## cr1841

## Source details

**Canonical URL:** [cr1841](https://www.imf.org/-/media/files/publications/cr/2018/cr1841.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2018/cr1841.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2018/cr1841.pdf.json)

---

### Enhancing the Monetary Policy Framework in Korea
- Experience with Inflation Targeting (IT):
  - IT adopted in 1998 contributed to low and more stable inflation in line with the target.
  - After the global financial crisis (GFC) inflation and inflation expectations fell below the bottom of the 2.5–3.5 percent inflation target range after 2012.
  - In 2016 the target range was replaced by a level inflation target of 2 percent (a reduction of a percentage point relative to the mid-point of the target range).
  - The reduction in 2016 weakened the credibility of the nominal anchor; credibility should recover gradually but can be accelerated by enhanced communication of how policy will achieve the new target.
  - Forward interest rate curves: after the GFC the slope anticipated hikes following sharp cuts; in the second cycle of cuts starting in 2012 forward curves were downward sloping after the first cut then reverted to upward sloping or flat paths that shifted down with each cut, suggesting market participants were not expecting additional cuts.
- Strengthening credibility:
  - Goal: more firmly anchor inflation expectations when inflation can deviate from target for prolonged periods.
  - Enhanced communication: characterize the central bank “policy reaction function” and explain how policy rates will be adjusted as circumstances change.
  - Inflation-forecast targeting (IFT):
    - Provide an inflation forecast and explain how the policy rate will be adjusted so inflation converges to the target over the medium term.
    - Does not require systematically achieving the target in realized data; requires communicating how policy will return inflation to target and close the output gap over the medium term.
    - Can be implemented without a numerical policy-rate path; qualitative discussion of likely responses to shocks can suffice.
  - BOK communication already includes elements of IFT which can be built upon.
- Challenges and risk-management approach:
  - Key uncertainties increasing policy complexity:
    - Size of the output gap: potential growth estimates declined from a range of 3.0-3.4 percent during 2011-15, to 2.8-2.9 percent in 2016-20.
    - Equilibrium real interest rate: difficulty estimating extent of decline complicates decisions.
    - High capital flow and currency volatility: need clearer, explicit commitment to exchange rate flexibility to avoid diverting monetary policy from inflation objective.
    - Financial stability: when objectives conflict, communication must clarify that policy-rate decisions are guided by price stability; macroprudential policy should target financial stability.
  - Risk-management recommendation:
    - Respond more strongly to larger shocks than to small ones, especially for large negative shocks where there is risk of hitting the Effective Lower Bound (ELB).
    - More aggressive rate cuts in ELB risk situations can lead to sharper depreciation, higher expected inflation, and a larger fall in the real interest rate, magnifying stimulatory effect.

### Policy tools and communication priorities (monetary)
- Strengthen forward-looking communication:
  - Explain the policy reaction function and likely adjustments in the policy rate as circumstances change.
  - Provide an inflation forecast and describe the policy path consistent with returning inflation to target over the medium term (inflation-forecast targeting).
  - Enhance market and public understanding without necessarily publishing a numerical policy-rate path.
- Clarify exchange rate role:
  - Signal a strong commitment to exchange rate flexibility and explain how flexibility supports monetary policy transmission and shock absorption.
- Coordinate with macroprudential policy:
  - Use macroprudential tools to address financial stability risks so monetary policy can prioritize price stability when objectives conflict.

### Forecasting and Policy Analysis System (FPAS) and model calibration
- FPAS purpose and components:
  - (i) a model-based macroeconomic forecast with an endogenous future path for the short-term interest rate;
  - (ii) incorporation of uncertainty into policy decision making;
  - (iii) assessment of risks to monetary policy based on alternative scenarios and other factors not captured by the model.
- Model used for illustrative calibration (Korea):
  - New-Keynesian, Dynamic Stochastic General Equilibrium with model consistent expectations.
  - Core equations for output gap, core inflation, policy interest rate, and exchange rate; output gap depends on lead/lag, foreign activity, and deviations of real longer-term interest rate and real exchange rate from equilibrium.
  - Core inflation: expectations-augmented Phillips curve; Phillips Curve is non-linear (quite flat when negative output gap is large).
  - Exchange rate linked via uncovered interest rate parity; additional linear equations for headline inflation, trade/financial linkages, and yield curve.
- Policy reaction functions:
  - Linear Taylor rule: policy rate responds to deviation between model forecast of inflation three-quarters ahead and the inflation target, and to the output gap.
  - Quadratic-loss (non-linear) reaction function: responds more strongly to larger shocks and incorporates interest rate smoothing.
- Model purposes:
  - Illustrate convergence of inflation back to target under alternative policies and reaction functions.
  - Assess macroeconomic risks via scenarios where shocks widen the output gap and cause inflation deviations.

### Illustrative simulations and scenarios (FPAS)
- 2013 easing-cycle simulation:
  - Historical episode: eight policy rate cuts totaling 200 basis points over three years.
  - Projections run for two reaction functions (linear IFB and quadratic loss-minimization).
  - Baseline starting 2013 assumes no new shocks; adding actual post-2013 historical shocks makes simulated economy much weaker and delays convergence to the inflation target beyond the figure’s time scale.
  - With historical shocks added, model projects a lower policy rate path with continued gradual reductions after initial cuts; inclusion of shocks leads to larger cumulative rate cut and exchange rate depreciation (noted for quadratic loss reaction function).
- Downside scenario with fiscal backstop:
  - Shock assumed to start in mid-2017: widening of the output gap by 1 percentage point and a 0.5 percentage points decline in inflation.
  - Shock chosen large enough for interest rates to hit the ELB when monetary policy alone is used.
  - Combining monetary easing with more expansionary fiscal policy:
    - Monetary easing can be more moderate; reduces probability that monetary policy will be constrained by the ELB.
    - Central bank may avoid cutting policy rate to the ELB and period of low rates is shorter.
    - Fiscal expansion magnifies positive effect because at the ELB the fiscal multiplier is larger due to little or no crowding out.

### Conclusions on monetary framework
- Credibility and framework:
  - Reduction in the inflation target by a percentage point to 2 percent in January 2016 weakened the nominal anchor.
  - Adoption of IFT with greater reliance on model-based forecasting and policy analysis can help rebuild credibility, strengthen expectations management, and improve transmission.
- Managing uncertainty and communications:
  - Clarify how policy will manage uncertainty related to output gap, equilibrium real interest rate, capital flow volatility and financial stability, and resolve conflicts among objectives.
  - Provide greater forward guidance to shape expectations and private-sector behavior.
- Limits and complementary policies:
  - Monetary policy foster low inflation and predictable environment but cannot address all macro challenges.
  - Fiscal policy can reinforce monetary policy; structural policies can boost potential growth; macroprudential policies can ensure financial stability and limit excessive household credit growth so monetary policy can focus on inflation.

### Key numeric points preserved (monetary section)
- Eight policy rate cuts totaling 200 basis points over three years.
- Downside scenario shock: output gap widened by 1 percentage point and inflation declined by 0.5 percentage points (assumed to start in mid-2017).
- Inflation target reduced by a percentage point to 2 percent in January 2016.

---

### Potential Growth, Aging, and Long-Term Fiscal Pressures
- Potential growth path (authors' extended multivariate filter estimates):
  - 2020-30: 2.2
  - 2030-40: 1.9
  - 2040-50: 1.5
  - 2050-60: 1.2
- MOSF (2015) comparison:
  - 2020-30: 2.6
  - 2030-40: 1.9
  - 2040-50: 1.4
  - 2050-60: 1.1
- Uncertainty sources: capital stock data limitations; potential structural shocks affecting productivity convergence.
- Potential growth is not policy invariant: structural reforms can raise productivity and labor participation.

### Fiscal impact of aging and long-term projections
- Projected public spending increase for pension and healthcare by about 10–16 percent of GDP by 2060.
- Annual primary expenditure assumed to increase by about 0.3 percentage point per year on average from 2023 through 2060, implying cumulative increase of around 13 percentage points by 2060 (mid-point of projected pension and health-care increases).
- Long-run nominal interest rate assumption:
  - Nominal interest rate on government debt assumed to increase gradually to 4.5 percent in 2060 (when the old-dependency ratio is projected to peak).
  - 4.5 percent long-run rate embodies an approximately 50 bps country-risk premium relative to the United States.
  - Inflation assumed at the 2 percent target.
  - Real interest rate on government bonds increases above real GDP growth since 2035; real interest rate > real growth rate (no-ponzi-game condition).

### Scenario: increased age-related spending with unchanged revenues
- Keeping total revenue fixed as a share of GDP results in explosive debt dynamics:
  - Consolidated fiscal balance would turn negative in 2024 and reach a deficit of about 1 4 percent of GDP in 2050 (text preserved exactly as presented).
  - Debt level would exceed 100 percent of GDP by 2050.
  - NABO (2016) projection under no policy change: consolidated central government deficit of about 8 percent of GDP and debt-to-GDP ratio of 111 percent in 2050.
  - NABO projects revenues-to-GDP ratio will drop by about 3 percentage points by 2060 under unchanged policies; MOSF (2015) projects revenues-to-GDP ratio to remain fairly stable between 2016 and 2060.

### Policy options to ensure long-term debt sustainability
- Options include:
  - Increase retirement ages in private and public sectors; current pension age details preserved: pension age currently 61 with at least ten years of contributions; reduced early pension from age 56; normal pension age gradually increased reaching 65 in 2033; early pension age to increase to 60. An additional hike in retirement age by 3 years by 2035 would reduce pension spending by an estimated 1 percentage point of GDP in 2050.
  - Increase revenues by broadening the tax base.
  - Increase payroll contribution rates currently very low at 9 percent (compared to a 20 percent average in advanced economies).
  - Increase selected tax rates, e.g., VAT currently at 10 percent.
- Given low public debt and low global interest rates, borrowing could finance rising age-related spending; government debt expected to remain below 40 percent of GDP for at least 15 years.

### Illustrative debt paths and fiscal scenarios
- IMF (2013) benchmark for advanced economies' public debt ceiling estimated at 85 percent of GDP.
- Illustrative path parameters:
  - Option to allow debt-to-GDP ratio to reach 45 percent in the long run.
  - Common assumptions: revenues kept constant for 10 years; fiscal surplus declines gradually; deficit arises in 2024; deficit peaks at about 1.5 percent of GDP in 2027; debt-to-GDP ratio declines until 2027 to about 30 percent.
  - "Permanent deficit" scenario:
    - Deficit stabilizes at 1.5 percent of GDP.
    - Debt reaches 40 percent of GDP in 2077 and achieves steady state of 45 percent of GDP past beyond 2100.
    - Revenues increase from 22 to 33 percent of GDP in 2060.
  - "Temporary deficit" scenario:
    - Revenues increase to nearly 34 percent of GDP in 2060.
    - Deficit declines to zero in 2084.
    - Primary balance turns into a surplus in 2040.
    - Public debt-to-GDP ratio peaks at 30 percent in 2056 and declines afterwards.
- Small differences in the speed of revenue increases have sizable impacts on debt-to-GDP.

---

### Fiscal Policy Simulations and Structural Reform Scenarios (FSGM / G20MOD)
- Model: G20MOD module of the Flexible System of Global Models (FSGM), multi-region forward-looking semi-structural model with stock-flow consistency.
- Simulation: permanent increase in targeted transfers of 0.75 percent of baseline GDP (to liquidity-constrained consumers), financed by:
  - gradual increase in consumption taxes, or
  - gradual increase in labor income taxes, or
  - gradual increase in capital income taxes,
  - plus a rise in deficit by 0.8 percentage points.
- Key findings:
  - Increasing consumption tax is the least harmful revenue source (low multiplier) and least allocative distortion.
  - Increasing capital taxation reduces investment; increasing labor income tax lowers employment; both lead to larger output losses relative to increasing consumption taxes.
  - In all three financing cases government-debt-to-GDP ratio increases by 15 percentage points above the baseline in the very long run and stabilizes there.
  - Private saving as a share of GDP declines by 1 percentage point in the long run, supporting stronger private consumption and investment.
  - Results: higher GDP growth, reduction in Korea’s current account surplus, and real exchange rate appreciation.
  - Strengthening social safety nets helps move economy toward consumption-led, inclusive growth but is not a permanent boost to output—needs productivity and labor-supply policies.

### Structural reforms and fiscal measures to raise potential growth
- Fiscal levers:
  - Efficient public investment (infrastructure) to raise productive capacity.
  - Social benefits for specific groups (childcare, ALMP) to increase labor supply and potential output.
  - Government spending on R&D or tax incentives to encourage private R&D.
  - Rebalance tax structure away from direct taxes on labor/capital toward indirect taxes on consumption (VAT) and property taxes.

---

### Reform package to boost labor supply and simulation results
- Reform package includes: easing product market regulation and employment protection legislation, raising share of consumption and property taxes in total tax revenues, increasing childcare spending, and strengthening ALMP.
- Quantitative assumptions:
  - Implement in 10 years 75 percent of product market and employment protection reforms to converge regulation to average of three OECD frontier economies.
  - Increase share of consumption and property taxes in total tax revenues by 3 percentage points, accompanied by decline in share of labor income tax.
  - Childcare spending expands by 0.25 percent of GDP.
  - ALMP spending increases by 0.5 percent of GDP.
  - Financing: higher VAT revenues and higher deficit.
- Simulation results:
  - Potential output could rise by more than 6 percent in the long run.
  - Average potential growth would increase by over 0.6 percentage point a year for a decade.
  - Additional childcare and ALMP spending would result in a rise in public debt in the very long run of 15 percentage points of GDP.
  - Ten-year deviations from baseline:
    - Real output increase in 10 years: about 6 percent.
    - Real consumption increase in 10 years: about 9 percent.
    - Current account surplus decline in 10 years: nearly 2 percent of GDP.
  - Overall fiscal impact: policy package would imply a 30 percentage points increase in the debt ratio relative to baseline and deficit increase by 1.5 percentage points of GDP relative to baseline.
- Illustrative fiscal paths:
  - "Temporary deficit" path:
    - Public debt-to-GDP would peak at 51 percent in 2061, then decline.
    - Deficit would peak at 3 percent of GDP in 2027 and decline afterwards.
  - "Permanent deficit" path:
    - Public debt would reach 70 percent of GDP in 2080 and stabilize at 75 percent past beyond 2100.
    - Deficit would stabilize at around 3 percent.
  - These levels remain below the 85 percent benchmark for advanced economies.

