Future Estimates of Public Long-Term Care Insurance Premiums in Japan and Their Considerations

Yutaka Taniguchi; Tadayoshi Otsuka

doi:10.1515/apjri-2025-0015

Article Open Access

Future Estimates of Public Long-Term Care Insurance Premiums in Japan and Their Considerations

Published/Copyright: February 24, 2026

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information Explore this Subject

From the journal Asia-Pacific Journal of Risk and Insurance

Abstract

In the context of a globally aging population, nursing care coverage has become a critical issue in insurance research. Japan, where population aging is advancing ahead of global trends, may serve as a reference for many other countries. The country’s public long-term care insurance system, introduced over two decades ago, is administered by 1,571 municipalities (as of 2020), representing a more localized governance structure compared to the 47 prefectures. Half of the system’s financial resources derive from long-term care insurance premiums, which vary across municipalities. Accurate projections of these premiums are essential for informing policies related to social security financing and public fiscal burdens. However, current estimates lack precision. This study projects the long-term care certification rate and insurance premiums at the municipal level for the next 20 years and analyzes regional disparities. Results indicate that nationwide premiums nationwide are expected to increase by 1.5–1.6 times over two decades, with disparities widening between municipalities. By 2040, half of the 20 municipalities with the highest premiums are projected to be in Osaka Prefecture, while substantial increases are also anticipated in commuter areas surrounding Tokyo. These findings underscore the need for targeted policy interventions to mitigate regional inequalities and ensure the financial sustainability of the long-term care system.

Keywords: long-term care insurance premiums; future estimates; municipal disparities

1 Introduction

The proportion of people aged 65 or older in the global population (the aging rate) increased from 5.1 % in 1950 to 9.4 % in 2020 and is projected to reach 18.7 % in 2060. Figure 1 presents the aging rates across countries. As of 2020, Japan has the most advanced aged population, and it is estimated to continue exhibiting one of the highest aging rates worldwide. In the context of global aging, the focus of insurance research has shifted from traditional protection for the working population, such as coverage for death and inability to work, to protection for retirement and long-term care, with the latter emerging as a particularly important area. Japan, where population aging is progressing ahead of the global trend, offers valuable insights into long-term care policy that likely to become increasingly relevant globally. Globally, the scope of coverage by public long-term care insurance is often only partial. Japan’s public long-term care insurance covers all citizens over the age of 40. Furthermore, Japan’s system also collects and publishes extensive data on long-term care certification rates and the utilization of nursing care services, from the preventive care stage to severe nursing care needs.

Figure 1:

Changes in aging rates (proportion of population aged 65 and over) in each country. (Source: UN, world population prospects: the 2022 revision “2024 Ageing Society White Paper” (Cabinet Office of Japan)).

There have been many studies on public long-term care insurance systems. Survey studies of public long-term care insurance systems include Review of evolution of the public long-term care insurance (LTCI) system in different countries: influence and challenge by Chen et al. (2020). Many previous studies have analyzed the financial sustainability of the public long-term care insurance system. Some studies have estimated future long-term care insurance premiums, which form the basis of finances. These are Yang et al.’s (2020) Current and projected future economic burden of Parkinson’s disease in the U.S. and Vanella et al.’s (2024) Long-Term Care in Germany in the Context of the Demographic Transition – An Outlook for the Expenses of Long-Term Care Insurance through 2050. Yang et al. (2020) are limited to the need for nursing care due to certain diseases such as Parkinson’s disease, and is therefore insufficient to grasp the overall trend in the need for nursing care. Vanella et al.’s (2024) prediction methods are simple and the accuracy of the predictions is not necessarily high. Comprehensiveness and accuracy are important when estimating future public long-term care insurance premiums.

Previous studies that have estimated future long-term care insurance premiums in Japan include Fukui (2016), Kikuchi (2019), and Mitsubishi Research Institute (2020). However, even in these previous studies, future estimates of the rate of long-term care certification are simple and have issues with accuracy. Furthermore, the characteristics of nursing care services in each region are not reflected in the nursing care costs. In contrast, we believe that highly accurate estimates of the rate of long-term care certification and future nursing care costs, which utilize the wide coverage and abundant data of Japan’s public long-term care insurance, will be useful in planning future measures for long-term care security in many countries.

The purpose of this paper is to make highly accurate estimates of future long-term care certification rates and estimate long-term care insurance premiums by municipality (insurer). Additionally, we analyze disparities in long-term care insurance premiums between municipalities. Following the author’s previous research, the analysis method uses the PLAT model, which can take cohort effects into account, and nursing care costs reflect the characteristics of the nursing care cost structure according to each insurer’s nursing care resources.

Our findings suggest that the nationwide premiums are expected to increase by 1.5–1.6 times over two decades, with disparities widening between municipalities. By 2040, half of the 20 municipalities with the highest premiums are projected to be in Osaka Prefecture. It was also confirmed that municipalities that will see a significant increase in premiums in 2040, 20 years from now, will be concentrated in Tokyo’s commuter towns.

2 Current Status of Japan’s Public Long-Term Care Insurance System

In Japan, the continuing rise in average life expectancy is accompanied by increasing long-term care insurance premiums. Municipalities serve as the insurers of the system. Table 1 compares Japan’s public long-term care insurance system with similar systems in other countries. Japanese municipalities contribute 70–90 % of long-term care service costs and finance operations by collecting premiums from two categories of insured individuals: Category 1 (aged 65 and older) and Category 2 (aged 40–64). The financial resources are divided equally among public funding and insurance premiums. The public share is distributed as follows: national government (20 %), municipalities (12.5 %), prefectural governments (12.5 %), and intermunicipal adjustments (5 %) (Figure 2). The insured pay the other half of the premiums monthly, with the amount determined by each municipality based on the projected cost of insurance benefits over three years. These projections are revised every three years when formulating long-term care insurance business plans.

Table 1:

Differences between Japan’s public long-term care insurance system and those of other countries.

Feature	Japan	Germany	Canada	France	Netherlands	Korea
System name	Long-term care insurance system	Pflegeversicherung	State programs	APA^a	WLZ^b	Long-term care insurance system
Year introduced	2000	1995	1995	2002	2015	2008
Insurer (operation)	Municipalities	Nursing care funds^c	Provinces & territories	Prefecture	National care administration	National health insurance corporation
Number of insurers	1,579	109	13	13	1	1
Eligibility	Category 1: aged ≥65 Category 2: aged 40–64 with specific conditions^d	All individuals with public or private health insurance (no age limit)	Not applicable (no premium-based system)	Not applicable (no premium-based system)	Residents or income earners in the Netherlands	All residents (no age limit)
Benefit requirements	Certified as requiring long-term care or support; 40–64 s must have qualifying diseases	Certified according to long-term care grade (no age limit)	None specified	Certified under a long-term care class (aged ≥60)	Must meet all of the following: physical or cognitive impairment; lifelong care needs; 24-h care requirement	Certified for long-term care; typically aged ≥65, but includes under-65 s with senile diseases
Funding	50 % public funding, 50 % insurance premiums	Insurance premiums	Federal and provincial funding	58 % prefectural funds, 42 % national solidarity bank (2018)	81 % insurance premiums, 19 % government subsidies (2017)	64 % premiums, 11 % subsidies, 23 % medical benefits (2017)
Payment type	In-kind benefits^e	In-kind and cash benefits	In-kind benefits	In-kind benefits	In-kind and cash benefits	In-kind benefits
Out-of-pocket expenses	10–30 %^f	Based on personal assets	Based on income	Based on income	15 % (home care), 20 % (institutional care)	Not specified

