## 1jpnea2020002 ## Source details **Canonical URL:** [1jpnea2020002](https://www.imf.org/-/media/files/publications/cr/2020/english/1jpnea2020002.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2020/english/1jpnea2020002.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2020/english/1jpnea2020002.pdf.json) --- ### Disappearing Cities — Key findings and regional housing-market patterns - Japan’s total population peaked in 2010 and is projected to continue to decline. - Old age dependency ratio exceeded 40 percent in 2014 and is expected to accelerate, reaching above 70 percent in the next 50 years. - Regional divergence: - Large cities (notably the Greater Tokyo area) experience net migration inflows; many other regions experience net migration outflows. - Greater Tokyo area prefectures (Tokyo, Saitama, Chiba, Kanagawa): population grew by about 10 percent in the past two decades. - Hokkaido lost about 6 percent of its population over the same period. - Prefecture-level population concentration: Tokyo’s population increased by 1.3 percentage points; Hokkaido decreased by 0.3 percentage points. - Housing-market channel and vacancies: - Dwelling-related liabilities (purchase of house and/or land) consist of about 75 to 90 percent of total household liabilities (Japan’s National Survey of Family Income and Expenditure, 2014). - Nearly 13 percent of Japan’s total dwellings are vacant. - In the next 15 years, the number of vacant houses will increase to 21.7 million houses, or about one-third of total dwellings in Japan. - Residential land price dynamics: - Since the beginning of Abenomics (noted as 2013/2014 context): prices increased by 2.7 percent since 2014 and by 0.7 percent in 2018; nationwide average increase of 1.9 percent since the beginning of Abenomics in 2013. - Prefecture-level extremes since 2014: Miyagi and Fukushima up by 18 percent; Tokyo up by 16 percent; Akita -6.5 percent; Shimane -5.5 percent; Yamanashi -4.8 percent. - Since 2002 (2002–2018): Tokyo is the only prefecture with an increase in house prices. - Household trends: - Number of households increased by 76 percent from 1970 to 2015; increase was 116 percent for large cities and 18 percent for rural areas. - Total number of households in 2018 stands at around 50 million. - Average family size has declined with increases in nuclear and one-person families. ### Mechanisms, modeling insights, and empirical strategy - Durable-housing model (Glaeser and Gyourko (2005)) predictions: - Nonlinear (asymmetric) relationship: house price decline associated with population loss is larger than price rise associated with population gain of the same size. - Rationale: housing durability and asymmetric supply elasticities (easier to add than to demolish), producing a kinked supply curve. - Sales by departing owners can increase supply and induce further price declines. - Empirical specification (panel regressions): - ∆P_{i,t} = α + β * POPLOSS_{i,t-1} + γ * POPGAIN_{i,t-1} + ε_{it} - P = publicly-assessed residential land price adjusted for inflation (Yen/meter^2). - POPLOSS_{i,t-1} = 0 if population grew; equals actual percentage decline if population fell. - POPGAIN_{i,t-1} = 0 if population fell; equals actual population growth rate if population grew. - Controls: lagged “old dependency ratio” and “vacancy rate” (both in percentage change with one lag). - Expected signs: β and γ positive, with β > γ. - Empirical correlations (1996–2018) support predicted non-linearity; Tokyo’s housing-price growth lies below the relationship predicted by other prefectures. ### Panel regression findings (selected numeric results) - Dependent variable: %Δ real land price. - Period labels: 1970-2015 | 1970-2000 | 2000-2015. - Selected coefficients and statistics (as presented): - HHLOSS 5.96112.4415.77*---24.01***23.85***23.92*** (standard errors shown in parentheses in source) - HHGAIN 3.091***1.818**3.011***7.004***3.7857.700**0.3570.4430.567 - Δ ln OLDDEP -1.516***-1.721***-3.047**-5.279*-0.177-0.165 - Δ ln Vacancy rate -0.305*-0.4250.0400 - Constant -0.145**0.165*0.727***-0.427***0.2440.938**-0.216***-0.188***-0.200*** - Observations 376 376 329 188 188 141 141 141 141 - R-squared 0.703 0.720 0.747 0.647 0.663 0.720 0.340 0.342 0.343 - Number of code 47 47 47 47 47 47 47 47 47 - Fixed effects Y Y Y Y Y Y Y Y Y - Time effects Y Y Y Y Y Y Y Y Y - Mechanism amplification: expectations of future price declines can induce selling, adding supply and creating a “vicious cycle” that produces larger-than-model-predicted price falls. ### Policy-relevant observations and recommendations (housing and regional policy) - Risks and concerns: - Demographic-driven oversupply in rural areas implies risks to household wealth and bank balance sheets, as land and real estate serve as strong collateral. - Larger magnitude of price declines with population loss implies potential for significant wealth erosion in declining regions. - Expectations of lower future house prices could trigger further outflows and disposals, amplifying declines. - Policy suggestions: - Close regional disparities by addressing drivers of uneven population growth. - Align housing supply and real estate investment decisions with demographic trends to avoid over-investment. - Mitigate risks from high vacancy rates and promote regional revitalization to counteract concentration in a few large cities. - Note: historical tax incentives that reduced land tax when structures remained contributed to high vacancy rates; new legislation enacted in 2014 to accelerate demolition of vacant houses. --- ### Baseline macroeconomic impacts of demographic decline and automation scenarios - Baseline (assuming total factor productivity growth unaffected by population decline): - The level of GDP will fall by 8 percent with respect to the levels observed between 2000 and 2010. - Fiscal spending will increase largely due to age-related spending. - VAT rates (or consumption tax rates) will have to increase permanently by 10 percentage points to maintain the debt-to-GDP ratio at the level observed in 2019 (that is, at about 230 percent). - Distributional effects: - Market income inequality rises: Gini coefficient rises by five points (market income). - Mechanisms: falling share of labor-income-determined households, rising share of asset/wealth-determined (older) households; wealth distribution is more unequal than wage distribution. - Automation (“Aging + Robots”) scenarios: - Automation might raise per-capita GDP by 4 percentage points above the average level seen during the period 2000–11. - Automation reduces but does not eliminate fiscal pressures: - Necessary permanent increase in VAT rates is 10 percentage points in the Aging scenario and 5 percentage points in the Aging + Robots scenario. - Distributional impacts: - Automation can reduce market income inequality relative to aging-only scenario. - Gini index using disposable income falls by 2 percentage points in the Aging + Robots scenario (compared to baseline). - Share of consumption for the bottom 20 percent and middle segment increases in the aging scenario, and increases are larger with automation. - Limits and sensitivity: - If future technological growth is only one half that observed between 2000–12 (also half the assumed rate), automation would not suffice to prevent the fall of per-capita GDP caused by demographic transition. - Policy implication: - Encourage spending on research and development; improve education to familiarize workers with new technologies and enhance labor productivity. - Strengthen social safety nets to protect workers (noting differential exposure of female and elderly/low-skill part-time workers). ### Japan’s fertility trends and stylized facts - Population projection: projected to decline by more than 25 percent in the next 40 years if current low fertility persists. - Total fertility rate: 1.4 children per woman in 2018. - Replacement rate: 2.1 children per woman. - Government “desired” fertility rate: 1.8. - NIPSSR survey: - Ideal number of children for a couple: 2.32 on average. - Couples plan to have on average 2.01 children (sum of number of children already born, 1.68, and number of additional children planned, 0.33). - Marriage and unmarried statistics: - Marriage rate halved from around 10 in the early 1970s to around 5 in 2015. - Unmarried rate of females aged 30–34 has risen to about 35 percent; 90 percent of unmarried females aged 18–34 intend to marry in the future. - Cost and opportunity factors: - 56 percent of couples cite cost of childbearing as top reason for not having ideal number of children; 40 percent cite reluctance to bear a child at advanced age. - Female labor force participation (FLFP) rose from 63 percent in 2012 to 71 percent in 2018 (ages 15–64). - Longitudinal survey: female workers share drops from 62 percent to 35 percent after giving birth; recovers to 61 percent about 5.5 years after giving birth. Most re-enter as part-time workers: full-time share drops from 38 percent to 25 percent after childbirth and remains ~26 percent even 7.5 years after birth. ### Empirical analysis of fertility drivers — PMG estimator results and policy implications - Data and model: - Panel dataset covers 47 prefectures for 2001–2015. - Dependent variable: prefecture-level total fertility rate. - Five explanatory variables: education costs (share of education expenses), FLFP (women 15–64), wage gap (female/male monthly wages), childcare facilities (capacity per female population 20–44), unemployment rate. - Estimator: Pooled Mean Group (PMG) — long-run coefficients common across prefectures; short-run coefficients allowed to differ. - Main PMG long-run coefficients (statistically significant where indicated): - Wage gap: long-run coefficient 1.039*** (standard error 0.198). - FLFP: long-run coefficient 4.003*** (standard error 0.583). - Education costs: long-run coefficient -1.170** (standard error 0.507). - Childcare facilities: long-run coefficient 3.153*** (standard error 0.811). - Unemployment rate: long-run coefficient 4.442*** (standard error 0.655). - Short-run dynamics and delta coefficients (selected): - Lag of total fertility rate coefficient 0.438*** (standard error 0.0340). - Error correction term -0.385*** (standard error 0.0353). - Change in FLFP -10.29*** (standard error 2.774). - Change in childcare facilities 1.015* (standard error 0.564). - Observations: 658. - Sargan test (p-value) for GMM-SYS: 0.000. - Interpretation: - Long-run: smaller male-female wage gap and higher FLFP associated with higher fertility; increased childcare facilities and lower education costs support higher fertility. - Short-run: some variables (wage gap, FLFP) show negative short-run impacts, reflecting opportunity-cost effects and labor-market frictions around childbirth. - Policy recommendations: - Continue Work Style Reform to improve long-term fertility