## wpiea2021093-print-pdf

## Source details

**Canonical URL:** [wpiea2021093-print-pdf](https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021093-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2021/english/wpiea2021093-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2021/english/wpiea2021093-print-pdf.pdf.json)

---

### INTRODUCTION and context
- Real interest rates have declined globally, including in Korea, in past decades.
- Korean short-term real interest rates declined from an average rate of 7.5 percent in the 1980s and 1990s to an average rate of just below 1 percent since 2000.
- Inflation fell from an average 7 to 2.5 percent over the same period and has averaged 1.25 percent since 2013, undershooting the Bank of Korea’s inflation target for much of that time.
- Markets are pricing in low rates for the foreseeable future, with an anticipated risk-free interest rate of 1.6 percent in 15 years’ time, which is around half the expected level priced in mid-2012.

### Definition and importance of the neutral real interest rate (r-star)
- The neutral rate (r-star) is defined as the rate of interest that sustains full employment (equivalently, output at potential) and stable inflation in the medium run when all transitory shocks dissipate.
- The level and likely future direction of r-star is key for assessing Korea’s monetary policy stance, future policy space and the risk of hitting the effective lower bound.
- With the nominal policy rate already low and assuming the nominal interest rate is subject to an effective lower bound, and the inflation target is a small positive number, there is limited scope for conventional monetary easing.
- A decline in the neutral rate would lower economic and social discount rates and raise the present value of future climate damages (social cost of carbon), strengthening the rationale for greening the economy.

### Key empirical findings (headline conclusions)
- The neutral rate in Korea has declined over the last 30 years, and by a greater magnitude than in other advanced economies.
- The neutral rate in Korea is highly sensitive to trend growth.
- All neutral rate estimates have fallen below zero in recent years.
- Since the Asian crisis, net savings has risen, and the capital risk premia has risen as demographic rotation has accelerated; these private sector forces have pushed down the neutral rate.
- When accounting for financial factors, the neutral real rate is more cyclical than standard estimates based solely on inflation and the output gap.
- Monetary policy can have potentially persistent effects on the neutral real rate by offsetting widening or compression of financial market risk premia; quantitative easing that compresses bond risk premia is likely to support real neutral rates, all else equal.
- Korea’s neutral rate is likely to remain low over the medium-term and could fall further given demographic rotation, higher corporate and household leverage, and possible rises in precautionary savings post-COVID.
- Korea’s high savings rate and safety attributes of domestic government debt imply fiscal and structural policy could push up the neutral rate by issuing more positive-yielding long-term safe assets and encouraging investment.

### Stylized quantitative facts and baseline numbers
- Bank of Korea inflation target: 2 percent (note: in 2016 the Bank of Korea’s inflation target was lowered from 3.5 percent to 2 percent).
- Laubach-Williams baseline: neutral rate declined from around 5.2 percent in the early 1990s to 1 percent on the eve of the Global Financial Crisis; the neutral rate hit zero during the Global Financial Crisis and remained at zero till 2014.
- The unrestricted coefficient on trend growth is 7.9, positive and statistically significant at the 1 percent level, implying annualized trend growth affects the natural rate almost 2 to 1.
- Trend growth fell from around 6 percent pre-Asian Crisis to just above 2.5 percent as of 2020Q2 for Korea.
- Across models and recent estimates, the real neutral rate has fallen below zero between 2014 and 2016.
- Table 1 parameter estimates (selected, exact):
  - 훽1 (hysteresis trend growth): 1.24*** (unrestricted model), 1.26*** (restricted model)
  - 휗1 (interest rate in IS curve): -0.17*** (unrestricted), -0.20*** (restricted)
  - 휗2 (slope of Phillips curve): 0.21*** (unrestricted), 0.19** (restricted)
  - 훽4 (trend growth and neutral rate link): 7.79*** (unrestricted), 4.00 (restricted)
  - Bond risk premia effect: -1.85*** (unrestricted), -1.90*** (restricted)
  - Term premia effect: -1.51 (unrestricted), -1.68 (restricted)
- The risk premium coefficient 훽5 = −1.9 (negative), indicating higher bond risk premia push down the neutral rate.

### Methodology and alternative estimators
- Baseline estimator: Laubach and Williams (2003) semi-structural three-equation model estimated with data from 1991Q1 to 2020Q2; uses the short-term ex ante risk-free real interest rate (푟푡).
- Model equations (exact forms as presented):
  - IS curve: 푦t − 푦t* = 훽1(푦t−1 − 푦t−1*) − 휗1(푟t − 푟t*) + 휀1t
  - Phillips curve: (휋t − 휋t*) = 훽2(휋t−1 − 휋t−1*) − 휗2(푦t − 푦t*) + 휀2t
  - Neutral rate evolution: 푟t* = 훽3푟t−1* + (1 − 훽4)(푧t−1 휌4 ∆푦t*) + 휀3t
- Robustness and alternatives:
  - Stock and Watson (1998) median unbiased estimator used to address the ‘pile-up problem’ in estimating signal-to-noise ratios.
  - Time-varying parameter model (TVAR) following Lubik and Matthes (2015) estimated as a reduced-form alternative, allowing parameters to vary over time and capturing nonlinear comovement.
  - Augmented models include leverage and debt service (finance-neutral specification) and models including term and bond risk premia.

