## ch2

## Source details

**Canonical URL:** [ch2](https://www.imf.org/-/media/files/publications/weo/2025/april/english/ch2.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/weo/2025/april/english/ch2.pdf.md)
- [Structured JSON version](/-/media/files/publications/weo/2025/april/english/ch2.pdf.json)

---

### Introduction — overview and objectives
- Unprecedented demographic changes this century: global population growth will slow from 1.1 percent per year before the COVID-19 pandemic to basically zero in 2080–2100.
- Average age of the world’s population projected to increase by 11 years between 2020 and the end of the century.
- The share of the older population (ages 65 and older) is increasing rapidly worldwide, driving the rise of the “silver economy.”
- Objectives:
  - Assess cohort-level healthy-aging trends and their impact on labor market outcomes.
  - Evaluate global economic implications of demographic shifts and healthy-aging trends.
  - Explore targeted policies to mitigate negative effects of population aging.
- Key questions:
  - How have demographic trends evolved globally? How fast and uneven is the pace of aging across countries?
  - Is there evidence that later-born cohorts are healthier than earlier-born cohorts at the same age? How do healthy-aging trends vary across countries and socioeconomic groups? Has healthy aging increased labor market attachment and productivity of older individuals?
  - What are likely implications for growth, interest rates, public finances, and external balances? To what extent can healthy aging offset challenges?
  - How can policies generate growth tailwinds to mitigate adverse economic impacts?
- Data and methods:
  - Microsurvey data from approximately 1 million individuals from 29 advanced and 12 emerging market economies over 2000–22 used to establish healthy-aging trends and associations with labor market outcomes.
  - A multicountry, overlapping-generations general equilibrium model covering 69 economies—representing about two-thirds of global output and the world’s population—used to assess economic implications through the end of the century.
  - Baseline projections under current policies and model experiments assessing targeted progrowth policies.

### Major findings — demographic trends and macroeconomic outlook
- Aging is widespread:
  - All advanced economies and the largest emerging markets will have crossed their demographic turning point by 2035.
  - By 2070, most low-income countries will have experienced similar shifts.
- Aging is accelerating beyond advanced economies and the window for low-income countries to reap demographic dividends is gradually closing.
- Healthy aging provides partial offset:
  - Ongoing increases in labor supply and improvements in older workers’ human capital because of healthy aging are expected to contribute about 0.4 percentage point annually to global GDP growth over 2025–50.
  - Despite this, average global annual output growth under current policies is projected to decline by 1.1 percentage points over 2025–50 compared with the 2016–18 average, with demographic trends accounting for almost three-fourths of this decline.
- Interest rates and public finances:
  - Increasing share of older individuals with higher accumulated savings in large economies projected to exert downward pressure on interest rates.
  - Most countries likely to face a worse interest-growth differential than in the recent past and many will need higher primary balances than in 2016–18 to keep debt ratios stable from 2030 onward.
- External positions: Uneven demographic trends are likely to exert widening pressure on external global positions through the end of the century.

### Healthy aging and labor market implications — empirical evidence
- Broad-based healthy-aging gains across physical, cognitive, and mental health indicators for individuals ages 50 and above.
- Cognitive-function improvements:
  - “The 70s are the new 50s”: data from a sample of 41 advanced and emerging market economies indicate that, on average, a person who was 70 in 2022 had the same cognitive ability as a 53-year-old in 2000.
  - Over a decade, this pace of improvement in cognitive abilities is associated with:
    - an increase of approximately 20 percentage points in the likelihood that individuals remain engaged in the labor market (working or actively seeking employment),
    - an increase of about six hours in average weekly hours worked,
    - a 30 percent rise in labor earnings, conditional on being employed.
- Frailty index improvements:
  - On average, the frailty of a 70-year-old person in 2022 corresponded to that of a person who was 56 in 2000.
- Cross-country heterogeneity:
  - Cognitive health is positively associated with GDP per capita but exhibits notable variation across countries (for example, Sweden, Denmark, Finland differences; United States lags behind Nordic countries on measured health indicators despite comparable GDP per capita and higher health care spending).
- Survey coverage and measures:
  - Microdata analysis uses surveys conducted over 2000–22 in 41 advanced and emerging market economies and includes measures such as grip strength, lung function, memory, orientation, verbal fluency, basic mathematics, incidence of 18 chronic diseases, and health behaviors.
- Determinants of functional capacity:
  - Lifestyle factors—levels of physical activity, body mass index, and smoking—are significant determinants of functional capacity of older individuals after controlling for age and socioeconomic characteristics.
- Causal estimates (instrumental-variables, rescaled to reflect health trends over 10 years):
  - Rises in labor earnings and labor productivity by about 30 percent.
  - Increase in likelihood of participating in the labor force by about 20 percentage points.
  - Higher numbers of average weekly hours worked by about six hours.
- Age gradient and occupation-level findings:
  - Impact of better health on labor force participation is significantly larger for individuals in their 50s than for those in their 60s and 70s, implying nonhealth constraints (skills obsolescence, pension incentives, age discrimination).
  - Older workers with college educations are relatively well positioned to benefit from AI because it complements their tasks and skills.

### Change in Health-Adjusted Life Expectancy, 2000–21 — measurement and disparities
- World average is population-weighted, based on 183 countries. “Frontier” = maximum life expectancy across countries.
- Cognitive health score: first principal component of cognitive indicators, standardized to mean zero, standard deviation one. Regression sample period is 2000–22.
- Regressions of cognitive health for individuals ages 50 and older include controls for age, gender, education, and household wealth; country fixed effects are used in cross-country regressions.
- Cross-country and within-country disparities:
  - Individuals from emerging market economies (EMs) have lower health scores than those from advanced economies (AEs) after controlling for socioeconomic characteristics.
  - Within countries, average cognitive health scores are significantly lower for individuals in rural locations; individuals with at most primary education; lower-wealth households.
  - EMs show faster improvements in healthy aging compared with AEs (some cross-country “catching up”), but pace of improvements across gender, location, education, and wealth groups has been similar, indicating persistent socioeconomic health disparities.

