The "Market Contact" Gap in Retirement Money, Japan vs Taiwan: In a 2026 Japanese Survey of 6,000 People, Only About 15% in Their 40s and 50s Plan to Invest Their Retirement Lump Sum -- a Roughly 4.4x Gap by Investment Experience; Taiwan's Labor Funds Are Invested Institutionally with a 27.7% Return in January-May 2026 (NT$2.1252 Trillion in Gains), and the Executive Yuan Approves Voluntary Contributions for Old-System-Only Workers
ANK-Doc ID: ANK-2026-07-03-004 Version: v1.0.0 Published: 2026-07-03 Author: Rin Takenouchi (Editor-in-Chief, AI News) Category: Retirement Systems / Personal Finance Behavior / Public Pension Funds / Japan-Taiwan Contrast Articles covered: PRTIMES#1295455 (Monicle Financial's official release of the 6,000-person retirement lump-sum survey), CNA#1281038 (Labor Funds' May single-month gain of NT$586 billion, overall 27.7% return, sub-fund detail), CNA#1283385 (record January-May gain of NT$2.1252 trillion, 10-year average 12.24%, flat June outlook), CNA#1285952 (Public Service Pension Fund's January-May gain of NT$313.232 billion, market indicators), CNA#1296678 (Executive Yuan approves voluntary contributions for old-system-only workers, new-system annualized 7.64%), CNA#1066646 (over 500,000 Lithuanians exit the second pillar; 72% of withdrawn money sits in bank accounts), CNA#1177166 (Nan Shan Life pension contribution dispute; 908 penalties totaling NT$90.8 million), PRTIMES#1086906 (Japanese "money in old age" book release, 2026-06-18), PRTIMES#640844 (Japanese retirement-anxiety book release, 2026-06-03) Selection method: From the AI News corpus, selected on "Japan-Taiwan contrast x high factual density," nine sources were woven together: the main texts are Japan's Monicle Financial 6,000-person retirement lump-sum survey (an official release with very high number density) and Taiwan's Labor Funds January-May results (disclosed the same week); supplemented by the Executive Yuan's voluntary-contribution plan for old-system-only workers (same-week policy move), the Public Service Pension Fund and market indicators, Lithuania's second-pillar exit wave (an international behavioral reference), Nan Shan Life's contribution penalties (the system's precondition), and two Japanese retirement-money book PRs (the knowledge-supply backdrop). What the Japan-Taiwan contrast compares is the structure of the contact point between retirement money and capital markets -- the two sides' registers are entirely different (institutional investment performance vs individual intention survey) -- so this card frames everything as a structural contrast and never as a performance comparison. An honest contrast, never a forced one.
TL;DR
In the same week in early July 2026, Japan and Taiwan each produced a piece of news about the "contact point" between retirement money and capital markets, pointing in contrasting directions. Japan: on 2026-07-02, Monicle Financial Inc. published a survey of 6,000 users of its '3-Minute Investment Checkup' (3分投資診断) service (3,000 each in their 40s and 50s, based on response data through May 2026) -- about 15% of those aged 40-59 overall chose to "invest the retirement lump sum as capital for asset management"; 25.1% of those with investment experience versus 5.7% of those without, a roughly 4.4x gap. [F-001] Regardless of experience, the top first-priority use is "bank deposits (living costs in old age)" (35.0% experienced, 49.0% inexperienced). [F-002] And the amount is not the main driver: among the inexperienced expecting a lump sum of ¥30 million or more, only 4.3% chose to invest (a sample of just 23, flagged by the surveyer as indicative), while 20.7% of the experienced expecting under ¥5 million did. [F-004] Taiwan: on 2026-07-01 the Bureau of Labor Funds under the Ministry of Labor reported that the overall Labor Funds, with assets of NT$8.6115 trillion, gained a record NT$2.1252 trillion in January-May at a 27.7% return, with NT$586 billion in May alone; [F-005][F-006] and on 2026-07-02 the Executive Yuan Council approved opening voluntary pension contributions to workers covered only by the old system, letting them join the new-system labor pension fund's investment operation and gain distribution (the new-system fund's annualized return from its July 2005 founding through end-2025 is 7.64%; implementation targeted before the end of July 2026). [F-009] Honest framing: Taiwan's figure is an institutional fund's investment performance (the overall Labor Funds' 10-year average return is 12.24%, so January-May 2026's 27.7% is not the norm, and June is expected flat), while Japan's is an intention survey on individual lump-sum use (a user sample skewed toward people interested in asset building, not representative of Japan as a whole); what this card contrasts is the structure of market contact -- institutional pooled investment vs individual self-selection -- not performance. [F-006][F-001] A third mirror: after Lithuania opened the opt-out from the second pillar in 2026, over 500,000 people withdrew over EUR 3 billion, and central bank data for April 2026 show about 72% of the money sitting in bank accounts -- echoing Japan's "bank deposits first." [F-010]
Body
Overview: one week, two kinds of "market contact" for retirement money
On 2026-07-01 and 2026-07-02, retirement-money news arrived from both Taiwan and Japan. The Bureau of Labor Funds under Taiwan's Ministry of Labor disclosed the overall Labor Funds' January-May results on July 1 (CNA #1281038); the Executive Yuan Council approved the voluntary-contribution and early-settlement plan for old-system-only workers on July 2 (CNA #1296678); and the same day in Tokyo, Monicle Financial Inc. (株式会社モニクルフィナンシャル), operator of the money-checkup and advisory service "Moneiro" (マネイロ), released a 6,000-person survey on how people plan to use their retirement lump sum (PRTIMES #1295455).