### Financial factors, leverage, risk premia, and convenience yield
- Financial cycles and risk premia have persistent effects on the neutral rate; the finance-neutral estimate can differ substantially from standard Laubach-Williams estimates.
- Korea’s household credit from banks rose from around 85 to 100 percent of GDP in recent years.
- The IS equation is augmented to include leverage measures; exact augmented IS equation:
  - 푦t − 푦t* = 훽1(푦t−1 − 푦t−1*) − 휗1(푟t − 푟t*) − 휗2 levt + 휀1t
- Leverage and debt-service dynamics included:
  - levt = 훽2 levt−1 + 훽3(r − r t*) + 훽4 dsr t−1 + 휀2t
  - dsr t = 훽5 dsr t−1 + 휀3t
- Bond risk premia and term premia are incorporated into the neutral rate equation as:
  - 푟t* = 훽3푟t−1* + (1 − 훽3)(푧t−1 휌4 ∆푦t*) + 훽4 tpt + 훽5 rpt + 휀4t
- Term premium measure: Kim and Wright (2005) three-factor arbitrage-free term structure model on the zero-coupon yield curve; bond risk premium measured as BAA corporate minus 10-year Korea government bond.
- Findings on premia:
  - The neutral rate is more countercyclical when accounting for shifts in risk premia.
  - The risk-adjusted neutral rate has, on average, been higher than the baseline Laubach-Williams estimate, especially during periods of faster growth.
  - A likelihood ratio test comparing models with and without bond risk premia yields a p-value of 0.24 (null that no bond premium is required cannot be rejected).

### Savings-investment balance, capital risk premia, and asset pricing puzzles
- Real risk-free rates have trended down over 30 years while returns on private capital have remained stable or increased, creating a widening wedge between safe rates and returns on risky capital.
- Observed patterns:
  - Stock market valuation ratios have increased only moderately.
  - Corporate earnings yields have risen.
  - The share of labor income has declined.
  - The savings-investment gap has widened, with much of the increase in savings driven by non-financial corporates; the non-financial corporate savings-investment deficit turned into a surplus since 2015.
- Safety and liquidity (convenience yield) appear to compress term premia and widen the wedge between risk-free rates and required returns on risky assets, consistent with strong demand for Korean government paper and the “savings glut” hypothesis.

### Box 1 — firm performance, market power, and distributional effects of low rates
- Mechanisms summarized:
  - Traditional view: lower long-term rates raise net present value of future cash flows, encouraging investment in productivity-enhancing technologies.
  - Alternate view: persistent low rates can encourage greater concentration in market power of large firms, potentially reducing aggregate productivity.
- Empirical evidence in Korea:
  - Non-chaebol firms have deleveraged while large chaebol-affiliated firms have maintained or increased leverage.
  - Samsung’s market capitalization is now around 30 percent of the KOSPI, up from around 20 percent a decade ago.
  - Firm performance (ROA) between chaebol and non-chaebol firms has widened.
  - Smaller corporates and SMEs have declining performance while a large share of SME debt is estimated at-risk.
- Panel regression (micro balance sheet data 2008–2018, over 3000 Korean firms):
  - Estimated equation (as presented):
    푅푂퐴_{푖푡}=훼∙푅푂퐴_{푖푡−1}+퐴∙푓푖푟𝑚_{푐ℎ푎푟_{푖푡−1}}+훽∙푚푎𝑐푟표_{푓푖𝑛𝑎𝑛𝑐𝑖𝑎𝑙_{𝑡−1}}+(훽_{7}퐹푖𝑟𝑚 푆𝑖𝑧𝑒∗훽_{8}𝑖𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝑟𝑎𝑡𝑒)+휐_{𝑖𝑡}
  - Reported estimated specification (as presented):
    푅푂퐴_{푖𝑡}=5.22+1.5∙표푝푒푟푎푡𝑖𝑛𝑔 푟𝑒𝑣𝑒𝑛𝑢𝑒+0.14∙ 푝𝑟표𝑓𝑖𝑡 𝑚𝑎𝑟𝑔𝑖𝑛−3.61∙푛𝑒𝑡 𝑑𝑒𝑏𝑡/𝑎𝑠𝑠𝑒𝑡+0.82∙푓𝑖𝑟𝑚 𝑠𝑖𝑧𝑒−0.42∙푓𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑐표𝑛𝑑𝑖𝑡𝑖표𝑛+0.91∙(퐹푖𝑟𝑚 푆𝑖𝑧𝑒∗푖𝑛𝑡𝑒𝑟𝑒𝑠푡 𝑟𝑎𝑡𝑒)+휐_{𝑖𝑡}
  - Interpretation: profitability is negatively associated with higher net debt/asset and positively associated with size, turnover and profit margins; the positive interaction implies larger firms benefit more from lower interest rates.

