightened aggressively to contain these risks and appear to be working.
A. Policy Challenges from High Household Debt
1. Recent rapid household credit growth has increased the household debt to GDP ratio, posing a risk to economic growth and financial stability (Figure). It reflects a broader financial cycle driven by monetary easing that reduced the policy rate to an historic low, and strong demand for housing. This contributed to rise in housing prices, especially for apartments in the Seoul region, although housing prices do not appear overvalued for the country overall. The current low level of interest rates also exacerbates the risk from a large rise in interest rates on household debt, which could lead to a deterioration in asset quality. This Selected Issues Paper assesses these risks and the design of policies to effectively target them.
2. A key challenge is to identify the level of household debt with which Korea can operate safely. If this is underestimated, it could lead to costly, unnecessary deleveraging. Conversely, overestimation would involve operating with excessive levels of system risk. Adding to this challenge is the potential for raising this level by strengthening the resilience of the financial system, which would generally be preferable to deleveraging. The experience of other advanced economies can serve as a benchmark here. A few of them operate safely with a household debt ratios well above the Korean level, raising the question of what institutional features of their financial systems facilitate this. These cases are possible models for how Korea could strengthen its institutional framework to enhance the resilience of their financial system.
Korea - Household Debt
(Pcrcentof GDP)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: CEICIMF World Economic Outlook: Staff calculations.Korea - Household Debt
(Pcrcentof GDP)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: CEICIMF World Economic Outlook: Staff calculations.Korea - Household Debt
(Pcrcentof GDP)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: CEICIMF World Economic Outlook: Staff calculations.3. Korea is at the forefront of countries using macroprudential policy. The recent tightening has been successful in containing systemic risk, slowing credit growth and stabilizing house prices. However, an easing of policies in 2013-14, when the economy weakened, also may have exacerbated the financial cycle.
B. The Financial Cycle and Risks from High Household Debt
4. Growth in the ratio of household debt to GDP accelerated in the last few years driven by a strong credit cycle. The debt to GDP ratio has risen to a high level, posing a risk to economic growth (Figure). The surge in household credit growth driving this rise has slowed to 9.5 percent (y. o. y.) in Q3 2017, its lowest pace in two years, suggesting the credit cycle – illustrated by applying an HP filter to credit – could be turning (Figure). Decomposition of credit growth into that for banks and nonbank financial institutions (NBFIs) shows that much of the acceleration in 2014 was driven by NBFIs (Figure). The latter were less strictly regulated than banks in the wake of the Global Financial Crisis (GFC) as bank regulation was tightened under the auspices of the Basel Committee. The decline in interest rates to records lows after the GFC also contributing to the credit cycle (Figure).
5. The financial cycle in housing prices has been more moderate, limiting risks. Recent house price increases have been concentrated in specific regions, with average prices stabilizing for the country overall (Figure). Apartment prices around Seoul are still registering significant increases, reflecting strong demand from household formation and the effect of record-low interest rates. There is also evidence of speculative demand, as reflected in multiple purchases of apartments by individuals, and purchases and resales before construction is completed.
Korea - Household Credit
(Year-ori-year growth)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources; Haver Staff calculations.Korea - Household Credit
(Year-ori-year growth)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources; Haver Staff calculations.Korea - Household Credit
(Year-ori-year growth)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources; Haver Staff calculations.
Housing Price Inflation: Apartments and Jeonse Rent Index
(year-on-year growth, percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources:CEIC; IMF staff calculations.Housing Price Inflation: Apartments and Jeonse Rent Index
(year-on-year growth, percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources:CEIC; IMF staff calculations.Housing Price Inflation: Apartments and Jeonse Rent Index
(year-on-year growth, percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources:CEIC; IMF staff calculations.
Credit to Households
(percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: HaveCredit to Households
(percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: HaveCredit to Households
(percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: Have6. The risk from high household debt could increase as decline in interest rates to record lows reverses (Figure). With the ratio of household debt to GDP now relatively high, there is a risk that a large, sharp rise in interest rates as the Bank of Korea and other major central banks starting to tighten monetary policy, could have a more substantial impact on household consumption and asset quality. This risk is evaluated through stress testing by the authorities. For banks, stress tests show that even for a very large shock of 300 basis points, the Basel capital ratio falls 1.4 percent to 13.7 percent—well above the regulatory minimum (Table 1). Interest rate stress tests for insurance companies of 150 basis points show that their solvency ratio remains well above its regulatory minimum. Other NBFIs show comparable levels of resilience.2 Overall, this suggests that risks from a jump in interest rates to financial stability are well contained.