### General equilibrium analysis — model design and baseline assumptions
- Model is an extension of the global overlapping-generations framework in Auclert and others (2024).
- Country coverage:
  - 21 advanced economies,
  - 4 emerging market economies (including China and India, which together account for almost 50 percent of emerging market economies’ GDP),
  - a bloc economy comprising 44 low-income countries (LICs) expected to pass their demographic turning points after 2040 (denoted LIC bloc).
  - Altogether, the model accounts for about two-thirds of the world economy and population.
- Healthy aging in the model:
  - Country-specific age-productivity profiles vary over time to integrate the impact of healthy aging on effective labor supply (proxied by labor earnings capturing effects on labor productivity and number of hours worked) documented in the empirical section.
  - Baseline assumes a continued—though moderating—improvement in the functional capacity of workers ages 50 and older over the next three decades, reflecting persistence of 2000–22 improvements as current young cohorts age.
- Productivity forces:
  - Growth of total factor productivity (TFP) at the global frontier.
  - Convergence toward the TFP frontier.
  - Impact of demographics on TFP growth through innovation and entrepreneurship channels.
- Global capital market assumptions:
  - Integration of China, India, and the LIC bloc into global capital markets is imperfect, producing a wedge between domestic and global interest rates that is assumed to decline gradually as reforms and integration proceed.
- Fiscal policy calibration and retirement ages:
  - Initial values for effective retirement rates, labor taxes, retirement replacement rates, and other public spending are calibrated to match country-specific targets.
  - In the baseline, effective retirement ages are assumed to increase by one month per year over 60 years in all countries (except for India and the LIC bloc, where they are assumed unchanged).
  - Labor taxes, replacement rates, and other public spending adjust period by period so that trajectories of debt-to-GDP ratios are aligned with WEO projections until 2029 and remain stable from then onward.

### Baseline projections: growth, interest rates, and primary balances
- Global GDP growth:
  - Projected to be 1.1 percentage points lower than the 2016–18 average over 2025–50, and 2 percentage points lower when the average over 2025–2100 is considered.
- Country examples:
  - Advanced economies with relatively older populations (example: Japan) are projected to see their economies shrink under baseline fertility and migration assumptions.
  - Advanced economies with projected stable or growing working-age populations (examples: Canada and the United States) will continue to grow, albeit more slowly over time.
  - China: projected deceleration of 2.7 percentage points in GDP growth over 2025–50 relative to the 2016–18 average.
  - India: projected deceleration of about 0.7 percentage point in GDP growth in 2025–50, intensifying over 2050–2100 as the demographic turning point is passed.
  - Low-income countries (LIC bloc): expected to see a sharper deceleration in growth in the second half of the century once demographic dividends turn into headwinds.
- Output per capita: world average is projected to be about 0.6 percentage point lower in 2025–50 and 1.8 percentage points lower toward the end of the century relative to 2016–18.
- Interest-growth differential (r − g):
  - Projected to be higher than the 2016–18 average for all economies except India and the LIC bloc over the next 25 years.
  - Average r − g for the world is projected to be 1 percentage point higher in 2025–50 than in 2016–18, moderating to about 0.5 percentage point toward the end of the century.
- Fiscal pressure implication:
  - About half of the model economies are projected to need higher primary-balance-to-GDP ratios than they had on average over 2016–18 to keep debt-to-GDP ratios stable from 2030 onward; this group includes China, Japan, and the United States.

### Net foreign assets and capital flow implications
- Large emerging market economies (China and India) would accumulate foreign assets, especially over 2050–2100.
- Many advanced economies would gradually draw down foreign assets throughout the projection horizon.
- The net foreign asset position for the LIC bloc would worsen through most of the projection period thanks to continued capital inflows, but this trend would slow and eventually reverse around 2070 as aggregate wealth increases with population aging.

### Contribution of healthy aging and demographics to growth
- Healthy aging contribution (world):
  - Projected to add about 0.4 percentage point to GDP growth, on average, over 2025–50.
  - If gains from healthy aging were abstracted from, global output growth would slow by 1.5 percentage points instead of 1.1 percentage points in 2025–50 relative to 2016–18.
  - The average contribution to world growth would be about 0.1 percentage point over 2050–75 and would decline further thereafter.
- Economy-specific healthy-aging contributions:
  - Positive and sizable, ranging from about 0.3 percentage point to 0.6 percentage point over 2025–50 for individual economies in the model; notably large for India and relatively lower for Japan.
- Demographics alone account for about half of the projected slowdown in GDP growth over 2025–2100 relative to 2016–18:
  - Demographic forces explain 1.1 percentage points out of a reduction of 2 percentage points in global GDP growth over 2025–2100.
  - Country examples: average contribution of demographic forces to GDP growth in 2025–2100 ranges from close to −2.8 percentage points in India to −0.4 percentage point in Finland and Slovenia.
- Sensitivity to fertility assumptions:
  - Under UNWPP high- and low-fertility scenarios, fertility is projected to remain 0.5 children above (high) or below (low) the medium-fertility scenario over most of the projection period.
  - Country-specific growth estimates under alternative fertility assumptions can vary, for instance, by 0.5 percentage point in Australia and 1.6 percentage points in China.