### Box 2 — Korea Treasury Bonds, convenience yield, and neutral-rate decomposition
- Convenience yield, safety, and liquidity:
  - Korea Treasury Bonds (KTBs) offer high liquidity and safety, producing low yields relative to Korean AA- corporate bonds.
  - Historically, a 1 percent rise in the public debt-to-GDP raised the 3-year KTB rate relative to the riskier corporate bond yield by 2.1 basis points.
- Quantitative decomposition:
  - Estimates show the nominal neutral rate fell by almost 7 percentage points between 1998 and 2020.
  - Approximately 5.7 percentage point of this decline is attributable to an increase in the convenience yield and is precisely estimated.
- Theoretical steady-state expression (as presented):
  - r* = 1/σ ⁄ · g_c + η + ρ
- Savings-investment determinants affecting r* emphasized:
  - Demographics and aging: working-age share decline, rising life expectancy increasing desired saving.
  - Real GDP and productivity growth (g_c): trend growth decline after the Asian crisis; hysteresis channels (deleveraging and labor market hysteresis) reduced potential output and neutral rates.
  - Historical co-movement: trend-decline in real interest rates and current account; widening spread between real interest rate and marginal product of capital (MPK).
- Extended empirical model augmentations for Korea:
  - Added trend labor force growth (lfg_t^*), trend labor productivity (lpg_t^*), labor market equations, nominal wage equation, current account equation, and corporate bond spreads; estimation sample 1999Q1 to 2020Q2.
- Comparative estimates:
  - Encompassing broader savings-investment determinants suggests Korea’s neutral rate may be currently somewhat less negative than indicated by an equilibrium determined purely by inflation and the output gap.
  - Since 2017 structural factors partly offset downward pressure and neutral rates differ by around 0.5 percent (from 2017 onward).

### Policy implications and recommendations
- Fiscal policy:
  - Issuing more positive-yielding long-term safe assets to absorb excess savings and fill the savings-investment gap could stimulate activity and raise the neutral rate; fiscal policy may need a more prominent stabilization role relative to monetary policy.
- Structural and regulatory policy:
  - Policies that encourage investment and reduce precautionary saving without sacrificing vital social objectives would help raise the neutral rate.
  - Regulatory policies encouraging investment for productive purposes (e.g., climate change mitigation, digitalization, deepening the social safety net) could help fill the savings-investment gap.
- Monetary policy and macroprudential interaction:
  - Monetary policy actions can impart persistent effects on real rates via financial channels; aggressive easing that is not reversed can bias nominal and real rates downward.
  - A lower neutral rate increases the probability that the effective lower bound becomes binding, suggesting a greater role for macroprudential policies tuned over the cycle.
  - Prudential intervention may put upward pressure on the neutral rate as a by-product of containing financial risk premia.
- Unconventional monetary policy:
  - Quantitative easing that compresses bond risk premia can support real neutral rates and act as a substitute when conventional rates hit the effective lower bound, but it raises questions about the independence of the neutral rate from policy actions.
- Other implications:
  - Increased asset price volatility and compressed neutral rates complicate monetary stabilization and can make collateral constraints more binding, increasing default risk during investment busts.

### Summary findings
- The Korean neutral rate has fallen over several decades and is currently in negative territory.
- Contributing factors: demographic rotation, declining productivity, slower trend growth, a rise in saving reflecting a wider capital risk premium, and demand for a convenience yield driven by liquidity and safety concerns.
- Equilibration of private-sector saving and investment may require a very low real rate over the medium term.
- Policy levers to raise the neutral rate include fiscal issuance of long-term safe assets, structural measures to raise investment or reduce precautionary saving, and macroprudential policies that contain risk premia.

*Italic source: IMF Working Paper (content unit: wpiea2021093-print-pdf).*

### INTRODUCTION

### INTRODUCTION

### Trends in real interest rates
- Real interest rates have declined globally, including in Korea, in past decades.
- Korean short-term real interest rates declined from an average rate of 7.5 percent in the 1980s and 1990s to an average rate of just below 1 percent since 2000.

### Inflation developments
- During the same period, inflation fell from an average 7 to 2.5 percent.
- Inflation has fallen further in recent years, averaging 1.25 percent since 2013, undershooting the Bank of Korea’s inflation target for much of that time.

### Market expectations
- Korean markets are pricing in low rates for the foreseeable future, with an anticipated risk-free interest rate of

*Source: wpiea2021093-print-pdf — INTRODUCTION*

### 1.6 percent in 15 years’ time, which is around half the expected level priced in mid-2012.

### 1.6 percent in 15 years’ time, which is around half the expected level priced in mid-2012.

### Definition and importance of the neutral real interest rate (r-star)
- The neutral rate (r-star) is defined as the rate of interest that sustains full employment (equivalently, output at potential) and stable inflation in the medium run when all transitory shocks dissipate.
- The level and likely future direction of r-star is key for assessing Korea’s monetary policy stance, future policy space and the risk of hitting the effective lower bound.
- With the nominal policy rate already low and assuming the nominal interest rate is subject to an effective lower bound, and the inflation target is a small positive number, there is limited scope for conventional monetary easing.
- A decline in the neutral rate would lower economic and social discount rates and raise the present value of future climate damages (social cost of carbon), strengthening the rationale for greening the economy.

### Key empirical findings (paper’s headline conclusions)
- The neutral rate in Korea has declined over the last 30 years, and by a greater magnitude than in other advanced economies.
- The neutral rate in Korea is highly sensitive to trend growth.
- All neutral rate estimates have fallen below zero in recent years.
- Since the Asian crisis, net savings has risen, and the capital risk premia has risen as demographic rotation has accelerated; these private sector forces have pushed down the neutral rate.
- When accounting for financial factors, the neutral real rate is more cyclical than standard estimates based solely on inflation and the output gap.
- Monetary policy can have potentially persistent effects on the neutral real rate by offsetting widening or compression of financial market risk premia; quantitative easing that compresses bond risk premia is likely to support real neutral rates, all else equal.
- Korea’s neutral rate is likely to remain low over the medium-term and could fall further given demographic rotation, higher corporate and household leverage, and possible rises in precautionary savings post-COVID.
- Korea’s high savings rate and safety attributes of domestic government debt imply fiscal and structural policy could push up the neutral rate by issuing more positive-yielding long-term safe assets and encouraging investment.