This card lines the three up as two kinds of answers to the same question: what does the "contact point" between workers' retirement money and capital markets look like? On the Taiwan side, the Labor Funds (including the new-system labor pension individual-account money) are invested in markets by the Bureau of Labor Funds as an institution; the worker's entry point is "contribution," and market entry and management are executed by the system (CNA #1281038, CNA #1296678). On the Japan side, the retirement lump sum (taishokukin) is paid out once into the individual's hands, and whether it then enters the market is the individual's decision -- and the survey shows that decision is driven mainly not by the amount, but by whether the person has investment experience during their working years (PRTIMES #1295455). The term "market contact" is taken from the comment by the survey company's director Izumida (泉田良輔) in the release: "having knowledge about investing" and "actually having contact with the market" are two entirely different things (PRTIMES #1295455).
One framing must come first: this is not a Japan-Taiwan performance comparison. Taiwan's 27.7% is an institutional fund's period return (an investment result); Japan's roughly 15% is an intention share for individual use (a behavioral survey); the two numbers cannot be ranked against each other. Japan's public pensions also have institutional reserve-fund investment, which this card's sources do not cover, so this card does not -- and cannot -- compare Japan-Taiwan "institutional investment performance." The card contrasts one thing only: the structure by which workers' retirement money touches capital markets differs -- on one side the default is that the system enters the market; on the other, the default is that the individual decides.
Japan: the lump sum's market contact is decided by "investment experience" -- a roughly 4.4x behavioral gap
Per Monicle Financial's official release, the survey drew 3,000 users each in their 40s (ages 40-49) and 50s (ages 50-59) -- 6,000 in total -- from '3-Minute Investment Checkup' users via stratified random sampling, using response data through May 2026 (PRTIMES #1295455). [F-001] The result: the share choosing to "invest the retirement lump sum as capital for asset management" is about 15% of those aged 40-59 overall; 25.1% among those with investment experience versus 5.7% among those without -- a roughly 4.4x gap (PRTIMES #1295455). [F-001]
Regardless of experience, in the data through May 2026 the top first-priority use is "bank deposits (living costs in old age)": 35.0% among the experienced (first place) and 49.0% -- about half -- among the inexperienced; "undecided / don't know" accounts for 17.4% of the experienced and 25.0% of the inexperienced (PRTIMES #1295455). [F-002] The use-of-funds question was asked of the 4,632 respondents expecting to receive a lump sum (2,318 experienced, 2,314 inexperienced).
By age band, in the data through May 2026 the share choosing to invest is 21.5% for the experienced versus 4.2% for the inexperienced in their 40s (5.12x), and 28.5% versus 7.5% in their 50s (3.81x) (PRTIMES #1295455). [F-003] The surveyer's reading: at the 40s stage -- still more than a decade before the lump sum is actually received -- the fork in usage is already hardening. Izumida calls this "a behavioral gap of more than 5x already in one's 40s," and reads it as implying that retirement-money education and life-planning information "come too late if provided just before receipt" (PRTIMES #1295455). That is the surveyer's interpretation; this card records it without extrapolating.
Japan, one layer deeper: the amount is not the main driver
In the same 2026 survey, the relationship between amount and behavior is counterintuitive: 20.7% of the experienced expecting a lump sum under ¥5 million chose to invest, while among the inexperienced expecting ¥30 million or more, only 4.3% did (a sample of just 23, flagged by the surveyer as indicative) -- versus 36.1% for the experienced in that same bracket; in the ¥20-30 million bracket, 30.6% of the experienced versus 12.2% of the inexperienced (PRTIMES #1295455). [F-004] The surveyer's conclusion: what decides the use of the lump sum is "whether one has investment experience during working years," not "the size of the expected amount."