prospects. - Policies to reduce involuntary exclusion of female workers after childbirth; enable regular workers to retain regular status. - Measures: (i) further increasing childcare availability; (ii) rewarding firms with high retention rates of female employees after childbirth; (iii) eliminating disincentives to regular and full-time work embedded in tax and social security systems. - Reduce direct costs of childbearing (education and childcare) and sustain coordinated, persistent policy packages given small individual policy impacts and negative short-run effects. --- ### BoJ monetary-policy framework, unconventional measures, communication, and recommendations - Evolution of framework and UMPs: - 2006 framework: Board members disclosed “understanding” of price stability (0 to 2 percent); “two-perspective” approach (near-term outlook and longer-term risks). - Post-GFC: ZLB response (2008), Comprehensive Monetary Easing (October 2010), QQE in 2013 (with 2 percent target and annual JGB purchases ~¥50 trillion, later increased to ¥80 trillion in October 2014). - 2016: Negative Interest Rate Policy (NIRP) introduced; resulted in yield-curve flattening and financial-sector side-effects. - 2016: Yield Curve Control (YCC) implemented — targeted short-term (NIRP) and long-term (10-year JGB yield); BoJ bought JGBs along the curve to support reflation while reducing JGB purchases. - Outcomes under YCC: - Since 2016: economic growth averaged above potential; core inflation stabilized at slightly below one percent; yield curve steepened relative to pre-YCC; BoJ’s JGB purchases fell markedly; little progress in permanently lifting inflation expectations; low bank profitability and search-for-yield risks remain. - Communication assessment and shortcomings: - Commitment in 2013 was an improvement but “likely too extreme,” reducing credibility by overselling the horizon and policy space. - Communication under QQE/NIRP/YCC was at times ambiguous and complicated (redundant quantity and rate guidance). - Decision-making showed tendency to place larger weight on current conditions than on forward-looking paths. - Inflation Forecast Targeting (IFT) rationale and proposed adjustments: - Adopt an IFT framework to provide structured policy making and clearer communication. - Strengthen decision process: staff constructs inflation and growth forecasts conditional on given policy rate-paths; Board votes on policy path that best fulfills mandate. - Publish the policy path and associated economic forecast in the Economic Outlook Report along with alternative scenarios and motivations. - Simplify communication: abandon quantity guidance on JGB purchases; de-link inflation overshooting commitment from the monetary base; de-emphasize redundant forward guidance if policy path is published. - Mitigation for transparency concerns: preserve dissenting views via Summary of Opinions. - Other recommendations: - Announce comprehensive review of the price stability objective to re-confirm or re-evaluate the inflation level consistent with price stability; consider introducing a range around the inflation target and emphasize medium- to long-term attainment. - Better communicate trade-offs between price stability and financial stability. --- ### Inbound tourism: trends, determinants, vulnerabilities, and policy priorities - Recent performance and size: - After turning positive in 2015, travel services surplus reached almost one half of one percent of GDP in 2018. - GDP share of tourism industry in 2016: 1.9 percent; OECD average: 4.8 percent. - Tourism employment share in Japan: 9.6 percent. - Number of inbound tourists tripled since 2012; domestic tourism industry contributes about two percent of nominal GDP (¥11 trillion in absolute terms). - Price and spending comparisons: - Visitors from America, Europe and Oceania spend $200-$600 more per capita in Thailand than in Japan. - Entertainment spending: 15 percent of tourists’ budget in Thailand vs. 3 percent in Japan. - Regional and source-market concentration: - Foreign tourists concentrated in Kanto (Tokyo), Kinki (Osaka/Kyoto), Hokkaido, and Okinawa. - Kanto and Kinki regions receive more than 70 percent of total foreign tourist expenditure. - Origin-country concentration: almost half of foreign visitors in Kyushu are South Korean; in Kanto and Kinki, 30 percent are Chinese. - Determinants (panel ARDL long-run effects — selected coefficients, statistical significance preserved): - Origin country GDP long-run coefficients: - China: 2.214*** - Emerging Asia: 3.936*** - Advanced Asia: 2.648*** - Non-Asia: 5.492*** - Bilateral real exchange rate long-run coefficients: - China: 1.385* - Emerging Asia: 0.679*** - Advanced Asia: 2.606*** - Non-Asia: 1.131*** - Multiple visa dummy long-run effects: - China: 1.210*** - Emerging Asia: 0.547*** - Visa requirement dummy (China): -1.340*** - Visa relaxation impacts: - Multiple-entry visas increased tourists from China by 121 percent and from other emerging Asian countries by 55 percent (long-run effects). - Visa-free travel for some emerging Asian countries boosted numbers by 134 percent (long-run effect). - Short-run vulnerability to natural disasters: - Disaster dummy simultaneous impact: - Emerging Asia: -0.143***. - Non-Asia: -0.107***. - One period after disaster: - Emerging Asia: -0.431*** (tourists could decrease by 43 percent in the following period). - Non-Asia: -0.272***. - Policy actions and budgets: - Since 2012: Tourism Strategy Promotion Council, updated 2016 strategy. - January 2019: international tourist departure tax introduced; budget set to almost double to ¥66.6 billion in FY 2019. - Over 40 countries granted relaxation of visa requirements during 2013–18. - Policy priorities to 2030: - Continued regional diversification: disseminate regional information; low-cost flight connections to regional airports; regional sightseeing corridors; coordination between JNTO and local DMOs. - Diversify source markets: target broader set of Asian emerging markets; further relax visa requirements and flight connections. - Shift to experience-oriented tourism and “Japan branding” to increase per-capita spending and reduce price sensitivity. - Disaster resilience: implement disaster information policy to provide efficient and accurate information and assist tourists in rescheduling. - Supply-side synchronization: increase labor and infrastructure supply (consider more foreign labor, higher female and elderly participation); build infrastructure (free Wi-fi, multilingual signage, cashless payments). --- ### Current account composition, income-balance response to REER, and external adjustment - Mechanical and economic channels of income-balance response to REER: - Mechanical: foreign assets often denominated in foreign currency; REER appreciation mechanically decreases foreign assets and income credits (as percent of GDP). - Share of foreign assets denominated in foreign currency in 2017: - around 70 percent in the United States; - 85 percent in Japan; - nearly 100 percent in the median emerging economy (EME). - Share of foreign liabilities denominated in domestic currency in 2017: - 85 percent for the United States; - 82 percent for the median G6 economy; - 67 percent for Japan. - Economic: REER appreciation may reduce domestic activity and profits, affecting income debits especially if domestic firms are partially foreign-owned. - Theoretical implications by NFA position: - Both income credits and debits expected to decrease following a REER appreciation. - In large net creditor countries (e.g., Japan), a REER appreciation likely decreases the income balance if income credits dominate, reinforcing the usual negative trade-balance response. - Empirical findings (panel of 40 countries, 1986–2018): - Income credit and income debit elasticities generally negative. - Income balance semi-elasticities significantly smaller than trade-balance semi-elasticities. - For Japan and other large net creditors, income-balance response marginally reinforces trade-balance response. - Policy recommendation to address high external income balance: - Promote inward FDI to boost income debits and reduce external income balance; steps include addressing corporate governance issues and regulatory/administrative barriers to foreign ownership. - Overall message: - External adjustment via the exchange rate mainly operates through the trade balance, with marginal support from the income balance; the compositional shift toward a larger income balance does not fundamentally modify external adjustment dynamics for Japan. --- ### GVAR analysis — international spillovers and Japan’s role - Model and data: - Global Vector Autoregression (GVAR) with 33 country/region-specific models; quarterly data primarily 1981Q2–2018Q2 (extended estimation sample 2013Q1–2018Q2 for certain analyses). - Variables include real GDP, inflation, primary balance to GDP ratio, public debt to GDP ratio, real exchange rate, short and long-term interest rates, index of financial market stress, and oil price. - Spillover magnitudes (median effects after one year): - One percent decline in China’s GDP: - Asia and Pacific: 0.3 percent decline. - Japan: 0.25 percent decline. - Europe and Americas: about 0.2 percent decline. - One percent decline in U.S. GDP: - North America: 0.65 percent decline. - Japan: 0.2 percent decline. - Asia: 0.2 percent decline. - One percent decline in Euro Area GDP: - Europe: 0.65 percent decline (median). - Japan: 0.2 percent decline. - One percent fall in global GDP: - Asia-Pacific median country: 1.5 percent reduction. - Japan: 1.3 percent reduction in real median output. - One standard deviation positive shock to advanced-economy FSI: - World output falls by around 0.2 percent on average over the first year. - Japan: 0.2 percent decline (median). - Outward spillovers from Japan: - One percent decline in Japan’s GDP generates: - Other Asian countries: around 0.2 percent output loss after one year (median). - Non-Asian regions: about 0.1 percent after one year. - Structural change: using 1984–86 trade weights, the impact of a Japan shock was larger historically (other Asian median output -0.4 percent), indicating Japan’s outward spillovers have declined over recent decades. - Interpretation: - Global sensitivity to China’s developments is higher than to Japan’s; Japan’s outward spillovers remain regionally important but smaller than those of China and the United States. *Source: IMF staff calculations and chapter text as presented in content unit 1jpnea2020002.