### Stylized quantitative facts and baseline numbers
- Bank of Korea inflation target: 2 percent (note: in 2016 the Bank of Korea’s inflation target was lowered from 3.5 percent to 2 percent).
- Laubach-Williams baseline: neutral rate declined from around 5.2 percent in the early 1990s to 1 percent on the eve of the Global Financial Crisis; the neutral rate hit zero during the Global Financial Crisis and remained at zero till 2014.
- The unrestricted coefficient on trend growth is 7.9, positive and statistically significant at the 1 percent level, implying annualized trend growth affects the natural rate almost 2 to 1.
- Trend growth fell from around 6 percent pre-Asian Crisis to just above 2.5 percent as of 2020Q2 for Korea.
- Across models and recent estimates, the real neutral rate has fallen below zero between 2014 and 2016.
- Table 1 parameter estimates (selected, exact):
  - 훽1 (hysteresis trend growth): 1.24*** (unrestricted model), 1.26*** (restricted model)
  - 휗1 (interest rate in IS curve): -0.17*** (unrestricted), -0.20*** (restricted)
  - 휗2 (slope of Phillips curve): 0.21*** (unrestricted), 0.19** (restricted)
  - 훽4 (trend growth and neutral rate link): 7.79*** (unrestricted), 4.00 (restricted)
  - Bond risk premia effect: -1.85*** (unrestricted), -1.90*** (restricted)
  - Term premia effect: -1.51 (unrestricted), -1.68 (restricted)
- The risk premium coefficient 훽5 = −1.9 (negative), indicating higher bond risk premia push down the neutral rate.

### Methodology and alternative estimators
- Baseline estimator: Laubach and Williams (2003) semi-structural three-equation model estimated with data from 1991Q1 to 2020Q2; uses the short-term ex ante risk-free real interest rate (푟푡).
- Model equations (exact forms as presented):
  - IS curve: 푦t − 푦t* = 훽1(푦t−1 − 푦t−1*) − 휗1(푟t − 푟t*) + 휀1t
  - Phillips curve: (휋t − 휋t*) = 훽2(휋t−1 − 휋t−1*) − 휗2(푦t − 푦t*) + 휀2t
  - Neutral rate evolution: 푟t* = 훽3푟t−1* + (1 − 훽4)(푧t−1 휌4 ∆푦t*) + 휀3t
- Robustness and alternatives:
  - Stock and Watson (1998) median unbiased estimator used to address the ‘pile-up problem’ in estimating signal-to-noise ratios.
  - Time-varying parameter model (TVAR) following Lubik and Matthes (2015) estimated as a reduced-form alternative, allowing parameters to vary over time and capturing nonlinear comovement.
  - Augmented models include leverage and debt service (finance-neutral specification) and models including term and bond risk premia.

### Financial factors, leverage, risk premia, and convenience yield
- Financial cycles and risk premia have persistent effects on the neutral rate; the finance-neutral estimate can differ substantially from standard Laubach-Williams estimates.
- Korea’s household credit from banks rose from around 85 to 100 percent of GDP in recent years.
- The IS equation is augmented to include leverage measures; exact augmented IS equation:
  - 푦t − 푦t* = 훽1(푦t−1 − 푦t−1*) − 휗1(푟t − 푟t*) − 휗2 levt + 휀1t
- Leverage and debt-service dynamics included:
  - levt = 훽2 levt−1 + 훽3(r − r t*) + 훽4 dsr t−1 + 휀2t
  - dsr t = 훽5 dsr t−1 + 휀3t
- Bond risk premia and term premia are incorporated into the neutral rate equation as:
  - 푟t* = 훽3푟t−1* + (1 − 훽3)(푧t−1 휌4 ∆푦t*) + 훽4 tpt + 훽5 rpt + 휀4t
- Term premium measure: Kim and Wright (2005) three-factor arbitrage-free term structure model on the zero-coupon yield curve; bond risk premium measured as BAA corporate minus 10-year Korea government bond.
- Findings on premia:
  - The neutral rate is more countercyclical when accounting for shifts in risk premia.
  - The risk-adjusted neutral rate has, on average, been higher than the baseline Laubach-Williams estimate, especially during periods of faster growth.
  - A likelihood ratio test comparing models with and without bond risk premia yields a p-value of 0.24 (null that no bond premium is required cannot be rejected).

### Savings-investment balance, capital risk premia, and asset pricing puzzles
- Real risk-free rates have trended down over 30 years while returns on private capital have remained stable or increased, creating a widening wedge between safe rates and returns on risky capital.
- Observed patterns:
  - Stock market valuation ratios have increased only moderately.
  - Corporate earnings yields have risen.
  - The share of labor income has declined.
  - The savings-investment gap has widened, with much of the increase in savings driven by non-financial corporates; the non-financial corporate savings-investment deficit turned into a surplus since 2015.
- Safety and liquidity (convenience yield) appear to compress term premia and widen the wedge between risk-free rates and required returns on risky assets, consistent with strong demand for Korean government paper and the “savings glut” hypothesis.