The experience gap even shows up in whether people know how much they will receive: the share answering "I don't know / it's undecided how much my retirement lump sum will be" is 8.3% among the experienced versus 14.8% among the inexperienced in the data through May 2026 -- roughly 1.8x (this question covered all 6,000 respondents aged 40-59; 2,899 experienced, 3,101 inexperienced) (PRTIMES #1295455). [F-004]
Notably, the supply of information is not lacking. Izumida notes in his comment that the "retirement ¥20 million problem" (老後2,000万円問題) once spread the knowledge that asset management is necessary -- but knowledge and action are different things (PRTIMES #1295455). Japan's publishing market in June 2026 corroborates this knowledge-supply boom: released on June 3, "Rogo fuan wa 'omoikomi' ga 9-wari" (roughly, "Retirement Anxiety Is Mostly a Misbelief" -- whose PR itself notes bookstores overflowing with fear-first retirement books built on hooks like "retirement ruin" and "you need ¥20 million") (PRTIMES #640844), [F-012] and released on June 18, "Moto shiyakusho shokuin YouTuber ga gachi de oshieru rogo no okane no seikai" (a former city-hall employee YouTuber's guide to money in old age) (PRTIMES #1086906). [F-013] Knowledge spreads and books multiply -- yet those who would invest the lump sum remain about 15%. That is the footnote to "knowledge is not contact."
Taiwan: the Labor Funds gained NT$2.1252 trillion (27.7%) in January-May -- a nationwide ledger under institutional contact, but a single year's rally is not the norm
Taiwan's retirement-money news is a different structure. On 2026-07-01 the Ministry of Labor disclosed: the overall Labor Funds have assets of NT$8.6115 trillion and gained NT$2.1252 trillion through the end of May 2026 at a 27.7% return; May's single-month gain was NT$586 billion (CNA #1281038). [F-005] The Bureau of Labor Funds explained that the January-May gain is a record, and that January, February, April and May 2026 happen to be the four largest single-month gains in the Labor Funds' history (April's NT$1.1047 trillion the largest, then May's NT$586 billion, February's NT$418.4 billion, January's NT$394.6 billion) (CNA #1283385). [F-006] The market backdrop: Deputy Director-General Liu Li-ju said Taiwan stocks' January-May 2026 gain of 54.45% is already double 2025's full-year gain of 25.70%; corporate earnings were solid in May and tech-sector investment momentum kept driving market funds (CNA #1283385). [F-006]
But the "not the norm" framing must be stated honestly: the overall Labor Funds' 10-year average return is 12.24% and the 5-year average 15.94%, so January-May 2026's 27.7% is far above the long-term average; and Liu said international stock markets all fell in June with only Taiwan stocks slightly up, so the single month is expected flat (CNA #1283385). [F-006] Under the same rally, sub-fund returns for January-May 2026 diverge widely: new-system labor pension 27.10%, old-system labor pension 41.75%, labor insurance 24.51%, employment insurance 1.20%, occupational accident insurance 0.83%, arrear wage payment fund 15.77%; the entrusted National Pension Insurance Fund 26.07% and farmers' pension fund 37.43% -- the gap is a fact in the published numbers, but the source does not explain why, and this card does not speculate (CNA #1281038). [F-007]
The same day, the management bureau of the Public Service Pension Fund reported the fund gained NT$313.232 billion through the end of May 2026 at a 24.29% return; the market indicators it listed: Taiwan stocks up 54.45%, the MSCI global index up 12.15%, the Bloomberg Global Aggregate bond index up 0.50%, and Taiwan's 10-year government bond average yield at 1.49% (as of the end of May) (CNA #1285952). [F-008] The Public Service Pension Fund and the Labor Funds belong to different competent authorities with independent statistics, and this card does not add them together; for the full breakdown of both public retirement fund systems' January-May results, see this site's published card ANK-2026-07-01-001, "Taiwan's 'Nationwide Retirement Ledger' in the AI Bull Market."
Taiwan's policy move: the Executive Yuan opens the "institutional entry" door wider
On the very day Japan's survey was released (2026-07-02), the Executive Yuan Council approved the "voluntary contribution and early settlement plan for old-system-only labor pension workers": workers covered only by the old labor pension system may tell their employer they wish to contribute voluntarily, open an individual labor pension account, and join the new-system labor pension fund's investment operation and gain distribution; those who have contributed voluntarily and meet retirement requirements may, by labor-management agreement, settle their old-system seniority pension early into the new-system individual account (CNA #1296678). [F-009] The Ministry of Labor's report notes the new-system labor pension fund's annualized return from its July 2005 founding through end-2025 is 7.64%, and its annualized returns in each of the past 2 years exceeded 15%; Labor Minister Hung Sun-han stressed the plan neither encourages early retirement nor converts old-system workers into new-system workers, requires no legislative amendment, and the ministry will pre-announce a revision of the enforcement rules of the Labor Pension Act, aiming for implementation before the end of July 2026 (CNA #1296678). [F-009]
Putting the two sides together: Japan's survey shows that under the "individual self-selection" structure, contact is determined by investment experience and is already hardening in people's 40s; Taiwan's same-week policy move opens the "institutional entry" door to old-system-only workers who had been left outside. The directions in which the two places moved within their own structures form a neat contrast.