Source: international comparative survey report on public long-term care systems, comprehensive policy survey and research fund for medical security (National Federation of Health Insurance Societies 2020); long-term care security in the World, Masanobu Masuda (Houritsubunkasha, 2014). ^aAPA, allocation Personnalisée d’Autonomie, a French long-term care allowance program. ^bWLZ, Wet langdurige zorg, the Dutch long-term care act for intensive care needs. ^c“Nursing care funds” (Pflegekassen) are managed by statutory and private health insurers in Germany. ^d“Category 2” insured persons (aged 40–64) in Japan are eligible if diagnosed with one of 16 age-related diseases (e.g., early-onset dementia, cerebrovascular disease). ^eIn-kind benefits include direct provision of services such as home care, nursing support, and institutional care. ^fJapan’s co-payment ranges from 10 % (low-income) to 30 % (high-income), depending on income and care level.

Figure 2:

Funding of Japan’s public long-term care insurance system. (Source: Created by the author from “about the long-term care insurance system (for those who have turned 40 years old)” (Ministry of Health, Labour and Welfare 2024)).

When the long-term care insurance system began in 2000, the national average premium (weighted by the population of each municipality) was 2,911 yen per month. By 2023, it had increased to 6,014 yen per month – 2.07 times the initial figure. During this period, the aging rate (the proportion of the population aged 65 years and over) rose from 17.4 % in 2000 to 28.6 % in 2020, an increase of 1.64 times. The disparity between these growth rates highlights that the rise in long-term care insurance premiums has outpaced the aging rate. Historical data illustrated in Figure 3 confirms this pattern and indicates that further premium increases are anticipated.

Figure 3:

Trends in the aging rate and long-term care insurance premiums. (Source: “White Paper on Aging Society” 2023 (Cabinet Office) on the aging rate, and “Recent Trends in the Nursing Care Sector” 2023 (Ministry of Health, Labour and Welfare Bureau)).

3 Research on Long-Term Care Certification Rates

In Otsuka and Yutaka 2019, Future Estimation of Healthy Life Expectancy and the Number of Persons Requiring Long-Term Care, a model from the family of probabilistic mortality prediction models (Generalized Age-Period Cohort [GAPC]) was used to estimate the future proportion of individuals requiring long-term care. The GAPC model consists of:

Random Component:
D x t ∼ Poisson E x t c μ x t or D x t ∼ Binomial E x t 0 , q x t
Systematic Component:
η x t = α x + ∑ i = 1 N β x i κ t i + β x 0 γ t − x
Link Function:
g E D x t E x t = η x t
If D _xt follows a Poisson distribution, g is a log link function; if D _xt follows a binomial distribution, g is a logit link function.
Constraints for a set of parameter constraints to be uniquely determined:

θ = α x , β x i , γ t − x

Here

D x t : Number of deaths at age x in year t

E x t c : Median risk exposure at age x in year t

E x t 0 : Risk exposure at age x in year t at the beginning of the year

E D x t / E x t c = μ x + t , E D x t / E x t 0 = q x + t

The Lee-Carter model, widely used for mortality estimation, has undergone various transformations and extensions, including the Renshaw and Haberman (2006) cohort component and the Cairns-Blake-Dowd (CBD) model (Cairns et al. 2006). The PLAT model, combining CBD and Lee-Carter features, was used in the study to estimate long-term care certification rates for 2040 (Plat 2009).

In view of the rapid spread of such probabilistic mortality models, attempts have been made to identify the commonalities between them. Currie (2016) demonstrated that many mortality models can be represented in terms of generalized linear models or generalized nonlinear models. The family of probabilistic mortality models organized by Currie (GAPC) led to the development and introduction of the R package StMoMo by Villegas et al. (2017) (Table 2).

Table 2:

Overview of stochastic mortality models (StMoMo) and their prediction formulas.

Model	Prediction formula
LC (Lee-Carter)^a	η_xt = α_x + β_x ⁽¹⁾ κ_t ⁽¹⁾
CBD (Cairns–Blake–Dowd)^b	η_xt = κ_t ⁽¹⁾ + (x − x̄) κ_t ⁽²⁾
APC (age-period-cohort)^c	η_xt = α_x + κ_t ⁽¹⁾ + γ_(t−x)
RH (Renshaw–Haberman)^d	η_xt = α_x + β_x ⁽¹⁾ κ_t ⁽¹⁾ + γ_(t−x)
M7 (Quadratic CBD)^e	η_xt = κ_t ⁽¹⁾ + (x − x̄) κ_t ⁽²⁾ + [(x − x̄)² − ŝ_x ²] κ_t ⁽³⁾ + γ_(t−x)
PLAT (Plat Model)^f	η_xt = α_x + κ_t ⁽¹⁾ + (x̄ − x) κ_t ⁽²⁾ + γ_(t−x)

^aLC, Lee–Carter model. ^bCBD, Cairns–Blake–Dowd model. ^cAPC, age–period–cohort model. ^dRH, Renshaw–Haberman model. ^eM7: quadratic extension of the CBD model with a cohort term. ^fPLAT: extension of the CBD model with age-modulated effects and cohort term. In each formula: •α_x is a static age function that captures the general shape of age-related mortality. •β_x ⁽¹⁾ is an age-adjustment term interacting with the time component. •κ_t ⁽ⁱ⁾ represents mortality trend(s) over time. •γ_(t−x) denotes the cohort effect. •x̄ is the mean of the age variable, and ŝ_x ² is the sample variance.

Otsuka and Yutaka 2019, used data from 2009 to 2015, obtained via the Fact-Finding Survey on Long-Term Care Benefits (Ministry of Health, Labour and Welfare), but estimated up to the next 20 years (2040) using the PLAT model. When comparing the AIC and BIC, which indicate the accuracy of estimating the percentage of individuals requiring long-term care, they adopted the PLAT model, in which both the AIC and BIC were small, and the estimation results showed reasonable values. The PLAT model combines the features of the CBD model, which expresses mortality in old age as a logit, and the Age- Period-Cohort (APC) model, which expresses the cohort, with the Lee-Carter model, which is generally used for mortality estimation, and is considered to be a mortality model suitable for each age group. However, in the previous study, the cohort effect was not sufficiently reflected in the model owing to the small number of years of data.

4 Methods of Analysis

4.1 Future Estimation of the National Total Long-Term Care Certification Rate

As the average life expectancy of the elderly increases every year, previous studies have indicated that the rate of aging is slowing due to variations in age at birth.^[1] Therefore, when estimating future long-term care certification rates, it is essential to conduct a cohort analysis that accounts for differences in birth age and chronological age. This study uses data spanning 14 years (October 2009 to 2022) from the Fact-Finding Survey on Long-Term Care Benefits (Ministry of Health, Labour and Welfare), characterized by gender and five-year age groups.