### Policy implications and tradeoffs
- Combined benefits: stronger social safety nets, structural reforms, and fiscal spending to boost labor supply would improve competitiveness, raise long-run growth, and rebalance toward consumption-led inclusive growth.
- Fiscal buffers: maintaining government debt at projected levels provides buffers for reunification fiscal costs (estimates range from around 4 percent to 25 percent of GDP each year for a decade).
- Long-term strategy: given low public debt and low interest rates, preferable to finance rising age-related spending with higher borrowing and additional revenues from broader tax base and selected tax increases; size of government will need to rise.
- Recommendation: implement comprehensive package of fiscal and structural measures (targeted social spending, fiscal support for employment and female labor force participation, structural reforms to raise productivity).

---

### Global Value Chains (GVC) Integration and Current Account Implications
- Literature insights:
  - GVC integration improves current account via productivity gains; residents save part of temporary income gains, raising private saving and improving current account transitorily.
  - Adjustment depends on form and pace of GVC integration and presence of structural impediments.
- Channels for Korea:
  - Trade channel:
    - Forward integration (exports of intermediate inputs) has risen from 13 to 20 percent of value added; recent expansion dominated by forward integration driven by outsourcing.
    - Forward integration initially reduces current account surplus (outsourcing shifts production abroad) but may reverse over medium term as competitiveness gains drive exports of intermediate goods.
    - Backward integration (imports of intermediates) has declined recently.
  - Investment/FDI channel:
    - Outsourcing leads firms to invest abroad (outward FDI), reducing domestic investment and widening current account surplus; data show outward FDI growth and declining domestic investment ratio.
  - Foreign investment income channel:
    - Outsourcing increases profitability of foreign operations; repatriated profits raise current account surplus; strong growth in foreign investment income inflows suggests significance of this channel.
  - Income-distribution channel:
    - Outsourcing can raise private saving by shifting income toward capital (lower MPC) and reducing labor share; Korea’s falling labor share and rising capital share is consistent with this mechanism.
- Net assessment:
  - GVC integration yields long-run benefits for Korea given firm strengths in key export industries; increases vulnerability to external shocks including protectionism.
- Empirical correlation:
  - Adding backward and forward participation measures to IMF EBA specification indicates forward integration has a positive, and backward integration a negative, correlation with the current account.
  - Effect of GVC integration on the current account shifted from negative to positive but remains small relative to the current account surplus (7 percent of GDP in 2016).

---

### Housing Prices, Household Debt, and Financial Stability
- Recent housing price developments:
  - Increases concentrated in specific regions; apartment prices around Seoul still registering significant increases.
  - Evidence of speculative demand: multiple purchases, purchases and resales before construction completion.
- Risks from high household debt and interest-rate shocks:
  - High household debt to GDP implies a large, sharp rise in interest rates could substantially impact household consumption and asset quality.
  - Stress test evidence:
    - For banks: a 300 basis points shock reduces Basel capital ratio by 1.4 percent to 13.7 percent—well above regulatory minimum.
    - Insurance companies: 150 basis points shock leaves solvency ratio well above regulatory minimum.
    - Other NBFIs show comparable resilience.
  - Overall assessment: risks from a jump in interest rates to financial stability are well contained.
- Actions and macroprudential policy:
  - Upgrading NBFI supervision; introducing Debt Service Ratio (DSR) covering all forms of household debt in 2018-19.
  - Mortgage shift: fixed-rate amortizing mortgages now make up 47 percent of the total.
  - Targeted initiatives: debt restructuring and repayment support, including writing off around $6 bn of debt of 1.6 million people earning less than $1,000 equivalent per month.
  - Macroprudential measures:
    - LTV and DTI reduced to record lows of 40 percent (from highs of 70 and 60 percent in August 2014).
    - Lower LTV of 30 percent for borrowers with multiple mortgages and in designated speculative regions.
    - DTI tightened in October 2017 by broadening range of debt subject to it.
    - New comprehensive DSR to be implemented for banks in mid-2018 and for NBFIs in 2019.
  - Evidence of effectiveness: household credit growth slowed; speculative pre-construction purchases diminished.

---

### Labor Market Duality and Flexicurity
- Nature and drivers:
  - Korea has very strict dismissal regulation for regular workers; high reinstatement orders and steep severance schedules incentivize non-regular hiring.
  - Large productivity gaps:
    - Labor productivity in the service sector is 45 percent of manufacturing (OECD average: 90 percent).
    - Small firms’ productivity is 24 percent of large firms (OECD average: 69 percent).
  - Insider-outsider dynamics and union characteristics:
    - Union density ~10 percent (OECD average: 17 percent).
    - Company welfare often larger than legal requirements; unions criticized for representativeness.
- Evidence on flows (Table 1 — Flow Probabilities by Employment Status, 2013-2014):
  - Regular → Regular: 92.5%
  - Regular → Temporary: 1.4%
  - Regular → Self-employed: 0.9%
  - Regular → Non-employed: 5.1%
  - Temporary → Regular: 4.8%
  - Temporary → Temporary: 81.4%
  - Temporary → Self-employed: 1.7%
  - Temporary → Non-employed: 12.2%
  - Self-employed → Regular: 1.3%
  - Self-employed → Temporary: 2.2%
  - Self-employed → Self-employed: 92.1%
  - Self-employed → Non-employed: 4.4%
  - Non-employed → Regular: 3.6%
  - Non-employed → Temporary: 4.1%
  - Non-employed → Self-employed: 0.9%
  - Non-employed → Non-employed: 91.4%
- Flexicurity case for Korea:
  - Move toward “protecting the workers, not the jobs”; combine flexible hiring/firing, strong safety nets, and active labor market policies (ALMPs).
  - Flexicurity “golden triangle” levers:
    - (i) Flexible rules for hiring and firing.
    - (ii) Strong safety net (unemployment insurance).
    - (iii) Active labor market policies.
  - Comparative spending (2015):
    - Denmark: ALMP = 1.7 percent of GDP; Unemployment insurance = 1.1 percent of GDP.
    - Korea: ALMP = 0.4 percent of GDP; Unemployment insurance = 0.3 percent of GDP.
- Recommendations:
  - Reduce excessive job protection in favor of worker protection (UI and activation).
  - Smooth severance and turnover cost schedules.
  - Strengthen ALMPs focusing on training and reallocation.
  - Expand targeted social programs and subsidies; calibrate reforms to fiscal capacity and social context.

---

### Modeling Duality: Search-and-Matching General Equilibrium Experiments
- Model: general equilibrium search-and-matching based on Dolado et al. (2017), calibrated to Korea.
- Mechanisms: employment protection and matching frictions generate duality and a “revolving door” spike in job destruction before temporary → permanent transitions.
- Policy experiments (five simulations): decrease firing costs; increase unemployment insurance (UI); increase matching productivity; increase training; combined flexicurity package.
- Table 2 — Selected outcomes (values preserved):
  - Unemployment rate:
    - Baseline: 7.50%
    - Decrease in Firing Cost: 6.90%
    - UI Increase: 7.80%
    - Increase Matching: 7.00%
    - Increase Training: 7.30%
    - Flexicurity Package: 6.70%
  - Output (index):
    - Baseline: 1.00
    - Decrease in Firing Cost: 1.00
    - UI Increase: 1.00
    - Increase Matching: 1.01
    - Increase Training: 1.08
    - Flexicurity Package: 1.09
  - Average Productivity (index):
    - Baseline: 1.00
    - Decrease in Firing Cost: 1.00
    - UI Increase: 1.00
    - Increase Matching: 1.00
    - Increase Training: 1.08
    - Flexicurity Package: 1.08
  - Average Income (index):
    - Baseline: 1.00
    - Decrease in Firing Cost: 1.01
    - UI Increase: 1.00
    - Increase Matching: 1.02
    - Increase Training: 1.08
    - Flexicurity Package: 1.11
  - Average Welfare (index):
    - Baseline: 1.00
    - Decrease in Firing Cost: 1.00
    - UI Increase: 1.02
    - Increase Matching: 1.03
    - Increase Training: 1.08
    - Flexicurity Package: 1.12
  - Relative job-turnover (temp to perm):
    - Baseline: 7.50
    - Decrease in Firing Cost: 2.29
    - UI Increase: 7.50
    - Increase Matching: 7.50
    - Increase Training: 8.00
    - Flexicurity Package: 4.29
  - Relative wage (temp to perm):
    - Baseline: 0.75
    - Decrease in Firing Cost: 0.83
    - UI Increase: 0.75
    - Increase Matching: 0.71
    - Increase Training: 0.77
    - Flexicurity Package: 0.87
  - Income Gini:
    - Baseline: 16.1
    - Decrease in Firing Cost: 15.4
    - UI Increase: 15.6
    - Increase Matching: 16.6
    - Increase Training: 16.3
    - Flexicurity Package: 15.0
- Findings:
  - Decrease in firing costs (≈60 percent reduction simulated): reduces revolving door, raises temp wages relative to permanent, reduces inequality and unemployment, small change in average income/output.
  - UI increase (15 percent simulated): lowers inequality and raises welfare but raises unemployment.
  - Increase matching productivity (20 percent): raises output/income/welfare via lower unemployment but does not reduce revolving door and may widen wage gap.
  - Increase training (10 percent): strongest single-measure impact on output/productivity/income/welfare but may exacerbate revolving door and inequality; effects reported as upper bound (costs not accounted).
  - Flexicurity package (combined): raises output, productivity, income, welfare; lowers revolving door and wage gap; reduces Income Gini to 15.0; raises income/welfare across quintiles.

---

### Youth Unemployment and Inactivity: Drivers and Policy Responses
- Overview and impacts:
  - Youth unemployment (age 15–24) in 2016: 10.7 percent (more than 2 percentage points lower than OECD average).
  - Adult unemployment (age 25–54) in 2016: 3.4 percent.
  - Youth unemployment is more than three times adult unemployment in Korea.
  - Marriage rate declined to 5.5 per 1000 people in 2016; fertility rate at 1.17 in 2016.
  - Youth not in employment, education or training (15–29): 18 percent as of 2013.
- Trends and composition:
  - Youth employment rate in 2016: 27 percent (13.5 percent below OECD average).
  - Field-of-study mismatch among youth: 33 percent.
  - Share of temporary employment for youth in 2016: 25.5 percent (OECD average: 24.6 percent).
  - Decline in youth population 2000–2012: 1.2 million; led to decline in employed youth by almost 900 thousand and employment rate fall from 60.1 percent to 58.1 percent.
  - Since 2012 youth cohort grew by 200 thousand until 2016; unemployment rose from 7.5 in 2012 to 9.8 in 2016; employment rate rose slightly to 58.3 percent in 2016.
- Drivers:
  - Cyclical: GDP growth explains about 20 percent of variation in youth unemployment for Korea; sensitivity largely driven by consumption growth.
  - Structural: skill and expectation mismatches; labor market duality; large education gap between elderly and youth.
  - Demographics: recent temporary youth cohort growth contributed to rising youth unemployment; cohort forecasted to cliff in 2019 and drop sharply thereafter.
- Policy levers and programmatic examples:
  - Supply-side: Meister schools (job placement rates at 90 percent; 41 schools by 2016); National Competency Standard (NCS); work-study dual system (introduced 2013: by August 2017 around 50000 youth and 10000 companies participated).
  - Demand-side: quotas and subsidies for youth hiring by public institutions and SMEs; 2017 subsidy KRW 20 million a year for three years for SMEs hiring three youth as regular workers.
  - Matching and information: subsidized 6-month internships since 2009; expanded on-campus employment centers and human resources network.
  - Active Labor Market Policies (ALMPs): spending 0.4 percent of GDP in 2015 (OECD average 0.5 percent); composition heavily on direct job creation (~56 percent) and training/employment incentives (~28 percent).
- Policy recommendations:
  - Combine demand- and supply-side measures; focus on skill and field-of-study mismatch, reduce labor market duality, reorient ALMPs toward quality and matching, expand pre-emptive measures (education plans, labor market forecasting).
  - Evaluate and monitor programs; prioritize demand-boosting fiscal measures given fiscal space and negative output gap.

---

### Minimum Wage Increase (2018) and Firm/Household Implications
- 2018 minimum wage decision:
  - Statutory minimum wage for 2018 set at KRW 7,530 per hour, a 16.4 percent rise compared to this year.
  - Government plans to pay part of additional burden estimated at KRW 3 trillion (0.2 percent of GDP) for 2018.
  - Previous minimum wage stood at $5.8 per hour (PPP); in 2018 it will increase to $7.2 per hour (PPP).
- General considerations and Korea-specific evidence:
  - In Korea about 7 percent of workers receive the minimum wage.
  - Historical econometric analysis: minimum wages have tended to follow average wages; a 1 percent increase in average real wages increased the real minimum wage a year later by about the same magnitude.
  - Minimum wages explain about 30 percent of variation in youth employment (corresponding to 1 percentage point), compared with around 10 percent for non-youth employment (0.2 percentage point).
  - Cautionary studies: Lee (2008) did not find statistically significant impact of Korea’s minimum wage increase on youth employment.
- Expected impacts of the 2018 hike:
  - The 2018 hike is unprecedented relative to recent productivity, wage growth, and inflation trends.
  - Staff expect the sharp increase to push overall wages up due to departure from historical pattern and size of the real-term hike.
  - Potential adverse effects:
    - Greater difficulty for low-skilled workers, youth, and the elderly to find employment.
    - Vulnerable firms (small service firms) may reduce employment, raise productivity, or close.
  - Policy implications:
    - Analyze the impact before further hikes; monitor youth, elderly, small firms, and low-skilled employment.
    - Boost ALMPs for workers who lose employment rather than subsidizing SME costs to avoid perpetuating incentives that prevent firm growth.
- Household demand effect:
  - Employment by monthly wage (all employed persons in 2016): 2016: 1260; 2017: 1352; 2018: 1573; 2022: 2090.
  - Key finding: 2018 minimum wage hike should support growth by boosting domestic demand as income shifts to lower-income workers with higher marginal propensity to consume, but some low-skilled workers may face job difficulty.