### Policy scenarios: labor supply measures and quantified impacts
- Three key policy levers examined:
  1. Healthy-aging policies (narrowing cross-country differences in functional capacity of workers ages 50 and older).
  2. Increasing effective retirement age (postponing effective retirement given healthy-aging improvements).
  3. Closing gender labor force participation gaps.
- Healthy-aging policy scenario:
  - Assumes narrowing of cross-country differences in functional capacity of workers ages 50 and older by one-fourth, equivalent to about 49 percent of the estimated gains over 2000–22.
  - For the world, average annual GDP growth over 2025–2100 would be about 0.2 percentage point higher than in the baseline, and 0.3 percentage point higher over 2025–50.
- Increasing effective retirement age scenario:
  - Assumes effective retirement ages increase faster than baseline in countries where life expectancy at retirement is 20 years or more; otherwise evolve as in baseline.
  - For the world, average annual GDP growth over 2025–2100 would be about 0.1 percentage point higher than in the baseline.
- Closing gender participation gaps scenario:
  - Assumes narrowing country-specific gender gaps in labor force participation by three-fourths by 2040.
  - For the world, average annual GDP growth over 2025–2100 would be 0.1 percentage point per year higher than in the baseline, and 0.3 percentage point higher over 2025–50.
- Combined policy package:
  - Implementing all three labor supply measures together would raise global average annual growth by 0.3 percentage point over 2025–2100 relative to baseline, reversing about one-third of the drop in growth attributable to demographic trends through the end of the century.
  - Over 2025–50, the combined package would boost growth by about 0.6 percentage point, offsetting close to three-fourths of the drag from demographics during that period.
  - Some countries (India, low-income countries, and some European economies) could reap even higher growth dividends.
  - Non-monetary benefits: keeping older workers engaged can offer societal benefits from improved well-being.

### Fiscal implications of labor supply policies and heterogeneous fiscal gains
- Direct fiscal channels:
  - Increased female labor force participation and employment would boost labor tax revenues.
  - Raising effective retirement ages would increase labor taxes and reduce transfer payments.
- Indirect fiscal channel via r − g:
  - Higher GDP growth from labor policies would help reduce r − g, easing fiscal pressures and lowering the primary balance needed to stabilize debt ratios—though labor supply policies could also put upward pressure on interest rates due to changes in aggregate savings and investment demand.
- Model simulation results:
  - Five model economies (Greece, India, Italy, Spain, and the LIC bloc) would see fiscal respite from lower r − g over 2025–50, reinforcing direct benefits on primary balances.
  - Under the combined policy scenario, all model economies would gain fiscal space, but gains are heterogeneous:
    - Equivalent fiscal gains, on average, would be more than 4 percentage points of GDP in Greece and Italy.
    - Less than 1 percent of GDP in China and the United Kingdom.
- Caveats:
  - Implementing some policies can entail direct budgetary costs (for example, active labor market policies), which could reduce net fiscal dividends.
  - Some healthy-aging policies (for example, taxes on alcohol and tobacco; preventive health policies) can generate revenue or future savings.
  - The model abstracts from quantifying direct implementation costs due to large uncertainty.

### Conclusions and policy implications — summary directives
- Demographic and healthy-aging summary:
  - Declining birth rates and increasing life expectancy are causing a sustained decline in population growth and significant changes in age structure.
  - Individuals ages 50 and older are, on average, aging in better health than previously; increased longevity has been accompanied by improvements in physical and cognitive capacities across subsequent cohorts.
  - Healthier aging has been associated with higher labor force participation rates, a higher likelihood of being employed, and higher labor earnings for individuals ages 50 and older.
  - Spending on health promotion and prevention accounts for only 1–6 percent of total health expenditure in member countries of the Organisation for Economic Co-operation and Development.
- Growth and fiscal outlook:
  - Ongoing gains from healthy aging are estimated to boost annual global growth by about 0.4 percentage point over 2025–50.
  - Under current policies, global output growth would decline on average by about 2 percentage points through the end of the century.
  - With lower growth prospects and historically high levels of public debt, many countries will need significant fiscal efforts to keep debt-to-GDP ratios stable beyond 2030.
  - Simulations suggest a combination of policies for boosting labor supply could attenuate the slowdown in global growth over 2025–50 resulting from demographic headwinds by almost three-fourths.
  - Progrowth policies could contribute to higher global interest rates but would provide substantial fiscal dividends and create additional fiscal space for many countries to finance critical spending; some economies would still require additional fiscal effort.
- Policy recommendations to extend healthy, productive working lives:
  - Health promotion and prevention targeting tobacco smoking, harmful alcohol use, physical inactivity, and unhealthy diets; measures include immunization, regular health checks, screenings, taxation (for example, on tobacco and unhealthy food), regulations (for example, to promote smoke-free environments), and access to mental health resources.
  - Pension reforms and retirement policy: change statutory retirement ages, reduce early retirement benefits, introduce incentives to postpone retirement, allow phased retirement; avoid purely age-based provisions.
  - Lifelong learning and workplace adjustments: lifelong upskilling and reskilling, flexible work arrangements, workplace adaptations, combat biases and discrimination against older individuals.
  - Policies to raise female labor force participation and support fertility: improved parental leave systems, expanded affordable childcare, flexible work arrangements.
  - Global integration and financial access: enhance access to international financial markets—credit and capital market reforms—and strengthen governance and institutions to enable low-income countries to reap demographic dividends.
- Technological progress and innovation:
  - Structural reforms to promote market competition, financial accessibility, and labor market flexibility can boost productivity growth by fostering innovation and a more efficient allocation of capital and labor.
  - AI-related technologies are complementary to labor in occupations more typical of older workers and can help older workers cope with functional decline.
  - Research and development in the scientific understanding of biological aging could further extend healthy longevity.
  - AI-based solutions in health care can scale up preventive practices (for example, automating screening and diagnostics and bringing expertise to underserved areas).