Interest Rate
(In percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Source: CEIC Data Company Ltd.Interest Rate
(In percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Source: CEIC Data Company Ltd.Interest Rate
(In percent)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Source: CEIC Data Company Ltd.Korea: Results of Interest Rate Stress Tests of Bank Capital Adequacy
Korea: Results of Interest Rate Stress Tests of Bank Capital Adequacy
| Size of Interest Rate Shock | Basel Capital Ratio after Shock | Impact of Shock on Basel Capital Ratio |
|---|---|---|
| + 100bp | 14,9 | −0.2 |
| +200bp | 14.4 | −0.7 |
| +300bp | 13.7 | −1.4 |
Korea: Results of Interest Rate Stress Tests of Bank Capital Adequacy
| Size of Interest Rate Shock | Basel Capital Ratio after Shock | Impact of Shock on Basel Capital Ratio |
|---|---|---|
| + 100bp | 14,9 | −0.2 |
| +200bp | 14.4 | −0.7 |
| +300bp | 13.7 | −1.4 |
C. Lessons from Other Countries with High Household Debt
7. Higher household debt/GDP ratios in other advanced economies provides insights on what can be sustained (Figure). Econometric analysis shows that higher household debt is typically associated with greater likelihood of financial crisis.3 However, countries such as Australia, New Zealand and Canada with higher household debt than Korea all weathered the GFC without a crisis; while others, like the U.K. and U.S., experienced crises (Figure). The experiences of these successful countries may provide lessons for Korea on how to maintain stability with high levels of household debt as an alternative to costly deleveraging.
Household Debt - 2016
(Percent of GDP)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank.Household Debt - 2016
(Percent of GDP)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank.Household Debt - 2016
(Percent of GDP)
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank.
Advanced Economies - Household Debt and Capital Ratio
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank; IMF World Economic Outlook.Advanced Economies - Household Debt and Capital Ratio
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank; IMF World Economic Outlook.Advanced Economies - Household Debt and Capital Ratio
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank; IMF World Economic Outlook.8. Strong institutional frameworks that build financial resilience enable countries to safely carry higher household debt. There is substantial anecdotal evidence from country studies for this but limited cross-country econometric evidence. The latter reflects the difficulty constructing good indicators of institutional quality and financial resilience. Widely used measures like the capital adequacy ratio do not really capture quality. Countries with demonstrated capacity to safely carry high household debt do not necessary have particularly high CARs. This is reflected in the lack of correlation between the CAR and household debt/GDP ratio (Figure). Measures of the quality of supervision work better empirically, indicating that countries with “strict supervision” are the ones able to safely operate with high debt (Figure).4 Estimates from a panel regression reported in the GFSR, Chapter 2, “Household Debt and Financial Stability” assesses the extent to which different structural indicators reduce the negative effect of high household debt on GDP growth and finds that the stricter supervision produces the most improvement in resilience.
Advanced Economies - Household Debt and Supervisory Power Index 1/
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank: IMF WTO.1/ Wriether the supervisory authorities have the authority to take specific actions to prevent and correct problems: index ranges from 0 (no powers) to 14 (most powers).Advanced Economies - Household Debt and Supervisory Power Index 1/
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank: IMF WTO.1/ Wriether the supervisory authorities have the authority to take specific actions to prevent and correct problems: index ranges from 0 (no powers) to 14 (most powers).Advanced Economies - Household Debt and Supervisory Power Index 1/
Citation: IMF Staff Country Reports 2018, 041; 10.5089/9781484341636.002.A004
Sources: World Bank: IMF WTO.1/ Wriether the supervisory authorities have the authority to take specific actions to prevent and correct problems: index ranges from 0 (no powers) to 14 (most powers).9. The Korean authorities are taking steps to build resilience. These include upgrading NBFI supervision to harmonize it with that for banks and promoting more rigorous bank credit assessment. New supervisory tool such as the Debt Service Ratio (DSR) covering all forms of household debt are being introduced in 2018-19. Mortgage contracts are being changed to insulate household from interest rate risk through a rapid shift from variable-rate bullet loans to fixed-rate amortizing mortgages, which now make up 47 percent of the total. As wealthy households generally have assets to cover debt repayment, new government initiatives are targeting lower-income, highly leveraged borrowers where risks are concentrated. These facilitate debt restructuring and support repayment in the event of distress; including, for example, writing off around $6 bn of debt of 1.6 million people earning less than $1,000 equivalent per month. The cumulative effect of these policies and initiatives should substantially improve financial resilience.