* ### References ____________________________________________________________________________ 15 ### References ### Disappearing Cities: Demographic Headwinds and Their Impact on Japan’s Housing Market — Key Findings - Japan’s total population peaked in 2010 and is projected to continue to decline. - The old age dependency ratio exceeded 40 percent in 2014 and is expected to accelerate, reaching above 70 percent in the next 50 years. - Regional divergence: large cities (notably the Greater Tokyo area) experience net migration inflows, while many other regions experience net migration outflows. - Housing-market channel: dwelling-related liabilities (purchase of house and/or land) consist of about 75 to 90 percent of total household liabilities (Japan’s National Survey of Family Income and Expenditure, 2014). - Vacancies and “Akiya”: nearly 13 percent of Japan’s total dwellings are vacant. In the next 15 years, the number of vacant houses will increase to 21.7 million houses, or about one-third of total dwellings in Japan. - Household trends: the number of households increased by 76 percent from 1970 to 2015; increase was 116 percent for large cities and 18 percent for rural areas. Total number of households in 2018 stands at around 50 million. Average family size has declined with increases in nuclear and one-person families. ### Regional Population and House Price Patterns (selected statistics) - Greater Tokyo area prefectures (Tokyo, Saitama, Chiba, Kanagawa): population grew by about 10 percent in the past two decades. - Hokkaido lost about 6 percent of its population over the same period. - Prefecture-level population concentration: Tokyo’s population increased by 1.3 percentage points; Hokkaido decreased by 0.3 percentage points. - Residential land price dynamics (public assessment value, Yen/m2): long swing since early 1990s with five periods identified (pre-bubble; post-bubble until 2001; mini-bubble 2002–0 8; pre-Abenomics; Abenomics). - Since the beginning of Abenomics (noted as 2013/2014 context): prices increased by 2.7 percent since 2014 and by 0.7 percent in 2018; nationwide average increase of 1.9 percent since the beginning of Abenomics in 2013. - Prefecture-level price extremes since 2014: Miyagi and Fukushima up by 18 percent (reconstruction-related), Tokyo up by 16 percent; large losses in Akita (-6.5 percent), Shimane (-5.5 percent), and Yamanashi (-4.8 percent). - Since 2002 (2002–2018): Tokyo is the only prefecture with an increase in house prices. ### Mechanisms and Modeling Insights - Durable-housing model (Glaeser and Gyourko (2005)) predicts a nonlinear (asymmetric) relationship between population change and housing price change: - The magnitude of house price decline associated with population loss is larger than the magnitude of house price rise associated with population gain of the same size. - Rationale: housing is durable; supply is more elastic when building new houses but inelastic when demolishing houses, producing a kinked supply curve and asymmetric price responses. - If supply increases due to sales by departing owners, further price declines can occur. - Empirical strategy: regressions following Glaeser and Gyourko (2005): - Specification: ∆P_{i,t} = α + β * POPLOSS_{i,t-1} + γ * POPGAIN_{i,t-1} + ε_{it} - P is publicly-assessed residential land price adjusted for inflation (Yen/meter^2) as proxy for housing prices. - POPLOSS_{i,t-1} = 0 if prefecture i’s population grew during period t; equals actual percentage decline if the prefecture lost population during the period. - POPGAIN_{i,t-1} = 0 if prefecture i experienced population loss during period t; equals actual population growth rate if the prefecture gained population. - Controls include lagged “old dependency ratio” and “vacancy rate” (both in percentage change with one lag). - Expected signs: β and γ positive, with β > γ, implying larger price declines for population loss than price increases for population gain of equal magnitude. - Simple correlations (1996–2018) between housing price change and net migration flows support the predicted non-linearity; Tokyo’s housing-price growth lies below the relationship predicted by other prefectures. ### Policy-relevant Observations and Recommendations (as presented) - Demographic-driven oversupply, especially in rural areas, implies risks to household wealth and bank balance sheets because land and real estate serve as strong collateral. - The larger magnitude of price declines associated with population loss suggests potential for significant wealth erosion in regions with declining populations. - Expectations of lower future house prices could trigger further population outflows and disposal of houses, amplifying price declines—model-based predictions likely understate potential falls. - Policy suggestions highlighted: - Close regional disparities by addressing drivers of uneven population growth. - Avoid potential over-investment by aligning housing supply decisions with demographic trends. - Consider policies that mitigate risks from high vacancy rates and promote regional revitalization to counteract concentration in a few large cities. ### Boxes, Figures, and Tables Referenced (structure of chapter) - BOX: 1. Recent Developments in Residential Property Prices in Japan - FIGURES: Population Change by Prefectures Between 1996-2015; Household Statistics; House Price Change by Prefecture; Population Growth and Housing Price Change; Changes in House Price and Net Migration Flows; Education and House Prices. - TABLE: 1. Empirical Analysis of Land Price and Population Changes *Prepared by Yuko Hashimoto (RES), Gee Hee Hong (APD), and Xiaoxiao Zhang (University of Cambridge).* ### 12. Panel regressions using prefecture-level housing prices and population dynamics ### 12. Panel regressions using prefecture-level housing prices and population dynamics ### Empirical results from panel regressions - Table 1 reports regression results based on panel data of the annual change in population and real land prices across three time periods: (i) 1970–2015, (ii) 1996–2015, and (iii) 2010–15. - Key qualitative findings: - During 1970–2015 (column 1), the estimated coefficient for population loss is higher than that of population gain, confirming the prediction by Glaeser and Gyourko (2005) that housing prices fall faster with declining population than they increase with a growing population. - In more recent periods, the correlation between housing price decline and population decline stays robust, while the relationship between housing price increase and population increase disappears—consistent with an overall decline in housing prices across regions since the early-1990s bubble burst, regardless of the extent of population growth. - Selected numeric entries from Table 1 (as presented): - HHLOSS 5.96112.4415.77*---24.01***23.85***23.92*** (4.446)(7.504)(8.005)(4.410)(4.235)(4.252) - HHGAIN 3.091***1.818**3.011***7.004***3.7857.700**0.3570.4430.567 (0.783)(0.819)(1.097)(1.559)(2.321)(3.203)(0.797)(0.851)(0.809) - Δ ln OLDDEP -1.516***-1.721***-3.047**-5.279*-0.177-0.165 (0.355)(0.391)(1.438)(2.641)(0.298)(0.317) - Δ ln Vacancy rate -0.305*-0.4250.0400 (0.153)(0.268)(0.196) - Constant -0.145**0.165*0.727***-0.427***0.2440.938**-0.216***-0.188***-0.200*** (0.0606)(0.0961)(0.0984)(0.116)(0.356)(0.460)(0.0375)(0.0495)(0.0714) - Observations 376 376 329 188 188 141 141 141 141 - R-squared 0.703 0.720 0.747 0.647 0.663 0.720 0.340 0.342 0.343 - Number of code 47 47 47 47 47 47 47 47 47 - Fixed effects Y Y Y Y Y Y Y Y Y - Time effects Y Y Y Y Y Y Y Y Y - Dependent variable: %Δ real land price - Period labels under table: 1970-2015 | 1970-2000 | 2000-2015 ### Mechanisms amplifying population-driven housing price declines - Empirical results suggest negative effects on housing prices from population decline may be larger than model-based predictions. - Mechanism described: - Households expecting future house price declines may sell and have less incentive to own, adding to oversupply and creating further downward pressure on prices. - This potential “vicious cycle” can lead to larger declines in house prices in regions with declining population than predicted by models that do not factor in expectations. ### Why are Japanese moving to large cities? - Concentration drivers identified: - Jobs and education for younger Japanese: large variation across prefectures in higher-education shares—rural prefectures such as Akita and Aomori about 5 percent with university degrees or higher, versus Tokyo or Kanagawa close to 20 percent. - Service amenities (health care, retail) concentrated in large cities; example: number of general hospitals (per 100 square kilometers of habitable area) — Japan average 6; Tokyo 42.4; Akita 1.7 (2016 data). - Skilled workers can generate growth in endogenous amenities, increasing housing values (citing Shapiro, 2006; Glaeser and Saiz, 2004). - Empirical associations: - Positive correlation between housing price changes and the share of the educated population (Figure 6, right chart). - Higher-education facilities concentrated in four major cities; number of colleges and universities by prefecture shown (Statistics Bureau of Japan, 2017). ### Policy implications - Addressing regional disparities is crucial to prevent excessive falls in housing prices in rural areas. - 2014 policy response: Abe administration established the headquarters for regional revitalization in the Cabinet Office to promote even growth and retain population and talents in rural areas. - Government measures: tax incentives for companies shifting core functions or expanding headquarters outside other parts of the country; tax incentives for households who moved outside the capital area. - Forward-looking housing supply and real estate investment policies needed to incorporate demographic trends to prevent further price declines: - Vacant houses create social issues and negative externalities—presence of vacant houses tends to bring down the value of other houses in the neighborhood. - A regional supply of housing that factors in future demographic trends would help avoid over-investment, decrease the number of vacant houses, and help place upward pressure on housing prices. - Historical tax policy note: - Tax rate on land used to be reduced to one sixth of the appraised value if there remained a residential structure on the land—initially to incentivize home construction when population grew—but this contributed to high vacancy rates by incentivizing owners not to demolish structures. New legislation enacted in 2014 to accelerate demolition of vacant houses. ### Box 1 — Recent developments in residential property prices in Japan - National and city-level price changes: - National average of residential land prices declined by 0.1 percent (annual survey by the Ministry of Land, Infrastructure, Transport and Tourism). - Prices rose in Tokyo, Osaka and Nagoya by 0.9 percent on the back of solid demand for condominiums and offices. - Condominium price dynamics: - Condominium prices have increased by 23 percent at the national level since 2013. - Prices for new condominiums in the Greater Tokyo area in 2018 averaged ¥71.4 million (about $650,000). - Potential drivers of recent appreciation: - Ultra-low interest rate environment incentivizing increased real estate lending. - Increase in tourism leading to investment in tourist destinations. - Inheritance tax considerations: anecdotal evidence that condominiums are evaluated below market value for tax assessment, making them attractive bequests compared to financial assets; strong tax incentive to hold real estate and take out housing loans (loan interest tax deductible at market value if to carry out a bequest). - Related empirical note: - IMF (2017) concluded condominium prices appear moderately overvalued in Tokyo, Osaka, and several outer regions, exceeding fundamentals by 5 to 10 percent. - Price indices presented: - Condominium Prices (2010=100, NSA) — series shown for Tokyo including suburbs, Osaka including suburbs, Hokkaido/Tohoku, Kyushu-Okinawa, Japan (nationwide). - Residential Property Price Index (2010=100, NSA) — series for Residential Land, Detached House, Condominiums. *Source: Authors' estimations.