The percentage of individuals certified as requiring support or care was calculated by dividing the number of certified individuals by the population for each age group and gender. Using the ratio of long-term care need by gender, age, and degree of care from the past 14 years, projections were made up to 2040 using the PLAT model, similar to the methodology in Future Estimation of Healthy Life Expectancy and the Number of Persons Requiring Long-Term Care (Otsuka and Yutaka 2019). While earlier research relied on only 7 years of data, which limited the reflection of cohort effects, this study uses twice the amount of data, enabling better incorporation of cohort effects in the PLAT model.

The accuracy of the PLAT model, as used in the previous study, is confirmed in the Appendix. Specifically, the model parameters were estimated using data from 2009 to 2018 to predict the long-term care certification rate from 2019 to 2022. These predictions were then compared to the observed certification rates from 2019 to 2022.

4.2 Future Estimates of Long-Term Care Certification Rates by Municipality

In Section 4.1, we estimated the nationwide rate of long-term care certification. Next, we estimate certification rates at the municipal level. However, directly applying the method used in Section 4.1 to municipalities results in unstable estimates for those with small populations. To address this, we apply the nationwide certification rate (calculated in Section 4.1) to future population estimates for municipalities, as provided by the National Institute of Population and Social Security Research. Previous studies, such as Taniguchi’s (2022) Consideration on the Sustainability of Public Long-Term Care Insurance and Taniguchi et al. (2022) Analysis of Regional Differences in the Nursing Care Certification Rate by EDA Method, suggest that certification rates are higher in metropolitan areas and lower in depopulated regions. Based on this, discrepancies between national totals and local rates were considered. To estimate the number of certified individuals and corresponding insurance premiums at the municipal level, the growth rate of certified individuals was applied instead of directly estimating local certification numbers. Owing to data limitations, Fukushima Prefecture was excluded from the calculation. As shown in the following formula, the first stage estimates the number of individuals certified as requiring long-term care in each municipality, but in the second stage, the actual value of the number of individuals in need of long-term care in each municipality is used as the initial value, and the estimation is estimated by applying the increase rate of the number of certified individuals as needing long-term care to the initial value.^[2]

The estimation process involves two stages:

Stage 1:

(1) Number of certified individuals requiring long term care in Region A every five years in the future , by gender , age , and degree of long term care required = National Institute of Population and Social Security Research estimates for Region A (every five years in the future) × Total rate of certification of long term care required nationwide based on the PLAT model (every five years in the future , by gender , age , and degree of care required)

Stage 2:

(2) Number of certified individuals requiring long term care in Region A (every five years in the future, by gender, age, and degree of long term care required) = (Number of certified individuals as requiring long term care in Region A as of 2020 according to the Long–Term Care Insurance Business Status Report (Reiwa 2nd Annual Report)) × (increase in the number of certified individuals requiring long term care in Region A based on 2020, as determined in the first stage (every five years in the future , by gender , age , and degree of care required))

4.3 Future Estimation of Long-Term Care Insurance Premiums by Municipality

Public long-term care insurance is funded equally by public expenses (50 %) and premiums (50 %). Premiums are paid by individuals aged 65 or older (Category 1) and those aged 40–64 (Category 2). The standard monthly premium for each municipality was calculated using the following formula^[3]:

(3) Standard monthly premium for Region A = Total long term care service costs in Region A × Number of individuals aged 65 years and over nationwide / Number of individuals aged 40 years and over nationwide × 50 % / Number of individuals aged 65 years and over in Region A / 12

Data for population estimates are sourced from the National Institute of Population and Social Security Research. Long-term care costs reflect regional characteristics rather than national averages, using data from the Long-Term Care Insurance Business Status Report (Reiwa 2nd Fiscal Year Annual Report).

The total service costs for Region A were estimated as follows:

(4) ∑ x ∈ 65 ～ 69 , 70 ～ 74 , 75 ～ 79 , 80 ～ 84 , 85 ～ 89 , 90 ～ , i ∈ L e v e l 1 r e q u i r i n g a s s i s t a n c e ∼ L e v e l 5 r e q u i r i n g c a r e Number o f certified i n d i v i d u a l s needing care level i in age group × in Region A × Number of home care service recipients with nursing care level i in Region A Number of certified i n d i v i d u a l s needing nursing care level i in Region A ( ＝ Home service utilization rate ) × ( Home care service costs per home care service recipient with care level i in Region A ) + Number of recipients of community based care services with level i of care required in Region A Number of certified i n d i v i d u a l s needing nursing care level i in Region A ＝ Rate of use of community based services × Cost of community based services per person receiving community based services with level i of care required in Region A + Number of facility service recipients with nursing care level i in Region A Number of certified i n d i v i d u a l s needing nursing care level i in Region A ( ＝ Facility service utilization rate ) × ( Facility service cost per person receiving facility services with nursing care level i in Region A ) × Nursing care service costs for insured persons of category 1 in Region A + nursing care service costs for insured persons of category 2 in Region A Nursing care service costs for insured persons of category 1 in Region A

Adjustments for municipal differences in income and elderly population proportions are reflected in subsidy correction factors. The correction factors ensure that premiums are equitable across municipalities accounting for disparities in age demographics and income levels. As shown in , the subsidy is adjusted so that (1) the benefit level of the municipality is the same, and (2) the insurance premium burden is identical if the income level of the insured person is the same. Therefore, assuming that the income level and unit cost of benefits are constant, if the proportion of elderly individuals in later stages of life is higher than the national average, the correction factor for such proportion will be less than one, the adjustment payment rate will exceed 5 %, and the long-term care insurance premium will be reduced. Conversely, if this proportion is lower than the national average, the correction factor will be greater than 1, the adjustment payment rate will fall below 5 %, and long-term care insurance premiums will rise. Although the formula for calculating the adjustment subsidy differs depending on the applicable year, in this study, to compare the long-term care insurance premiums for 2020 and the long-term care insurance premiums for the future in 2040, the formula for the adjustment grant as of 2020 – the 7th period (2018–2020) – was also applied to future estimates. Specifically, the “correction coefficient for the proportion of elderly individuals in later stages of life” in Equation (5) and the “correction factor for income level” in Equation (6) are multiplied by the “standard amount of long-term care insurance premiums for Region A (monthly)” in Equation (3) to calculate the standard amount (monthly) of long-term care insurance premiums for Region A, after adjusting for adjustment subsidies Equation (7) (Figure 4).

Figure 4:

Conceptual diagram of the adjustment grant. (Source: created and updated to 2020 figures by the author based on Mecha nism of Adjustment Grants (Ministry of Health, Labour, and Welfare 2004)).