### Policy implications and recommendations
- Fiscal policy: Issuing more positive-yielding long-term safe assets to absorb excess savings and fill the savings-investment gap could stimulate activity and raise the neutral rate, suggesting a more prominent role for fiscal policy for stabilization vis-à-vis monetary policy.
- Structural and regulatory policy: Policies that encourage investment and reduce precautionary saving without sacrificing vital social objectives would help raise the neutral rate.
- Monetary policy and macroprudential interaction:
  - Monetary policy actions can impart persistent effects on real rates via financial channels; aggressive easing in crises that is not reversed can bias nominal and real rates downward.
  - A lower neutral rate increases the probability that the effective lower bound becomes binding, suggesting a greater role for macroprudential policies, finely tuned over the cycle.
  - Prudential intervention may put upward pressure on the neutral rate as a by-product of containing financial risk premia.
- Unconventional monetary policy: Quantitative easing that compresses bond risk premia can support real neutral rates and act as a substitute when conventional rates hit the effective lower bound, but it raises questions about the independence of the neutral rate from policy actions.

*Italic source: IMF Working Paper (content unit: wpiea2021093-print-pdf).*

### Box 1). Rising risk premia should be reflected in relatively lower prices of risky assets such as

### Box 1). Rising risk premia should be reflected in relatively lower prices of risky assets such as equities; while the Korean equity prices has generally trended upward, they have tended to be priced at a ‘discount’. A key question also is why investment has not picked up, given the low safe interest rate and relatively higher return on capital.

### Asset returns, risk premia, and interest rates
- Markets are pricing in low risk free rates over the medium-term.
- The earnings yield has been stable, resulting in a widening equity risk premia.
- Progressive fall in rates has sustained bond returns.
- Rising risk premia means the Korean equity market trades at a discount compared to peers.
- Reflected by a rising dividend-price ratio despite falling long-term real rates since the financial crisis.
- The negative correlation between equities/bonds returns suggests compressed risk premia and flight-to-quality.
- Negative correlations of equity and bond returns have tended to coincide with sharp increases in implied volatility of assets, sometimes overlapping with recessionary periods.

### Korea saving-investment dynamics
- The rise in the Korean savings rate has been driven by NFCs.
- There has been no corresponding increase in private investment spending.
- As a result, the S-I gap for the HH and NFC corporate sector has steadily moved in positive balance.
- During this time income going to labor has fallen, despite growing GVA from NFCs.
- Korea’s savings and investment ratios remain higher than peers.

### Box 1. Low for Long Interest Rates, Korean Firm Performance and Market Power — summary of mechanisms
- Traditional view: when long-term rates fall the net present value of future cash flows increases, encouraging firms to invest in productivity-enhancing technologies; low interest rates thus have an expansionary effect through stronger firm productivity growth.
- Alternate view: persistent low rates can encourage greater concentration in market power of large firms because larger firms can better take advantage of low rates; dominance of large corporations can impact aggregate corporate performance via lower productivity (they can choke off markets to increase profits).
- Together, these imply persistent low rates may act as a drag on overall corporate performance and economic growth, a concern in Korea given significant firm market concentration and low productivity for smaller firms.

### Empirical evidence and firm-level patterns in Korea
- Korean micro firm data: non-chaebol firms have deleveraged (measured by firm debt-to-asset ratio) while much larger chaebol-affiliated firms have maintained or increased leverage by taking advantage of lower borrowing costs.
- The market capitalization of the largest chaebols has grown; Samsung’s market capitalization is now around 30 percent of the KOSPI, up from around 20 percent a decade ago.
- Firm performance (measured by ROA) between chaebol and non-chaebol affiliated firms has widened, reflecting growing market concentration and rising corporate profits.
- Smaller corporates and SMEs have seen declining performance while a large share of SME debt is estimated at-risk.
- Larger corporates have maintained their leverage ratios and seen their performance rise.
- Smaller firms are the largest source of employment in Korea; further small firm balance sheet weakness would pose significant indirect risks to household balance sheets.

### Panel regression linking firm characteristics, interest rates, and ROA (micro balance sheet data 2008–2018, over 3000 Korean firms)
- Estimated equation (as presented):
  푅푂퐴_{푖푡}=훼∙푅푂퐴_{푖푡−1}+퐴∙푓푖푟푚_{푐ℎ푎푟_{푖푡−1}}+훽∙푚푎푐푟표_{푓푖𝑛𝑎𝑛𝑐𝑖𝑎𝑙_{𝑡−1}}+(훽_{7}퐹푖𝑟𝑚 푆𝑖𝑧𝑒∗훽_{8}𝑖𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝑟𝑎𝑡𝑒)+휐_{𝑖𝑡}
- Reported estimated coefficients and specification (as presented):
  푅푂퐴_{푖푡}=5.22+1.5∙표푝푒푟푎푡𝑖𝑛𝑔 푟𝑒𝑣𝑒𝑛𝑢𝑒+0.14∙ 푝𝑟표𝑓𝑖𝑡 𝑚𝑎𝑟𝑔𝑖𝑛−3.61∙푛𝑒𝑡 𝑑𝑒𝑏𝑡/𝑎𝑠𝑠𝑒𝑡+0.82∙푓𝑖𝑟𝑚 𝑠𝑖𝑧𝑒−0.42∙푓𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑐표𝑛𝑑𝑖𝑡𝑖표𝑛+0.91∙(퐹푖𝑟𝑚 푆𝑖𝑧𝑒∗푖𝑛𝑡𝑒𝑟𝑒𝑠푡 𝑟𝑎𝑡𝑒)+휐_{𝑖𝑡}
- All coefficients are statistically significant and imply:
  - Profitability is negatively associated with a firm capital structure that carries more debt, and positively associated with size, turnover and profit margins.
  - The positive coefficient value for the interaction term implies the bigger the firm, the greater (more positive) the effect of interest rates on firm performance — i.e., lower rates have been more likely to benefit larger firms.