* ### 8. In the baseline scenario, the long-term ### 8. In the baseline scenario, the long-term ### Baseline macroeconomic impacts of demographic decline - Long-term macroeconomic costs of Japan’s declining and shrinking population are substantial. - Assuming total factor productivity growth is not affected by a decline in population growth: - The level of GDP will fall by 8 percent with respect to the levels observed between 2000 and 2010. - Less workers and less demand due to an aging and shrinking population will lead to a decline in total output. - Fiscal spending will increase largely due to age-related spending. - VAT rates (or consumption tax rates) will have to increase permanently by 10 percentage points to maintain the debt-to-GDP ratio at the level observed in 2019 (that is, at about 230 percent). ### Distributional effects of demographic transition - Market income inequality rises dramatically under aging: - Gini coefficient rises by five points (market income). - Mechanisms driving higher market-income inequality: - Labor income becomes slightly more equally distributed by wage level. - The share of those whose total income is mainly determined by labor income (young households) falls, while the share of those whose total income is mainly determined by asset and wealth (older households) increases. - Distribution of wealth is much more unequal than distribution of wages; period-by-period differences in wages compound through time, amplifying inequality. ### Automation (“Aging + Robots”) scenarios and impacts - If technological change continues at its current speed, automation could offset some demographic effects: - Automation might raise per-capita GDP by 4 percentage points above the average level seen during the period 2000–11. - Automation reduces but does not eliminate fiscal pressures: - Necessary permanent increase in VAT rates is 10 percentage points in the Aging scenario and 5 percentage points in the Aging + Robots scenario. - Distributional impacts of automation: - Automation can reduce market income inequality relative to the aging-only scenario. - Under automation: - The share of market income for the top 20 percent increases, but less than in the aging-only scenario. - The share of market income for the middle segment declines, but the decline is smaller than in the aging-only scenario. - Through facilitating financing of the social security system and requiring a smaller VAT increase, automation can reduce disposable income (and consumption) inequality: - Gini index using disposable income falls by 2 percentage points in the Aging + Robots scenario (compared to baseline). - The share of consumption for the bottom 20 percent and middle-segment of the population increases in the aging scenario, and increases are larger with automation. ### Limits, uncertainty, and sensitivity to the pace of technological change - The extent to which automation mitigates demographic challenges depends on future speed and breadth of technological progress. - Empirical and historical evidence suggests technological progress may not accelerate and could slow: - IMF (2017) finds the growth rate of cyclically-adjusted total factor productivity for advanced economies between 2008–14 is about half the rate of growth observed between 2003–07, and approximately one fifth the rate of growth observed between 1990 and 2002. - If the future growth rate of technological change is only one half that observed between 2000–12 (also half the assumed rate underlying the analysis), automation would not suffice to prevent the fall of per-capita GDP caused by the demographic transition. - Policy implication: encourage spending on research and development, and improve education levels to familiarize workers with new technologies and enhance labor productivity. ### Distributional policy considerations - Policies to address distributional consequences of automation are important: - Automation may differentially affect female workers and elderly/low-skill part-time workers: - Hamaguchi and Kondo (2018) estimate female workers in Japan are exposed to higher risk of being replaced by ‘computerization’ than male workers. - Elderly workers often enter the workforce as part-time workers in low-skill sectors and are more likely to be replaced by automation. - Strong and effective social safety nets will be crucial to protect displaced or vulnerable workers during the demographic transition. ### Japan’s fertility rate: projections and stylized facts - Japan’s population is projected to decline by more than 25 percent in the next 40 years, if the current low fertility rate persists. - Japan’s total fertility rate: - 1.4 children per woman in 2018. - Population replacement level is 2.1 children per woman. - Government “desired” fertility rate is 1.8 (the rate expected if people had their desired number of children). - NIPSSR survey: ideal number of children for a couple is 2.32 on average. - NIPSSR survey: couples plan to have on average 2.01 children (sum of number of children already born, 1.68, and number of additional children planned, 0.33). - Demographic and social indicators: - Marriage rate halved from around 10 in the early 1970s to around 5 in 2015. - Unmarried rate of females aged 30–34 has risen to about 35 percent; 90 percent of unmarried females aged 18–34 intend to marry in the future. - Cost and opportunity factors limiting fertility: - Top reason cited by 56 percent of couples for not having their ideal number of children is the cost of childbearing; 40 percent cite reluctance to bear a child at an advanced age. - Female labor force participation rates (FLFP) rose from 63 percent in 2012 to 71 percent in 2018 (for those aged 15 to 64). - Opportunity cost effects: - Longitudinal survey of Japanese newborns in 2010: female workers share drops from 62 percent to 35 percent after giving birth; total share recovers to 61 percent about 5.5 years after giving birth. - Most re-entering female workers do so as part-time workers: share of full-time workers drops from 38 percent to 25 percent after childbirth and remains around 26 percent even 7.5 years after giving birth. - Enduring implications for lifetime income and employment benefits when full-time female workers give up full-time status after childbirth. ### Empirical analysis: data, methodology, and explanatory variables - Panel dataset: - Covers 47 prefectures for the period 2001 to 2015. - Dependent variable: prefecture-level total fertility rate. - Five explanatory variables included: - Education costs: share of education expenses in total expenditures for non-single households (proxy for direct cost of child-rearing). Expected sign: negative. - Female labor force participation rate (women 15–64). Expected sign: positive (based on prior literature). - Wage gap: ratio of female to male monthly wages (average monthly contractual cash earnings). Expected sign could be positive in Japan’s context because a smaller wage gap yields better income prospects after childbirth and may increase affordability of childbearing; however, interpretations vary. - Availability of childcare facilities: capacity (sum of authorized quotas of children) divided by female population 20–44 years old. Expected sign: positive. - Unemployment rate: prefecture-level unemployment (model-based estimates). Expected sign: negative. - Purpose: study prefecture-level drivers of Japan’s fertility rate and assess whether public policy can help raise fertility by removing obstacles. *Source: IMF Staff calculations and IMF text.* ### 9. The Pooled-Mean Group estimator is the preferred model. The simple pooled OLS ### 9. The Pooled-Mean Group estimator is the preferred model. ### Methodology and estimation approach - The simple pooled OLS regression model does not allow for prefecture-specific effects. - The fixed effects estimator accounts for prefecture-specific effects but fails to deal with endogeneity of explanatory variables with respect to the total fertility rate. - The PMG (Pooled Mean Group) estimator proposed by Pesaran et al. (1999) is used here, following D'Addio and Mira d'Ercole (2005). - PMG distinguishes long-run and short-run dynamics: long-run coefficients are assumed identical across prefectures; short-run coefficients are allowed to differ. - A GMM (Generalized Method of Moments)-System estimator (Arellano and Bover (1995) and Blundell and Bond (1998)) was considered but is less preferred because the post-estimation Sargan test rejected the null hypothesis that over-identifying restrictions are valid, indicating potential misspecification. - Prefectural data covers the period 2001-15 for 47 prefectures. - Note on data construction: As the National Census is a quinquennial survey, values for gap years are filled by linear interpolation. ### Main long-run and short-run results (PMG estimates) - Every explanatory variable has a statistically-significant impact on the prefectural fertility rate in the long-run (PMG long-run coefficients): - Wage gap: long-run coefficient 1.039*** (standard error 0.198) - Interpretation: A rise in the female wage relative to the male wage (a smaller gap between male and female wages) has a positive impact on fertility rates in the long-run; short-run sign is negative (see short-run coefficient below). - FLFP (Female labor force participation rate): long-run coefficient 4.003*** (standard error 0.583) - Interpretation: The long-run result indicates that a one-percentage point increase in the FLFP rate is associated with a 0.04 increase in the fertility rate (see Table 1). - Education costs: long-run coefficient -1.170** (standard error 0.507) - Interpretation: A reduction in education costs has a positive impact on the fertility rate in the long run (note: not confirmed in GMM estimates). - Childcare facilities: long-run coefficient 3.153*** (standard error 0.811) - Interpretation: An increase in childcare facilities has a positive impact on fertility in both short run and long run. - Unemployment rate: long-run coefficient 4.442*** (standard error 0.655) - Interpretation: Long-run positive correlation could imply a low opportunity cost of childbearing when unemployment rates are high; coefficient is statistically insignificant in GMM estimates and short-run sign is negative. - Short-run PMG coefficients (where reported) and dynamics: - Wage gap short-run coefficient 0.402*** (standard error 0.0815) — the PMG table displays positive short-run coefficient but text notes the sign is negative in the short-run; the table lists 0.402*** under column "(short-run coefficients)". - FLFP short-run coefficient 0.787*** (standard error 0.214) — text notes short-run sign is negative; table lists 0.787*** under "(short-run coefficients)". - Education costs short-run coefficient -0.0518 (standard error 0.226) — not statistically significant. - Childcare facilities short-run coefficient 1.151*** (standard error 0.211). - Unemployment rate short-run coefficient -0.116 (standard error 0.266) — not statistically significant. - Lag of total fertility rate coefficient 0.438*** (standard error 0.0340). - Error correction term -0.385*** (standard error 0.0353). - Short-run changes (PMG delta coefficients): - Change in wage gap -0.355*** (standard error 0.0835). - Change in FLFP -10.29*** (standard error 2.774). - Change in education costs 0.135 (standard error 0.172). - Change in childcare facilities 1.015* (standard error 0.564). - Change in unemployment rate -0.888* (standard error 0.465). - Constant terms: PMG long-run constant -0.873*** (standard error 0.0814); PMG short-run constant -0.116 (standard error 0.0999). ### Table 1 key statistics and diagnostics - Observations: 658 - Sargan test (p-value) for GMM-SYS: 0.000 - Note: Standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1 ### Interpretation of economy and policy-relevant mechanisms - Wage gap: - Long-run: a smaller male-female wage gap is associated with higher fertility, consistent with lower opportunity costs of childbirth over the long-run. - Short-run: a smaller wage gap could lower fertility in the short-run due to larger foregone income during maternity leave (text describes negative short-run effect). - Female labor force participation: - Long-run: higher FLFP associated with higher fertility; magnitude described as a one-percentage point increase in FLFP associated with a 0.04 increase in fertility rate. - Short-run: negative sign discussed in text, possibly reflecting opportunity cost effects. - Education and childcare costs: - Reduction in education costs has positive long-run impact (not confirmed in GMM). - Increases in childcare facilities have positive impacts in both short-run and long-run, indicating effectiveness of child-friendly policies. - Unemployment: - Long-run positive coefficient could reflect lower opportunity cost of childbearing when unemployment is high; short-run negative coefficient noted; GMM estimates show insignificance. ### Policy implications and recommendations - Japan’s Work Style Reform, intensified since 2016, could have a positive impact on fertility over the long-term. - Policies to reduce the unwilling exclusion of female workers after childbirth could raise fertility over the long-term; focus on enabling female regular workers to retain regular-worker status after childbirth. - Potential measures for authorities: - (i) further increasing childcare availability; - (ii) rewarding firms with high retention rates of female employees after childbirth; - (iii) eliminating disincentives to regular and full-time work embedded in the tax and social security systems. - Measures to alleviate direct costs of childbearing could reinforce fertility gains, such as lowering education and childcare costs. - Public policies supporting fertility should be implemented in a coordinated and sustained manner: - Each policy impact is relatively small; a wide array of mutually-reinforcing policies is needed to make a meaningful impact. - Negative short-run effects of FLFP and the wage gap on total fertility rate mean persistence is required—sustaining public policies even if short-run fertility falls. *Source: Author's calculations and analysis as presented in content unit 1jpnea2020002.* ### 5. The BoJ also introduced a new monetary policy framework to improve policy ### 5. The BoJ also introduced a new monetary policy framework to improve policy predictability while preserving flexibility. ### New framework (2006) - The framework consisted of two components: - BoJ Board members disclosed their “understanding” of price stability (ranging from 0 to 2 percent). - Policy decisions were guided by a “two perspective” approach: - First perspective: assess whether the near-term outlook (1-2 years) followed a path of sustainable growth under price stability. - Second perspective: examine various risks to the outlook over the longer term, including financial stability risks. ### BoJ response to crises and unconventional monetary policies (UMP) - After the Global Financial Crisis (GFC): - Summer of 2008: economic activity depressed and inflation dropped; overnight rate lowered to the ZLB and measures taken to strengthen financial institutions and market functioning. - October 2010: BoJ reverted to quantitative easing by introducing Comprehensive Monetary Easing (CE) framework, re-introducing ZIRP and an asset purchase program including JGBs and risky assets to reduce term and risk premia. - CE response was more protracted and smaller in size compared to the U.S. Federal Reserve and the ECB. - 2013: QQE (Quantitative and Qualitative Easing) introduced shortly after raising the price stability target to two percent: - BoJ committed to increase its annual purchase JGBs holdings by about ¥50 trillion per year. - Initial results: inflation rose, exchange rate depreciated, growth picked up; deterioration occurred in second half of 2014 due to fall in oil prices and weak demand after the April 2014 consumption tax rate hike. - October 2014 response: BoJ raised the annual increase of JGB holdings from ¥50 trillion to ¥80 trillion. - 2016: Negative Interest Rate Policy (NIRP) introduced to lower interest on excess reserves into negative territory: - Intention: put downward pressure on short-term interest rates and raise inflation expectations. - Result: larger-than-anticipated impact on yields, significant flattening of the yield curve and compression of term spreads with associated financial sector side-effects (increased risk taking and further decline in profitability of financial institutions). - 2016: Yield Curve Control (YCC) implemented as new operational framework to support a protracted reflation strategy: - Targeted both short-term interest rate (NIRP) and long-term interest rate (10-year JGB yield). - BoJ bought JGBs along the entire yield curve to prevent the long end from falling while keeping the short end unchanged. - Intended benefits: sustain monetary accommodation while reducing JGB purchases and addressing concerns about running out of JGBs. ### Outcomes under YCC and ongoing risks - Since 2016: - Economic growth has averaged above potential. - Core inflation stabilized at slightly below one percent. - Yield curve steepened compared to levels just before YCC implementation. - BoJ’s purchases of JGBs have fallen markedly. - Little progress in permanently lifting inflation expectations. - Low bank profitability and search for yield by financial institutions remain significant medium-term financial stability risks. ### Communication strategies and evolution - Pre-Abenomics: - Communication focused on exploring the “duration effect”; example: under QEP monetary easing would “continue until the CPI (excluding perishables) registers stably a zero percent or an increase year on year.” - The 2006 “two-perspective approach” anchored policy guidance in a medium-term numerical “understanding” of price stability. - With the two percent inflation target (2013): - Communication aimed at quickly re-anchoring inflation expectations. - BoJ promised to achieve the price stability target “at the earliest possible time, with a time horizon of about two years,” deviating from previous medium- to long-term emphasis. - BoJ committed to massive increase in JGB purchases to back verbal commitment with strong policy actions. - 2016 shift under YCC: - Time horizon for achieving the target was de-emphasized; BoJ began highlighting financial side-effects. - BoJ committed to expand the monetary base until the inflation target was achieved (the “overshooting commitment”). - Policy guidance became more complicated: operational shift to interest rate targeting while retaining quantitative guidance on JGB purchases, creating a growing discrepancy between quantity guidance and actual JGB purchases. - 2018–2019: - In summer 2018, in response to upward pressure on the 10-year JGB yield, BoJ increased variability range around the zero percent yield target and introduced forward guidance for policy rates. - Initial forward guidance was time-dependent (to cover the scheduled October 2019 consumption tax increase). - Fall 2019: shifted to state-based guidance — keep short- and long-term interest rates low “as long as it is necessary to pay close attention to the possibility that the momentum toward achieving the price stability target will be lost.” ### Lessons learned and way forward — key findings - Lack of stable and clear policy objectives complicated implementation and hampered reflation: - Early large weight on financial stability plus absence of a clear price stability target may have contributed to insufficient stimulus and premature normalization (2000 and 2006). - Large relative weight on price stability and over-optimism during QQE/NIRP/YCC likely damaged credibility and kept inflation expectations persistently low. - Emphasis on achieving price stability “as soon as possible” and unrealistic inflation forecasts were problematic given limited policy space, clogged monetary transmission, and rising financial stability costs. - Decision-making shortcomings: - Tendency to place larger weight on current rather than future conditions, despite monetary policy working with a lag. - Policy deliberations often focused on current policy settings rather than the entire future policy path. - Even after the 2006 “two-perspective approach” and clarifications, it is unclear how policy decisions are systematically guided by the BoJ Board’s economic forecasts and the two percent inflation target. ### Policy recommendations (specific measures) - Strengthen and clarify commitment to the inflation target while increasing policy flexibility to address financial stability concerns: - Announce a comprehensive review of the price stability objective (similar to reviews in 2000, 2006, and 