Correction factor for the proportion of elderly individuals in later stages of long-term care insurance premiums in Region A:

(5) 0.5 × (National average proportion of elderly individuals in early stages of life) × (National average rate of nursing care certification for early stages of life) ＋（ National average proportion of elderly individuals in the late elderly age group ) × （ National average rate of nursing care certification for late stages of life ) (Proportion of early elderly individuals among elderly population in Region A ) × ( National average rate of nursing care certification for early stages of life ) ＋（ Proportion of late elderly individuals among elderly population in Region A ) × （ National average rate of nursing care certification for late stages of life ) + 0.5 × ( National average proportion of elderly individuals in early stages of life ) × ( National average rate of nursing care certification for early stages of life ) ＋（ National average proportion of elderly individuals aged 75 − 84 ) × （ National average proportion of nursing care certification rate for those aged 75 − 84 ) ＋（ National average proportion of elderly individuals aged 85 or more ) × （ National average proportion of nursing care certification rate for those aged 85 or more ) ( Proportion of early elderly individuals among elderly population in Region A ) × ( National average rate of nursing care certification for early stages of life ) ＋（ Proportion of elderly individuals aged 75 − 84 in Region A ) × （ National average proportion of nursing care certification rate for those aged 75 − 84 ) ＋（ Proportion of elderly individuals aged 85 or more in Region A ) × （ National average proportion of nursing care certification rate for those aged 85 or more )

Correction factor for the income level of the long-term care insurance premium base in Region A:

(6) 1 − 0.5 × Percentage of Stage 1 insured persons in Region A − Percentage of Stage 1 insured persons on the national average + 0.25 × Percentage of Stage 2 insured persons in Region A − Percentage of Stage 2 insured persons on the national average + 0.25 × Percentage of Stage 3 insured persons in Region A − Percentage of Stage 3 insured persons on the national average + 0.1 × Percentage of Stage 4 insured persons in Region A − Percentage of Stage 4 insured persons on the national average − 0.2 × Percentage of Stage 6 insured persons in Region A − Percentage of Stage 6 insured persons on the national average − 0.3 × Percentage of Stage 7 insured persons in Region A − Percentage of Stage 7 insured persons on the national average − 0.5 × Percentage of Stage 8 insured persons in Region A − Percentage of Stage 8 insured persons on the national average − 0.7 × Percentage of Stage 9 insured persons in Region A − Percentage of Stage 9 insured persons on the national average

(7) Standard amount of long term care insurance premiums for Region A after adjustment for subsidy monthly = Standard long term care insurance premiums for Region A in Equation 3 monthly . Correction factor for the proportion of elderly individuals in later stages of life × Equation 5 Correction factor for income level × Equation 6

The calculation method for the Future Estimation of Long-Term Care Insurance Premiums by Insurer (After Adjustment of Adjustment Grant) is as follows: First, based on the actual value of long-term care insurance premiums by insurers in 2020, the adjustment amount due to the adjustment subsidy – that is, the adjustment for the proportion of elderly individuals in later stages of life and the adjustment for income level – was calculated according to Equations (8) and (9). By deducting these adjustments from the actual value of long-term care insurance premiums, it becomes “long-term care insurance premiums by insurer in 2020 (before adjustment of adjustment subsidies).” Here, using the long-term care insurance premiums for each insurer in 2020 and 2040 (before consideration of adjustment subsidies) obtained in Equation (3), the increase rate from 2020 to 2040 is multiplied by “long-term care insurance premiums by insurer in 2020 (before adjustment subsidies)” to calculate “long-term care insurance premiums by insurer in 2040 (before adjustment subsidies).” Finally, Equation (7) derives “long-term care insurance premiums by insurer in 2040 (before adjustment subsidy)” to “long-term care insurance premiums by insurer in 2040 (after adjustment subsidy).”

Adjustment of the proportion of elderly individuals in later stages of the adjustment subsidy to the standard amount of long-term care insurance premiums in Region A:

(8) Standard amount of nursing care insurance premium in Region A after adjustment for adjustment subsidy （ Correction coefficient for the proportion of elderly individuals ) × Correction coefficient for income level × （ Correction coefficient for the proportion of elderly individuals − 1 ) + （ Correction coefficient for the proportion of elderly individuals − 1 ) × （ Correction coefficient for income level − 1 ) × 0.5

Adjustment of the income level of the adjustment grant from the standard amount of long-term care insurance premiums in Region A:

(9) Standard amount of nursing care insurance premium in Region A after adjustment for adjustment subsidy （ Correction coefficient for the proportion of elderly individuals ) × Correction coefficient for income leve × （ Correction coefficient for income leve − 1 ) + （ Correction coefficient for the proportion of elderly individuals − 1 ) × （ Correction coefficient for income leve − 1 ) × 0.5

5 Trial Calculation Results and Their Analysis

5.1 Basic Statistics of Long-Term Care Insurance Premiums for 2000, 2020, and 2040

The long-term care insurance premiums by municipality are listed in Table 3.^[4] The figures for 2000 and 2020 are actual values, while the figures for 2040 are future estimates derived using the methods described in Section 4. The upward trend in long-term care insurance premiums is expected to persist. The weighted average of premiums, calculated using the number of Category 1 insured persons in each municipality, increased from 2,911 yen in 2000 to 5,866 yen in 2020 (a 2.0-fold increase) and is projected to rise to 9,277 yen by 2040 (a 1.6-fold increase from 2020). Although the rate of increase in the next 20 years is expected to be slower than in the past 20 years, premiums are anticipated to continue rising.^[5]

Table 3:

Descriptive statistics of long-term care insurance premiums in Japan (2000–2040).

Statistic	2000 (¥)	2020 (¥)^c	2040 (¥)^c	2020/2000 (ratio)	2040/2020 (ratio)
Minimum value	1,500	3,000^b	3,457	2.0	1.2
Simple average value	NA^a	5,758	8,343	NA^a	1.4
Weighted average	2,911	5,866	9,277	2.0	1.6
Maximum value	4,000	8,700^b	13,329	2.2	1.5
Standard deviation	NA^a	681	1,227	NA^a	1.8
Maximum/minimum (%)	267 %	290 %	386 %	1.1	1.3
Maximum − minimum (¥)	2,500	5,700	9,872	2.3	1.7
Coefficient of variation	NA^a	0.118	0.147	NA^a	1.2

^a“NA” indicates data were not available or not applicable for the corresponding year. ^bMinimum and maximum values for the year 2000 are approximate estimates. ^cData from Fukushima Prefecture are excluded from the 2020 and 2040 statistics.

Additionally, the disparity in premiums set by each municipality is expected to widen. The maximum premium increased from approximately 4,000 yen in 2000 to 8,700 yen in 2020 (a 2.2-fold increase) and is projected to reach 13,329 in 2040 (a 1.5-fold increase from 2020). This trend is consistent with the weighted average. In contrast, the minimum premium rose from about 1,500 yen in 2000 to 3,000 yen in 2020 (a 2.0-fold increase) and is projected to reach 3,457 yen in 2040 (a 1.2-fold increase from 2020). The increase in the minimum premium is comparatively modest. The difference between the maximum and minimum premiums widened from 2,500 yen in 2000 to 5,700 yen in 2020 and is expected to reach 13,329 in 2040. The difference between the minimum and maximum is 2.3 times from 2,500 yen to 5,700 yen in the past 20 years and is expected to increase 1.7 times from 5,700 yen to 9,872 yen in the next 20 years. The maximum and minimum values are 267 % in 2000, 290 % in 2020, and 386 % in 2040, a difference of approximately four times.

The coefficient of variation (standard deviation/national simple average) was 0.118 in 2020 and is projected to rise to 0.147 in 2040 (a 1.2-fold increase). This indicates that the disparity in premiums is not limited to specific municipalities but is widening across all municipalities.