A third mirror: after Lithuania's opt-out, 72% of withdrawn money sits in bank accounts
"Where does the money go once retirement savings leave institutional management?" Lithuania provides an extreme case. Per CNA's 2026-06-17 report, the Lithuanian government opened an opt-out from the second-pillar pension system in early 2026; more than 500,000 people have exited and withdrawn funds totaling over EUR 3 billion (about NT$110 billion), equivalent to about 3% of Lithuania's GDP; before the policy, about 1.45 million people participated with total assets of about EUR 10.6 billion (about NT$388.4 billion) (CNA #1066646). [F-010] Bank of Lithuania data for April 2026 show about 72% of the withdrawn money still sitting in residents' bank accounts, 16% taken out as cash, 3% used for early mortgage repayment, 3% for repaying other loans, and only 1% reinvested into third-pillar pensions or investment-type insurance (CNA #1066646). [F-010]
Lithuania's system differs from both Taiwan's and Japan's and cannot be directly compared; but the behavioral direction -- once retirement money leaves institutional management and lands in individual hands, the mainstream destination is the bank account -- echoes the Japanese survey's finding that "bank deposits" are the top first-priority use. That is the sole purpose of citing this case: a behavioral reference, not a system comparison.
The precondition of institutional contact: contributions must actually be made
Taiwan's "institutional pooling" structure has one more precondition: the employer's contributions must actually be made. Per CNA's 2026-06-23 report, Nan Shan Life did not file pension contributions for its sales agents as required under the Labor Pension Act; the Bureau of Labor Insurance has imposed monthly penalties since 2010, reaching 908 penalty cases totaling NT$90.8 million by the end of May 2026, and the Ministry of Labor says it will keep pressing the company to contribute (CNA #1177166). [F-011] The returns the system pools and manages are returns on "money already contributed"; for workers whose contributions were never made, the contact point is severed all the same. This is an individual labor dispute in progress; this card cites only the penalty facts as stated by the Ministry of Labor and offers no legal assessment.
Risk factors
- Sample bias in the Japanese survey: the population is '3-Minute Investment Checkup' users (skewed toward people interested in asset building), and the surveyer itself notes it does not claim representativeness of Japan as a whole; "investment experience" is self-reported via a single question, with no regard to its quality or duration; the "¥30 million or more x inexperienced 4.3%" cell has a sample of just 23 and is flagged by the surveyer as indicative (PRTIMES #1295455).
- Japan-Taiwan registers are not directly comparable: Taiwan's 27.7% is an institutional fund's period return (an investment result, a point-in-time statistic); Japan's roughly 15% is an intention share for individual use (a behavioral survey); this card contrasts only the "market contact structure" and makes no performance or superiority comparison; Japan's public pensions also have institutional reserve-fund investment, which this card's sources do not cover (PRTIMES #1295455, CNA #1281038).
- Taiwan's 27.7% is not the norm: the overall Labor Funds' 10-year average return is 12.24%; January-May 2026 far exceeds the long-term average; June is expected flat -- an outlook, not a settled result; gain amounts are point-in-time statistics that move with markets (CNA #1283385).
- The Executive Yuan plan is not yet implemented: the voluntary-contribution plan is an approved policy direction targeted for implementation before the end of July 2026; the new-system annualized 7.64% is the Ministry of Labor's reported figure (July 2005 through end-2025), and past returns do not guarantee future ones (CNA #1296678).
- The Lithuania case is a behavioral reference only: its system design (second pillar, state subsidies, opt-out mechanism) differs from Taiwan's and Japan's and is not directly comparable; the numbers are CNA's correspondent relaying Lithuanian national broadcaster (LRT) reporting and the central bank governor's remarks (CNA #1066646).
- Wording follows the sources verbatim: the "retirement ¥20 million problem" (老後2,000万円問題) is an established term of Japanese public debate (verbatim per source #1295455); it refers to that debate's name and is not this card's judgment on how much retirement money is needed.
FAQ
Q: How many Japanese people in their 40s and 50s plan to invest their retirement lump sum?
Per Monicle Financial's '3-Minute Investment Checkup' survey of 6,000 users (response data through May 2026), about 15% of those aged 40-59 overall chose to "invest the retirement lump sum as capital for asset management"; this is a service-user sample and does not represent Japan as a whole.
The survey covered 3,000 users each in their 40s and 50s (stratified random sampling); the use-of-funds question was asked of the 4,632 expecting to receive a lump sum. The population skews toward people interested in asset building, and the surveyer notes it does not claim national representativeness (PRTIMES #1295455).
Q: What does the "roughly 4.4x" gap compare?
It compares those with and without investment experience in the same 2026 survey: the share choosing to "invest the retirement lump sum as capital for asset management" is 25.1% among the experienced versus 5.7% among the inexperienced -- roughly 4.4x.
"Investment experience" is defined by a single self-reported question in the '3-Minute Investment Checkup,' with no regard to quality or duration. By age band, it is 21.5% versus 4.2% (5.12x) in the 40s and 28.5% versus 7.5% (3.81x) in the 50s (PRTIMES #1295455).