### Quantified simulations and scenario results (selected figures and calibrations)
- Healthy aging contribution: boost to annual global growth of about 0.4 percentage point over 2025–50.
- Long-run demographic drag: global output growth decline on average about 2 percentage points through the end of the century.
- Combined policy scenario: average fiscal gains reported over 2025–2100, with country-level variation.
- Pension reform instrument sizes needed to reverse the aging-induced increase in the public-debt-to-GDP ratio over 75 years (calibrated outcomes):
  - Single-Instrument Reform (Immediate / Delayed):
    - Retirement Age (Years): +6 / +8
    - Replacement Rate (%): −25 / −35
    - Contribution Rate (%): +18 / +34
  - Mix (Immediate / Delayed):
    - Retirement Age (Years): +2 / +2.7
    - Replacement Rate (%): −8.3 / −11.7
    - Contribution Rate (%): +6 / +11.3

### Low-income countries: financial integration and migration scenarios (selected quantitative findings)
- Status quo scenario: net foreign liabilities peak at about 13 percent of GDP.
- Enhanced financial integration scenario:
  - Net foreign liabilities reach about 180 percent of GDP by 2070–80.
  - Stock of capital and output significantly higher than in the status quo; both GDP and GDP per capita are about 19 percentage points higher than in the status quo scenario in the long term.
  - Long-run increase in gross national income per capita is smaller than that for GDP but still sizable at about 7 percentage points.
- Enhanced financial integration plus migration scenario:
  - Aggregate GDP in the LIC bloc is about 5 percentage points lower relative to the enhanced financial integration scenario but still 14.5 percentage points higher than in the status quo scenario.
  - GDP per capita in the LIC bloc is about 1.2 percentage points lower than in the enhanced financial integration scenario but almost 18 percent higher than in the status quo scenario.
- Implication: Financial sector reforms and strengthened governance and institutions are important to enable low-income countries to capture demographic dividends and offset potential output losses from migration outflows.

### Box 2.2 — Intergenerational considerations in pension reforms
- Main findings on reform design and timing:
  - Under a combined reform scenario, the increase in retirement age could be less, at two years.
  - Consumption losses from reforms that rely on a single instrument are significantly larger than those when a mix of instruments is used, at least for one of the generations.
  - The size of required fiscal measures is more profound and aggregate consumption losses are larger when the reforms are postponed for 10 years compared with those in a scenario in which they are implemented immediately.
    - Containing the rise in public debt induced by aging requires a 6-year increase in the retirement age if the reform is carried out immediately, but an 8-year increase is needed if the reform is postponed by 10 years.
    - The consumption losses from postponing reforms usually fall disproportionately on the young compared with the old.
- Evidence from the consumption-impact exercise:
  - Figure context: "Figure 2.2.1. Average Change in Consumption, 2025–65 (Deviation from 2025, percentage points)"
    - Solid bars denote average consumption losses or gains over a period of 40 years from a reform implemented immediately.
    - Markers denote consumption losses or gains if instead the reform is delayed by 10 years.
    - The red bars show the impact of aging only, and the rest of the bars show the impact of aging and the respective reforms.
  - Reform scenarios considered in the figure: No reform; Lower replacement rate; Higher retirement age; Higher contribution rate; Mix of reforms.
  - Key qualitative takeaway: Combining the three measures helps ensure the burden is shared across the young and old, potentially contributing to the acceptability and feasibility of reforms.
- Policy recommendations:
  - Policymakers should act sooner rather than later.
  - Use a combination of tools to ensure a fairer distribution of the burden across generations and by doing so enhance the feasibility and acceptability of pension reforms.
  - Although the exercise is calibrated for a typical advanced economy with a population that has already aged significantly, the lessons are even more pertinent for emerging market economies and low-income developing countries, which currently have lower old-age dependency ratios but will experience a faster pace of population aging and thus have less time to react.

*Source: ch2 - Introduction; 2. Change in Health-Adjusted Life Expectancy, 2000–21; Baseline projections; Conclusions and Policy Implications; Box 2.2 (PDF chapter).*

### Introduction

### ch2 - Introduction

### Overview
- Unprecedented demographic changes this century: global population growth will slow from 1.1 percent per year before the COVID-19 pandemic to basically zero in 2080–2100.
- Average age of the world’s population projected to increase by 11 years between 2020 and the end of the century.
- The share of the older population (ages 65 and older) is increasing rapidly worldwide, driving the rise of the “silver economy.”

### Chapter objectives and central questions
- Objectives:
  - Assess cohort-level healthy-aging trends and their impact on labor market outcomes.
  - Evaluate global economic implications of demographic shifts and healthy-aging trends.
  - Explore targeted policies to mitigate negative effects of population aging.
- Key questions:
  - Global demographic transition: How have demographic trends evolved globally? How fast and uneven is the pace of aging across countries?
  - Healthy aging: Is there evidence that later-born cohorts are healthier than earlier-born cohorts at the same age? How do healthy-aging trends vary across countries and socioeconomic groups? Has healthy aging increased labor market attachment and productivity of older individuals?
  - Economic implications: What are likely implications for growth, interest rates, public finances, and external balances? To what extent can healthy aging offset challenges?
  - Role of policies: How can policies generate growth tailwinds to mitigate adverse economic impacts?

### Data and methods
- Microsurvey data from approximately 1 million individuals from 29 advanced and 12 emerging market economies over 2000–22 used to establish healthy-aging trends and associations with labor market outcomes.
- A multicountry, overlapping-generations general equilibrium model covering 69 economies—representing about two-thirds of global output and the world’s population—used to assess economic implications through the end of the century.
- Baseline projections under current policies and model experiments assessing targeted progrowth policies.