5.2 Municipalities with High and Low Long-Term Care Insurance Premiums

This section examines future estimates of long-term care insurance premiums across municipalities. As shown in Section 5.1, premiums in many municipalities increased between 2020 and 2040, and disparities between municipalities widened. The Spearman rank correlation coefficient for premiums ranked in descending order for 2020 and 2040 was 0.40680, indicating a weak correlation. This suggests that increases in premiums are driven more by individual municipality-specific factors than by structural factors common across municipalities.

Tables 4a and b presents the 2020 actual premiums and 2040 estimated premiums for municipalities ranked in descending order. For the top 20 municipalities with the highest premiums in 2040, a circle is noted under “2020 applicable” if they were also in the top 20 in 2020. Similarly, a circle is placed under “increase rate” if they were among the top 20 municipalities with the largest rate of increase. The same approach is applied for the bottom 20 municipalities.

Table 4a:

Municipalities with the highest long-term care insurance premiums.

Rank	2020 prefecture	2020 municipality	Premium in 2020 (¥)	2040 prefecture	2040 municipality	Premium in 2040 (¥)	Applicable in 2020^a	Rate of increase^b
1	Tokyo	A Village	8,700	Osaka	A City	13,329	✓
2	Akita	B Town	8,400	Tokyo	B Ward	13,146		✓
3	Aomori	C Town	8,380	Aomori	C Town	12,651	✓
4	Iwate	D Town	8,100	Iwate	D Town	12,533	✓
5	Osaka	E City	7,927	Osaka	E City	12,423
6	Akita	F Town	7,900	Osaka	F Union	12,392
7	Wakayama	G Town	7,829	Osaka	G City	12,093
8	Aomori	H Town	7,760	Osaka	H City	11,942
9	Nara	I Village	7,700	Hokkaido	I Town	11,846
10	Wakayama	J Town	7,700	Nara	J City	11,825
11	Kagoshima	K Town	7,700	Akita	K Town	11,657	✓
12	Wakayama	L Town	7,650	Saitama	L Town	11,598		✓
13	Aomori	M Town	7,620	Nara	M Village	11,463
14	Kagoshima	N Town	7,600	Osaka	N City	11,379		✓
15	Aomori	O Village	7,500	Osaka	O City	11,350
16	Tokyo	P Village	7,500	Aomori	P Town	11,322	✓
17	Nara	Q Village	7,500	Osaka	Q Village	11,273
18	Okayama	R Village	7,500	Osaka	R Town	11,257		✓
19	Kagawa	S Town	7,500	Akita	S Town	11,240
20	Aomori	T Town	7,480	Osaka	T City	11,171

^a✓ = Indicates applicable. ^bRate of increase between 2020 and 2040.

Table 4b:

Municipalities with the lowest long-term care insurance premiums.

Rank	2020 prefecture	2020 municipality	Premium in 2020 (¥)	2040 prefecture	2040 municipality	Premium in 2040 (¥)	Applicable in 2020^a	Percentage decrease^b
1	Hokkaido	A Village	3,000	Tokyo	A Village	3,457	✓	✓
2	Gunma	B Town	3,300	Kochi	B Town	4,428	✓
3	Tokyo	C Village	3,374	Hokkaido	C Village	4,538	✓
4	Hokkaido	D Town	3,800	Kochi	D Village	4,605		✓
5	Miyagi	E Town	3,900	Yamanashi	E Town	4,730		✓
6	Chiba	F Town	3,900	Hokkaido	F Village	4,797		✓
7	Hokkaido	G Town	4,000	Nagano	G Village	4,904		✓
8	Hokkaido	H Village	4,000	Tokyo	H Town	4,996		✓
9	Saitama	I Town	4,000	Gunma	I Town	5,022	✓
10	Aichi	J City	4,040	Okinawa	J Village	5,066		✓
11	Aichi	K Town	4,041	Kagoshima	K Village	5,107		✓
12	Miyagi	L Town	4,050	Chiba	L Town	5,138
13	Hokkaido	M Town	4,180	Miyagi	M Town	5,223	✓
14	Hokkaido	N Town	4,200	Hokkaido	N Town	5,242	✓
15	Saitama	O Town	4,261	Kagoshima	O Town	5,328
16	Hokkaido	P City	4,300	Hokkaido	P Town	5,402
17	Hokkaido	Q City	4,300	Miyagi	Q Town	5,405	✓
18	Ibaraki	R City	4,300	Tochigi	R Town	5,419
19	Yamanashi	S City	4,300	Yamanashi	S City	5,431	✓
20	Gifu	T Town	4,300	Yamanashi	T Village	5,488
20	Aichi	U Town	4,300
20	Kochi	V Town	4,300

^a✓ = Indicates applicable. ^bPercentage decrease between 2020 and 2040.

Of the top 20 municipalities in 2040, 15 were not in the top 20 in 2020, representing the majority. Nine of these municipalities are located in Osaka Prefecture. Conversely, five municipalities that remained in the top 20 between 2020 and 2040 are located in Aomori, Akita, Iwate, and Osaka prefectures. The aging population in the Tohoku region explains the trend there, while Osaka’s high utilization rate of nursing care services contributes to its ranking. For example, in Osaka City, Osaka Prefecture, the high proportion of elderly individuals living alone increases reliance on home care services, given the lack of family support. Additionally, the proportion of low-income residents is high,^[6] limiting premium collections compared to other municipalities.^[7]

Among the bottom 20 municipalities in 2040, more than 12 (over half) were not in the bottom 20 in 2020. These municipalities are primarily towns and villages. As of 2020, aging populations in these regions often correspond to smaller increases in premiums. For instance, villages and islands in Hokkaido, Yamanashi, and Miyagi prefectures demonstrate low premium growth due to limited nursing care service supply and health maintenance efforts. For example, in a village in Hokkaido, the absence of long-term care facilities necessitates relocation for severely dependent individuals, reducing local costs (Silver Industry Newspaper 2021). In Gunma Prefecture, initiatives such as employing elderly individuals in local hot spring facilities contribute to better health and lower care costs (Sankei News 2018). The Spearman rank correlation coefficient results reinforce that premium increases are primarily influenced by individual circumstances. The significant ranking changes over 20 years for the top 20 municipalities in terms of growth rate suggest these factors are largely municipality-specific.

5.3 Municipalities with Large and Small Rates of Growth in Long-Term Care Insurance Premiums

Table 5 lists the municipalities with the largest and smallest growth rates in long-term care insurance premiums between 2020 and 2040. The municipalities with the largest increases are concentrated in Saitama, Chiba, Ibaraki, and Osaka prefectures, where the population of individuals aged 90 and older is projected to triple by 2040.

The numerator for long-term care insurance premiums (care costs) is expected to double, while the population of individuals aged 65–74 years, (contributing to premium income) is declining. Conversely, municipalities with the smallest growth rates are mostly villages and islands, where population aging is already advanced. Previous studies often attribute disparities in premiums to service provision levels (Ministry of Health and Labour 2017; Taniguchi 2022). However, this section suggests a very strong relationship between the growth rate of premiums and changes in elderly population dynamics.