---

### Corporate Structure, Investment Determinants, and Vulnerabilities
- Data sample and chaebol context:
  - Firm-level consolidated accounts for 200 very large Korean companies (2006–2015); sample of 196 firms analyzed; 98 identified as chaebols affiliated; 55 part of top four Chaebols (Samsung, Hyundai, LG, SK).
- Stylized facts:
  - Size and growth:
    - Average asset size of top four chaebols affiliates nearly double other large firms as of 2015.
    - Average assets of top four affiliates grew over 50 percent from 2011 to 2015; other groups <40 percent growth.
  - Profitability:
    - Average return on assets typically higher among top 4 affiliates; profitability declined after GFC; since 2013 higher on average in large non-conglomerate firms.
    - Chaebols excluding top four recorded lowest profitability; share of firms with negative returns rose steeply and was as high as 30 percent in 2015.
  - Liquidity, solvency, leverage:
    - Solvency ratio much lower and leverage higher for chaebols excluding top four; interest cover ratio low for this group; Tobin’s Q lower.
  - R&D and investment:
    - Top four chaebols spend double on R&D (share of operating revenues) compared to other companies; chaebols excluding top four spend least.
    - Investment growth slightly downward trend; weaker in chaebols excluding top four.
- Determinants of investment (panel regression 2007–2015):
  - Tobin’s Q significant determinant of investment for overall sample; cash flow not significant for overall sample.
  - Adding leverage: leverage has significant negative coefficient.
  - Heterogeneity:
    - Tobin’s Q significant only for chaebols (both top four and others).
    - Cash flow positive and significant for large non-chaebol companies, but not for chaebols.
  - Interpretation:
    - Chaebols’ investment driven by growth opportunities and insensitive to internal financing; other large firms face financing constraints driven by internal funds.
- Policy implications:
  - Polarization between top four chaebols and other conglomerates suggests need for corporate restructuring, especially among chaebols excluding top four.
  - Ensure fair allocation of credit and capital market financing to non-chaebol firms based on consistent project evaluation.

*International Monetary Fund. Selected Issues chapter and associated analyses (cr1841).*

### References ______________________________________________________________________________ 14

### ENHANCING THE MONETARY POLICY FRAMEWORK IN KOREA

### Experience with Inflation Targeting in Korea
- Adoption of inflation targeting (IT) in 1998 contributed to low and more stable inflation in line with the target.
- After the global financial crisis (GFC) inflation and inflation expectations fell below the bottom of the 2.5–3.5 percent inflation target range after 2012, as activity slowed.
- In 2016 the target range was replaced by a level inflation target of 2 percent, a reduction of a percentage point relative to the mid-point of the target range.
- The reduction in the inflation target in 2016 created monetary policy challenges by weakening credibility of the nominal anchor; credibility should recover gradually but monetary policy can accelerate this by enhancing communication of how policy will achieve the new target.
- Forward interest rate curves provide a measure of IT effectiveness: after the GFC the slope of forward curves anticipated policy rate hikes following sharp crisis-era cuts; in the second cycle of rate cuts starting in 2012, forward curves were downward sloping after the first cut and then reverted to an upward sloping or flat path that shifted down with each policy rate cut, suggesting market participants were not expecting additional cuts and weakening transmission.

### Strengthening the Credibility of the Inflation Target
- Goal: more firmly anchor inflation expectations to the target in an environment where inflation can deviate from target for prolonged periods.
- Enhanced communication of policy rate intentions can strengthen credibility by characterizing the central bank “policy reaction function” and explaining how policy rates will be adjusted as circumstances change.
- Inflation-forecast targeting (IFT) can be used to provide an inflation forecast and explain how the policy rate will be adjusted so inflation converges to the target over the medium term.
  - IFT does not require systematically achieving the target in realized data; it requires communicating how policy will return inflation to target and close the output gap over the medium term.
  - IFT can be implemented without a numerical policy-rate path; qualitative discussion of likely responses to shocks can suffice and evolve with experience.
- BOK communication already includes elements of IFT which can be built upon; evidence suggests IFT strengthens the nominal anchor provided by the inflation target.

### Challenges Facing Korean Monetary Policy
- Key sources of uncertainty increasing the complexity of monetary policy:
  - The size of the output gap.
    - Uncertainty about slowdown in potential growth complicates output gap estimation.
    - A recent estimate reports potential growth declining from a range of 3.0-3.4 percent during 2011-15, to 2.8-2.9 percent in 2016-20.
    - When the output gap is uncertain and thought to be small, a cautious approach (holding the policy rate steady until new data) is advisable; when a large shock widens the gap significantly, a forceful policy-rate response is appropriate.
  - The equilibrium real interest rate.
    - Difficulty estimating the extent of decline in the equilibrium real interest rate complicates decisions; when the policy rate is thought to be close to equilibrium, caution is needed in raising rates.
  - High capital flow and currency volatility.
    - Communication of the role of the exchange rate becomes more challenging; a clearer, explicit commitment to exchange rate flexibility is needed to signal that monetary policy will not be diverted from the inflation objective to dampen FX volatility.
    - The flexible exchange rate helps insulate domestic monetary conditions from foreign monetary shocks and limit pass-through of U.S. rate hikes into domestic interest rates.
  - Financial stability.
    - When price stability and financial stability objectives conflict, communication must clarify that policy-rate decisions are guided by price stability.
    - The BOK’s accountability for both objectives requires explaining how macroprudential policy can be deployed to target financial stability, allowing monetary policy to focus on price stability.
- Risk-management approach recommended:
  - Respond more strongly to larger shocks than to small ones, particularly for large negative shocks where there is a risk of hitting the Effective Lower Bound (ELB).
  - In such situations more aggressive rate cuts can lead to sharper depreciation, higher expected inflation, and a larger fall in the real interest rate, magnifying the stimulatory effect.

### Policy Tools and Communication Priorities
- Strengthen forward-looking communication of how the inflation target will be achieved by:
  - Explaining the policy reaction function and likely adjustments in the policy rate as economic circumstances change.
  - Providing an inflation forecast and describing the policy path consistent with returning inflation to target over the medium term (inflation-forecast targeting).
  - Enhancing market and public understanding of policy intentions without necessarily publishing a numerical policy-rate path.
- Clarify the role of the exchange rate:
  - Signal a strong commitment to exchange rate flexibility and explain how flexibility supports monetary policy transmission and shock absorption.
- Coordinate with macroprudential policy:
  - Use macroprudential tools to address financial stability risks so monetary policy can prioritize price stability when objectives conflict.

*Based on “Enhancing the Monetary Policy Framework in Korea,” Selected Issues chapter (text provided).*

### 11.      Inflation-forecast targeting central banks typically rely on a model-based forecasting

### cr1841 - 11.      Inflation-forecast targeting central banks typically rely on a model-based forecasting

### Forecasting and Policy Analysis System (FPAS): purpose and components
- FPAS provides a more forward-looking, model-based approach that recognizes and takes into account the limitations of models and avoids short-term bias from less-structured, data-driven approaches.
- Main elements of FPAS:
  - (i) a model-based macroeconomic forecast with an endogenous future path for the short-term interest rate;
  - (ii) incorporation of uncertainty into policy decision making;
  - (iii) an assessment of risks to monetary policy based on alternative scenarios, as well as other factors not captured by the model.

### Model used for illustrative calibration (Korea)
- Model type and core equations:
  - A new-Keynesian, Dynamic Stochastic General Equilibrium with model consistent expectations.
  - Core equations for the output gap, core inflation, the policy interest rate, and the exchange rate.
  - Output gap depends on its own lead and lag, foreign activity, and the deviation from equilibrium of the real longer-term interest rate and the real exchange rate.
  - Core inflation: expectations-augmented Phillips curve where the output gap drives short-term changes in inflation; the Phillips Curve is non-linear (becomes quite flat when the negative output gap is large).
  - Exchange rate linked to interest rate via uncovered interest rate parity.
  - Additional linear equations for headline inflation; trade and financial linkages with the rest of the world; and the yield curve.
- Policy reaction functions in the model:
  - Linear Taylor rule: policy rate responds to the deviation between the model forecast of inflation three-quarters ahead and the inflation target, and to the output gap.
  - Quadratic-loss (non-linear) reaction function: responds more strongly to larger shocks and incorporates interest rate smoothing for gradual policy-rate adjustment.
- Model purposes:
  - Provide a forecast from initial conditions to illustrate convergence of inflation back to target under alternative macro policies and reaction functions.
  - Assess macroeconomic risks by implementing scenarios where shocks widen the output gap and cause deviations of inflation from target.

### Illustrative simulations and historical comparisons
- 2013 easing-cycle simulation:
  - Model applied to conditions starting in 2013 to compare projected policy-rate and macro-variable paths with actual outcomes.
  - Historical episode included eight policy rate cuts totaling 200 basis points over three years; projections run for two reaction functions (linear IFB and quadratic loss-minimization).
- Incorporating historical shocks:
  - Baseline model scenario starting in 2013 assumes no new shocks; adding the sequence of actual post-2013 historical shocks makes the simulated economy much weaker and delays convergence to the inflation target (beyond the time scale shown in the referenced figure).
  - With historical shocks added, the model projects a lower policy rate path with continued gradual reductions after an initial series of cuts; inclusion of shocks leads to a larger cumulative rate cut and exchange rate depreciation (noted for the quadratic loss reaction function).
  - Final policy decisions would also account for factors not captured by the model.
- Downside scenario with fiscal backstop:
  - Assumed shock starts in mid-2017: widening of the output gap by 1 percentage point and a 0.5 percentage points decline in inflation.
  - Shock chosen to be large enough for interest rates to hit the ELB when monetary policy alone is used.
  - When combined with a more expansionary fiscal policy, monetary easing can be more moderate, reducing the probability that monetary policy will be constrained by the ELB; the central bank may avoid cutting the policy rate to the ELB and the period of low rates is shorter.
  - Fiscal expansion magnifies positive effect because when the interest rate is at the ELB the fiscal multiplier is larger in the model due to little or no crowding out (which normally would push up the interest rate).

### Conclusions and policy implications
- Credibility and framework:
  - Reduction in the inflation target by a percentage point to 2 percent in January 2016 weakened the nominal anchor.
  - Adoption of inflation-forecast targeting (IFT) with greater reliance on model-based forecasting and policy analysis can help rebuild credibility more rapidly, strengthen expectations management, and improve monetary policy transmission.
  - IFT involves providing a macroeconomic forecast of inflation and communicating the central bank’s reaction function and how it would react to shocks to achieve its inflation target.
- Managing uncertainty and communications:
  - Clarify how policy will manage uncertainty related to the output gap, the equilibrium real interest rate, capital flow volatility and financial stability, and how conflicts among objectives will be resolved.
  - Provide greater forward guidance to shape inflationary expectations and private-sector behavior to enhance monetary policy effectiveness.
- Limits of monetary policy and the role of complementary policies:
  - Effective, credible monetary policy cannot address all macroeconomic challenges; it fosters robust growth with low inflation and provides a predictable environment for other policies.
  - Fiscal policy can reinforce monetary policy (as illustrated by model scenarios), structural policies can boost potential growth, and macroprudential policies can ensure financial stability and limit excessive household credit growth so monetary policy can focus on the inflation target.

### Key numeric points and projections preserved from the chapter
- Eight policy rate cuts totaling 200 basis points over three years (historical episode).
- Downside scenario shock: widening of the output gap by 1 percentage point and a 0.5 percentage points decline in inflation (assumed to start in mid-2017).
- Inflation target reduced by a percentage point to 2 percent in January 2016.
- Korea fiscal and demographic facts (from subsequent sections in the same source):
  - Government debt below 40 percent.
  - Average fiscal surplus of the consolidated central government of 1.2 percent of GDP since 2010.
  - Old-age dependency ratio expected to rise by 50 percentage points in the next 50 years.
  - Pension and health related public spending will increase by 10-16 percent of GDP by 2060.
  - Personal saving rate has increased by 8 percentage points since the early 2000s.
  - Share of private consumption to GDP has fallen by 6 percentage points.
  - Korea’s total factor productivity is around 60 percent that of the U.S.
  - Long-run steady-state for the Korea-U.S. productivity ratio projected at 68 percent.
  - Participation rate assumed to stay at the current level of 63 percent.
  - Unemployment rate assumed to gradually return to 3.5 percent in the long run.
  - Potential growth projected to steadily fall until 2058.

*International Monetary Fund. Selected Issues chapter on Inflation-Forecast Targeting and FPAS (cr1841).*

### 12.      With these   assumptions, potential growth will gradually slow down to 1.2 percent. Our

### cr1841 - 12.      With these   assumptions, potential growth will gradually slow down to 1.2 percent. Our

### Potential growth: estimates and uncertainty
- With the stated assumptions, potential growth will gradually slow down to 1.2 percent.
- Authors' average annual potential growth estimates (extended multivariate filter):
  - 2020-30: 2.2
  - 2030-40: 1.9
  - 2040-50: 1.5
  - 2050-60: 1.2
- MOSF (2015) estimates for comparison:
  - 2020-30: 2.6
  - 2030-40: 1.9
  - 2040-50: 1.4
  - 2050-60: 1.1
- Significant uncertainty in estimates due to:
  - difficulty obtaining reliable and timely capital stock data,
  - potential structural shocks affecting productivity convergence across countries.
- Potential growth is not policy invariant; it can be raised through structural reforms to boost productivity and labor force participation.

### Fiscal impact of aging: projected spending pressures
- Projected increase in public spending for pension and healthcare by about 10–16 percent of GDP by 2060.
- Pension spending projections:
  - MOSF (2015): increase from 3.0 to 5.2 to 7.2 to -10.6 to 7.5 (table format preserved as in source)
  - National Pension Research Institute (2013): 1.2; 2.5; 4.1; 5.7; 6.9; 5.7 (table format preserved)
  - NABO (2012): 2.4; 4.3; 5.8; 7.7; 8.9; 6.5 (table format preserved)
- Health spending projections:
  - MOSF (2015): 0.6; 0.9; 0.9; 0.8; 0.2 (table format preserved)
  - OECD (2015): ranges presented (table indicates ---8.6-13.2 4.6-9.2) (table format preserved)
  - IMF (2016): 3/4.4 6.1 7.6 9.0 10.0 5.6 (table format preserved)
- Annual primary expenditure as a share of GDP is assumed to increase by about 0.3 percentage point per year on average from 2023 through 2060, implying a cumulative increase of around 13 percentage points by 2060 (mid-point of projected pension and health-care increases).

### Long-term interest rate and no-ponzi-game assumption
- Nominal interest rate on government debt assumed to increase gradually to 4.5 percent in 2060 (when the old-dependency ratio is projected to peak).
- The 4.5 percent long-run interest rate embodies an approximately 50 bps country-risk premium relative to the United States.
- Inflation is assumed at the 2 percent target.
- The real interest rate on government bonds increases above real GDP growth since 2035; the assumption that the real interest rate is greater than the real growth rate is characterized as a no-ponzi-game condition.

### Scenario: Increased age-related spending with unchanged revenues
- Keeping total revenue fixed as a share of GDP would result in explosive debt dynamics.
- Consolidated fiscal balance would turn negative in 2024 and reach a deficit of about 1 4 percent of GDP in 2050 (text preserved exactly as presented).
- Debt level would exceed 100 percent of GDP by 2050.
- NABO (2016) projection under no policy change: consolidated central government deficit of about 8 percent of GDP and debt-to-GDP ratio of 111 percent in 2050.
- Note: NABO projects revenues-to-GDP ratio will drop by about 3 percentage points by 2060 under unchanged policies; MOSF (2015) projects revenues-to-GDP ratio to remain fairly stable between 2016 and 2060.