Main Macroprudential Policy Actions
Main Macroprudential Policy Actions
| Date | Impact | Action |
|---|---|---|
| 2005 | Tightening | Introduced DTI |
| Summer 2008 | Eating | Raised LTV, eased registration, property transfer taxes, End urban resale limits except in 3 regions. |
| Summer 2009 | Tightening | Lowered LTV. |
| Summer 2010 | Eating | Cut registration, property and transfer taxes, |
| Winter 2011 | Eating | Raised LTV, eased registration, property transfer taxes, and urban resale limits in three regions. |
| Summer 2014 | Eating | Raised DTI and LTV, canceled transfer tax for owners with multiple homes. |
| Spring/sumrrer 2016 | Tightening | Tighten mortgage lending standards. |
| Summer 2017 | Tightening | Lowered LTV and DTI, raited transfer tax in region with speculative activity. |
| Fall 2017 | Tightening | Expended DTT coverage of debt and introduced DSR, |
Main Macroprudential Policy Actions
| Date | Impact | Action |
|---|---|---|
| 2005 | Tightening | Introduced DTI |
| Summer 2008 | Eating | Raised LTV, eased registration, property transfer taxes, End urban resale limits except in 3 regions. |
| Summer 2009 | Tightening | Lowered LTV. |
| Summer 2010 | Eating | Cut registration, property and transfer taxes, |
| Winter 2011 | Eating | Raised LTV, eased registration, property transfer taxes, and urban resale limits in three regions. |
| Summer 2014 | Eating | Raised DTI and LTV, canceled transfer tax for owners with multiple homes. |
| Spring/sumrrer 2016 | Tightening | Tighten mortgage lending standards. |
| Summer 2017 | Tightening | Lowered LTV and DTI, raited transfer tax in region with speculative activity. |
| Fall 2017 | Tightening | Expended DTT coverage of debt and introduced DSR, |
D. Role of Macroprudential Policies
10. Macroprudential policies are being extensively used to curb risks from high household debt and credit growth. A broad range of macroprudential instruments that have been tightened and new ones introduced (Table 2). The loan-to-value (LTV) and debt-to-income (DTI) ratios were reduced to record lows of 40 percent, and are now well below recent highs of 70 and 60 percent, respectively, to which they were increased in August 2014. And, a lower level of 30 percent was set for borrowers with multiple mortgages and in designated regions of speculative activity, mostly around Seoul. In October 2017, the DTI was effectively tightened further by broadening the range of debt subject to it. Also announced is a new, debt-service ratio (DSR) with comprehensive coverage of all household debts, which will be implemented for banks in mid-2018; and then for NBFIs at the start of 2019.
11. Evidence suggests that this macroprudential tightening will be effective. The growth in credit to households has slowed significantly over the last few months. Moreover, speculative purchases of apartments before construction is has diminished. An event study analysis by Federal Reserve Board economists finds that hikes in LTVs and DTIs have been effective in slowing credit growth and housing price increases.5 New cross-country panel regression analysis show that use of LTVs and DTIs is effective in reducing real household credit growth across 34 advanced and emerging market economies, including Korea.6
Prepared by R. Sean Craig (APD).
Financial Stability Report “Examination of Nonbank Financial Institutions’ Interest Rate Risk,” pages 114-23, June 2017, Bank of Korea
This finding is documented in the October 2017 GFSR Chapter 2 “Household Debt and Financial Stability,” which estimates a panel regression covering 34 countries.
This indicator was produced in Barth, James R., Gerard Caprio Jr., and Ross Levine. 2013. “Bank Regulation and Supervision in 180 Countries from1999 to 2011.” Journal of Financial Economic Policy (5).
Akinci, Ozge, and Jane Olmstead-Rumsey “How Effective are Macroprudential Policies? An Empirical Investigation,” Board of Governors of the Federal Reserve System, International Finance Discussion Papers, No. 1136, May 2015.
Reported in the October 2017 GFSR Chapter 2, “Household Debt and Financial Stability,” Box 2.5.; which also discussed findings in other cross countries studies.