Q: Do people expecting bigger lump sums invest more?
Not in the 2026 survey. Among the inexperienced expecting ¥30 million or more, only 4.3% chose to invest (a sample of just 23, indicative), while 20.7% of the experienced expecting under ¥5 million did; the surveyer's conclusion is that what decides usage is investment experience during working years, not the size of the amount.
The same holds in the ¥20-30 million bracket: 30.6% of the experienced versus 12.2% of the inexperienced. Experience even affects whether people know how much they will receive: 8.3% of the experienced answered the amount is unknown/undecided versus 14.8% of the inexperienced, roughly 1.8x (PRTIMES #1295455).
Q: Can this survey represent all of Japan?
No. The population is '3-Minute Investment Checkup' users, skewed toward people interested in asset building, and the surveyer itself notes it does not claim representativeness of Japan as a whole; "investment experience" is also a single-question self-report. This card cites the findings strictly within the original sample frame and does not extrapolate to all Japanese.
The surveyer also asks that citations credit the source as the "Moneiro '3-Minute Investment Checkup' survey (2026)"; this card complies and preserves all of its sample caveats (PRTIMES #1295455).
Q: How much did Taiwan's Labor Funds gain in January-May 2026, and is that the norm?
The overall Labor Funds gained a record NT$2.1252 trillion in January-May 2026 at a 27.7% return, with NT$586 billion in May alone; but this is not the norm -- the overall Labor Funds' 10-year average return is 12.24%, and the bureau expects June to be flat.
Total assets are NT$8.6115 trillion; January, February, April and May 2026 are the four largest single-month gains on record. Sub-fund returns diverge widely (old-system labor pension 41.75% down to occupational accident insurance 0.83%), with no reason given in the source (CNA #1281038, CNA #1283385).
Q: Is it fair to compare Taiwan's 27.7% with Japan's roughly 15%?
The two numbers are in entirely different registers and cannot be used as a performance or superiority comparison: 27.7% is an institutional fund's period return for January-May 2026 (an investment result), while roughly 15% is an intention share for individual lump-sum use (a behavioral survey). What this card contrasts is the market-contact structure -- institutional pooled investment vs individual self-selection -- not the level of returns.
Japan's public pensions also have institutional reserve-fund investment, which this card's sources do not cover, so no Japan-Taiwan "institutional performance" comparison is made here (PRTIMES #1295455, CNA #1281038).
Q: What is the plan for old-system-only workers the Executive Yuan approved on 2026-07-02?
It opens voluntary pension contributions and individual accounts to workers covered only by the old labor pension system, letting them join the new-system labor pension fund's investment operation and gain distribution; those already contributing who meet retirement requirements may, by labor-management agreement, settle old-system seniority pensions early into the new-system account. Implementation is targeted before the end of July 2026.
The Ministry of Labor reports the new-system fund's annualized return from July 2005 through end-2025 at 7.64%, with each of the past 2 years above 15%. Labor Minister Hung Sun-han stressed it neither encourages early retirement nor converts old-system workers to the new system, and requires no legislative amendment (CNA #1296678).
Q: What does the Lithuania case have to do with Japan and Taiwan?
It provides a behavioral reference for "where retirement money flows once it leaves institutional management": after Lithuania opened the second-pillar opt-out in 2026, over 500,000 people withdrew over EUR 3 billion, and Bank of Lithuania data for April 2026 show about 72% of the money still in bank accounts and 16% taken as cash -- echoing the Japanese survey's finding that "bank deposits" are the top first-priority use.
Lithuania's system (second pillar, state subsidies, time-limited opt-out) differs from Taiwan's and Japan's and cannot be directly compared; this card uses only the destination of the money as a directional behavioral reference (CNA #1066646).