### Policy options to ensure long-term debt sustainability
- Options include:
  - Contain long-term pension spending pressures by increasing the retirement ages in both the private and public sectors. (Current pension age details preserved: pension age currently 61 with at least ten years of contributions; reduced early pension from age 56; normal pension age gradually increased reaching 65 in 2033; early pension age to increase to 60. An additional hike in the retirement age by 3 years by 2035 would reduce pension spending by an estimated 1 percentage point of GDP in 2050.)
  - Increase revenues by broadening the tax base with both efficiency and equity gains.
  - Increase payroll contribution rates that are currently very low, at 9 percent, compared to a 20 percent average in advanced economies.
  - Increase selected tax rates, for example the VAT rate, which is currently at only 10 percent.
- Given Korea’s low public debt and the low global interest rate environment, rising age-related spending could also be financed through borrowing; government debt expected to remain below 40 percent of GDP for at least 15 years, leaving room to increase debt-to-GDP over the long term.

### Debt ceiling context and illustrative debt paths
- IMF (2013) benchmark for advanced economies' public debt ceiling estimated at 85 percent of GDP (signal approach).
- Multiple paths can keep Korea’s public debt well below that threshold while accommodating aging-related spending.
- Illustrative path parameters and outcomes:
  - Allow debt-to-GDP ratio to reach 45 percent in the long run as one option.
  - Common assumptions in illustrative scenarios:
    - Revenues are kept constant for 10 years; fiscal surplus declines gradually; deficit arises in 2024; deficit peaks at about 1.5 percent of GDP in 2027.
    - Debt-to-GDP ratio declines until 2027 to about 30 percent.
  - "Permanent deficit" scenario:
    - Deficit stabilizes at 1.5 percent of GDP.
    - Debt reaches 40 percent of GDP in 2077 and achieves the steady state of 45 percent of GDP past beyond 2100.
    - Revenues increase from 22 to 33 percent of GDP in 2060.
  - "Temporary deficit" scenario:
    - Revenues increase to nearly 34 percent of GDP in 2060.
    - Deficit declines to zero in 2084.
    - Primary balance turns into a surplus in 2040.
    - Public debt-to-GDP ratio peaks at 30 percent in 2056 and declines afterwards.
- Small differences in the speed of revenue increases have sizable impacts on debt-to-GDP by affecting the primary balance and overall debt dynamics.

### Fiscal policy simulations (FSGM / G20MOD): strengthening social safety nets and boosting labor supply
- Model used: G20MOD module of the Flexible System of Global Models (FSGM), multi-region forward-looking semi-structural model with stock-flow consistency.
- Permanent increase in targeted transfers (to liquidity-constrained consumers) of 0.75 percent of baseline GDP simulated, financed by:
  - gradual increase in consumption taxes, or
  - gradual increase in labor income taxes, or
  - gradual increase in capital income taxes,
  - plus a rise in deficit by 0.8 percentage points.
- Key simulation findings:
  - Increasing consumption tax is the least harmful revenue source in macroeconomic impact (low multiplier) and allocative distortions.
  - Increasing capital taxation reduces investment; increasing labor income tax lowers employment; both lead to larger output losses relative to increasing consumption taxes.
  - In all three financing cases, government-debt-to-GDP ratio increases by 15 percentage points above the baseline in the very long run and stabilizes there.
- Effects on saving, consumption, investment, and external balance:
  - OECD (2011) finding preserved: an increase in public health care spending by 1 percent of GDP is associated with a decline in the saving rate by 1.9 percentage points.
  - In the FSGM simulation, private saving as a share of GDP declines by 1 percentage point in the long run, supporting stronger private consumption and investment.
  - Resulting effects include higher GDP growth, a reduction in Korea’s current account surplus, and real exchange rate appreciation (stronger imports).
  - Strengthening the social safety net helps move the economy toward a consumption-led, inclusive growth model and reduces vulnerability to external shocks, but it is not a permanent boost to output and needs to be combined with productivity- and labor-supply-raising policies.

### Structural reforms and fiscal measures to raise potential growth
- Fiscal policy can support long-term growth through:
  - Efficient public investment, especially in infrastructure, to raise productive capacity.
  - Social benefits for specific groups (e.g., public spending for childcare and active labor market policies, ALMP) to increase labor supply and potential output.
  - Government spending on research and development (R&D) or tax incentives to encourage private R&D to enhance productivity.
  - Fiscal measures to support labor and product market reforms to mitigate distributive effects and facilitate productivity gains.
  - Rebalancing the tax structure away from direct taxes on labor and capital toward indirect taxes on consumption (VAT) and property taxes to improve allocative efficiency.

*Source: Authors' estimates.*

### 27.      Simulations are carried out to assess the impact of fiscal measures to boost labor

### Simulations are carried out to assess the impact of fiscal measures to boost labor

### Reform package and scenario assumptions
- Reform package includes: easing product market regulation and employment protection legislation, raising the share of consumption and property taxes in total tax revenues, increasing childcare spending, and strengthening active labor market policies (ALMP).
- Quantitative scenario assumptions:
  - Korea implements in 10 years 75 percent of the product market and employment protection reforms that would allow regulation to converge to the average of the three OECD economies at the frontier.
  - The share of consumption and property taxes in total tax revenues is assumed to increase by 3 percentage points, accompanied by a decline in the share of labor income tax that boosts labor force participation.
  - Childcare spending expands by 0.25 percent of GDP.
  - ALMP spending increases by 0.5 percent of GDP.
  - Financing: combination of higher VAT revenues and higher deficit.

### Simulation results: macroeconomic impacts
- Potential output and growth:
  - Potential output could rise by more than 6 percent in the long run.
  - On average, potential growth would increase by over 0.6 percentage point a year for a decade.
  - Note: "In the model the impact of an increase in spending on active labor market policies could be under estimated, as it is assumed to affect equilibrium employment, but does not have an impact on the participation rate."
- Public finances and spending:
  - The additional spending on childcare and ALMP would result in a rise in public debt in the very long run of 15 percentage points of GDP.
- Output, consumption, and current account (10-year deviations from baseline):
  - Real output would increase by about 6 percent in 10 years.
  - Real consumption would increase by about 9 percent in 10 years.
  - The current account surplus would decline by nearly 2 percent of GDP in 10 years.

### Fiscal outcomes and scenarios
- Overall fiscal impact of the comprehensive policy package:
  - The policy package would imply a 30 percentage points increase in the debt ratio relative to the baseline scenario.
  - Deficit would increase by 1.5 percentage points of GDP relative to the baseline.
- Example paths:
  - "Temporary deficit" path (illustrative):
    - With additional fiscal measures to boost social protection and labor supply, public debt-to-GDP would peak at 51 percent in 2061, and would decline afterwards.
    - Deficit would peak at 3 percent of GDP in 2027 and decline afterwards.
  - "Permanent deficit" path (illustrative):
    - Public debt would reach 70 percent of GDP in 2080 and stabilize at 75 percent past beyond 2100.
    - This level would still be below the 85 percent benchmark for advanced economies.
    - The deficit would stabilize at around 3 percent.

### Policy implications, tradeoffs, and broader considerations
- Combined benefits:
  - Stronger social safety nets, structural reforms, and increased fiscal spending to boost labor supply would improve competitiveness, raise long-run growth, and rebalance the economy toward a more inclusive and consumption-driven model, thereby reducing vulnerability to external shocks.
- Fiscal buffers and reunification risk:
  - Maintaining government debt at the projected levels would provide buffers to deal with the fiscal costs of possible reunification.
  - Great uncertainty surrounds timing and modalities of reunification; estimates of possible fiscal costs range from around 4 percent to 25 percent of South Korea’s GDP each year for a decade.
  - With careful policy responses, integration would also bring benefits—an increase in the labor force and higher potential growth over time.
- Long-term fiscal strategy:
  - Rapidly aging population and structural impediments worsen long-term fiscal outlook and potential growth.
  - Under current baseline assumptions, keeping fiscal revenue to GDP constant would result in an unsustainable long-term path of public debt.
  - Given Korea’s low public debt and low interest rates, it would be preferable to finance rising age-related spending with higher borrowing and additional revenues obtained by broadening the tax base and selected tax increases.
  - Overall, the size of government will need to rise.
- Policy recommendation summary:
  - Implement a comprehensive package of fiscal and structural measures: increase targeted social spending to reduce poverty and bolster private consumption; adopt fiscal measures to support employment and boost female labor force participation; pursue structural reforms to raise productivity growth and efficiency.
  - Illustrative simulations indicate that, even accounting for projected increases in age-related expenditure, higher social safety net spending, and fiscal measures to bolster labor supply, Korea can stabilize the debt-to-GDP ratio to levels well below the estimated dangerous threshold.

*Republic of Korea — International Monetary Fund (authors' estimates and chapter text).*

### 8.      In Korea, forward

### 8.      In Korea, forward

### B. Insights from the Literature on GVC Integration
- Integration into GVCs will impact the current account through a variety of channels.  
- Economic theory implies that GVC integration improves the current account. The productivity gain from GVC integration will lead to a transitory, relative improvement in competitiveness that should be gradually eroded as competing countries achieve similar improvements by following suit.  
- Residents recognize that the income gain from the productivity shock will be temporary and, therefore, save part of it to smooth consumption. This rise in the private saving rate contributes to a temporary improvement in the current account (Brumm et.al., 2017).  
- How this occurs in practice will depend on the form and pace o f GVC integration, and whether the adjustment process occurs smoothly, consistent with theory, or is slowed by structural impediments.  
- Theoretical literature focuses largely on the trade channel, using general-equilibrium microeconomic models of trade that capture employment impact and effect on income accruing to factors of production. Many models build on Melitz (2003) and extensions (e.g., Demidova et. al. (2013)) to cover trade in intermediate inputs.  
- These models typically assume that adjustment through other macroeconomic channels occurs smoothly and rapidly; when structural impediments prevent smooth adjustment, these other macroeconomic channels become qualitatively important.  
- Empirical literature largely uses panel estimation exploiting improved data on trade in value-added (intermediate) goods; these yield average cross-country, cross-industry effects. Country-specific case studies (such as Korea) are essential complements because each country’s experience depends on its economic structure.

### C. Channels through which GVC Integration Impact Korea
- Overview: GVC integration impacts the current account through multiple channels: the trade channel (immediate), and second-round macroeconomic channels including domestic investment/FDI, labor income, foreign investment income, and distribution of income. These channels affect private saving and investment, and thus the current account. The impact depends on Korea’s differential effects vis-à-vis other countries, consistent with multilateral consistency.
- Trade channel:
  - Effect depends on whether GVC integration is forward or backward participation. For most countries, including Korea, these two channels are relatively balanced (Korea lies almost on the 45º line in the Figure).
  - Increase in forward integration = rise in exports of intermediate inputs. When associated with outsourcing, the short-term net effect on the trade balance should be negative: additional exports of intermediate inputs for use in production of the outsourced good will be more than offset by the fall in exports of this (outsourced) good. Over time, this negative effect should reverse as exports of intermediate inputs increase strongly with productivity gains. In Korea, recent GVC integration has been driven by forward integration, with exports of intermediate inputs rising steadily from 13 to 20 percent of value added (Figure).
  - Increase in backward integration = rise in imports of intermediate inputs due to lower cost of imported inputs relative to domestic inputs. The rise in imports should be offset by a positive effect on exports from lower input costs increasing competitiveness. In countries where backward integration is associated with outsourcing of production to that country, the net effect on the trade balance will be positive as exports of the outsourced goods are likely to exceed imports of intermediate goods used in their production. In contrast to the rise in forward integration, backward integration has declined over the last few years (Figure). This outsourcing has been concentrated in sectors where Korean firms are particularly strong, where productivity and competitiveness gains are likely to be substantial.
- Investment/FDI channel:
  - Outsourcing in GVCs leads local firms to invest abroad through FDI rather than at home, which tends to reduce domestic investment (Figure). The associated weakness in domestic demand widens the current account surplus. Over time, this effect can be partially reversed as increased exports of intermediate inputs necessitate new investment to ramp-up production, and as redeployment of displaced labor into other domestic activities could require new investment. Data for Korea are consistent with this channel, with outward FDI growing strongly and the domestic investment ratio declining steadily (Figure).
- Foreign investment income channel:
  - Outsourcing should increase productivity, generating higher profits in Korean corporates’ foreign operations. These profits will be repatriated through the investment income balance, which raises the current account surplus directly. The strong growth in foreign investment income inflows following the surge in FDI associated with outsourcing suggests that this channel is significant. This positive effect on the current account is likely to be partly offset as this income is received by residents who spend part of it, boosting demand.
- Income-distribution channel:
  - Outsourcing can temporarily worsen income distribution, raising the aggregate private saving rate. Higher corporate profits accrue to shareholders/investors/managers while domestic employment falls (initially only partly offset by employment for additional production of intermediate exports). The result is a fall in the labor share relative to capital income; because labor has a higher propensity to consume than capital, this increases the private saving rate. As labor is redeployed, employment and labor income should recover, reversing this effect. In Korea, the falling share of labor income and rising capital share is consistent with this mechanism.

### D. Implications of GVC Integration for Korea’s Current Account
- Channel-by-channel impacts:
  - Trade channel: dominated by recent expansion in forward integration driven by outsourcing. The impact effect reduces the current account surplus as outsourcing shifts production of exports abroad, reducing net exports. Over the medium term, the trade channel effect could reverse as competitiveness gains drive strong growth in exports of intermediate goods.
  - The three macroeconomic channels (Investment/FDI, Foreign investment income, Income-distribution) tend to widen the current account surplus. While net effects through these channels should ultimately be temporary, duration depends on how rapidly labor is redeployed into higher value added activities and on erosion of Korea’s competitiveness gains as other countries improve productivity.
- Saving-investment balance:
  - GVC integration alters the saving-investment balance, reflected in a recent widening in the gap between saving and investment (Figure). Domestic investment has been weakened by outsourcing as firms invest abroad through FDI. In parallel, outsourcing raises the private saving rate via distributional shifts: weaker labor income growth and higher corporate earnings reduce labor’s income share (high MPC) and raise capital’s share (lower MPC), increasing private saving. Over time, these effects should reverse as labor is redeployed and domestic investment expands for higher value added intermediate production. The wider gap between saving and investment has been associated with a larger current account surplus (Figure).
- Empirical correlation:
  - Adding backward and forward participation in GVCs to the IMF’s multilaterally consistent model of current account determination (EBA specification) indicates forward integration has a positive, and backward integration a negative, correlation with the current account (Figure). Results are indicative because including a component of the trade balance (part of the dependent variable) as a regressor can give rise to endogeneity problems. The estimates show the effect of GVC integration on the current account has shifted from negative to positive, although this effect remains quite small relative to the size of the current account surplus (7 percent of GDP in 2016).
- Net assessment:
  - Integration into GVCs will yield long run benefits for Korea. GVC integration reflects the capacity to take advantage of developments in information and logistics technology facilitating coordination of production across countries. Korean firms lead in key export industries where GVC participation can yield large productivity gains by outsourcing lower value added parts (e.g., final assembly) to lower labor cost countries while retaining higher value added parts (e.g., sophisticated microprocessors) in Korea.
  - While increased global integration yields benefits, it also increases vulnerability to external shocks, including protectionism.