2013) to: - Re-confirm or re-evaluate the inflation level viewed as consistent with price stability; clarify that the price stability target will be achieved over the medium- to long-term. - Introduce a range around the inflation target to enable a more gradual reflation process consistent with realistic inflation projections and provide flexibility to address financial side-effects. - Better communicate trade-offs between financial stability and price stability objectives and clarify that other objectives, including financial stability, matter for monetary policy. - Communication of target re-evaluation and greater flexibility must be carefully managed: - A comprehensive review announcement would allow clear rationale and motivation for any changes. - BoJ could replicate the YCC communication strategy that successfully reduced JGB purchases without triggering policy normalization concerns. - Emphasize that a target range and longer time horizon are consistent with practices of several major central banks. *Source: https://www.imf.org/-/media/files/publications/cr/2020/english/1jpnea2020002.pdf* ### 19. The communication strategy under Abenomics has at times been overly ambitious, ### 19. The communication strategy under Abenomics has at times been overly ambitious, ### Assessment of BoJ communication under Abenomics - The commitment to achieve the inflation target in 2013 was an improvement over the pre-Abenomics period but "likely too extreme," which reduced policy credibility. - By communicating an unrealistic target horizon and overselling available policy space, "the public gradually come to discount the BoJ’s ability to achieve the inflation target." - Communication during QQE/NIRP/YCC did not adequately ensure policy predictability and "probably added to policy uncertainty and ambiguity." - The surprise implementation of the NIRP illustrates poor predictability. - The BoJ kept redundant guidance when changing framework (i.e., keeping both quantitative and interest rate guidance under YCC), adding to ambiguity. - There is a need for the BoJ to "better explain how its policy strategy can achieve the price stability target based on current and forecast economic conditions." - Anchoring long-run inflation expectations at the "2 percent" target depends on whether the public believes the BoJ’s strategy and actions are consistent with achieving the target. ### Inflation Forecast Targeting (IFT): rationale and proposed adjustments - IFT provides "an intuitive and structured approach to policy making" enabling clearer communication and expectation management; originally proposed by Svensson (1997). - Proposed adjustments for the BoJ under IFT: - Strengthen the decision-making process: - In preparation for a monetary policy meeting (MPM), BoJ staff could construct inflation and growth forecasts based on given policy rate-paths and the latest economic information. - These forecasts and associated policy paths would be presented to the Board at the MPM, and Board members would vote on which policy path best fulfills the BoJ’s mandate. - This contrasts with current practice: the Board’s decision would account for the entire future path of policy decisions, not just changes to current policy tools. - Publish the policy path and associated economic forecast: - Following the policy decision, the policy path and corresponding economic forecast should be published in the BoJ’s Economic Outlook Report. - A detailed discussion to motivate the policy path and the forecast is crucial to make them credible. - The outlook report should discuss alternative scenarios to clarify implications of selected shocks and to reinforce that the baseline forecast of the policy path is conditional on economic developments. - Simplify the communication strategy: - Abandon the quantity guidance on JGB purchases. - De-link the inflation overshooting commitment from the monetary base. - If the BoJ publishes the policy path associated with the forecast, then the forward guidance currently applied to policy rates would be redundant. - Operational notes: - Svensson (2019) discusses an exogenous policy rate path, while Clinton et al (2015) advocate an endogenously determined policy rate path via a reaction function. - The forecast is a key input but the process does not exclude individual Board members from incorporating their own judgment. ### Adoption experience and implications - The BoJ currently publishes economic projections by individual Board members that "take into account the effects of past policy decisions and financial market expectations regarding future policy." These forecasts do not incorporate the current policy decision nor the Board’s view on future policy settings. - Adopting IFT would imply discontinuing the publication of individual Board member forecasts in favor of publishing a staff forecast consistent with the Board’s view on current and future policy decisions. - Concern: removing individual Board projections could be seen as reducing transparency (loss of observed diversity in views). - Mitigation: allow dissenting views to be reflected in the Summary of Opinions, currently published shortly after an MPM. - The guiding principles behind IFT are similar to the BoJ’s existing guidelines for conduct of monetary policy; publishing the baseline staff forecast with alternative risk scenarios could quantify the BoJ’s "two-perspective approach." - IFT has been adopted by a number of central banks in both emerging and advanced countries; "most central banks publish both the baseline forecast as well as the assumed underlying policy path." ### Cross-country publication practices (high-level summary of Table 2) - Practices noted among selected central banks (excerpted): - Reserve Bank of Australia (RBA): Staff forecast published quarterly; No policy path assumption published; policy path is in line with financial markets; uncertainty bands and risks discussed. - Bank of Canada (BoC): Staff forecast published quarterly; No policy path assumption published; policy path endogenously determined within the model; risks discussed in detail, sometimes quantified. - Bank of England (BoE): Staff forecast published quarterly; Policy path assumption published; two assumptions used (in line with markets or constant); uncertainty bands and risks discussed. - Bank of Japan (BoJ): Forecasts of individual Board members published; No policy path published; policy path is in line with financial markets; general assessment of risks to the outlook. - European Central Bank (ECB), RBNZ, Riksbanken: staff forecasts published with policy path assumptions and published uncertainty/risk assessments in varying detail. - U.S. Federal Reserve (U.S. Fed): No monetary policy/inflation report; projections by Board members (including individual policy path projections) published four times per year. ### Conclusions and policy recommendations (from chapter) - Three main lessons: - Lack of stability and clarity of the BoJ’s price stability objective has complicated policy implementation and hampered reflation efforts. - Policy decisions have not been sufficiently forward-looking and could have been better tied to policy goals. - The BoJ’s communication strategy has at times been overly ambitious, ambiguous, and complicated. - To address weaknesses the BoJ could consider: - Increasing policy flexibility by introducing an inflation range around the target while emphasizing the medium- to long-term nature of achieving the price stability objective. - Adopting Inflation Forecast Targeting to improve policy credibility and predictability. - Simplifying its policy guidance by abandoning the quantity guidance on JGB purchases and de-linking the inflation overshooting commitment from the monetary base. *IMF chapter text provided in the source content.* ### 8. Inbound tourism also contributes to Japan’s current account surplus. After turning ### 1jpnea2020002 - 8. Inbound tourism also contributes to Japan’s current account surplus. After turning ### Inbound tourism and the current account - After turning positive for the first time in 2015, Japan’s travel services surplus reached almost one half of one percent of GDP in 2018. - The travel services surplus reflects: - A steep upward trend of travel credits since 2011. - The erosion of travel debits, dominated by a fall in the overseas spending of Japanese tourists rather than a decline in their annual headcount, which has fluctuated around 17 million persons. - Structural change: Compared to 10 years ago, the travel balance transformed from the largest deficit factor into the second-largest surplus factor of the services account, after income from intellectual property (MLIT, 2019). ### Tourism’s size, productivity potential, and international comparisons - GDP share of Japan’s tourism industry in 2016: 1.9 percent. - OECD average tourism GDP share in 2016: 4.8 percent. - Tourism employment share in Japan: 9.6 percent (noted as among the highest in the OECD). - The divergence between GDP share (1.9 percent) and employment share (9.6 percent) implies likely potential for labor productivity increases to alleviate industry growth constraints from labor shortages. ### Potential to boost inbound tourist spending - International comparison with Thailand: - Visitors from America, Europe and Oceania tend to spend $200-$600 more in per capita expenditure in Thailand than in Japan. - Composition differences: entertainment spending is 15 percent of tourists’ budget in Thailand vs. 3 percent in Japan. - Policy implication: Promoting entertainment services should continue to be a priority, along with attracting tourists from high-income countries to boost per capita tourist spending. - Reference: entertainment spending has a high marginal effect on tourist expenditure (MLIT, 2019). ### Regional concentration of the tourism boom - Foreign tourist concentration: highly concentrated in and around Tokyo (Kanto region), Osaka/Kyoto area (Kinki region), Hokkaido, and Okinawa. - Origin-country concentration: - Almost half of foreign visitors in Kyushu are South Korean. - In Kanto and Kinki, 30 percent are Chinese. - Spending and investment concentration: - Kanto and Kinki regions receive more than 70 percent of total foreign tourist expenditure. - Accommodation investment, though larger in all regions, remains concentrated in hotspots. ### Determinants of inbound tourism (empirical results) - Methodology: Panel ARDL regressions using quarterly data for Japan inbound tourists (Pesaran et al., 1999 approach). - Long-run effects (Table 1, Panel A — selected coefficients): - Origin country GDP: - China: 2.214*** - Emerging Asia: 3.936*** - Advanced Asia: 2.648*** - Non-Asia: 5.492*** - Bilateral real exchange rate: - China: 1.385* - Emerging Asia: 0.679*** - Advanced Asia: 2.606*** - Non-Asia: 1.131*** - Multiple visa dummy (long-run effect): - China: 1.210*** - Emerging