### Implications and risks
- Persistent low interest rates have been more beneficial to larger firms, potentially increasing their market power.
- Falling rates have helped contain corporate debt-at-risk, but low interest rates could further weaken small corporate performance and erode their debt servicing capacities.
- Weaker small firm balance sheets pose indirect risks to household balance sheets because small firms are the largest source of employment.
- Economic volatility in Korea has declined while aggregate savings have risen; however, strongly negative income growth shocks (e.g., Asian Crisis) may have increased perceived tail risks, raising desired savings rates despite lower volatility.
- The negative correlation between Korea bond and equity returns reflects flight-to-quality during periods of slower growth and heightened uncertainty; such negative correlations also occur in periods of high equity market volatility unrelated to recessions, coinciding with heightened perception of risk from global financial instability.

*Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021093-print-pdf.pdf*

### Box 2. Korea Government Bonds and Safety and Liquidity

### Box 2. Korea Government Bonds and Safety and Liquidity

### Convenience yield, KTBs, and asset pricing
- Private agents value liquidity and safety attributes of government debt, driving down yields on government bonds relative to other assets when supply is scarce.  
- Korea Treasury Bonds (KTBs) offer high liquidity and safety (the “Treasury convenience yield”), producing low yields relative to Korean AA- corporate bonds that offer less liquidity and safety.  
- The opposite effect occurs when Treasury supply is high.

### Empirical evidence on supply and yields
- Historically, a 1 percent rise in the public debt-to-GDP raised the 3-year KTB rate relative to the riskier corporate bond yield by 2.1 basis points.  
- The small increase in KTB yields relative to riskier assets implies investors continue to view Korean government bonds as offering safety and liquidity even when their supply increases.

### Convenience yield, neutral rate decomposition, and key quantitative findings
- Following Krishnamurthy and Vissing-Jorgensen (2012), the convenience yield ((1+CY_t)) of the risk-free rate can be decomposed into liquidity ((1+CY_t^l)) and safety ((1+CY_t^s)) parts.  
- The pricing equation presented: Et[(1+ r_t)(1+CY_{t+1}) M_{t+1}] = 1, where M_{t+1} is the stochastic discount factor, (1+ r_t) is the pecuniary return, and (1+CY_{t+1}) is the convenience yield.  
- An increase in the convenience yield depresses the safe real rate of return because investors accept a lower pecuniary return in exchange for higher convenience.  
- Estimates show the nominal neutral rate fell by almost 7 percentage points between 1998 and 2020.  
- Approximately 5.7 percentage point of this decline is attributable to an increase in the convenience yield and is precisely estimated.  
- Conclusion drawn: a shortage of safe assets, relative to less safe assets, has resulted in a decline in the risk-free rate.

### Ramsay growth model expression for equilibrium (neutral) rate
- The Euler/steady-state expression cited:
  r* = 1/σ ⁄ · g_c + η + ρ
  (as presented in the source text; where r* is the equilibrium/neutral rate, σ is the elasticity of intertemporal substitution, g_c is growth rate in per capita consumption, η and ρ are parameters as in the original formulation).
- Key theoretical implications:
  - r* moves one-for-one with the discount rate (ρ).
  - r* depends on population growth (n) and growth in per capita consumption (g_c).
  - In a risk-neutral world σ = 1 and ρ = 0.

### Savings-investment determinants affecting Korea’s neutral rate
- Demographics and aging (n):
  - Link from working-age population share to the neutral rate is conceptually ambiguous via life-cycle saving behavior.
  - Korea had a high working-age share (reflecting high population growth and fertility in the 1970s and 1980s); UN projections suggest working-age population will decline as fertility has fallen below Japan levels.
  - Rising life expectancy tends to increase desired saving, pushing up desired saving for a given working-age population.
  - Despite predicted higher savings supply, returns on investment in Korea have remained stable as the risk-free rate has fallen, resulting in a wider capital risk premia.
- Real GDP and productivity growth (g_c):
  - The neutral rate depends on intertemporal elasticity of substitution and consumption growth; by extension, GDP or productivity growth.
  - High growth bids up interest rates given available savings. Korea’s pre-Asian-crisis growth was high; post-crisis trend growth shifted downward.
  - Two hysteresis channels after the Asian crisis: deleveraging (reducing investment demand) and labor market hysteresis (from higher unemployment and lower participation) reducing potential output and neutral rates.
- Historical co-movement:
  - Since the early 1980s Korea’s structural excess saving is reflected in a trend-decline in real interest rates and the current account.
  - The marginal product of capital (MPK), dependency ratio, and real rate have trended down; the spread between the real interest rate and MPK (a proxy for capital risk premium) has widened, particularly since the Asian crisis.
  - The relative price of capital has remained stable in Korea and thus is ruled out as the driver of the rise in non-financial corporate savings.

### Extended empirical model (Laubach-Williams augmentations for Korea)
- Model augmentations added include:
  - Trend growth rate equation: trend labor force growth (lfg_t^*) and trend labor productivity (lpg_t^*).  
  - Labor market equation: labor market hysteresis and wages to capture trend productivity.  
  - Nominal wage equation: wages as function of trend inflation (π_t^*) and trend labor productivity growth.  
  - Current account equation: links business cycle and non-oil current account (cab_t) relative to trend; includes trading partners’ cyclical positions.  
- Additional included variables: unemployment rate (U_t), manufacturing wages (w_t), trend force labor growth (lfg_t^*), trend labor productivity (lpg_t^*), non-oil current account (cab_t), Bank of Korea’s inflation target (π_t^*), and corporate bond spreads to account for shifts in risk aversion.  
- Estimation sample: 1999Q1 to 2020Q2 (data limitations).