### Main findings — demographic trends and macroeconomic outlook
- Aging is widespread:
  - All advanced economies and the largest emerging markets will have crossed their demographic turning point by 2035.
  - By 2070, most low-income countries will have experienced similar shifts.
- Aging is accelerating beyond advanced economies and the window for low-income countries to reap demographic dividends is gradually closing.
- Healthy aging provides partial offset:
  - Ongoing increases in labor supply and improvements in older workers’ human capital because of healthy aging are expected to contribute about 0.4 percentage point annually to global GDP growth over 2025–50.
  - Despite this, average global annual output growth under current policies is projected to decline by 1.1 percentage points over 2025–50 compared with the 2016–18 average, with demographic trends accounting for almost three-fourths of this decline.
- Interest rates and public finances:
  - Increasing share of older individuals with higher accumulated savings in large economies projected to exert downward pressure on interest rates.
  - Most countries likely to face a worse interest-growth differential than in the recent past and many will need higher primary balances than in 2016–18 to keep debt ratios stable from 2030 onward.
- External positions: Uneven demographic trends are likely to exert widening pressure on external global positions through the end of the century.

### Healthy aging and labor market implications
- Evidence of broad-based healthy-aging gains across physical, cognitive, and mental health indicators for individuals ages 50 and above.
- Cognitive-function improvements particularly large:
  - “The 70s are the new 50s”: data from a sample of 41 advanced and emerging market economies indicate that, on average, a person who was 70 in 2022 had the same cognitive ability as a 53-year-old in 2000.
  - Over a decade, this pace of improvement in cognitive abilities is associated with:
    - an increase of approximately 20 percentage points in the likelihood that individuals remain engaged in the labor market (working or actively seeking employment),
    - an increase of about six hours in average weekly hours worked,
    - a 30 percent rise in labor earnings, conditional on being employed.
- Frailty index improvements:
  - On average, the frailty of a 70-year-old person in 2022 corresponded to that of a person who was 56 in 2000.
- Cross-country heterogeneity:
  - Cognitive health is positively associated with GDP per capita but exhibits notable variation across countries (for example, Sweden, Denmark, Finland differences; United States lags behind Nordic countries on measured health indicators despite comparable GDP per capita and higher health care spending).
- Survey coverage:
  - Microdata analysis uses surveys conducted over 2000–22 in 41 advanced and emerging market economies and includes measures such as grip strength, lung function, memory, orientation, verbal fluency, basic mathematics, incidence of 18 chronic diseases, and health behaviors.

### Policy implications and recommended approaches
- Multifaceted policy mix to increase labor supply, boost growth, and ease fiscal pressures amid population aging:
  - Lifelong policies to support human capital of workers in late adulthood (people between age 50 and retirement age), including health promotion and prevention measures, can significantly counter the effect of population aging on growth.
  - Gradually increase the effective retirement age in line with improvements in life expectancy to raise labor force participation among the 65-and-older age group.
  - Close gender gaps in labor force participation where they remain large.
  - Expand access to international financial markets through credit and capital market reforms and strengthen governance and institutions to allow younger, low-income countries to capture demographic dividends before the window closes and to offset losses from migration outflows to older, labor-scarce economies.
- Quantified policy impacts:
  - A combination of labor supply policies could boost global annual output growth by about 0.6 percentage point over the next 25 years, offsetting almost three-fourths of the drag from demographics during that period.
  - Fiscal dividends from progrowth policies would enable many countries to rebuild buffers and create space for critical spending needs.

### Limitations and scope exclusions
- Analysis abstracts from:
  - Shifts in consumer demand and sectoral reallocations driven by aging.
  - Implications for the financial sector, house prices, and urbanization.
  - Endogenous technological responses to aging—such as automation and artificial intelligence (AI)—which could mitigate some negative growth effects.

*Source: ch2 - Introduction (PDF chapter).*

### 2. Change in Health-Adjusted Life Expectancy, 2000–21

### 2. Change in Health-Adjusted Life Expectancy, 2000–21

### Trends and measurement
- World average is population-weighted, based on 183 countries. “Frontier” = maximum life expectancy across countries.
- Cognitive health score: first principal component of cognitive indicators, standardized to mean zero, standard deviation one. Regression sample period is 2000–22.
- Regressions of cognitive health for individuals ages 50 and older include controls for age, gender, education, and household wealth; country fixed effects are used in cross-country regressions.

### Cross-country and within-country disparities
- Individuals from emerging market economies (EMs) have lower health scores than those from advanced economies (AEs) after controlling for socioeconomic characteristics.
- Within countries, average cognitive health scores are significantly lower for:
  - individuals in rural locations,
  - individuals with at most primary education,
  - lower-wealth households.
- T-tests indicate that the differences in means are statistically significant for all socioeconomic categories shown.
- Heterogeneity over time: although EMs show faster improvements in healthy aging compared with AEs (some cross-country “catching up”), the pace of health improvements across gender, location, education, and wealth groups has been similar, indicating persistent socioeconomic health disparities.

### Determinants of functional capacity
- Lifestyle factors—levels of physical activity, body mass index, and smoking—are significant determinants of functional capacity of older individuals after controlling for age and socioeconomic characteristics (referenced analyses in Online Annex 2.2).

### Policy implications for narrowing healthy-aging gaps
- Strengthen health care quality and expand access, particularly for preventive care and disadvantaged groups.
- Provide incentives for healthy lifestyles.
- Example policy levers cited in Singapore’s experience: subsidizing healthier food options, regulating sugar content in beverages, building widespread public fitness centers, introducing automobile congestion charges, and subsidizing housing in proximity to family to promote intergenerational social connections.