Asia: 0.547*** - Visa requirement dummy (long-run effect) for China: -1.340*** - Short-run effects (Panel B — disasters): - Disaster dummy simultaneous impact: - Emerging Asia: -0.143*** (statistically significant) - Non-Asia: -0.107*** (statistically significant) - One period after disaster: - Emerging Asia: -0.431*** (tourists could decrease by 43 percent in the following period) - Non-Asia: -0.272*** - Two periods after disaster: effects vary by group; e.g., Advanced Asia shows 0.0479**. - Interpretation: - Visa relaxation policies for Asian countries increased tourist numbers: multiple-entry visas increased tourists from China by 121 percent and from other emerging Asian countries by 55 percent (long-run effects); visa-free travel for some emerging Asian countries boosted numbers by 134 percent. - Income levels in source countries and bilateral real exchange rates important for all origin countries; price elasticities highest for advanced Asia. - High price elasticities imply large exposure to exchange rate and price changes; safe-haven status of the yen and possibility of sudden yen appreciation could hinder inbound tourism. ### Short-run vulnerability to natural disasters - Natural disasters have large and prolonged negative impacts on tourism demand: - Emerging Asia tourists could decrease by 11–14 percent in the period a disaster occurs. - The impact becomes larger in the following period, with a 43 percent fall for emerging Asia. - For non-Asian tourists, negative impact lasts until the second lag, with tourist numbers 3.5 percent lower than before the disaster (lasting even after 9 months). ### Tourism promotion policies and spending - Policy actions since 2012 under Abenomics: - Establishment of a Tourism Strategy Promotion Council; strategy updated in 2016. - Policy measures aim to enhance regional attractiveness, industry competitiveness, and travel comfort. - Budget and institutional measures: - January 2019: an international tourist departure tax introduced as financing for tourism budget. - With the new income, the budget is set to almost double to ¥66.6 billion in FY 2019. - Over 40 countries were granted relaxation of visa requirements (multiple-entry visas or visa exemptions) during 2013–18. - Infrastructure and facilitation measures: - Increase in budgets of relevant agencies, more tax-free shops, more direct flights, and relaxation of visa requirements. - Major events attracted: 2019 Rugby World Cup, 2020 Olympic and Paralympic Games, EXPO 2025 — expected significant income multiplier effects. ### Conclusions and policy recommendations - Performance summary: - Number of inbound tourists tripled since 2012. - Domestic tourism industry contributes about two percent of nominal GDP (¥11 trillion in absolute terms). - Tourism impact remains concentrated in a few hotspots and tourists are predominantly from only three Asian source markets. - Priorities to achieve 2030 targets: - Continued regional diversification within Japan: - Broader and more accessible information dissemination about regional attractions. - More low-cost flight connections to regional airports. - Creation of regional sightseeing corridors. - Enhanced coordination between national and sub-national stakeholders (e.g., JNTO and local DMOs). - Diversification of tourism source markets: - Target broader set of Asian emerging markets through additional relaxation of visa requirements and flight connections. - More experience-oriented tourism and “Japan branding”: - Transition from shopping-oriented to experience-oriented tourism to differentiate Japan’s brand, decrease price sensitivity, incentivize longer stays, repeat visits, and higher spending. - Disaster resilience: - Implement disaster information policy to provide efficient and accurate information during and after natural disasters to avoid adverse tourist reactions and assist tourists in rescheduling visits. - Supply-side synchronization: - Increase supply of tourism services including labor and infrastructure to match demand growth. - Consider more foreign labor and increased female and elderly labor market participation as the labor force shrinks. - Continue building tourism infrastructure: free Wi-fi, multilingual signage, and cashless payment systems to address tourist discomforts identified in JTA survey (lack of public Wi-fi, poor non-Japanese language skills, lack of multilingual signage, lack of cutting-edge payment settlement methods). *Source: IMF staff synthesis of Japan tourism chapter (provided content).* ### Annex II. 2016 Tourism Strategy ### Annex II. 2016 Tourism Strategy ### Maximizing the attractiveness of tourism resources in order to make tourism the base of regional revitalization - Allow domestic and international visitors entry to “publicly owned heritage sites” - Shift the balance of heritage policy from “an over-emphasis on preservation only” to allow a greater understanding of the sites by tourists - Turn the current “national parks” into world-class “national parks” - Create “landscaping plans” for major tourism areas to improve townscapes - Complete “regional revitalization corridors” to allow comfortable travel to every corner of Japan - Functional enhancement of local airports and the promotion of Low-Cost Carrier (LCC) services - Renew and revitalize hot spring resorts and local towns through better management ### Foster innovation in the tourism industry to boost its international competitiveness and develop it into a core industry - Review regulations and restrictions in order to make the tourism industry more productive - Develop new longer-stay markets - Create and develop world-class Destination Management/Marketing Organizations (DMOs) - Promotion of MICE (Meetings, Incentives, Conferencing, Exhibitions) - Strategic relaxation of visa requirements - Promotion of information dissemination in multiple languages - Reforming the system of “work days” and “days off” towards realizing a more vibrant society - Comprehensive design promotion activities for the 2020 Tokyo Olympic and Paralympic Games ### Ensure all visitors can enjoy a satisfying, comfortable and stress-free sightseeing experience - Greatly improve hard and soft infrastructure to realize the most pleasant accommodation environment in the world *Source: Council for a Tourism Vision to Support the Future of Japan (2016).* ### 11. Such an offsetting pattern between the trade balance and the income balance is also ### 11. Such an offsetting pattern between the trade balance and the income balance is also ### D. Does the Change in Current Account Composition Towards Income Balance Affect its Responsiveness to the Real Exchange Rate? - Decomposition of the income balance response to REER movements: - Mechanical effect: - For most countries, foreign assets tend to be denominated in foreign currency, implying a REER appreciation mechanically decreases foreign assets and income credits (expressed as percentage of GDP). - In 2017 the share of foreign assets denominated in foreign currency was: - around 70 percent in the United States - 85 percent in Japan - nearly 100 percent in the median emerging economy (EME) - The share has been lower in the median G6 country since 1999 (at around 50 percent) following the creation of the euro. - Currency denomination of foreign liabilities is more heterogeneous: - 85 percent of the United States’ foreign liabilities were denominated in domestic currency in 2017 - 82 percent for the median G6 economy in 2017 - 67 percent for Japan in 2017 - EME borrowing in foreign currency (original sin) implies a mechanical decrease in foreign liabilities and income debits when the REER appreciates: - 80 percent of EME’s foreign liabilities in 1990 were denominated in foreign currency - that share declined to 40 percent in 2017 - Economic effect: - For small open economies and countries with low outward spillovers, a REER appreciation is unlikely to significantly affect income credits (except mechanically). - A REER appreciation may reduce domestic economic activity and profits (especially for exporters), possibly lowering income debits (expressed as percentage of GDP) to the extent domestic firms are at least partially foreign-owned. - Theoretical implications depending on net foreign asset (NFA) position (summary from Table 2): - Both income credits and debits are expected to decrease following a REER appreciation. - In large net creditor countries (e.g., Japan), a REER appreciation would likely lead to a decrease in the income balance if the income credit channel dominates (income credits > income debits), reinforcing the usual negative trade balance response. - In large net debtor countries, the income debit channel may dominate and a REER appreciation would likely lead to an increase in the income balance, partially counterbalancing the trade balance response. - Empirical approach and preliminary findings: - Data and sample: - Quarterly and annual IMF BoP data on a panel of 40 countries over 1986–2018. - Estimation: - Income credit and income debit elasticities to the real exchange rate estimated separately, controlling for the size of foreign assets (resp. liabilities) and related yields. - Country-specific income balance semi-elasticities derived using ratios of income credit and debit to GDP. - Results: - Both income credit and income debit elasticities are generally found to be negative (estimates vary by specification). - For all 40 studied countries, income balance semi-elasticities are significantly smaller than corresponding trade balance semi-elasticities (gross income flows are smaller than gross trade flows). - For Japan and other large net creditor countries, evidence indicates the income balance response to REER changes would marginally reinforce the trade balance response. - Overall, the current account response to REER in Japan is estimated to be marginally larger when including the income balance response in addition to the trade balance response. ### E. Conclusions - Key drivers of Japan’s increasing income balance: - The gradual increase primarily reflects net revenues from an increasingly positive NFA position. - Japan’s high income balance results from a highly asymmetric income account and has been increasing in line with: - a larger NFA position reflecting past current account surpluses - the increase in corporate saving - When compared with G6 peers, Japan’s relatively high income balance is due to: - higher yields on investment abroad, especially on FDI (from “better” geographical positioning and possibly profit shifting), which more than offset a somewhat lower stock of foreign assets; - much lower FDI liabilities (due to possible measurement issues, corporate governance, or regulatory and administrative issues) and lower portfolio debt liabilities (due to strong home bias and corporate deleveraging); - lower yields on portfolio debt liabilities (linked to extremely accommodative monetary policy and low credit risk in Japan). - Interconnectedness and offsetting patterns: - Offsetting patterns between trade and income balances highlight interconnectedness and the blurring impact of globalization and multinational firms on income attribution. - Gross trade flows remain significantly larger than gross income flows. - Japan’s current account balance is highly and positively correlated with its trade balance; correlation with its income balance is much weaker. - The income balance is negatively correlated with the trade balance, possibly reflecting: - (i) aging - (ii) an income effect - (iii) market pressures - (iv) offshoring - (v) profit shifting through transfer pricing - Exchange rate adjustment and policy implications: - The income balance response to REER movements in Japan is estimated to reinforce the trade balance response, although only marginally; income credits and debits tend to decrease with a REER appreciation, with a dominant income credit channel for large net creditor countries like Japan. - The income balance response is significantly smaller than the trade balance response due to relatively low gross income flows. - Policy recommendation to address Japan’s high external income balance: - Continued efforts to promote inward FDI would likely boost income debits and reduce the relatively high external income balance. - Steps to boost inward FDI could include: - addressing corporate governance issues (Japan’s corporate governance described as an “insider system” with limited power for shareholders, potentially discouraging foreign ownership) - addressing regulatory and administrative issues, including reducing the cost of doing business - These steps align with IMF advice to reform product markets by reducing barriers to entry in some industries and accelerating deregulation of agricultural and professional services sectors to foster growth and investment. - Overall message on external adjustment via the exchange rate: - Adjustment of Japan’s external current account via the exchange rate is estimated to mainly operate through the trade balance, with marginal support from the income balance. - The compositional change toward a larger income balance does not fundamentally modify the external adjustment process via the exchange rate; in Japan the income balance response may actually amplify the trade channel response. ### JAPANESE BUSINESS CYCLES, EXTERNAL SHOCKS, AND SPILLOVERS — GVAR analysis (selected findings) - Model and data: - A Global Vector Autoregression (GVAR) model integrates a quarterly macro-econometric model for Japan within a compact model of the world economy (including the global oil market). - Framework comprises 33 country/region-specific models. - The model includes real and financial variables: real GDP, inflation, primary balance to GDP ratio, public debt to GDP ratio, real exchange rate, short and long-term interest rates, an index of financial market stress, and the price of oil. - All data are quarterly, for the period 1981Q2 to 2018Q2. - Spillover magnitudes and transmission: - Following a one percent decline in China’s GDP: - economic activity falls by about 0.3 percent for the median Asian economy - about 0.25 percent for Japan - 0.2 percent for the median economy in Europe and the Americas - The median Asian economy’s fall in response to a one percent decline in United States GDP is about 0.2 percent after one year. - Adverse spillovers from the Euro Area slowdown are modest (assuming limited financial stress in the Euro Area). - In response to a financial stress shock, real GDP growth slows worldwide by 0.2 percentage points on average. - Outward spillovers from Japan output shocks have reduced over time but remain important for Asia-Pacific economies and the global economy. - Interpretation: - All regions are more sensitive to developments in China than to output shocks in the Euro Area, the United States, or Japan, reflecting evolving trade patterns and China’s emergence as a key global driver. - Stress in global financial markets can amplify spillovers from growth shocks in systemic economies. - Japan’s outward spillovers are smaller than those of China and the United States but remain regionally significant for Asia-Pacific partners. *Source: IMF chapter text (excerpts provided).* ### 5. The GVAR is a modelling framework of the world economy designed to explicitly ### 5. The GVAR is a modelling framework of the world economy designed to explicitly model economic and financial interdependencies across markets and countries at national and international levels ### Overview of the GVAR framework - Originally proposed by Pesaran et al. (2004) and further developed by Dees et al. (2007). - Links individual country-specific models into a coherent global modelling framework using time series, panel data, and factor analysis techniques. - Applied in: bank stress testing; analysis of China’s growing importance for the rest of the world economy; international macroeconomic transmission of weather shocks; impact of commodity price shocks; other real and financial shocks; and forecasting applications. - For an extensive survey of developments in GVAR modelling, see Chudik and Pesaran (2016). ### Model specification and dataset - The GVAR model includes 33 economies, which together cover more than 90 percent of world GDP. - Quarterly data for the period 1979Q1–2016Q4 are obtained from Mohaddes and Raissi (2018b) on oil prices, domestic macroeconomic variables (treated as endogenous), and country-specific foreign variables (taken to be weakly exogenous) for the 33 economies, and updated to 2018Q2. - The modelling framework includes the following regions/countries (as listed in the source): Asia and Pacific, North America, Europe, South America, Middle East and Africa (full country list preserved in source). ### Financial stress and oil-price treatment - The modelling framework includes an index of financial stress (FSI_t) in advanced economies. - The FSI for advanced countries is constructed by Cardarelli et al. (2009) as an average of: the beta of banking sector stocks; TED spread; the slope of the yield curve; corporate bond spreads; stock market returns; time-varying stock return volatility; and time-varying effective exchange rate volatility. - FSI is modelled as a common variable and included as a weakly exogenous variable in each of the 33 country/region-specific models, while allowing for feedback effects from macro variables to FSI. - Nominal oil prices in U.S. dollars are included in the country-specific models. - Oil prices are modelled as determined in global commodity markets (separately) and then introduced as a weakly exogenous variable in all countries, allowing both demand and supply conditions to influence the oil price directly. ### Spillover analysis: scope - The analysis examines inward spillovers to Japan from: - (i) macroeconomic shocks in other systemic economies (China, the Euro Area, and the United States), - (ii) a potential global growth slowdown, and - (iii) stress in global financial markets. - It also studies outward spillovers from Japan to other Asian countries and globally. - The estimation sample of the GVAR model is extended from 2013Q1 to 2018Q2. ### China slowdown: findings - China’s real GDP growth slowed from an average of about 10 percent over the period 1980–2013 to an average of 6½ percent between 2017 and 2019. - A one percent negative GDP shock in China translates into lower global economic growth. - Effects by region (median effects after one year): - Asia and Pacific: 0.3 percent decline. - Europe: 0.2 percent decline. - North America: 0.2 percent decline. - South America: 0.2 percent decline. - Middle East and Africa: 0.4 percent decline. - Japan: 0.25 percent decline. ### United States slowdown: findings - A one percent negative U.S. output shock median effects after one year: - North America: 0.65 percent decline (median output falling by about 0.65 percent). - Europe: 0.2 percent decline. - Asia: 0.2 percent decline. - Japan: 0.2 percent decline. - Channels: trade links, lower commodity prices, third-market effects, and increased financial integration amplified by the global role of the U.S. dollar. ### Euro Area slowdown: findings - A one percent negative GDP shock in the Euro Area leads to: - European countries: median output falls by 0.65 percent after one year. - South America: annual output elasticity of about 0.3. - Japan: 0.2 percent decline after one year. - Impacts beyond Europe vary by country depending on trade and commodity-price channels and assumed financial stress in the Euro Area. ### Global growth slowdown: findings - Context: trade policy uncertainty, geopolitical tensions, and idiosyncratic stress in key emerging market economies weighed on global growth in the second half of 2019. - Following a one percent fall in global GDP, median impacts after one year: - Asia-Pacific region: 1.5 percent reduction in GDP for the median country. - Middle East and Africa: even larger median impact (driven by demand-driven reduction in global commodity prices). - Japan: 1.3 percent reduction in real median output. ### Stress in global financial markets: findings - A one standard deviation positive shock to the FSI in advanced economies: - Is two-thirds of the shock that occurred during the European sovereign debt crisis, and one-tenth of the shock that occurred during the Global Financial Crisis. - Translates into world output falling by around 0.2 percent below the pre-shock level on average over the first year. - Regional median impacts after one year following a negative shock to the financial stress index: - South America: greatest impact among regions (largest growth spillover). - Asia (median country): about 0.1 percent decline. - Japan: 0.2 percent decline. ### Outward spillovers from Japan: findings and structural change - A one percent decline in GDP of Japan (using 2014-16 trade weights) generates: - Other Asian countries: around 0.2 percent output loss after one year (median). - Non-Asian regions: median effect of about 0.1 percent after one year. - Structural change over time: - Re-estimation using trade weights averaged over 1984 to 1986 shows the impact of a Japan output shock has declined significantly over the past thirty years. - Using 1984-86 trade weights, following a negative Japan GDP shock: - Other Asian countries: real median output declines by about 0.4 percent after one year. - Other regions: output losses of around 0.3 percent after one year. - Interpretation: the influence of negative Japan output shocks on the global economy has become smaller in recent decades. *Source: IMF chapter on GVAR modelling and spillover analysis (author’s estimates and cited references).* --- _Source: https://www.imf.org/-/media/files/publications/cr/2020/english/1jpnea2020002.pdf_