*Source: IMF staff estimates and analysis as presented in the chapter "8.      In Korea, forward" from the IMF country report.*

### 5.       The financial cycle in housing prices

### 5.       The financial cycle in housing prices

### Recent housing price developments
- Recent house price increases have been concentrated in specific regions, with average prices stabilizing for the country overall.
- 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 evidence of speculative demand, as reflected in multiple purchases of apartments by individuals, and purchases and resales before construction is completed.

### Risks from high household debt and interest-rate shocks
- With the ratio of household debt to GDP now relatively high, a large, sharp rise in interest rates as the Bank of Korea and other major central banks start to tighten monetary policy could have a substantial impact on household consumption and asset quality.
- Authorities evaluate this risk through stress testing:
  - For banks, stress tests show that for a shock of 300 basis points, the Basel capital ratio falls 1.4 percent to 13.7 percent—well above the regulatory minimum.
  - 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.
- Overall assessment: risks from a jump in interest rates to financial stability are well contained.

### Lessons from other countries with high household debt
- Econometric analysis shows that higher household debt is typically associated with greater likelihood of financial crisis (panel regression covering 34 countries reported in GFSR Chapter 2).
- Cross-country experiences:
  - Australia, New Zealand and Canada (with higher household debt than Korea) weathered the GFC without a crisis.
  - U.K. and U.S. experienced crises.
- Implication: successful countries provide lessons on how to maintain stability with high household debt as an alternative to costly deleveraging.

### Institutional frameworks and resilience
- Strong institutional frameworks that build financial resilience enable countries to safely carry higher household debt.
- Cross-country econometric evidence is limited due to difficulty constructing good indicators of institutional quality and financial resilience.
- Widely used measures like the capital adequacy ratio do not fully capture quality; lack of correlation between CAR and household debt/GDP ratio is noted.
- Measures of supervision quality work better empirically: countries with “strict supervision” are better able to safely operate with high debt.
- Panel regression in GFSR Chapter 2 finds that stricter supervision produces the most improvement in resilience.

### Actions by Korean authorities to build resilience
- Upgrading NBFI supervision to harmonize it with that for banks and promoting more rigorous bank credit assessment.
- New supervisory tools such as the Debt Service Ratio (DSR) covering all forms of household debt are being introduced in 2018-19.
- Mortgage contract changes to insulate households from interest rate risk:
  - Rapid shift from variable-rate bullet loans to fixed-rate amortizing mortgages, which now make up 47 percent of the total.
- Targeted initiatives for lower-income, highly leveraged borrowers:
  - Policies facilitate debt restructuring and support repayment in distress, including writing off around $6 bn of debt of 1.6 million people earning less than $1,000 equivalent per month.
- Expected outcome: cumulative effect of these policies and initiatives should substantially improve financial resilience.

### Role and effectiveness of macroprudential policies
- Macroprudential policies extensively used to curb risks from high household debt and credit growth; a broad range of instruments have been tightened and new ones introduced.
- Specific measures and parameters:
  - Loan-to-value (LTV) and debt-to-income (DTI) ratios were reduced to record lows of 40 percent, down from recent highs of 70 and 60 percent, respectively (those highs were reached in August 2014).
  - A lower LTV 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.
  - A new, comprehensive debt-service ratio (DSR) was announced: to be implemented for banks in mid-2018; and then for NBFIs at the start of 2019.
- Evidence of effectiveness:
  - Growth in credit to households has slowed significantly over the last few months.
  - Speculative purchases of apartments before construction have diminished.
  - 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.
  - Cross-country panel regression analysis across 34 advanced and emerging market economies, including Korea, shows LTVs and DTIs are effective in reducing real household credit growth.

*cr1841 - 5.       The financial cycle in housing prices*

### 11.      Korea has particularly strong restrictions on dismissal of regular workers. The OECD

### 11.      Korea has particularly strong restrictions on dismissal of regular workers. The OECD

### Dismissal regulations and incentives
- OECD EPL measures indicate Korea has very strict dismissal regulation in international comparison.
- Dismissal for economic reasons: firms must prove they actively tried to avoid dismissal and have exhausted “all means” (OECD, 2013).
- Dismissal for managerial reasons: must demonstrate an “urgent managerial necessity” (clear definition lacking) (Kim, 2015a).
- Dismissal above a certain number of employees must be reported to the MOEL in advance (OECD, 2013).
- Korea has a high frequency of reinstatement orders in cases of unfair dismissal.
- Severance payments follow a steep schedule and there is a significant gap in labor costs, incentivizing firms to hire non-regular employees.

### Product market segmentation and productivity gaps
- Labor productivity in the service sector was only 45 percent of that in manufacturing—far below the OECD average of 90 percent.
- Small firms’ productivity is only 24 percent of that of large firms, compared to an OECD average of 69 percent.
- Large productivity gaps have likely contributed to labor market duality as low productivity firms are unable to offer high-quality regular jobs (Ha and Lee, 2013; OECD, 2013).
- KLIPS micro data show a “double duality”:
  - (i) Temporary employees generally receive lower wages and have less social coverage than regular employees.
  - (ii) Temporary workers in SMEs are even worse off than temporary workers in large companies.

- Table 1. Flow Probabilities by Employment Status (2013-2014) — From → To (percent)
  - Regular → Regular: 92.5%
  - Regular → Temporary: 1.4%
  - Regular → Self-employed: 0.9%
  - Regular → Non-employed: 5.1%
  - Temporary → Regular: 4.8%
  - Temporary → Temporary: 81.4%
  - Temporary → Self-employed: 1.7%
  - Temporary → Non-employed: 12.2%
  - Self-employed → Regular: 1.3%
  - Self-employed → Temporary: 2.2%
  - Self-employed → Self-employed: 92.1%
  - Self-employed → Non-employed: 4.4%
  - Non-employed → Regular: 3.6%
  - Non-employed → Temporary: 4.1%
  - Non-employed → Self-employed: 0.9%
  - Non-employed → Non-employed: 91.4%

### Insider-outsider dynamics
- Insider-outsider dynamics arise due to labor turnover costs; insiders gain bargaining power and negotiate better contracts (Lindbeck and Snower, 2002).
- Evidence in Korea:
  - Company welfare in certain firms significantly larger than legally required for benefits and retirement allowance (OECD, 2013; Kim, 2016).
  - Union density is low at about 10 percent, compared to an OECD average of 17 percent.
  - Unions have been criticized for lacking representativeness and for negotiating high benefits for members (Eichengreen et al., 2015).
  - Unions largely represent regular workers, especially those in large companies.

### Global trend of market polarization (Korea-specific findings)
- Labor market polarization: employment shifting away from middle-wage tasks towards low- and high-wage jobs has been documented globally (Autor, 2010; Reijnders and de Vries, 2017).
- For Korea, evidence is mixed:
  - Reijnders and de Vries (2017) do not find evidence for polarization of employment shares.
  - Lee and Lee (2015) find offshoring contributed to larger wage inequality between temporary and regular workers in manufacturing.
- Changes in employment shares by wage levels (1993–2016):
  - 1993 to 1999: clear polarization with employment shares moving from medium-wage largely to high-wage occupations.
  - 2009 to 2016: further polarization with employment shares moving from medium- to low-wage occupations.

### Key labor market reforms (Box 1) — timeline and effects
- Pre-Asian crisis: high level of employment protection for regular workers (notification procedures, definition of justified/unfair dismissal, high possibility of reinstatement). Temporary work was very restrictive in types and maximum duration of fixed-term contracts.
- During the Asian crisis:
  - Employment protection for regular and temporary employment became significantly less strict.
  - Recognition of collective dismissal in case of “urgent managerial need.”
  - Dispatched Workers Act introduced: allowed temporary work agency employment in 26 specified occupational areas; did not set limits on duration of fixed-term contracts or renewals (Kim and Skott, 2016).
  - Social expenditure increased from an average of 2.9 percent of GDP (1990–1996) to 4.8 percent (1997–2003).
  - Increased flexibility of employment protection cited as a driver behind increases in non-regular employment and gaps in wages/benefits (Cho and Keum, 2009; Ha and Lee, 2013; OECD, 2013).

- Post-2007 measures:
  - Fixed-Term and Part-Time Employees Act and Dispatched Employees Act (2007): restricted employment period for fixed-term workers to two years and prohibited discrimination against non-regular workers performing similar tasks.
  - Criticisms: laws applied only to firms with at least five workers and very specific non-regular employment types, leading to avoidance behavior by firms (Cooke and Brown, 2015). Cap on length of fixed-term employment may have been counterproductive (firms fired/replaced to avoid threshold) (OECD, 2016).
  - Outcomes: around one-third of workers converted to regular employment; some literature suggests a modest decline in temporary workers (OECD, 2013; Ha and Lee, 2013).

- 2010–2014 enhancements:
  - 2010: Trade Union and Labor Relations Adjustment Act introduced framework allowing multiple unions within a firm and set cost limits for union representatives that employers must cover (OECD, 2013).
  - 2011: Comprehensive Non-regular Workers Initiatives expanded social insurance, strengthened minimum wage enforcement, and provided vocational training for non-regular workers (OECD, 2016).
  - 2012: Guidelines to reduce discrimination in wages/working conditions; penalties for inadequate use of temporary agency contracts strengthened (OECD, 2013).
  - 2014: Temporary Workers Act and Dispatch Workers Act amended to give greater protection to non-regular workers regarding discriminatory treatment and overtime compensation (Cooke and Brown, 2015).
  - Regulation of regular contracts: shortening of notice period in case of dismissal; replacing penal provisions with financial penalties in case of unfair dismissal (OECD, 2013).
  - September 2015: new tripartite agreement announced with main goal to alleviate dual structure of the labor market; critics say it later lost support from key actors (IMF, 2016).

- Social support and targeted programs:
  - Government increased social care programs targeted at the vulnerable: parental leave, childcare support, subsidies to the elderly, vocational training.
  - Share of social expenditure targeted at families increased from about 2.2 percent in 1998 to 11.5 percent in 2014.
  - Duru Nuri Social Insurance Subsidy Program provides targeted subsidies to low-wage workers at small businesses for social insurance contributions.

- New administration policy pillars to reduce duality:
  - (i) Increase households’ disposable income through a higher minimum wage and expanding eligibility for basic social security recipients.
  - (ii) Promote decent employment by restricting application of temporary contracts and reducing legal limit of weekly working hours.
  - (iii) Advance fair competition between large corporations and SMEs by promoting profit sharing and removing unfair practices.

### Policies to address market duality and the case for flexicurity
- Labor market duality in Korea is entrenched and multifaceted; policies can play a critical role but must address labor and product markets to tackle the “double duality.”
- Recommendation: consider moving towards Korea’s own form of “flexicurity” — shift to “protecting the workers, not the jobs.”
- Channels through which flexicurity would address duality:
  - Increasing productivity: flexicurity would enhance resource allocation and training, improving matches and lowering product market segmentation and duality (Blanchard et al., 2013).
  - Raising labor force participation: could increase participation, particularly among inactive individuals and women; adding a second earner may relax insiders’ demand for job-security.
  - Preparing for structural change: large education gap between youth and elderly and rapid aging make protecting workers rather than jobs critical to avoid mismatches and productivity losses.
  - Enhancing equity: reduce high protection for regular employees via a smoother schedule; combine with higher unemployment benefits and activation policies to protect the vulnerable.

- Flexicurity “golden triangle” levers:
  - (i) Flexible rules for hiring and firing (employment protection legislation).
  - (ii) Strong safety net in the form of unemployment insurance.
  - (iii) Active labor market policies.
- Principle: protect workers mainly through unemployment insurance rather than strict employment protection legislation; ensure active labor market policies maintain work incentives (Blanchard et al., 2013).

### Flexicurity country examples and best practices (Box 2)
- Country examples:
  - Denmark: leading example; significant spending on unemployment insurance and active labor market policies. Denmark’s OECD indicator for strictness of employment protection places it in the middle of OECD countries (length of mandated notification periods drives the indicator).
  - Austria: 2003 severance pay reform increased flexibility while ensuring income security; led to a substantial increase in job mobility (Kettemann et al., 2017).
  - Portugal: reforms between 2011 and 2015 reduced severance pay, eased the definition of fair dismissal, widened safety net, strengthened activation; OECD called reforms “a move in the right direction” with employment and unemployment rates improving significantly (OECD, 2017).

- Best practices and calibration:
  - The three pillars of flexicurity are interdependent and must be calibrated together.
  - Employment protection legislation should be limited; avoid turnover costs due to judicial uncertainty.
  - Legislation should focus on monetary transfers from the firm to the worker and increase smoothly with tenure, avoiding jumps that cause duality.
  - If unemployment insurance is raised, accompany it with effective active labor market policies (Blanchard et al., 2013).
  - Initial conditions matter: trust between firms and workers, “public spiritedness,” low unemployment rate, and healthy fiscal position increase likelihood of success.
  - Continuous calibration and attention to policy detail are critical; choice and design of active labor market policies must be country-specific.

- Comparative spending and fiscal considerations:
  - Denmark spending (2015): Active labor market policy = 1.7 percent of GDP; Unemployment insurance = 1.1 percent of GDP.
  - Korea spending (2015): Active labor market policy = 0.4 percent of GDP; Unemployment insurance = 0.3 percent of GDP.
  - Korea spends the largest share on direct job creation; Denmark focuses on training and sheltered employment/rehabilitation.
  - Fiscal balance: high spending on flexicurity can widen the labor tax wedge and discourage employment creation; Denmark reduced its tax wedge since 2000 and is now close to OECD average. Korea’s tax wedge is comparatively low.