### Labor market implications of healthy aging
- Simple correlations: higher scores on health indicators are associated with increased total labor earnings, higher labor productivity (proxied by hourly earnings), higher labor force participation, and more hours worked (see Online Annex Table 2.2.3).
- Instrumental-variables (two-stage least squares) approach uses exogenous health shocks proxied by development of chronic diseases and controls for smoking, poor nutrition, physical inactivity, and excessive alcohol use.
- IV estimates (rescaled to reflect health trends over 10 years) imply that average cognitive health gains observed for older-age individuals over a decade are associated with:
  - rises in labor earnings and labor productivity by about 30 percent,
  - an increase in likelihood of participating in the labor force by about 20 percentage points,
  - higher numbers of average weekly hours worked by about six hours.
- Better health is also associated with later retirement, working more weeks per year, and a lower probability of being unemployed; similar relationships hold for other health indicators.
- Age gradient: the impact of better health on labor force participation is significantly larger for individuals in their 50s than for those in their 60s and 70s, suggesting nonhealth constraints (skills obsolescence, pension incentives, age discrimination) can limit labor attachment.
- Occupation-level evidence: older workers with college educations are relatively well positioned to benefit from AI because it complements their tasks and skills.

### General equilibrium analysis: model design and baseline assumptions
- Model extension of the global overlapping-generations framework in Auclert and others (2024).
- Country coverage:
  - 21 advanced economies,
  - 4 emerging market economies (including China and India, which together account for almost 50 percent of emerging market economies’ GDP),
  - a bloc economy comprising 44 low-income countries (LICs) expected to pass their demographic turning points after 2040 (denoted LIC bloc).
  - Altogether, the model accounts for about two-thirds of the world economy and population.
- Healthy aging in the model:
  - Country-specific age-productivity profiles vary over time to integrate the impact of healthy aging on effective labor supply (proxied by labor earnings capturing effects on labor productivity and number of hours worked) documented in the empirical section.
  - Baseline assumes a continued—though moderating—improvement in the functional capacity of workers ages 50 and older over the next three decades, reflecting persistence of 2000–22 improvements as current young cohorts age.
- Productivity forces at the country level are driven by:
  - growth of total factor productivity (TFP) at the global frontier,
  - convergence toward the TFP frontier,
  - impact of demographics on TFP growth through innovation and entrepreneurship channels.
- Global capital market assumptions:
  - Integration of China, India, and the LIC bloc into global capital markets is imperfect, producing a wedge between domestic and global interest rates that is assumed to decline gradually as reforms and integration proceed.
- Fiscal policy calibration and retirement ages:
  - Initial values for effective retirement rates, labor taxes, retirement replacement rates, and other public spending are calibrated to match country-specific targets.
  - In the baseline, effective retirement ages are assumed to increase by one month per year over 60 years in all countries (except for India and the LIC bloc, where they are assumed unchanged).
  - Labor taxes, replacement rates, and other public spending adjust period by period so that trajectories of debt-to-GDP ratios are aligned with WEO projections until 2029 and remain stable from then onward.

*Source: ch2 - 2. Change in Health-Adjusted Life Expectancy, 2000–21 (PDF chapter).*

### 1.1 percentage points lower than the average over

### ch2 - 1.1 percentage points lower than the average over

### Baseline projections: growth, interest rates, and primary balances
- Global GDP growth: projected to be 1.1 percentage points lower than the 2016–18 average over 2025–50, and 2 percentage points lower when the average over 2025–2100 is considered.
- Advanced economies with relatively older populations (example: Japan) are projected to see their economies shrink under baseline fertility and migration assumptions.
- Advanced economies with projected stable or growing working-age populations (examples: Canada and the United States) will continue to grow, albeit more slowly over time.
- China: projected deceleration of 2.7 percentage points in GDP growth over 2025–50 relative to the 2016–18 average.
- India: projected deceleration of about 0.7 percentage point in GDP growth in 2025–50, intensifying over 2050–2100 as the demographic turning point is passed.
- Low-income countries (LIC bloc): expected to see a sharper deceleration in growth in the second half of the century once demographic dividends turn into headwinds.
- Output per capita: world average is projected to be about 0.6 percentage point lower in 2025–50 and 1.8 percentage points lower toward the end of the century relative to 2016–18.
- Interest-growth differential (r − g):
  - Projected to be higher than the 2016–18 average for all economies except India and the LIC bloc over the next 25 years.
  - Average r − g for the world is projected to be 1 percentage point higher in 2025–50 than in 2016–18, moderating to about 0.5 percentage point toward the end of the century.
- Fiscal pressure implication: about half of the model economies are projected to need higher primary-balance-to-GDP ratios than they had on average over 2016–18 to keep debt-to-GDP ratios stable from 2030 onward; this group includes China, Japan, and the United States.

### Net foreign assets and capital flow implications
- Large emerging market economies (China and India) would accumulate foreign assets, especially over 2050–2100.
- Many advanced economies would gradually draw down foreign assets throughout the projection horizon.
- The net foreign asset position for the LIC bloc would worsen through most of the projection period thanks to continued capital inflows, but this trend would slow and eventually reverse around 2070 as aggregate wealth increases with population aging.

### Contribution of healthy aging and demographics to growth
- Healthy aging contribution (world):
  - Healthy aging is projected to add about 0.4 percentage point to GDP growth, on average, over 2025–50.
  - If gains from healthy aging were abstracted from, global output growth would slow by 1.5 percentage points instead of 1.1 percentage points in 2025–50 relative to 2016–18.
  - The average contribution to world growth would be about 0.1 percentage point over 2050–75 and would decline further thereafter.
- Economy-specific healthy-aging contributions: positive and sizable, ranging from about 0.3 percentage point to 0.6 percentage point over 2025–50 for individual economies in the model; notably large for India and relatively lower for Japan.
- Demographics alone account for about half of the projected slowdown in GDP growth over 2025–2100 relative to 2016–18:
  - Demographic forces explain 1.1 percentage points out of a reduction of 2 percentage points in global GDP growth over 2025–2100.
  - Country examples: average contribution of demographic forces to GDP growth in 2025–2100 ranges from close to −2.8 percentage points in India to −0.4 percentage point in Finland and Slovenia.
- Sensitivity to fertility assumptions:
  - Under UNWPP high- and low-fertility scenarios, fertility is projected to remain 0.5 children above (high) or below (low) the medium-fertility scenario over most of the projection period.
  - Country-specific growth estimates under alternative fertility assumptions can vary, for instance, by 0.5 percentage point in Australia and 1.6 percentage points in China.