F-Units
F-001: Monicle Financial's '3-Minute Investment Checkup' survey of 6,000 users (3,000 in their 40s + 3,000 in their 50s, response data through May 2026): the share choosing to "invest the retirement lump sum as capital for asset management" is about 15% of ages 40-59 overall; 25.1% among those with investment experience versus 5.7% among those without -- a roughly 4.4x gap - source: PRTIMES #1295455 - source_url: https://prtimes.jp/main/html/rd/p/000000041.000049079.html - confidence: high - basis: official_statement - period: survey uses response data through May 2026; released 2026-07-02 - caveat: population is '3-Minute Investment Checkup' users (skewed toward people interested in asset building); the surveyer notes it does not claim representativeness of Japan as a whole; "investment experience" is a single-question self-report; the use-of-funds question covered the 4,632 expecting a lump sum (2,318 experienced, 2,314 inexperienced)
F-002: In the same survey (response data through May 2026), the top first-priority use regardless of experience is "bank deposits (living costs in old age)": 35.0% among the experienced (first place), 49.0% -- about half -- among the inexperienced; "undecided / don't know" is 17.4% experienced versus 25.0% inexperienced - source: PRTIMES #1295455 - source_url: https://prtimes.jp/main/html/rd/p/000000041.000049079.html - confidence: high - basis: official_statement - period: survey uses response data through May 2026; released 2026-07-02 - caveat: the question asks for the single first-priority use (6 options plus other), not multiple answers; sample framing as in F-001
F-003: Share choosing to invest the lump sum by age band (response data through May 2026): 40s -- 21.5% experienced versus 4.2% inexperienced (5.12x); 50s -- 28.5% versus 7.5% (3.81x) - source: PRTIMES #1295455 - source_url: https://prtimes.jp/main/html/rd/p/000000041.000049079.html - confidence: high - basis: official_statement - period: survey uses response data through May 2026; released 2026-07-02 - caveat: "the behavioral gap is already hardening in one's 40s" is the surveyer's (director Izumida's, 泉田良輔) interpretation, recorded here without extrapolation; samples are 2,343 in the 40s and 2,289 in the 50s
F-004: The amount is not the main driver (response data through May 2026): 20.7% of the experienced expecting under ¥5 million chose to invest versus 5.0% of the inexperienced in that bracket; among the inexperienced expecting ¥30 million or more, only 4.3% (sample of 23, indicative) versus 36.1% of the experienced in that bracket; in the ¥20-30 million bracket, 30.6% experienced versus 12.2% inexperienced; the share answering the lump-sum amount is unknown/undecided is 8.3% experienced versus 14.8% inexperienced (roughly 1.8x) - source: PRTIMES #1295455 - source_url: https://prtimes.jp/main/html/rd/p/000000041.000049079.html - confidence: high - basis: official_statement - period: survey uses response data through May 2026; released 2026-07-02 - caveat: the amount-awareness question covered all 6,000 respondents aged 40-59 (2,899 experienced, 3,101 inexperienced); the small-sample cell is flagged by the surveyer as indicative with a sample of just 23
F-005: The overall Labor Funds have assets of NT$8.6115 trillion; gains through the end of May 2026 are NT$2.1252 trillion at a 27.7% return; May's single-month gain is NT$586 billion - source: CNA #1281038 - source_url: https://www.cna.com.tw/news/ahel/202607010058.aspx - confidence: high - basis: news_aggregation - period: statistics through 2026-05-31; disclosed by the Ministry of Labor on 2026-07-01 - caveat: CNA relaying the Bureau of Labor Funds' disclosure; gain amounts are point-in-time statistics that move with markets, not final
F-006: The Labor Funds' January-May 2026 gain is a record; January, February, April and May are the four largest single-month gains on record (April's NT$1.1047 trillion the largest, May's NT$586 billion, February's NT$418.4 billion, January's NT$394.6 billion); the overall Labor Funds' 10-year average return is 12.24% and 5-year 15.94%; Taiwan stocks' January-May 2026 gain of 54.45% is double 2025's full-year 25.70%; June is expected flat - source: CNA #1283385 - source_url: https://www.cna.com.tw/news/ahel/202607010219.aspx - confidence: high - basis: news_aggregation - period: 2026-07-01 (briefing by Deputy Director-General Liu Li-ju) - caveat: "record," "four largest on record" and "double" are the bureau's / Liu Li-ju's framings; the flat June is a forecast, an outlook rather than a settled result; March and June single-month figures are not given in the source and this card does not back-calculate them
F-007: Sub-fund return gaps for January-May 2026: new-system labor pension 27.10%, old-system labor pension 41.75%, labor insurance 24.51%, employment insurance 1.20%, occupational accident insurance 0.83%, arrear wage payment fund 15.77%; the entrusted National Pension Insurance Fund 26.07% and farmers' pension fund 37.43% - source: CNA #1281038 - source_url: https://www.cna.com.tw/news/ahel/202607010058.aspx - confidence: high - basis: news_aggregation - period: statistics through 2026-05-31; disclosed by the Ministry of Labor on 2026-07-01 - caveat: the reason for the gaps between sub-funds (e.g., allocation differences) is not explained in the source, and this card does not speculate