### Policies to address Korea’s labor market duality (summary of policy implications)
- Move toward flexicurity-style reforms that:
  - Reduce excessive protection of jobs in favor of stronger protection for workers (unemployment insurance and activation).
  - Smooth severance and turnover cost schedules to avoid sharp incentives for non-regular hiring.
  - Strengthen active labor market policies, focusing on training and reallocation to productive firms/sectors.
  - Expand targeted social programs and subsidies (e.g., social insurance subsidies for low-wage workers at small businesses).
  - Ensure reforms are calibrated to Korea’s fiscal capacity and social context; continuous adjustment is necessary.

*Source: IMF staff summary of chapter content (cr1841).*

### 18.      A    general equilibrium search-and-matching model is applied to model duality and

### 18.      A    general equilibrium search-and-matching model is applied to model duality and 

### Model description and calibration
- Model type: general equilibrium search-and-matching model based on Dolado et al. (2017).
- Key endogenous outcomes: labor market duality (temporary vs. permanent workers), unemployment, and the “revolving door” spike in job destruction just before temporary workers transition to permanent status.
- Mechanisms captured:
  - Employment protection legislation and search-and-matching frictions generate duality.
  - Differences in wages and job security between temporary and permanent workers.
  - Employment protection legislation that jumps after a certain tenure induces the revolving door.
  - Severance pay and dismissal costs provide insurance value because workers are risk averse and have only partial private insurance.
- Baseline calibration: calibrated to the Korean economy.

### Calibrated model behavior (qualitative)
- Revolving door: significant spike in job-destruction just before temporary workers become permanent and face firing costs.
- Wages:
  - Temporary workers accept lower wages due to lower bargaining power and possible future retention as permanent workers.
  - Wage profile of regular (permanent) workers increases with tenure as bargaining power rises with accumulation of severance pay benefits.
  - Older workers accept lower wages than younger workers because they have lower ability to search for new employment once laid off.
- Distributional implications: reforms produce winners and losers across the income distribution.

### Policy experiments (five simulations) — modeling approach
- Pillars and modeling:
  - More flexible hiring and firing: modeled through a decline in the firing costs.
  - Stronger safety net: modeled through an increase in unemployment benefits.
  - Active labor market policies (ALMPs): modeled as an increase in the productivity parameter in the matching function and an increase in productivity of initial matches (training raises productivity).
- Experiments: each pillar introduced separately, and then a full flexicurity package combining measures.
- Results summarized in Table 2 (values reported exactly as in source).

### Table 2 — Korea: Policy Experiments (selected outcomes)
- Baseline; Decrease in Firing Cost; UI Increase; Increase Matching; Increase Training; Flexicurity Package
- Unemployment rate
  - Baseline: 7.50%
  - Decrease in Firing Cost: 6.90%
  - UI Increase: 7.80%
  - Increase Matching: 7.00%
  - Increase Training: 7.30%
  - Flexicurity Package: 6.70%
- Output (index)
  - Baseline: 1.00
  - Decrease in Firing Cost: 1.00
  - UI Increase: 1.00
  - Increase Matching: 1.01
  - Increase Training: 1.08
  - Flexicurity Package: 1.09
- Average Productivity (index)
  - Baseline: 1.00
  - Decrease in Firing Cost: 1.00
  - UI Increase: 1.00
  - Increase Matching: 1.00
  - Increase Training: 1.08
  - Flexicurity Package: 1.08
- Average Income (index)
  - Baseline: 1.00
  - Decrease in Firing Cost: 1.01
  - UI Increase: 1.00
  - Increase Matching: 1.02
  - Increase Training: 1.08
  - Flexicurity Package: 1.11
- Average Welfare (index)
  - Baseline: 1.00
  - Decrease in Firing Cost: 1.00
  - UI Increase: 1.02
  - Increase Matching: 1.03
  - Increase Training: 1.08
  - Flexicurity Package: 1.12
- Relative job-turnover (temp to perm)
  - Baseline: 7.50
  - Decrease in Firing Cost: 2.29
  - UI Increase: 7.50
  - Increase Matching: 7.50
  - Increase Training: 8.00
  - Flexicurity Package: 4.29
- Relative wage (temp to perm)
  - Baseline: 0.75
  - Decrease in Firing Cost: 0.83
  - UI Increase: 0.75
  - Increase Matching: 0.71
  - Increase Training: 0.77
  - Flexicurity Package: 0.87
- Income Gini
  - Baseline: 16.1
  - Decrease in Firing Cost: 15.4
  - UI Increase: 15.6
  - Increase Matching: 16.6
  - Increase Training: 16.3
  - Flexicurity Package: 15.0

### Findings by reform pillar
- Decrease in firing costs (about 60 percent reduction simulated)
  - Substantially reduces the revolving door by lowering job-destruction among temporary workers.
  - Raises temporary workers’ wages relative to permanent workers.
  - Significantly reduces inequality and the unemployment rate (less turnover → less time unemployed for temporary workers).
  - Little change in average income, welfare, productivity, and total output.
  - Distributional: income gains accrue to the bottom 40 percent of the income distribution (especially the young); the upper 40 percent lose income.
- Increase in unemployment benefits (simulated increase of 15 percent)
  - Lowers inequality (Gini declines).
  - Raises welfare.
  - No effect on duality and no significant impact on productivity, output, or average income.
  - Raises the unemployment rate due to higher incentives to remain unemployed.
  - Policy caveat: introducing unemployment benefits alone can address inequality but risks higher unemployment.
- Enhance matching productivity (increase matching productivity by 20 percent)
  - Increases output, income, and welfare through lowering unemployment.
  - No reduction in the revolving door; may exacerbate inequality by widening the wage gap between temporary and permanent workers (firms more ready to separate and post new vacancies).
- Increase training (increase productivity by 10 percent simulated)
  - Strongest impact among single measures on output, productivity, income, and welfare.
  - Exacerbates the revolving door effect and raises inequality because training raises productivity of the outside applicant pool.
  - Simulation does not account for costs of these measures; reported effects are an upper bound.
- Flexicurity package (all three pillars combined)
  - Raises output, average productivity, income, and welfare significantly.
  - Substantially lowers the revolving door and reduces the wage gap between temporary and permanent workers.
  - Lowers income inequality (Gini declines to 15.0).
  - Distributional analysis: the package raises income and welfare across all quintiles, both young and old.

### Conclusions and policy recommendations
- Nature of the problem:
  - Labor market duality in Korea is multifaceted, prominent among SMEs and the service sector, and disproportionately affects women, the elderly, and youth.
  - Likely macro consequences: increasing inequality, low fertility rates, and declining productivity growth.
  - Main drivers: employment protection legislations and large productivity differentials in the product market; insider-outsider dynamics, globalization, and technological progress also contribute.
- Policy guidance:
  - Korea should consider implementing a tailored version of flexicurity combining all three pillars (flexibility, safety net, active labor market policies) to reduce duality, lower inequality, and raise productivity and welfare.
  - Complementary components needed: flexibility at the intrinsic margin (flexible jobs to fit heterogeneous needs), education promoting responsibility and independent thinking, and a culture of life-long learning.
  - Social dialogue and trust: ownership by all social partners is critical, including non-unionized workers, SMEs, and the self-employed.
  - Implementation principles: measures should be carefully designed and targeted, gradually implemented, and constantly revised.
  - Specific design choices: unemployment benefits can be increased/expanded/extended in various ways; rigidity in regular contracts can be reduced by simplifying dismissal procedures or smoothing severance pay schedules; careful design and targeting of ALMPs is essential, with attention to women, youth, and the elderly.

*REPUBLIC OF KOREA — INTERNATIONAL MONETARY FUND*

### 2.      Youth unemployment and inactivity are having a negative impact on Korea’s economy

### 2.      Youth unemployment and inactivity are having a negative impact on Korea’s economy

### Overview and societal impacts
- Youth unemployment and inactivity have a scarring effect: youth who experienced periods of inactivity (i.e., not in employment, education or training) had a lower rate of employment, and higher rates of unemployment and inactivity later in life; they were more likely to be non-regular workers with lower wages (Nam and Kim, 2013).
- Marriage and fertility trends linked to youth labor-market conditions:
  - Marriage rates declined to a record-low of 5.5 per 1000 people in 2016.
  - Fertility rate at 1.17 in 2016, the lowest level since 2009.
- Public opinion surveys suggest growing pessimism about upward mobility and the return on effort and diligence (Kim, 2014).

### Trend and current state of Korea’s youth (un)employment
- Korea’s youth unemployment (age 15 to 24) and comparative position:
  - Youth unemployment rate in 2016: 10.7 percent (age 15 to 24), more than 2 percentage points lower than the OECD average.
  - Adult unemployment rate in 2016 (age 25 to 54): 3.4 percent.
  - Youth unemployment is more than three times as high as adult unemployment in Korea.
  - Korea is one of the few countries where youth unemployment increased since 2012; it had the third highest increase in the OECD after Turkey and Norway.
- Composition and breadth of the problem:
  - More than a third of the young unemployed have tertiary education; almost all have finished upper secondary education.
  - Youth not in employment, education or training (15 to 29): 18 percent as of 2013 (acknowledged as possibly overestimated due to data limitations on programs between upper secondary and post-secondary education).
- Employment quality and mismatches:
  - Korea’s employment rate for youth: 27 percent in 2016, 13.5 percent below the OECD average.
  - Field-of-study mismatches: 33 percent among the youth (type of mismatch among employed youth).
  - Share of temporary employment for youth in 2016: 25.5 percent (OECD average: 24.6 percent).
- Demographic-driven changes (age 20 to 29 focus where data allow):
  - Youth population declined by 1.2 million between 2000 and 2012.
  - This translated predominantly into a decline in employed youth by almost 900 thousand and an employment rate decline from 60.1 percent to 58.1 percent.
  - Unemployment rate remained around 7.5 percent during 2000–2012.
  - Since 2012 youth cohort grew by 200 thousand until 2016; unemployment rate rose from 7.5 in 2012 to 9.8 in 2016; employment rate rose only slightly to 58.3 percent in 2016.
- Recent job creation characteristics:
  - Increase in youth employment since 2012 concentrated in the service sector (hotels and restaurants, wholesale and retail trade) and in SMEs; jobs in manufacturing declined.
  - Many of the jobs created had monthly wages from 1.9 to 2.7 million won (which translates into 67 percent to 96 percent of the average wage in 2016).
- Short-term vs long-term unemployment:
  - Very low share of long-term unemployed (unemployed for one year or longer) among those aged 15 to 24: 0.5 percent.

### Drivers of youth (un)employment — summary
- Three main categories of drivers discussed: (i) cyclical factors, (ii) structural drivers, and (iii) policies.
- Empirical summary:
  - Cyclical factors, especially consumption growth, have a significant impact on youth unemployment in Korea, but the effect is smaller than in most other countries.
  - Structural factors (skill and expectation mismatches, labor market duality) significantly contribute to the level of youth (un)employment.
  - Demographic changes (temporary growth in youth cohorts, large education gaps between elderly and youth) likely contributed to the recent uptick in youth unemployment.
  - Policy variables that have affected youth (un)employment include high protection of regular workers, educational quality, and a focus of active labor market policies on direct job creation.

### Cyclical factors
- Theoretical channel:
  - Economic downturns reduce job vacancies and raise competition; youth are more sensitive to output growth and shocks because they have less firm-specific tenure and more fragile employment conditions.
- Korea’s empirical case:
  - Regression analysis based on Okun’s law: real GDP growth can explain about 20 percent of the variation in the youth unemployment rate for Korea.
  - Youth unemployment in Korea appears less elastic to GDP than in most other countries; possible reasons include a large share of highly educated youth and a tendency for youth to become inactive rather than unemployed.
  - Sensitivity to GDP components is largely driven by consumption growth (consistent with employment concentration in non-tradeable services such as hotels and restaurants).
- Sectoral evidence:
  - A smaller share of youth employment in manufacturing and construction (lower sensitivity to exports and investment); a larger share in non-tradeable service sectors explains higher sensitivity to consumption.
  - Temporary employment share for youth: 25 percent in 2016 versus 16 percent for ages 25 to 54.

### Structural factors
- Skill mismatch
  - Theoretical channel: mismatches in education level, specific skills, or field of study lower employment probability and/or employment quality; overeducation can create traps that retard transitions to adequate jobs.
  - Korea’s case:
    - Education mismatch has been increasing since 2006; Dao et al. (2014) estimate that a 10 percent decrease in mismatch could raise employment rates by about 6 to 14 percentage points.
    - Significant number of jobs remained unfilled in SMEs in 2016, especially for lower skill levels requiring only high school graduation.
    - Korea has by far the largest educational gap between elderly and youth; jobs vacated by older cohorts are often not suited to high-skilled youth.
    - Evidence that formal education falls short of businesses’ demands (Kim, 2015; OECD, 2012).
    - Field-of-study mismatch found to be high in Korea at 50 percent, compared to a country average of 38 percent (Montt, 2015).
- Expectation mismatch
  - Theoretical channel: higher-educated youth may choose unemployment or inactivity due to high reservation wages or expectations; in segmented/rigid labor markets, waiting for better jobs can be optimal.
  - Korea’s case:
    - 2015 national survey: 44 percent of youth aged 15–29 desired a career in government and public enterprises; 25 percent desired large companies.
    - Labor market duality: large wage premia for regular versus non-regular workers and for employees in large versus small companies; premia exceed those for college graduates.
    - Extensive preparation for preferred jobs (recruitment exams, extra diplomas/certificates) delays market entry.
    - More than 80 percent of young people who left their first job did so voluntarily; 47 percent left due to dissatisfaction with working conditions (Keum and Yi, 2016), supporting the presence of expectation mismatch.
- Demographics
  - Theoretical channels include capital-labor ratio effects, cohort-size effects on vacancies, and choices to stay in education; empirical evidence typically finds that larger cohorts lower wages and increase youth unemployment, though results vary by study.
  - Korea’s observed demographic influences:
    - Decline in youth population between 2000 and 2012 (1.2 million) led to large employment declines; subsequent cohort growth since 2012 contributed to rising unemployment between 2012 and 2016.

### Policy-relevant observations (from the source)
- Cyclical policies that support consumption growth can materially affect youth unemployment, given the higher youth employment share in consumption-sensitive service sectors.
- Structural policy levers to address youth (un)employment include:
  - Addressing skill and field-of-study mismatches between education outputs and labor-market demand.
  - Tackling labor market duality (high protection of regular workers, large wage premia in large firms and for regular contracts) to reduce incentives for queuing and prolonged job search among youth.
  - Reassessing active labor market policies that focus on direct job creation to ensure they improve job quality and match for youth.