### Policy scenarios: labor supply measures and growth tailwinds
- Three key policy levers examined:
  1. Healthy-aging policies (narrowing cross-country differences in functional capacity of workers ages 50 and older).
  2. Increasing effective retirement age (postponing effective retirement given healthy-aging improvements).
  3. Closing gender labor force participation gaps.
- Healthy-aging policy scenario:
  - Assumes narrowing of cross-country differences in functional capacity of workers ages 50 and older by one-fourth, equivalent to about 49 percent of the estimated gains over 2000–22.
  - For the world, average annual GDP growth over 2025–2100 would be about 0.2 percentage point higher than in the baseline, and 0.3 percentage point higher over 2025–50.
- Increasing effective retirement age scenario:
  - Assumes effective retirement ages increase faster than baseline in countries where life expectancy at retirement is 20 years or more; otherwise evolve as in baseline.
  - For the world, average annual GDP growth over 2025–2100 would be about 0.1 percentage point higher than in the baseline.
- Closing gender participation gaps scenario:
  - Assumes narrowing country-specific gender gaps in labor force participation by three-fourths by 2040.
  - For the world, average annual GDP growth over 2025–2100 would be 0.1 percentage point per year higher than in the baseline, and 0.3 percentage point higher over 2025–50.
- Combined policy package:
  - Implementing all three labor supply measures together would raise global average annual growth by 0.3 percentage point over 2025–2100 relative to baseline, reversing about one-third of the drop in growth attributable to demographic trends through the end of the century.
  - Over 2025–50, the combined package would boost growth by about 0.6 percentage point, offsetting close to three-fourths of the drag from demographics during that period.
  - Some countries (India, low-income countries, and some European economies) could reap even higher growth dividends.
  - Non-monetary benefits: keeping older workers engaged can offer societal benefits from improved well-being.

### Fiscal implications of labor supply policies
- Direct fiscal channels:
  - Increased female labor force participation and employment would boost labor tax revenues.
  - Raising effective retirement ages would increase labor taxes and reduce transfer payments.
- Indirect fiscal channel via r − g:
  - Higher GDP growth from labor policies would help reduce r − g, easing fiscal pressures and lowering the primary balance needed to stabilize debt ratios—though labor supply policies could also put upward pressure on interest rates due to changes in aggregate savings and investment demand.
- Model simulation results:
  - Five model economies (Greece, India, Italy, Spain, and the LIC bloc) would see fiscal respite from lower r − g over 2025–50, reinforcing direct benefits on primary balances.
  - Under the combined policy scenario, all model economies would gain fiscal space (able to increase spending or reduce taxes while maintaining public debt targets), but gains are heterogeneous:
    - Equivalent fiscal gains, on average, would be more than 4 percentage points of GDP in Greece and Italy.
    - Less than 1 percent of GDP in China and the United Kingdom.
- Caveats:
  - Implementing some policies can entail direct budgetary costs (for example, active labor market policies), which could reduce net fiscal dividends.
  - Some healthy-aging policies (for example, taxes on alcohol and tobacco; preventive health policies) can generate revenue or future savings.
  - The model abstracts from quantifying direct implementation costs due to large uncertainty.

*Source: https://www.imf.org/-/media/files/publications/weo/2025/april/english/ch2.pdf*

### Conclusions and Policy Implications

### ch2 - Conclusions and Policy Implications

### Demographic trends and healthy aging
- Declining birth rates and increasing life expectancy are causing a sustained decline in population growth and significant changes in age structure.
- Individuals ages 50 and older are, on average, aging in better health than previously; increased longevity has been accompanied by improvements in physical and cognitive capacities across subsequent cohorts.
- Healthier aging has been associated with higher labor force participation rates, a higher likelihood of being employed, and higher labor earnings for individuals ages 50 and older.
- Spending on health promotion and prevention accounts for only 1–6 percent of total health expenditure in member countries of the Organisation for Economic Co-operation and Development.

### Growth and fiscal outlook
- Ongoing gains from healthy aging are estimated to boost annual global growth by about 0.4 percentage point over 2025–50.
- Under current policies, global output growth would decline on average by about 2 percentage points through the end of the century.
- With lower growth prospects and historically high levels of public debt, many countries will need significant fiscal efforts to keep debt-to-GDP ratios stable beyond 2030.
- Simulations suggest a combination of policies for boosting labor supply could attenuate the slowdown in global growth over 2025–50 resulting from demographic headwinds by almost three-fourths.
- Progrowth policies could contribute to higher global interest rates but would provide substantial fiscal dividends and create additional fiscal space for many countries to finance critical spending; some economies would still require additional fiscal effort.