### Comparative neutral-rate estimates and recent divergence
- Encompassing broader savings-investment determinants suggests Korea’s neutral rate may be currently somewhat less negative than indicated by an equilibrium determined purely by inflation and the output gap.  
- Since the Asian crisis, the neutral rate with structural factors was below the Laubach-Williams estimate; they closely tracked each other after the Global Crisis but diverged from 2017.  
- Based on the most recent estimates, the neutral rates differ by around 0.5 percent (from 2017 onward structural factors partly offset downward pressure from subdued inflation and persistent output gap).

### Implications for monetary policy and financial stability
- A low neutral rate implies episodes when monetary policy is constrained at the effective lower bound are likely to be more frequent and longer lasting; more frequent use of unconventional policy measures, such as quantitative easing, may be warranted.  
- Wu (2017) suggests a negative neutral rate could reflect a shadow policy rate; Korea’s shadow policy rate has closely followed the actual policy rate, reflecting lesser use of quantitative easing by the Bank of Korea compared to the Federal Reserve.  
- Balanced Approach Taylor rule approximation used (as presented):
  i_t = r_t^* + π_{t+1}^e + 0.75(π_t − π_t^*) + 0.25(y_t − y_t^*)
- Korea data over the last 20 years implies the nominal short-term rate has not, on average, fully offset inflation. A Taylor rule coefficient on inflation less than one raises the risk of the economy getting stuck in a negative (sometimes unstable) equilibrium.  
- Since 2012 Korean data shows a persistent output gap and inflation persistently below target. From early-2017, implied policy rates based on neutral rates declined and began to converge; most recent data suggests monetary policy is restricted by the zero-lower bound.  
- Increased asset price volatility (partly due to compression in the neutral rate) complicates monetary stabilization: since 2015 the significant fall in the neutral rate has coincided with a rise in asset return volatility. If sustained, collateral constraints may become more binding, increasing periodic risk of defaults triggered by investment booms and busts.  
- Greater asset return volatility exacerbates the lower bound problem and emphasizes a greater role for macroprudential policies; prudential intervention can put upward pressure on risk-free rates as a by-product of containing systemic risk.

### Summary findings and policy implications
- The Korean neutral rate has fallen over several decades and is currently in negative territory.  
- Contributing factors: demographic rotation, declining productivity, slower trend growth, a rise in saving reflecting a wider capital risk premium, and demand for a convenience yield driven by liquidity and safety concerns.  
- Equilibration of private-sector saving and investment may require a very low real rate over the medium term. Historical episodes (Asian and Global Financial Crises) show little sign of rapid reversion; Jorda, Singh and Taylor (2020) evidence suggests real rates stay depressed years following a pandemic. Korea data also show expected riskier (equity) returns have fallen.  
- Policy levers and implications highlighted:
  - Structural measures that reduce saving or promote investment are desirable to regain more monetary policy space.  
  - Fiscal policy will likely need a more prominent stabilization role than monetary policy going forward.  
  - Regulatory policies encouraging investment without sacrificing social objectives could be beneficial.  
  - Issuing more positive-yielding long-term safe assets for productive investment (e.g., climate change mitigation, digitalization, deepening the social safety net) could help fill the savings-investment gap and raise the neutral rate.  
- Market projections: long-term risk-free rates are priced below projected growth rates over the medium term, while the marginal product of capital currently remains above growth rates—together providing an incentive for investment spending.

*Box 2. Korea Government Bonds and Safety and Liquidity — wpiea2021093-print-pdf*

### References

### References

### Natural rate of interest and r*
- Akinci, O., G. Benigno, M. Del Negro, and A. Queralto, 2020, ‘The Financial (In)Stability Real Interest Rate, R**’, Federal Reserve Bank of New York Staff Reports, No. 946  
- Borio, C. & P. Disyatat & P Rungcharoenkitkul, 2019, ‘What anchors for the natural rate of interest?’ BIS Working Papers 777, Bank for International Settlements.  
- Del Negro, M., D. Giannone, M. P. Giannoni, and A. Tambalotti, 2017, Safety, Liquidity, and the Natural Rate of Interest. Brookings Papers on Economic Activity, 48 (1(Spring)): 235–316.  
- Hakkio, C. S. and L. Smith, 2017, ‘Bond Premiums and the Natural Real Rate of Interest’, Economic Review, Federal Reserve Bank of Kansas City, issue Q I, pages 5–39.  
- Hamilton, J. D., E. S. Harris, J. Hatzius and K. D. West, 2015, The Equilibrium Real Funds Rate: Past, Present and Future, NBER Working Paper No. 21476.  
- Holston, Laubach, and Williams, 2017, “Measuring the Natural Rate of Interest: International Trends and Determinants,” Journal of International Economics 108, supplement 1 (May): S39–S75.  
- Laubach, Thomas and John C. Williams, 2003, Measuring the Natural Rate of Interest. The Review of Economics and Statistics, 85(4).  
- Lubik, Thomas A. And Christian Matthes, 2015, Calculating the Natural Rate of Interest: A Comparison of Two Alternative Approaches. Economic Brief EB15-10, Federal Reserve Bank of Richmond.  
- Rungcharoenkitkul, R., 2020, ‘R-star Decline and Monetary Hysteresis’, SUERF Policy Note, Issue No. 137.  
- Kozlowski, J., L. Veldkamp and V. Venkateswaran, 2018, ‘The Tail that Keeps the Riskless Rate Low’, NBER Working Paper No. 24362.  
- Curdia, V., and M. Woodford, 2016, ‘Credit Frictions and Optimal Monetary Policy’, Journal of Monetary Economics, vol. 84, pp. 30–65.  