Furthermore, we analyzed the municipalities with the largest increase in long-term care insurance premiums in Tables 5a, b, and 6. In most of the top 50 municipalities with the highest rates of increase in long-term care insurance premiums, these premiums, as of 2020, were below the national simple average. Thus, even in municipalities in which concerns about long-term care insurance premiums have not been expressed as of 2020, these premiums are expected to increase significantly by 2040. These municipalities are concentrated in Saitama, Chiba, and Ibaraki prefectures, which are commuter areas around Tokyo. These areas developed into Tokyo’s bed-downs in the 1970s, when Japan’s economic growth was undergoing large-scale development. At that time, individuals in their 20s and 30s lived in the area and continued to live there. In 2020, they were in their 70s, and in 2040, they will be in their 90s. These areas are populous, with 25 million people living in the four prefectures. This is equivalent to approximately 20 % of the Japanese population.

Table 5a:

Insurers with the highest rates of increase in long-term care insurance premiums (2020–2040).

Rank	Prefecture	Municipality	Premium increase 2020 → 2040 (%)	Numerator	Denominator	65–69 age (%)	70–74 age (%)	75–79 age (%)	80–84 age (%)	85–89 age (%)	90+ age (%)
1	Saitama	A Town	247 %	261 %	113 %	147 %	95 %	73 %	77 %	143 %	344 %
2	Saitama	B City	217 %	240 %	104 %	106 %	73 %	70 %	93 %	165 %	350 %
3	Saitama	C Town	217 %	273 %	134 %	180 %	106 %	79 %	103 %	186 %	407 %
4	Ibaraki	D Town	215 %	205 %	77 %	74 %	40 %	41 %	83 %	189 %	356 %
5	Saitama	E Town	213 %	218 %	84 %	75 %	45 %	44 %	100 %	188 %	254 %
6	Osaka	F Town	208 %	211 %	80 %	56 %	42 %	54 %	93 %	177 %	330 %
7	Ibaraki	G City	205 %	209 %	96 %	103 %	68 %	63 %	83 %	167 %	370 %
8	Saitama	H City	204 %	231 %	124 %	164 %	103 %	82 %	85 %	161 %	432 %
9	Saitama	I City	204 %	224 %	122 %	160 %	105 %	91 %	83 %	138 %	366 %
10	Mie	J Town	204 %	220 %	92 %	78 %	54 %	82 %	84 %	179 %	287 %
11	Hyogo	K City	204 %	201 %	104 %	129 %	88 %	75 %	74 %	113 %	275 %
12	Tokyo	L Ward	203 %	211 %	117 %	137 %	110 %	108 %	92 %	96 %	192 %
13	Kanagawa	M City	203 %	215 %	109 %	137 %	86 %	68 %	79 %	158 %	386 %
14	Saitama	N Town	202 %	207 %	98 %	106 %	70 %	65 %	80 %	163 %	335 %
15	Saitama	O City	201 %	204 %	99 %	109 %	73 %	66 %	82 %	146 %	303 %
16	Chiba	P City	201 %	228 %	115 %	137 %	89 %	77 %	86 %	154 %	388 %
17	Saitama	Q City	201 %	250 %	114 %	118 %	76 %	73 %	109 %	215 %	431 %
18	Yamaguchi	R Town	201 %	170 %	81 %	82 %	57 %	58 %	84 %	100 %	138 %
19	Osaka	S City	200 %	226 %	118 %	142 %	93 %	88 %	97 %	134 %	264 %
20	Chiba	T City	200 %	218 %	112 %	133 %	90 %	78 %	86 %	142 %	327 %

Table 5b:

Insurers with the lowest rates of increase in long-term care insurance premiums (2020–2040).

Rank	Prefecture	Municipality	Premium increase 2020 → 2040 (%)	Numerator	Denominator	65–69 age (%)	70–74 age (%)	75–79 age (%)	80–84 age (%)	85–89 age (%)	90+ age (%)
1	Tokyo	A Town	79 %	82 %	87 %	50 %	53 %	58 %	91 %	122 %	227 %
2	Yamanashi	B Town	80 %	45 %	65 %	80 %	63 %	79 %	55 %	42 %	79 %
3	Okinawa	C Village	84 %	95 %	104 %	72 %	81 %	119 %	120 %	118 %	267 %
4	Kochi	D Village	85 %	46 %	57 %	33 %	64 %	100 %	94 %	41 %	42 %
5	Tokyo	E Village	91 %	85 %	81 %	39 %	59 %	74 %	89 %	103 %	145 %
6	Kagoshima	F Village	91 %	91 %	97 %	70 %	88 %	141 %	100 %	100 %	145 %
7	Okinawa	G Town	93 %	109 %	111 %	86 %	90 %	140 %	158 %	151 %	101 %
8	Kagoshima	H Village	95 %	92 %	85 %	53 %	59 %	104 %	108 %	121 %	118 %
9	Akita	I Village	98 %	97 %	93 %	67 %	71 %	108 %	126 %	112 %	110 %
10	Hokkaido	J Village	100 %	108 %	101 %	71 %	79 %	113 %	83 %	118 %	161 %
11	Nagano	K Village	100 %	78 %	75 %	53 %	62 %	84 %	79 %	77 %	101 %
12	Nagano	L Village	100 %	71 %	83 %	65 %	107 %	92 %	117 %	56 %	56 %
13	Nagano	M Village	101 %	102 %	89 %	64 %	56 %	101 %	100 %	126 %	138 %
14	Yamagata	N Village	101 %	97 %	81 %	46 %	52 %	95 %	106 %	120 %	138 %
15	Nara	O Village	101 %	70 %	61 %	34 %	38 %	96 %	76 %	80 %	73 %
16	Miyazaki	P Town	102 %	77 %	70 %	47 %	50 %	82 %	86 %	92 %	82 %
17	Kumamoto	Q Village	102 %	141 %	122 %	95 %	78 %	143 %	155 %	172 %	190 %
18	Yamanashi	R Village	102 %	61 %	60 %	30 %	69 %	100 %	63 %	56 %	58 %
19	Tokyo	S Village	102 %	123 %	127 %	148 %	91 %	119 %	179 %	156 %	152 %
20	Niigata	T Town	103 %	92 %	81 %	57 %	56 %	91 %	92 %	89 %	128 %

Table 6:

Long-term care insurance premiums in 2020 for the top 50 insurers with the highest rate of increase in long-term care insurance premiums.