*Source: IMF staff analysis from chapter "2. Youth unemployment and inactivity are having a negative impact on Korea’s economy" (cr1841).*

### 20.      Korea’s case: As discussed above, Korea

### 20.      Korea’s case: As discussed above, Korea

### Youth cohort trends and labor market outcomes
- Korea experienced a period of almost 20 years in which the youth cohort declined.
- Since 2014, youth population (age 20 to 29) has started increasing again, which has coincided with increases in youth unemployment and some absolute increase in employment.
- Most of the increased employment has taken place in SMEs and the service sector.
- This recent increase is temporary and Korea is forecasted to reach a cliff in 2019 after which the youth cohort will drop off sharply.
- Risk: the current youth cohort faces the risk of becoming a “lost generation”, given the potential scarring effect of unemployment spells at a young age.

### Structural changes in labor market demand
- Theoretical channel: Technological change and integration into global value chains (GVCs) can drive polarization of the job distribution and the loss of certain routine jobs (Autor, 2010; Acemoglu and Autor, 2011).
- Youth may be disadvantaged if the education system does not prepare them for new job types; alternatively, youth may learn and adjust faster than older cohorts.
- Korea’s case:
  - Employment for youth in manufacturing and large companies has recently declined, which could be related to structural changes.
  - Manufacturing sector breakdown shows significant differences: the chemicals sector has added jobs for youth while being highly integrated in GVCs; electronics and computers has lost the most jobs for youth and is also highly integrated in GVCs.
  - Change in employment in manufacturing sectors, 2013-2016 (Age 15 to 29, thousand) shows large negative changes for Electronic Components, Computer, etc., and positive changes in Chemicals and Pharmaceuticals (sources: Kosis).

### Policy variables affecting youth (un)employment
- Significant policy variables in Korea:
  - High protection of regular workers.
  - Educational quality.
  - Focus of active labor market policies on direct job creation.
- Standard policy variables (tax wedge, minimum wage, generosity of unemployment benefits) are comparatively moderate in international comparison and are likely not strong factors driving the level and recent change in youth (un)employment.

### Labor costs
- Theoretical channel: Higher labor costs raise the relative price of labor and can lower labor demand (Banerji et al., 2014). Minimum wages and tax wedges can raise labor costs.
- Korea’s case:
  - Korea’s tax wedge was 22 percent in 2016, significantly below the OECD average of 36 percent.
  - Enterprise Labor Cost Survey (2015): 79 percent of labor costs were direct labor costs; severance pay accounted for 9 percent; legally required benefits accounted for 7 percent.
  - Other analysis suggests cost of dismissal, not hiring or employing, limits availability of regular employment (OECD, 2013; Kim 2015).
  - Korea lies above the OECD average for the minimum wage if measured relative to median wage.
  - Minimum wage history:
    - Introduced in 1988, applied to all workers since 2000.
    - Full minimum wage only recently expanded to workers below the age of 18.
    - Government has struggled with and recently tried to improve enforcement (OECD, 2013).
  - Most youth below the age of 30 received wages significantly above the minimum wage, but significant hikes could affect wages and potentially employment.
  - Youth employment by monthly wage (Below age 30, employed persons in 2016) distribution shown by education levels (sources: Kosis, Staff calculations).
  - Employment counts:
    - 2016: 1260
    - 2017: 1352
    - 2018: 1573
    - 2022: 2090

### Opportunity costs (unemployment benefits)
- Theoretical channel: High unemployment benefits may reduce search effort and increase reluctance to accept lower wages; adequate insurance can allow better job matching (Banerji et al., 2014).
- For youth, eligibility is a major issue as first-time job seekers are usually not part of unemployment benefit schemes and often rely on unemployment assistance benefits (ILO, 2012; Carcillo et al., 2015).
- Korea’s case:
  - Net replacement rate in the initial phase of unemployment was 56 percent, slightly below the OECD average of 63 percent.
  - Unemployment benefits require being insured under the Employment Insurance System for at least 180 days; no benefits if the employee quit voluntarily (OECD, 2007).
  - Young Koreans are less likely to be eligible for these benefits and often leave jobs voluntarily.
  - Government introduced an “Employment Success Package Program” in 2009 offering allowances to job seekers conditional on job-seeking activity or participation in trainings.

### Active Labor Market Policies (ALMPs)
- Theoretical channel: Activation policies (trainings, conditional allowances) are expected to increase employability and incentives to search for work; effectiveness depends on program type and implementation (Banerji et al., 2014).
- Korea’s case:
  - Spending on ALMPs was 0.4 percent of GDP in 2015, compared to an OECD average of 0.5 percent.
  - ALMP spending also falls below the OECD average if measured per unemployed.
  - Composition of ALMP spending in 2015: about 56 percent on direct job creation; training and employment incentives made up about 28 percent.
  - Cross-country research indicates direct job creation tends to be the least effective.
  - The new government plans to expand various of these measures.

### Key policies on youth (un)employment (summary of Box 1)
- Supply-side measures:
  - Meister schools (introduced 2010) based on German master craftsperson method; job placement rates at 90 percent; number increased to 41 schools by 2016.
  - National Competency Standard (NCS) developed in 2013; new curricula based on NCS introduced to high schools, Meister schools and vocational colleges.
  - Work-study dual system (introduced 2013): firms hire young job-seekers who study and earn NCS certificates with government financial support; Jeon et al. (2015) estimate net benefit of one position to be 8.9 million KRW. By August 2017, around 50000 youth and 10000 companies had participated.
  - Stepping stone program (2016): job training and education in public institutions and large enterprises.
  - Employment Success Package Program targets young unemployed age 18 to 34, financially supports job-seeking plans or vocational training (MOEL, 2016).
  - Improvements needed: ensure quality of placements, increase ownership by businesses, expand reach of programs (Kim, 2014).
- Demand-side measures:
  - Increasing public employment quotas for youth since President Roh; since 2013, government agencies and public enterprises have a quota for youth of 3 percent.
  - Subsidies to SMEs employing youth aged 15 to 29: tax credit of five million won for each permanent young job-seeker employed; 150 percent tax deductions for increases in wages paid for young permanent workers (MOSF, 2015).
  - 2017 subsidy: KRW 20 million a year for three years to SMEs that hire three youth workers as regular workers.
  - Support for voluntary improvements in SMEs working conditions (MOEL, 2016).
  - Youth entrepreneurship programs: Young Entrepreneurs Start-up Academy, Youth Development Fund, “Youth Business 1000” program.
  - K-Move program helps youth find jobs overseas; in 2016 it secured close to five thousand jobs (HRD, 2017).
  - To address quality of youth employment, critical to tackle dualities between regular and non-regular and small and large companies through reforms in labor and product market.
- Matching and information measures:
  - Since 2009, government subsidizes 6-month internships for unemployed youth in SMEs; initiative refocused on high-growth SMEs.
  - Expanded on-campus youth employment centers and a human resources computer network to improve information exchange (MOEL, 2016).
  - Suggestion: introduce pre-emptive measures to help youth adapt to a changing labor market and support career planning at an early stage.
- New administration plans (supplementary budget for 2017 and policy program):
  - Ask public institutions to fill more than 5 percent of their positions with youth.
  - Provide SMEs with extra incentives for hiring youth and promote overseas employment.
  - Pledge to reform public education and introduce innovative job training programs.
  - Supplementary budget plan for 2017 included: increase in the young startup fund by 500 billion won; adoption of a young job seeker allowance; increase in young SME employees matching funds to their asset building accounts.

### Education and vocational training
- Theoretical channel: Higher educational attainment tends to lower labor force participation as more time is spent in education; higher education generally increases productivity and employability. Quality and applicability of skills are crucial (Banerji et al., 2014).
- Korea’s case:
  - High social value of university education.
  - Gross enrolment ratio of tertiary education at 93 percent in 2015.
  - Very large private expenditure on higher education.
  - Korean students perform well on PISA; high school students study intensively for CSAT.
  - Universities chosen for network-enhancing role rather than teaching/research performance; curricula and lack of creative and critical skill development lead to job mismatches and too few technical specialists (Kis and Park, 2012; Eichengreen et al., 2015).
  - Vocational training has a negative image; some programs low quality and weak links to business (Kis and Park, 2012).
  - Low enrolment in vocational education and training at upper secondary level: 18 percent in 2014, compared to OECD average of 44 percent.
  - Government has undertaken significant steps to address these issues (see Box 1).

### Best practices (section D)
- Successful policies need to be comprehensive and tailored, with recurring impact evaluations and adjustments.
- Key design features:
  - Formulate and implement as a comprehensive policy framework embedded in national development plans.
  - Consider structural and cyclical aspects; combine demand- and supply-side interventions.
  - Recognize individual characteristics of youth and specific barriers they face.
  - Include qualitative targets with regular monitoring and evaluation to ensure adjustments.
  - Ensure coordination and coherence across stakeholders and include affected youth in design and implementation (ILO, 2012; ILO, 2015).
- Active labor market policies:
  - Should address individual characteristics and be well-targeted to the most vulnerable.
  - Design should avoid deadweight losses and substitution of existing employment.
  - Coordination with passive labor market policies (unemployment insurance and social assistance) provides incentives for participation.
  - Examples: Netherlands and Denmark conditionalities; youth guarantee schemes provide commitments to place youth in education, training or work programs (Banerji et al., 2014).

*International Monetary Fund. Chapter content on Korea’s youth employment (extracted from provided PDF content).*

### 34.      Well-designed fiscal policy can play an incentivizing role to increase private sector

### 34.      Well-designed fiscal policy can play an incentivizing role to increase private sector

### Fiscal policy and labor demand (paragraphs 34, 43)
- Counter-cyclical policies can reduce the negative impact of volatility on youth employment (ILO, 2012).
- Fiscal policy can stimulate demand through a reduction in labor taxes or social security contributions (SSC).
- Evidence and design guidance:
  - Cuts targeted to low-wage earners are most effective (Hammermesh, 1993; IMF, 2012).
  - Employer SSC cuts and application in more rigid labor markets have been effective (IMF, 2014).
  - Targeting of cuts can reduce fiscal costs but requires careful incentive design; cuts conditional on wage levels should be phased out gradually to avoid a “low-pay trap” (IMF, 2014).
  - Hiring subsidies should be short-term, carefully targeted and closely monitored (Kluve et al., 2016; IMF, 2014).
  - Targeting should be based on workers instead of firms (IMF, 2012) and avoid concentration on characteristics such as age and diploma to prevent substitution effects (Carcillo et al., 2015).
  - Program linkages to on-the-job training are critical (ILO, 2012).
- Macro recommendation:
  - Given youth unemployment sensitivity to the business cycle, address the current negative output gap and weak consumption growth; expansionary fiscal policy can help increase domestic demand and boost employment growth (paragraphs 43, 44).
  - Prioritize demand-boosting measures while cushioning policies that could have short-term contractionary effects (paragraph 44).

### Public employment programs (paragraph 35)
- Empirical findings:
  - Public employment programs have largely shown no or even negative effects on youth employment (Sianesi, 2001; Caliendo et al., 2011; Dorsett, 2006; Card et al., 2010).
- Potential reasons for weak or negative impacts:
  - Lack of transferable skills acquired.
  - Potential for stigma.
  - Reduction in search efforts.
  - Crowding out of private employment and the creation of applicant pools waiting for permanent public-sector jobs (IMF, 2012; Carcillo et al., 2015; ILO, 2015).
- Design improvements to enhance impact:
  - Clear policy objectives and careful targeting.
  - Ensure interlinkages with other activation strategies and skills training that increase employability beyond the public sector (ILO, 2015).

### Structural policies and labor market design (paragraph 36)
- Product and capital market reforms can encourage job growth (IMF, 2012).
- Labor market reforms can reduce duality between temporary and permanent contracts and improve youth transitions (OECD, 2010).
- Some countries use a separate minimum wage for youth set below adults to encourage employment; empirical results on this approach have been inconclusive (ILO, 2012).

### Supply-side measures: skills, apprenticeships, entrepreneurship (paragraphs 37–38)
- Skills and education alignment:
  - Supply-side measures must be based on rigorous analysis of skill demand and forecasting.
  - Forecasts should adjust curricula and inform counselling and guidance (ILO, 2012).
- Work-experience programs:
  - Apprenticeships, internships and labor market training should combine theoretical training with work experience.
  - Programs must provide recognized qualification or certification and market-based design; integration with formal schooling and a gateway to good quality jobs is crucial (ILO, 2015).
  - Incentives for apprenticeships have included tax credits, scholarships or exemptions (OECD, 2014); business colleges and applied universities can supply market-oriented skills (Schmid, 2013).
- Youth entrepreneurship:
  - Programs should identify barriers faced by youth and offer comprehensive packages: theoretical classes, financial support, business advisory services, mentoring and follow-up (ILO, 2012; ILO, 2015).
  - Introduce entrepreneurship into secondary and tertiary curricula to increase awareness and reputation of the profession (ILO, 2012; ILO, 2015).

### Transition and matching policies (paragraph 39)
- Public employment services must be personalized and provide accurate labor market information.
- Key elements:
  - Design an individual employment plan early in unemployment spells.
  - Use profiling systems based on accurate and timely labor market information and client personal situations.
  - Reach diverse groups and target those most in need via broad partnerships with educational institutions (ILO, 2012; ILO, 2015).
- Example — Denmark:
  - Quick, intensive, individualized services for youth of all ages.
  - Lower secondary pupils prepare education plans with parents, school and guidance center.
  - Older youth receive individual interviews, job-search training and educational opportunities or work placements quickly after welfare benefit applications.
  - A database providing an overview of all youth and their education is used to identify and target vulnerable youth (OECD, 2010).

### Korea-specific policy implications and recommendations (paragraphs 40–44)
- Context and drivers:
  - Korea’s youth unemployment rate is low in international comparison but has recently increased; share of inactive youth is exceptionally high (paragraph 40).
  - Contributing factors include cyclical, structural and policy variables: weak consumption, a temporary increase in the youth cohort, expectation and skill mismatches, educational quality, focus on direct job creation, and high protection of regular workers leading to labor market duality.
- Existing measures:
  - Meister schools, the work-study dual system and internships are important and can have long-run positive effects; strengthen business ownership, expand reach, and ensure quality of placements (paragraph 41).
  - Continuously evaluate and monitor impact and costs; adjust programs as necessary; retain a comprehensive policy framework for coordination.
  - Consider national media campaigns to encourage participation and stakeholder involvement; use direct job creation carefully and link it to services not adequately provided by the private sector (paragraph 41).
- Pre-emptive engagement:
  - Expand pre-emptive measures such as education plans in lower secondary education and skill-demand forecasts to guide educational investments.
  - Consider a database to identify vulnerable youth and provide ALMPs to discouraged youth, following Denmark’s example (paragraph 42).
- Broad-based structural reforms:
  - Product market reform to increase productivity in SMEs and services to provide decent jobs and reduce segmentation.
  - Reduce duality in employment contracts by simplifying dismissal rules for regular workers and adjusting contracts and benefits to a smoother schedule (paragraph 43).
- Comprehensive youth strategy:
  - Tackle youth (un)employment as part of a comprehensive, coordinated package addressing labor market duality, product market segmentation and sluggish domestic demand.
  - Encompass a move towards flexicurity, stronger social safety nets, product market structural reforms, and an increase in fiscal spending to boost labor supply (paragraph 44).
  - Implement policies in a coordinated and gradual way; prioritize demand-boosting measures given significant fiscal space and a still negative output gap (paragraph 44).