### Global real rates, safe assets, and macro-finance
- Caballero, R. J., Farhi, E., and Gourinchas, P.-O., 2017, The Safe Assets Shortage Conundrum. Journal of Economic Perspectives, 31(3):29-46  
- Farhi, E. and F. Gourio, 2018, ‘Accounting for Macro-Finance Trends: Market Power, Intangibles and Risk Premia’, Brookings Papers on Economic Activity Fall: 147–250  
- Gourinchas, O. and R. Hélène, 2019, ‘Global Real Rates: A Secular Approach’, BIS Working Paper No. 793.  
- Krishnamurthy, Arvind and A. Vissing-Jorgensen, 2012,‘The Aggregate Demand for Treasury Debt’, Journal of Political Economy, 120(2):, 233–267  
- Miranda-Agrippino, S. and R. Hélène Rey, 2020, ‘The Global Financial Cycle after Lehman’, AEA Papers and Proceedings, 110: 523–28.  
- Krishnamurthy, Arvind and A. Vissing-Jorgensen, 2012,‘The Aggregate Demand for Treasury Debt’, Journal of Political Economy, 120(2):, 233–267

### Monetary policy, zero lower bound, hysteresis, and macroprudential policy
- Eggertsson, G.. B. and P. R. Krugman, 2012, ‘Debt, Deleveraging, and the Liquidity Trap: A Fisher-Minsky-Koo Approach’, Quarterly Journal of Economics, 127(3): 1469–1513.  
- Summers, L., 2014, ‘U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero Lower Bound’, Business Economics, Vol. 49(2).  
- Wu, J. C. and F. D. Xia, 2016, ‘Measuring the Macroeconomic Impact of Monetary Policy at the Zero Lower Bound’, Journal of Money, Credit, and Banking, 48(2-3), 253–291.  
- Stein, J. C. and A. Sunderam, 2018, ‘The Fed, the Bond Market, and Gradualism in Monetary Policy’, Journal of Finance, 73(3): 1015–1060.  
- Rungcharoenkitkul, R. & C. Borio & P. Disyatat, 2019, ‘Monetary policy hysteresis and the financial cycle," BIS Working Papers 817, Bank for International Settlements.  
- Van der Ghote, 2020, ‘Benefits of Macro-Prudential Policy in Low Interest Rate Environments’, ECB Working Paper No. 2498.  
- Juselius, M. & C. Borio & P. Disyatat & M. Drehmann, 2017, ‘Monetary Policy, the Financial Cycle, and Ultra-Low Interest Rates’, International Journal of Central Banking, International Journal of Central Banking, vol. 13(3), pages 55–89, September.  

### Credit, housing, indebtedness, and financial history
- Adam, K., O. Pfaeuti and T. Reinelt, 2020, ‘Falling Natural Rates, Rising Housing Volatility and the Optimal Inflation Target’, mimeo.  
- Mian, A. & L. Straub & A. Sufi, 2020, ‘Indebted Demand’, NBER Working Papers 26940.  
- Jorda, Oscar., Schularick, Maurice., and Taylor, Alan. M., 2016, Macro Financial History and the New Business Cycle Facts. In NBER Macroeconomics Annual 2016, Volume 31, NBER Chapters, pages 213{263. National Bureau of Economic Research, Inc.  
- Jorda, Oscar., Knoll, K., Kuvshinov, D., Schularick, Maurice., and Taylor, A. M., 2017, The Rate of Return on Everything, 1870{2015. Working Paper Series 2017-25, Federal Reserve Bank of San Francisco.  
- Jorda, Oscar Taylor, Sanjay R. Singh and Alan. M., 2020, ‘The Long-Run Effects of Monetary Policy’, Federal Reserve Bank of San Francisco, Working Paper 2020–01.  
- Liu, E., A. Mian, A. Sufi, 2019, ‘Low Interest Rates, Market Power, and Productivity Growth’, NBER Working Paper No. 25505.  
- Kim, D. H and J. H. Wright, 2005, ‘An Arbitrage-Free Three-Factor Term Structure Model and the Recent Behavior of Long-Term Yields and Distant-Horizon Forward Rates’, Finance and Discussion Papers, Federal Reserve Board, 2005-33.  

### Methods, priors, and estimation approaches
- Giannone, D., M. Lenza and G. E. Primiceri, 2019, ‘Priors for the Long Run’, Journal of the American Statistical Association, Taylor & Francis Journals, vol. 114(526), pages 565–580, April.  
- Smets, F., and R. Wouters, 2007, ‘Shocks and Frictions in US Business Cycles: A Bayesian DSGE Approach’, The American Economic Review, vol. 97, no. 3, pp. 586–606.  
- Williams, P.D., Y. Abdih and and E. Kopp, 2020, ‘Reading the Stars’, IMF Working Paper No. 20/136.  

*Content derived from the "References" section of wpiea2021093-print-pdf*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021093-print-pdf.pdf_