Highest order	Insurer		Percentage increase	2020 Long-term care insurance premiums	Lower than average	Denominator: increase rate of 90 years old and over
1	Saitama	A Town	247 %	4,700	✓	344 %
2	Saitama	B City	217 %	4,837	✓	350 %
3	Saitama	C Town	217 %	4,800	✓	407 %
4	Ibaraki	D Town	215 %	4,650	✓	356 %
5	Saitama	E Town	213 %	4,000	✓	254 %
6	Osaka	F Town	208 %	5,412	✓	330 %
7	Ibaraki	G City	205 %	4,800	✓	370 %
8	Saitama	H City	204 %	4,825	✓	432 %
9	Saitama	I City	204 %	4,971	✓	366 %
10	Mie	J Town	204 %	5,216	✓	287 %
11	Hyogo	K City	204 %	4,690	✓	275 %
12	Tokyo	L Ward	203 %	6,470		192 %
13	Kanagawa	M City	203 %	4,857	✓	386 %
14	Saitama	N Town	202 %	4,880	✓	335 %
15	Saitama	O City	201 %	4,621	✓	303 %
16	Chiba	P City	201 %	5,270	✓	388 %
17	Saitama	Q City	201 %	4,500	✓	431 %
18	Yamaguchi	R Town	201 %	5,481	✓	138 %
19	Osaka	S City	200 %	5,700	✓	264 %
20	Chiba	T City	200 %	5,000	✓	327 %
21	Osaka	U City	199 %	5,360	✓	317 %
22	Miyagi	V Town	199 %	5,400	✓	168 %
23	Saitama	W City	198 %	4,700	✓	375 %
24	Chiba	X City	198 %	4,500	✓	356 %
25	Saitama	Y City	197 %	4,700	✓	373 %
26	Nara	Z Town	197 %	5,186	✓	295 %
27	Chiba	AA Town	196 %	3,900	✓	386 %
28	Ibaraki	AB City	195 %	4,800	✓	365 %
29	Osaka	AC City	195 %	6,210		337 %
30	Hyogo	AD Town	195 %	5,400	✓	283 %
31	Kyoto	AE City	194 %	5,098	✓	321 %
32	Saitama	AF City	194 %	4,501	✓	339 %
33	Osaka	AG City	194 %	5,083	✓	293 %
34	Saitama	AH City	194 %	4,851	✓	364 %
35	Osaka	AI Village	194 %	5,811		294 %
36	Saitama	AJ City	194 %	4,973	✓	340 %
37	Chiba	AK City	193 %	4,700	✓	380 %
38	Saitama	AL City	193 %	4,600	✓	362 %
39	Kyoto	AM City	193 %	5,250	✓	355 %
40	Chiba	AN City	193 %	5,600	✓	342 %
41	Saitama	AO Town	192 %	4,261	✓	382 %
42	Osaka	AP City	192 %	6,211		283 %
43	Saitama	AQ City	192 %	4,950	✓	443 %
44	Chiba	AR City	192 %	4,745	✓	333 %
45	Kyoto	AS Town	192 %	5,591	✓	289 %
46	Saitama	AT City	192 %	4,844	✓	431 %
47	Chiba	AU City	192 %	5,300	✓	317 %
48	Hiroshima	AV Town	191 %	5,696	✓	345 %
49	Aichi	AW City	191 %	4,783	✓	271 %
50	Kanagawa	AX City	191 %	5,800		314 %

6 Conclusions

This study estimated long-term care insurance premiums at the municipal level over the next 20 years and examined disparities in premiums across municipalities. The following key findings were identified: the ratio of maximum to minimum municipal premiums was 267 % in 2000, 290 % in 2020, and is projected to reach 386 % in 2040. This indicates that the gap between municipalities with the highest and lowest long-term care insurance burdens is expected to continue widening. Additionally, the coefficient of variation (standard deviation divided by the national simple average) is projected to increase 1.2-fold between 2020 and 2040, suggesting that disparities among all municipalities – not only between the extremes – will intensify. By 2040, 10 of the top 20 municipalities with the highest long-term care insurance premiums are expected to be concentrated in Osaka Prefecture. Municipalities with the steepest projected increases are concentrated in Tokyo’s commuter areas, including Saitama, Chiba, and Ibaraki prefectures. Many of these municipalities have not expressed significant concerns about long-term care insurance premiums as of 2020, yet premiums are projected to rise considerably over the next 20 years.

The growing disparities in long-term care insurance premiums between municipalities can be attributed to factors such as facility capacity ratios, as identified by Ando (2008), and the uneven distribution of day care and facility services, as highlighted by Taniguchi (2022) and the Ministry of Health, Labour and Welfare (2017). However, the most significant factor appears to be demographic shifts within municipalities, including the increase in the population aged 90 years and older and the decrease in the population aged 65–74 years. These changes are not evenly distributed but are concentrated in specific areas. The large populations in these municipalities make it difficult to formulate measures to address these issues comprehensively. Establishing a system that continuously provides appropriate long-term care services to residents of Tokyo’s commuter areas and Osaka, who are likely to bear the burden of high long-term care insurance premiums in the future, is an urgent administrative issue. Policies for providing outpatient and facility-based services in these regions will become increasingly critical.

Despite this, municipalities may struggle to fundamentally resolve the disparities in long-term care insurance premiums caused by demographic changes. Social insurance encompasses health, long-term care, old-age pension, and unemployment insurance. Since 2018, the management of health insurance has shifted from municipalities to prefectural governments, leading to a trend toward standardizing insurance premiums. The health insurance premiums of the health insurance provider, the Association Health Insurance, are calculated by multiplying the standard monthly remuneration (monthly salary) by the health insurance premium rate, but the most recent health insurance premium rates by prefecture for 2024 are a maximum of 10.42 % and a minimum of 9.35 %, yielding a maximum/minimum value of 111 %. These regional difference are driven by factors such as medical expenses, proportion of elderly people, and income levels. Conversely, old-age pension and unemployment insurance premiums are based on uniform insurance rates nationwide, without consideration of regional differences. With long-term care insurance premiums projected to quadruple by 2040, it is time to reconsider who should oversee their management.

Finally, some limitations should be acknowledged. This study did not analyze the long-term sustainability of Japan’s public long-term care insurance system. Furthermore, we did not compare the taxation frameworks of public long-term care insurance systems in other countries or conduct a comparative study of their sustainability. Future research addressing these issues would offer valuable insights for both domestic and international policy development.

Corresponding author: Tadayoshi Otsuka, Gradutate School of Accountancy, Waseda University, Tokyo, Japan, E-mail: otsukata@waseda.jp

Research ethics: Not applicable.
Informed consent: Not applicable.
Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
Use of Large Language Models, AI and Machine Learning Tools: None Declared.
Conflict of interest: The authors state no conflict of interest.
Research funding: Waseda Univetsity.
Data availability: Not applicable.

Appendix

Confirmation of validity of future estimates of long-term care certification rates

Using the results from 2009 to 2018, we predicted the certification rate of long-term care for the next four years, from 2019 to 2022, and compared it with the results for 2019–2022. Figure A is a graph comparing the actual and forecast long-term care certification rate for 2019–2022 by the level of care required, gender, and age group, with the solid line being the actual and the dotted line being the predicted values. Although the predicted value was larger than the actual value in the relatively young age group of mild long-term care required, such as Support Required 1, Support Required 2, and Long-Term Care Required 1, the proportion of these long-term care costs was low and the impact was minimal. Conversely, since it can be confirmed that the accuracy of predicting the degree of severe long-term care required is high, the accuracy of long-term care insurance premiums as a whole is considered to be sufficient. Furthermore, the accuracy of the model is higher as the forecast for the future of long-term care insurance premiums in this report is based on the results of 14 years, from 2009 to 2022, including the period of 2019–2022 used in this verification.

Figure A:

Support required 1. Support required 2. Nursing care required 1. Nursing care required 2. Nursing care required 3. Nursing care required 4. Nursing care required 5.