### Policies to extend healthy, productive working lives
- Health promotion and prevention
  - Target behavioral risk factors across the life course: tobacco smoking, harmful alcohol use, physical inactivity, and unhealthy diets.
  - Examples of measures: immunization, regular health checks, screenings for chronic diseases, campaigns to prevent substance abuse, taxation (for example, on tobacco and unhealthy food), regulations (for example, to promote smoke-free environments), and providing access to mental health resources.
  - Many such measures span beyond the health care sector, are not necessarily costly, and evidence suggests many are cost-effective and can produce savings by reducing expenditure on health intervention down the road.
- Pension reforms and retirement policy
  - Complement health interventions with pension reforms, training, and workplace adaptations to increase effective retirement ages in line with improvements in life expectancy.
  - Instruments include changing statutory retirement ages, reducing early retirement benefits, introducing incentives to postpone retirement, and allowing phased retirement.
  - Age-based provisions in policies are likely to be inefficient and should be reconsidered given heterogeneity in health and capacities among older workers.
- Lifelong learning and workplace adjustments
  - Lifelong upskilling and reskilling programs to keep individuals employable as they age, especially important in the context of a potential AI revolution where skilled older workers may complement AI.
  - Flexible work arrangements and workplace adaptations to improve the age-friendliness of jobs.
  - Combat biases and discrimination against older individuals to prevent premature exits from the labor force.
- Policies to raise female labor force participation and support fertility
  - Improve work-life balance for women through improved parental leave systems, expanding affordable childcare options, and promoting flexible work arrangements.
- Global integration and financial access
  - Enhance access to international financial markets—credit and capital market reforms—and strengthen governance and institutions to enable low-income countries to reap demographic dividends.
  - Deeper financial integration can offset labor losses from migration outflows toward advanced economies.

### Technological progress and innovation
- Structural reforms to promote market competition, financial accessibility, and labor market flexibility can boost productivity growth by fostering innovation and a more efficient allocation of capital and labor.
- AI-related technologies are complementary to labor in occupations more typical of older workers and can help older workers cope with functional decline.
- Research and development in the scientific understanding of biological aging could further extend healthy longevity.
- AI-based solutions in health care can scale up preventive practices (for example, automating screening and diagnostics and bringing expertise to underserved areas).

### Simulations and scenario results (selected quantitative findings)
- Healthy aging contribution: boost to annual global growth of about 0.4 percentage point over 2025–50.
- Long-run demographic drag: global output growth decline on average about 2 percentage points through the end of the century.
- Combined policy scenario: average fiscal gains reported over 2025–2100 (figure note), with country-level variation.
- Pension reform instrument sizes needed to reverse the aging-induced increase in the public-debt-to-GDP ratio over 75 years (calibrated outcomes):
  - Single-Instrument Reform (Immediate / Delayed):
    - Retirement Age (Years): +6 / +8
    - Replacement Rate (%): −25 / −35
    - Contribution Rate (%): +18 / +34
  - Mix (Immediate / Delayed):
    - Retirement Age (Years): +2 / +2.7
    - Replacement Rate (%): −8.3 / −11.7
    - Contribution Rate (%): +6 / +11.3

### Low-income countries: financial integration and migration scenarios (selected quantitative findings)
- Status quo scenario: net foreign liabilities peak at about 13 percent of GDP.
- Enhanced financial integration scenario:
  - Net foreign liabilities reach about 180 percent of GDP by 2070–80.
  - Stock of capital and output significantly higher than in the status quo; both GDP and GDP per capita are about 19 percentage points higher than in the status quo scenario in the long term.
  - Long-run increase in gross national income per capita is smaller than that for GDP but still sizable at about 7 percentage points.
- Enhanced financial integration plus migration scenario:
  - Aggregate GDP in the LIC bloc is about 5 percentage points lower relative to the enhanced financial integration scenario but still 14.5 percentage points higher than in the status quo scenario.
  - GDP per capita in the LIC bloc is about 1.2 percentage points lower than in the enhanced financial integration scenario but almost 18 percent higher than in the status quo scenario.
- Implication: Financial sector reforms and strengthened governance and institutions are important to enable low-income countries to capture demographic dividends and offset potential output losses from migration outflows.

*Source: International Monetary Fund, ch2 - Conclusions and Policy Implications (chapter text).*

### Box 2.2. Intergenerational Considerations in Pension Reforms

### Box 2.2. Intergenerational Considerations in Pension Reforms

### Main findings on reform design and timing
- Under a combined reform scenario, the increase in retirement age could be less, at two years.
- The consumption losses from reforms that rely on a single instrument are significantly larger than those when a mix of instruments is used, at least for one of the generations.
- The size of required fiscal measures is more profound and aggregate consumption losses are larger when the reforms are postponed for 10 years compared with those in a scenario in which they are implemented immediately.
  - Containing the rise in public debt induced by aging requires a 6-year increase in the retirement age if the reform is carried out immediately, but an 8-year increase is needed if the reform is postponed by 10 years.
  - The consumption losses from postponing reforms usually fall disproportionately on the young compared with the old.

### Evidence from the consumption-impact exercise
- Figure context: "Figure 2.2.1. Average Change in Consumption, 2025–65 (Deviation from 2025, percentage points)"
  - Solid bars denote average consumption losses or gains over a period of 40 years from a reform implemented immediately.
  - Markers denote consumption losses or gains if instead the reform is delayed by 10 years.
  - The red bars show the impact of aging only, and the rest of the bars show the impact of aging and the respective reforms.
- Reform scenarios considered in the figure: No reform; Lower replacement rate; Higher retirement age; Higher contribution rate; Mix of reforms.
- Key qualitative takeaway: Combining the three measures helps ensure the burden is shared across the young and old, potentially contributing to the acceptability and feasibility of reforms.

### Policy implications and recommendations
- Policymakers should act sooner rather than later.
- Use a combination of tools to ensure a fairer distribution of the burden across generations and by doing so enhance the feasibility and acceptability of pension reforms.
- Although the exercise is calibrated for a typical advanced economy with a population that has already aged significantly, the lessons are even more pertinent for emerging market economies and low-income developing countries:
  - As of today, their old-age dependency ratios are lower than those of advanced economies.
  - They will experience a faster pace of population aging than did the latter, which means they will have less time to react.

*Source: IMF staff calculations and Box 2.2, World Economic Outlook: A Critical Juncture Amid Policy Shifts (April 2025).*

---


_Source: https://www.imf.org/-/media/files/publications/weo/2025/april/english/ch2.pdf_