*Source: https://www.imf.org/-/media/files/publications/cr/2018/cr1841.pdf*

### 1. The new administration of President Moon announced in July 2017 its economic policy

### 1. The new administration of President Moon announced in July 2017 its economic policy

### Policy announcement and 2018 minimum wage decision
- The new administration’s platform aims to shift the “growth paradigm from large production and fast-follower-led growth to sustainable growth.”
- A key pillar is income-led growth, which includes raising the minimum wage.
- The Minimum Wage Commission increased the statutory minimum wage for 2018 to KRW 7,530 per hour, a 16.4 percent rise compared to this year.
- To partially compensate small businesses for the increased labor cost, the government plans to pay part of the additional burden, which is estimated at KRW 3 trillion (0.2 percent of GDP) for 2018.
- Last year (pre-2018), Korea’s minimum wage stood at $5.8 per hour (PPP for private consumption); in 2018 it will increase to $7.2 per hour.
- OECD context: Korea’s 2016 minimum wage ranked 16th out of 27 OECD countries; the 2018 minimum wage would place Korea in the middle of the group, assuming no change in other economies’ minimum wages.
- The hike will push the ratio of minimum to average wage slightly above the OECD average.

### General considerations on minimum wage effects
- Employment
  - A minimum wage increase can cause some low-skilled workers to lose employment, some to keep jobs with higher wages, or lead to non-compliance or reductions in non-wage benefits; the net effect is ultimately empirical.
  - Neumark and Wascher: a 1 percent increase in the minimum wage would lower youth employment by around 0.12 percent in the short run and 0.21 percent in the longer run.
  - IMF (2016) for CESEE: a 10 percent increase in minimum wage is associated with a youth employment reduction of 2 percent.
  - OECD (2015): “at reasonable levels, increases in the minimum wage are unlikely to cause substantial job loss,” but negative impacts can be larger for youth and vulnerable groups; country- and time-specific factors matter.
  - Case studies: New Jersey—no systemic impact; Seattle—significant negative impact on low-skilled labor; France—high minimum wage relative to average wage can lower overall employment.
- Overall wage levels and pass-through
  - A minimum wage increase mechanically raises the overall wage level and can have a “ripple” effect.
  - U.S. evidence: a 1 percent increase in the minimum wage is estimated to increase payments to workers at or just above the minimum wage by 0.8 percent and payments to workers further up the wage scale by around 0.25 – 0.4 percent.
  - IMF work on CESEE: up to 0.15 percent pass-through to general wages over two years in response to a 1 percent increase in the minimum wage.
- Non-compliance
  - Widespread non-compliance dampens the intended effects: examples include two-thirds of garment factories in California and more than 10 percent of workers in the U.K. social care sector paid below statutory minimums.
- Inequality
  - Minimum wage is an important tool to manage income inequality and, unlike many other measures, does not involve direct fiscal costs.
  - Broad consensus: minimum wages lower inequality provided they are not high enough to significantly lower employment (especially for youth and low-skilled).
  - Autor, Manning, and Smith (2016): during 1979–89, decline in the real value of the minimum wage accounted for 30 to 55 percent of the rise in lower-tail inequality.
- Competitiveness: firm responses
  - Firms can: pass costs to consumers, accept lower profits, or raise productivity.
  - Empirical examples:
    - Pass-through: Aaronson (2001) finds restaurant prices rise when wage bills rise.
    - Lower profits: Bell and Machin (2015) find unexpected minimum wage increases reduced stock values of low-wage firms in the U.K.
    - Higher productivity: Riley and Bodibene (2015) show firms with many low-skilled workers raised productivity in response.
  - Net impact on competitiveness depends on country and sector characteristics.
- Growth and inflation
  - Impact on growth is contested: proponents argue redistribution could boost consumption and growth; opponents argue competitiveness and employment could suffer.
  - Onaran and Galanis (2012): decline in wage share tends to lower growth in a number of countries (including Korea) and raise growth in others.
  - Bassanini and Venn (2008): positive impact of minimum wage on productivity.
  - Sabia (2015): no evidence that minimum wage increases change output, but output shifts from low- to higher-skilled industries.
  - Cyclical context matters: with slack and moderate minimum wages, hikes can boost demand without significant employment loss; with closed or positive output gap, hikes can raise inflation and prompt monetary tightening.
  - Lemos (2004): a 10 percent increase in the minimum wage in the U.S. would increase overall prices by no more than 0.4 percent (food prices increase by 4 percent).

### The case of Korea — key factual findings
- Coverage and historical prevalence
  - In Korea, about 7 percent of workers receive the minimum wage.
  - Comparative figures: less than 5 percent in the U.S. and the U.K.; around 10 percent in France and Germany.
- Historical econometric analysis (staff)
  - A co-integrating regression found real minimum wage, real average wages, and employment are co-integrated; exogenous controls included change in terms of trade, productivity growth, and population growth; a dummy accounted for the Asian crisis.
  - Historically the minimum wage in Korea has followed, rather than led, average wages.
  - A 1 percent increase in average real wages increased the real minimum wage a year later by about the same magnitude.
  - Short-run: employment is positively correlated with average and minimum wages (consistent with average wages leading minimum wages).
  - Long-run: total employment declines with higher minimum wages.
  - Minimum wages explain about 30 percent of the variation in youth employment (corresponding to 1 percentage point), compared with around 10 percent for non-youth employment (corresponding to 0.2 percentage point); estimations control for working age population growth and other exogenous factors.
  - Caution: Lee (2008) did not find statistically significant impact of Korea’s minimum wage increase on youth employment; possible reasons include high youth education levels (60 percent of unemployed youth are university graduates) and job mismatches with youth population increase.
- Income distribution and competitiveness historically
  - Because minimum wages have tended to follow average wages and not materially change overall employment historically, increases in the minimum wage are unlikely to have materially changed overall income distribution after initial establishment in late 1980s.
  - Yoon (2015) offers an alternative view: rising working poor and concentration of minimum wage workers in the lowest income group imply minimum wage increases can have considerable impact on the working poor.
  - Export sector competitiveness: most exporters use little low-skilled labor and have remained strong; large wage increases could reduce competitiveness because exporters generally cannot pass on cost increases to prices.
  - Service sector and small firms: bulk of minimum wage earners are in services; no clear historical evidence that minimum wage hikes unduly impacted these firms, but more analysis is needed.
- Historical growth and inflation evidence in Korea
  - No clear historical evidence that minimum wage increases affected Korea’s growth and inflation—likely because minimum wages historically followed productivity growth and wages.

### The 2018 minimum wage hike — specific implications and risks
- Unprecedented nature
  - The 2018 hike is unprecedented: substantially higher than recent productivity growth, wage growth, and inflation; occurring when youth unemployment is high and increasing.
  - The only comparable episode was 2001, which followed strong productivity and wage growth and declining youth employment—conditions not present in 2018.
- Expected wage and employment effects
  - Staff expect the sharp increase to push overall wages up due to departure from historical pattern and the size of the real-term hike.
  - Potential adverse effects:
    - Increased difficulty for low-skilled workers, youth, and the elderly in finding employment.
    - Firms that cannot pass costs to prices may reduce employment and increase productivity, or face closure.
    - Elderly owners/employees of small mom-and-pop shops are vulnerable to closures.
  - Final impact depends on the economy’s ability to generate employment in other areas.
- Firm-level balance of forces
  - Two opposing forces for firms: higher domestic demand from higher wages versus higher labor costs.
  - Korea’s firms are overall in good health (reflected in strong stock market increases), and average wage aligns broadly with average value added per worker compared to other OECD countries.
  - Vulnerable firms: those that cannot pass on costs or raise productivity, typically small service firms employing low-skilled workers (mom-and-pop shops).
- Indicators and staff visuals (as presented)
  - Youth employment by monthly wage (below age 30, employed persons in 2016) — graphical distributions presented.
  - Numerical youth employment markers shown in figure: 2016: 1260; 2017: 1352; 2018: 1573; 2022: 2090.

*International Monetary Fund staff summary of IMF Republic of Korea chapter content.*

### 6000.0 and over

### 6000.0 and over

### Minimum wage hike and household demand (2018)
- Employment by monthly wage (all employed persons in 2016): 
  - 2016: 1260
  - 2017: 1352
  - 2018: 1573
  - 2022: 2090
- Key finding:
  - The 2018 minimum wage hike should support growth by boosting domestic demand because income will shift to lower-income workers, who have lower saving and higher consumption rates.
  - However, some low-skilled workers may face greater difficulty finding jobs.

### Policy implications for the 2018 minimum wage hike
- Analyze the impact of the 2018 wage hike before additional hikes:
  - Monitor impacts on youth, the elderly, smaller firms, and firms that mostly employ low-skilled labor.
  - Literature suggests a “moderate” minimum wage would not be detrimental, but determining “moderate” is crucial.
  - With the 2018 hike, Korea’s minimum wage level will be at many of the other advanced economies; another large hike on the heels of 2018 may stress labor markets and firms.
- Boost active labor market policies:
  - Provide training and job-matching support for workers who lose employment as a result of the 2018 hike.
  - Channel resources to active labor market policies rather than subsidizing SME costs, to avoid perpetuating incentives for firms not to grow (the “Peter Pan Syndrome”).

### Corporate landscape: chaebols and data overview
- Data and sample:
  - Firm-level consolidated balance sheets and profit and loss accounts from 2006 to 2015 for 200 very large Korean companies (Orbis database).
  - Inclusion criteria for very large companies: operating revenues at least US$ 100 million; total assets at least US$ 260 million; number of employees at least 1000; listed on the stock exchange.
  - Sample composition: 196 firms in analysis, of which 98 were identified as chaebols affiliated; 55 of these were part of the top four Chaebols (Samsung, Hyundai, LG and SK).
- Definitions and context:
  - “Chaebol” denotes groups of formally independent firms under common administrative and financial control of one dominant family, often interlocked by circular shareholdings.

### Stylized facts on size, profitability, and balance-sheet indicators
- Size and growth:
  - The average asset size of chaebols affiliated companies is much higher than that of other large Korean companies.
  - As of 2015, the average size of the 55 affiliated companies of the top four chaebols in the sample was nearly double that of other large Korean firms.
  - The average assets of the top four chaebols affiliated companies recorded a growth rate of over 50 percent from 2011 to 2015, compared to less than 40 percent growth rates in the other two firm groups.
- Profitability:
  - Average return on assets has typically been higher among affiliates of the top 4 chaebols than in other conglomerates.
  - Profitability appears to have declined after the global financial crisis; since 2013 it appears, on average, higher in large companies not part of conglomerates.
  - Since 2009, chaebols excluding the top four have recorded the lowest profitability; the share of firms with negative returns increased steeply in this group since 2010 and was as high as 30 percent in 2015.
- Liquidity, solvency, and leverage:
  - Liquidity ratio (current assets/current liabilities) shows polarization: top four chaebols and other large companies on one hand, chaebols excluding top four on the other.
  - Solvency ratio (shareholder funds / liabilities) is much lower for chaebols excluding the top four compared to the other two groups.
  - Leverage (debt-to-equity and debt-to-asset ratios) is more elevated in chaebols excluding the top four.
  - Industry-adjusted debt-to-equity (firm debt-to-equity minus sector average) is higher for chaebols excluding the top 4.
  - Interest cover ratio (operating profits / interest paid) is low for chaebols excluding the top four, indicating potential challenges in meeting interest payment obligations.
  - These balance-sheet weaknesses are mirrored in lower Tobin’s Q for chaebols excluding the top four.

### R&D and investment patterns
- R&D spending:
  - The top four chaebols are spending double as much on R&D (as a share of their operating revenues) than the other companies.
  - R&D spending is lowest among chaebols excluding the top four.
  - R&D spending has increased somewhat since 2006 across the three groups.
- Investment growth:
  - Investment growth has been on a slight downward trend in all three firm groups.
  - In recent years, investment growth has typically been weaker in chaebols excluding the top four.

### Determinants of investment (econometric results, 2007–2015)
- Model setup:
  - Panel regression on 200 Korean firms over 2007 to 2015.
  - Specification follows a neoclassical investment model: investment related to expectations of future profitability (proxied by Tobin’s Q) and cash flow (liquid assets); sector dummies and time effects included.
- Main findings:
  - Tobin’s Q (expectations of future profitability) is an important and significant determinant of overall investment behavior.
  - Cash flow (as percent of assets) is not statistically significant for the overall sample.
  - When leverage (debt-to-equity) is added, leverage has a significant and negative coefficient.
  - Profitability and interest coverage ratio are not statistically significant in the investment regressions.
  - Sector dummies are not statistically significant, suggesting similarity of the investment function across sectors.
- Heterogeneity across firm groups:
  - Interaction with firm-group dummies shows Tobin’s Q is significant only for chaebols (both top four and others), but not for other large companies.
  - Cash flow is positive, large, and statistically significant for large companies that are not chaebols, but not for chaebols.
  - Results hold when leverage is added as a regressor.
- Interpretation:
  - Chaebols’ investment is driven by growth opportunities and is insensitive to availability of internal financing.
  - Other large companies that are not part of conglomerates are driven by availability of funds (internal cash flow) rather than market-perceived future growth opportunities, suggesting they face more binding financing constraints.

### Conclusions and policy implications on corporate vulnerabilities
- Summary of vulnerabilities:
  - Polarization between the top four chaebols and the other conglomerates: indicators of profitability, solvency, liquidity, leverage and market valuation are all weaker for chaebols excluding the top four compared to the top four and other large firms.
  - Sizable and increasing share of firms with negative returns, elevated leverage, and low interest coverage among chaebols excluding the top four are sources of vulnerability and suggest a need for corporate restructuring.
- Policy implications:
  - Corporate restructuring, while challenging in the short term, can deliver medium-term benefits including increased investment, capital productivity and employment.
  - Econometric evidence suggests non-chaebol firms face stronger financing constraints; policy should ensure credit and capital market financing allocation is fair to all firms and based on consistent project evaluation criteria.

*Source: cr1841 - 6000.0 and over (IMF country report content).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2018/cr1841.pdf_