Figure A:

(continued)

Figure A:

(continued)

Figure A:

(continued)

References

Ando, Michito. 2008. “An Empirical Analysis of Regional Differences in Provision and Premiums of Long-Term Care Insurance.” The Quarterly of Social Security Research 44 (1): 94–109. [in Japanese].Search in Google Scholar

Brouhns, N., M. Denuit 2002. “A Poisson Log-Bilinear Regression Approach to the Construction of Projected Lifetables." Insurance: Mathematics and Economics 31 (3): 373–93.10.1016/S0167-6687(02)00185-3Search in Google Scholar

Cairns, A. J. G., D. Blake, and K. Dowd. 2006. “A Two-Factor Model for Stochastic Mortality with Parameter Uncertainty: Theory and Calibration.” The Journal of Risk and Insurance 73 (4): 687–718.10.1111/j.1539-6975.2006.00195.xSearch in Google Scholar

Chen, Linhong, Lu Zhang, and Xiaocang Xu. 2020. “Review of Evolution of the Public Long-Term Care Insurance (LTCI) System in Different Countries: Influence and Challenge.” BMC Health Services Research 20 (1): 1–21. https://doi.org/10.1186/s12913-020-05878-z.Search in Google Scholar

Currie, I. D. 2016. “On Fitting Generalized Linear and non-linear Models of Mortality.” Scandinavian Actuarial Journal 2016 (4): 356–83.10.1080/03461238.2014.928230Search in Google Scholar

Fukui, Tadashi. 2016. “Future Estimation of Long-Term Care Insurance Finance by Prefecture.” Journal of Kyoto Sangyo University 33: 61–80. [in Japanese].Search in Google Scholar

Kikuchi, Masahiko. 2019. “Sustainability of the Public long-term Care Insurance System and Differences Between Local Governments.” Journal of Insurance Studies 647: 101–31. [in Japanese].10.5609/jsis.2019.647_101Search in Google Scholar

Lee, R. D., and L. R. Carter. 1992. “Modeling and Forecasting U.S. Mortality.” Journal of the American Statistical Association 87 (419): 659–71.10.1080/01621459.1992.10475265Search in Google Scholar

Masuda, Masanobu. 2014. In Long-Term Care Security in the World, 2nd ed. Location: Houritsubunkasha. [in Japanese].Search in Google Scholar

Ministry of Health, Labour, Welfare. 2017. "Supporting Future Social Security Together with the Local Community." Ministry of Health, Labour and Welfare White Paper for 2017. https://www.mhlw.go.jp/wp/hakusyo/kousei/05/ (accessed January 1, 2025). [in Japanese]Search in Google Scholar

Mitsubishi Research Institute. 2020. “Survey and Research Project Report on the Ideal Long-Term Care Insurance System for 2040.” https://pubpjt.mri.co.jp/pjt_related/roujinhoken/jql43u00000000s3-att/R1_007_2_report.pdf (accessed January 1, 2025). [in Japanese]Search in Google Scholar

National Federation of Health Insurance Societies. 2020. “International Comparative Survey Report on Public Long-Term Care Systems.” https://www.kenporen.com/include/outline/pdf/kaigai_r01_03.pdf (accessed January 1, 2025). [in Japanese].Search in Google Scholar

Otsuka, Tadayoshi, and Yutaka Taniguchi. 2019. “Future Estimation of Healthy Life Expectancy and the Number of Persons Requiring Long-Term Care.” Journal of the Society for Life Economic 49: 91–112. [in Japanese].Search in Google Scholar

Plat, R. 2009. “On Stochastic Mortality Modeling.” Insurance: Mathematics and Economics 45 (3): 393–404.10.1016/j.insmatheco.2009.08.006Search in Google Scholar

Renshaw, A., and S. Haberman. 2006. “A Cohort-Based Extension to the Lee-Carter Model for Mortality Reduction Factors.” Insurance: Mathematics and Economics 38 (3): 556–70.10.1016/j.insmatheco.2005.12.001Search in Google Scholar

Sankei News. 2018. “Long-Term Care Insurance Premium Standard Amount is the Second Lowest in Japan, and the Trend is to Focus on ‘Prevention’.” Sankei News. https://www.sankei.com/article/20180728-NGK6LKUBINLBZAO4AZTS3OTF44/#:∼:text=%E4%BB%8B%E8%AD%B7%E4%BF%9D%E9%99%BA%E6%96%99%E5%9F%BA%E6%BA%96%E9%A1%8D%E3%81%AF%E3%80%81%E4%BB%8B%E8%AD%B7%E3%82%B5%E3%83%BC%E3%83%93%E3%82%B9%E7%B5%A6%E4%BB%98%E8%B2%BB,%E3%81%A7%EF%BC%92%E7%95%AA%E7%9B%AE%E3%81%AB%E4%BD%8E%E3%81%84%E3%80%82 (accessed January 1, 2025). [in Japanese].Search in Google Scholar

Silver Industry Newspaper. 2021. “Efforts to Suppress the Upper Limit on the Amount of Long-Term Care Insurance Premiums Used.” Silver Industry Newspaper. https://www.care-news.jp/news/RFn9m (accessed January 1, 2025). [in Japanese].Search in Google Scholar

Suzuki, Takao, Yukiko Nishita, Seungwon Jeong, Hiroyuki Shimada, Rei Otsuka, Katsunori Kondo, et al.. 2021. “Are Japanese Older Adults Rejuvenating? Changes in Health-Related Measures Among Older Community Dwellers in the Last Decade.” Rejuvenation Research 24 (1): 37–48. https://doi.org/10.1089/rej.2019.2291.Search in Google Scholar

Taniguchi, Yutaka. 2022. “Consideration on the Sustainability of Public Long-Term Care Insurance: An Examination of the ‘Proposal for Fiscal Consolidation’.” Journal of Insurance Science 2022 (658): 21–40. [in Japanese].10.5609/jsis.2022.658_21Search in Google Scholar

Taniguchi, Yutaka, Takafumi Owada, and Tadayoshi Otsuka. 2022. “Analysis of Regional Differences in the Nursing Care Certification Rate by EDA Method.” Journal of the Japanese Association of Risk, Insurance and Pensions 13 (1): 1–25. [in Japanese].Search in Google Scholar

Vanella, Patrizio, Christina Benita Wilke, and Moritz Heß. 2024. “Long-Term Care in Germany in the Context of the Demographic Transition—An Outlook for the Expenses of Long-Term Care Insurance Through 2050.” Econometrics 12 (4): 1–20.10.3390/econometrics12040028Search in Google Scholar

Villegas, Andres M., Pietro Millossovich, and Vladimir K. Kaishev. 2017. “StMoMo: An R Package for Stochastic Mortality Modeling.” https://cran.r-project.org/web/packages/StMoMo/vignettes/StMoMoVignette.pdf (accessed January 1, 2025).Search in Google Scholar

Yang, Wenya, Jamie L. Hamilton, Catherine Kopil, James C. Beck, Caroline M. Tanner, Roger L. Albin, et al.. 2020. “Current and Projected Future Economic Burden of Parkinson’s Disease in the U.S.” npj Parkinson’s Disease 6 (1): 1–9. https://doi.org/10.1038/s41531-020-0117-1.Search in Google Scholar

Received: 2025-04-18

Accepted: 2026-01-08

Published Online: 2026-02-24

This work is licensed under the Creative Commons Attribution 4.0 International License.

https://doi.org/10.1515/apjri-2025-0015

Keywords for this article

long-term care insurance premiums; future estimates; municipal disparities

Creative Commons

BY 4.0