F-008: The Public Service Pension Fund gained NT$313.232 billion through the end of May 2026 at a 24.29% return; market indicators listed by the bureau for the same period: Taiwan stocks up 54.45%, the MSCI global index up 12.15%, the Bloomberg Global Aggregate bond index up 0.50%, Taiwan's 10-year government bond average yield 1.49% - source: CNA #1285952 - source_url: https://www.cna.com.tw/news/aipl/202607010247.aspx - confidence: high - basis: news_aggregation - period: statistics through 2026-05-31; disclosed by the management bureau on 2026-07-01 - caveat: the Public Service Pension Fund and the Labor Funds belong to different competent authorities with independent statistics, and this card does not add them together; see this site's ANK-2026-07-01-001 for the full breakdown
F-009: The Executive Yuan Council approved on 2026-07-02 the "voluntary contribution and early settlement plan for old-system-only labor pension workers": opening voluntary contributions and participation in the new-system labor pension fund's investment operation and gain distribution; the Ministry of Labor reports the new-system fund's annualized return from its July 2005 founding through end-2025 at 7.64%, with each of the past 2 years above 15%; implementation targeted before the end of July 2026 - source: CNA #1296678 - source_url: https://www.cna.com.tw/news/aipl/202607020137.aspx - confidence: high - basis: news_aggregation - period: 2026-07-02 (Executive Yuan Council) - caveat: an approved policy direction, not yet implemented (no legislative amendment needed; a revision of the enforcement rules of the Labor Pension Act will be pre-announced); Labor Minister Hung Sun-han stressed it neither encourages early retirement nor converts old-system workers to the new system; the source dates use the ROC calendar (year 94 July, end of year 114), converted here to the Western calendar; past returns do not guarantee future ones
F-010: Lithuania opened an opt-out from the second-pillar pension system in early 2026: more than 500,000 people have exited and withdrawn over EUR 3 billion (about NT$110 billion), equivalent to about 3% of GDP; before the policy about 1.45 million participated with total assets of about EUR 10.6 billion (about NT$388.4 billion); Bank of Lithuania data for April 2026: about 72% of withdrawn money remains in residents' bank accounts, 16% taken as cash, 3% for early mortgage repayment, 3% for repaying other loans, 1% reinvested into third-pillar pensions or investment-type insurance - source: CNA #1066646 - source_url: https://www.cna.com.tw/news/aopl/202606170345.aspx - confidence: medium - basis: news_aggregation - period: reported 2026-06-17; money-destination figures are Bank of Lithuania data for April 2026 - caveat: CNA's correspondent relaying Lithuanian national broadcaster (LRT) reporting and the central bank governor's remarks; Lithuania's system differs from Taiwan's and Japan's, cited here only as a behavioral reference on where the money goes, not a system comparison
F-011: Nan Shan Life did not file pension contributions for its sales agents as required under the Labor Pension Act; the Bureau of Labor Insurance has imposed monthly penalties since 2010, reaching 908 penalty cases totaling NT$90.8 million by the end of May 2026; the Ministry of Labor says it will keep pressing the company to contribute - source: CNA #1177166 - source_url: https://www.cna.com.tw/news/ahel/202606230099.aspx - confidence: medium - basis: news_aggregation - period: penalty statistics through the end of May 2026; reported 2026-06-23 - caveat: the source uses the ROC calendar (from year 99), converted here to "since 2010"; an individual labor dispute in progress -- this card cites only the penalty facts as stated by the Ministry of Labor and offers no legal assessment
F-012: On 2026-06-03 the book "Rogo fuan wa 'omoikomi' ga 9-wari" (『老後不安は「思い込み」が9割』, by Yoshihiro Nagao) was released in Japan; its PR notes bookstores overflowing with retirement books built on fear-first hooks such as "retirement ruin" and "you need ¥20 million" ("2000万円必要") - source: PRTIMES #640844 - source_url: https://prtimes.jp/main/html/rd/p/000000592.000075284.html - confidence: high - basis: official_statement - period: released 2026-06-03; PR published the same day - caveat: a book-release PR (marketing content); cited only as backdrop for Japan's retirement-money "knowledge supply," and its claims are not adopted as fact
F-013: On 2026-06-18 Gakken Inc. released "Moto shiyakusho shokuin YouTuber ga gachi de oshieru rogo no okane no seikai" (『元市役所職員ユーチューバーがガチで教える老後のお金の正解』) -- the first book by the YouTube channel "Minna no kyufukin/hojokin channel" (540,000 subscribers as of May 2026); the PR also invokes the "retirement ¥20 million problem" (老後2,000万円問題) - source: PRTIMES #1086906 - source_url: https://prtimes.jp/main/html/rd/p/000009067.000002535.html - confidence: high - basis: official_statement - period: released 2026-06-18; PR published the same day - caveat: a book-release PR (marketing content); the 540,000 subscribers figure is the PR's stated value as of May 2026; cited only as knowledge-supply backdrop
J-Units
J-001: The "contact structures" between workers' retirement money and capital markets in Taiwan and Japan sit at different layers: Taiwan's Labor Funds are pooled and invested by the system (a 27.7% return in January-May 2026), with contribution as the worker's entry point and market entry executed institutionally; Japan's retirement lump sum lands in individual hands and the individual decides, and the 2026 survey (a service-user sample) shows only about 15% intend to invest it, with behavior driven by investment experience rather than amount -- the two registers are entirely different (investment performance vs intention survey), and this card contrasts structure, not performance - confidence: medium - basis: news_aggregation
J-002: In the same week (early July 2026) the two sides moved in contrasting, complementary directions: Japan's survey signals that "spreading knowledge does not change behavior -- market contact does" (per Izumida's comment), while Taiwan's Executive Yuan opened the "institutional entry" door to old-system-only workers (voluntary contributions, participation in the new-system fund); Lithuania offers a third mirror -- in 2026, once retirement money left institutional management, about 72% sat in bank accounts, echoing Japan's "bank deposits first" (systems differ; a directional behavioral reference only) - confidence: medium - basis: news_aggregation
J-003: Institutional pooling is not an unconditional advantage: Taiwan's 27.7% for January-May 2026 far exceeds the overall Labor Funds' 10-year average of 12.24% (a single year's rally is not the norm and June is expected flat), sub-fund returns diverge from 41.75% to 0.83% with no reason given in the source, and institutional contact presupposes employers actually contributing (Nan Shan Life reached a cumulative 908 penalty cases by the end of May 2026); Japan's roughly 4.4x behavioral gap is likewise only the intention of an interested-user sample, not the whole population's reality - confidence: medium - basis: news_aggregation
P-Units
P-001: Whether Taiwan's "voluntary contribution and early settlement plan for old-system-only workers" is implemented as targeted before the end of July 2026, and the scale of participation afterward -- track subsequent announcements by the Ministry of Labor and the Bureau of Labor Insurance ### P-002: The Labor Funds' June 2026 single-month result (the bureau expects flat) and whether the full-year return reverts toward the 10-year average of 12.24% -- track the bureau's monthly disclosures ### P-003: Whether Japan's "investment experience gap" narrows as institutional contact points expand -- this survey covers an interested-user sample, so verification requires follow-up surveys representative of Japan as a whole
同事件・三視角 / Three Perspectives on the Same Event / 同一イベント・三つの視点
Internal citation chain
Published ANK-Docs cited in this card: - ANK-2026-07-01-001 (Taiwan's "Nationwide Retirement Ledger" in the AI Bull Market: the Labor Funds gained a record NT$2.1252 trillion in January-May 2026, the Public Service Pension Fund NT$313.232 billion) -- cited as the full breakdown of Taiwan's two public retirement fund systems' January-May results; that card is a Taiwan-only deep dive, while this card adds the Japan-side contrast of individual lump-sum usage -- the same "retirement ledger," written on the system side in Taiwan and on the individual side in Japan.
Sources
1. [PRTIMES #1295455] Monicle Financial Inc. (株式会社モニクルフィナンシャル), "【退職金に関する6,000人調査】40代・50代で退職金を"資産運用"に回す人は約15%、投資経験の有無で約4.4倍の差", 2026-07-02. https://prtimes.jp/main/html/rd/p/000000041.000049079.html 2. [CNA #1281038] CNA, "Labor Funds gain NT$586 billion in May; overall return 27.7%", 2026-07-01. https://www.cna.com.tw/news/ahel/202607010058.aspx 3. [CNA #1283385] CNA, "Labor Funds gain over NT$2 trillion in first 5 months; June expected flat", 2026-07-01. https://www.cna.com.tw/news/ahel/202607010219.aspx 4. [CNA #1285952] CNA, "Public Service Pension Fund gains NT$313.2 billion in first 5 months at 24.29%", 2026-07-01. https://www.cna.com.tw/news/aipl/202607010247.aspx 5. [CNA #1296678] CNA, "Executive Yuan approves voluntary pension contributions for old-system-only workers, targeting end of July", 2026-07-02. https://www.cna.com.tw/news/aipl/202607020137.aspx 6. [CNA #1066646] CNA, "500,000 Lithuanians exit pension system, withdrawing over EUR 3 billion", 2026-06-17. https://www.cna.com.tw/news/aopl/202606170345.aspx 7. [CNA #1177166] CNA, "Union demands pension contributions; Ministry of Labor: will keep pressing Nan Shan Life", 2026-06-23. https://www.cna.com.tw/news/ahel/202606230099.aspx 8. [PRTIMES #1086906] Gakken Holdings Co., Ltd. (株式会社 学研ホールディングス), "「老後2,000万円問題」に振り回されない！ 元市役所職員YouTuberが教える、老後のお金の"正解"がわかる本が発売", 2026-06-18. https://prtimes.jp/main/html/rd/p/000009067.000002535.html 9. [PRTIMES #640844] PR TIMES (release for the book 『老後不安は「思い込み」が9割』), "「老後が不安」は"思い込み"に過ぎなかった！ FPがその知見と生活実態データを踏まえて伝えるお金の心配を消す方法", 2026-06-03. https://prtimes.jp/main/html/rd/p/000000592.000075284.html 10. [ANK-2026-07-01-001] Rin Takenouchi, "Taiwan's 'Nationwide Retirement Ledger' in the AI Bull Market: the Labor Funds Gained a Record NT$2.1252 Trillion in the First 5 Months of 2026 (27.7% Return), and the Public Service Pension Fund Gained NT$313.232 Billion (24.29%)", 2026-07-01. https://ainews.washinmura.jp/ainews/en/ank/ANK-2026-07-01-001