The Taiwan-Japan Retail-Money "Temperature Gap" — A Survey of 250 Wealthy Japanese: 36.8% of Respondents Keep 50% or More in Cash and Rank "Preservation" First at 34.0%; Japan's NISA Accounts Topped 28 Million at End-2025 (23% of the Population, per a Rakuten Securities Release); Contrast with Taiwan's Record 14,332,896 Cumulative Brokerage Accounts and NT$32.579 Billion Single-Month Regular-Savings Investment in May 2026 (Citing This Site's Published Cards)
ANK-Doc ID: ANK-2026-07-03-007 Version: v1.0.0 Published: 2026-07-03 Author: Rin Takenouchi (Editor-in-Chief, AI News) Category: Personal Finance / Asset Management / NISA / Taiwan-Japan Contrast Articles covered: PRTIMES#1283286 (LIS LLC, survey of 250 wealthy Japanese with net financial assets of ¥100 million or more), PRTIMES#1282501 (Rakuten Securities: third straight overall No. 1 in the NISA brokerage satisfaction ranking, plus the cited 28-million-plus NISA account count), PRTIMES#1282493 (Oricon: NISA brokerage/bank usage data from 7,545 users), PRTIMES#1282495 (Oricon: 2026 NISA brokerage/bank satisfaction rankings), PRTIMES#1295068 (Clabo Inc.: portfolio-strategy survey of 992 crypto-asset investors), PRTIMES#1295455 (Monicle Financial: retirement-money survey of 6,000 people) Selection method: Selected from the AI News corpus on "same-week high factual density x Taiwan-Japan contrast." The anchor article is LIS LLC's survey of 250 wealthy individuals (the defensive end's dense sample numbers: cash ratios, preservation orientation, the crypto generational fault line, the offshore-asset polarization). This is chained to the same week's (July 1-2, 2026) releases on Japan's mass-market layer — the Rakuten Securities release citing 28 million-plus NISA accounts and 23% of the population, Oricon's 7,545-user usage data and satisfaction rankings, Clabo's diversified portfolios of 992 crypto-experienced investors, and Monicle Financial's 6,000-person retirement survey showing a 4.4x experience gap — to assemble a layered thermometer of Japanese retail money. The Taiwan side honestly contrasts via this site's published cards (ANK-2026-06-01-001 on the nationwide rush into the market; ANK-2026-06-16-001 on the great retail-money migration). Weak links in the pack were cut: the JDN wealthy-inbound-travel consortium (consumption, not asset management), Kansai Mirai Bank's single-region award, and the expert-panel ranking (duplicative of the user-satisfaction ranking).
TL;DR
A cluster of surveys and statistics released in late June and early July 2026 sketches the "temperature gap" in Taiwan-Japan retail money. Japan's "defensive" end: per an online survey released on July 1, 2026 by LIS LLC (合同会社LIS) of 250 individuals with net financial assets of ¥100 million or more (conducted June 22-23, 2026), 36.8% of respondents keep 50% or more of their financial assets in cash and deposits, and 24.0% keep 80% or more (the source also phrases this as "one in four"); the top priority in asset management is "not losing assets (preservation)" at 34.0%, followed by "preparing for inflation and yen depreciation" at 19.2%. [F-001][F-002] Current crypto-asset ownership splits sharply by generation: 51.6% in their 30s and 40.0% in their 20s (n=20, reference value) versus 6.6% for those 60 and over. [F-006] Japan's "entering" end: a Rakuten Securities release of July 1, 2026 cites that Japan's domestic NISA accounts topped 28 million at end-2025, equivalent to 23% of the total population; per Oricon's 2026 usage survey of 7,545 NISA users, the top reason for starting to invest is "putting idle cash to work" (65.6% for brokerages, 64.9% for banks), and the most common monthly regular-contribution band is "¥10,000 to under ¥30,000." [F-009][F-010] Experience forms a second fault line: Monicle Financial's survey of 6,000 service users in their 40s and 50s finds 25.1% of experienced investors would put retirement money into investment versus 5.7% of the inexperienced — a roughly 4.4x gap. [F-013] Taiwan contrast (citing this site's published cards): Taiwan's cumulative brokerage accounts hit a record 14,332,896 in May 2026, single-month regular-savings investment reached NT$32.579 billion, and those aged 30 and under made up 51.8% of new accounts — a nationwide rush into the market shown by official statistics. But the Japan side is sample surveys and issuer releases while the Taiwan side is official whole-population statistics; the populations and yardsticks differ, so this card juxtaposes rather than compares on a single scale.
Body
Overview: two thermometers in the same week
This card consists of two thermometers. The first is inserted into the "layered structure" of Japanese retail money: at the top, LIS LLC's survey of 250 wealthy individuals with net financial assets of ¥100 million or more (released July 1, 2026); at the mass-market layer, the same week's releases — the cited NISA account count, Oricon's 7,545-user usage data, Clabo's survey of 992 crypto-experienced investors, and Monicle Financial's 6,000-person retirement survey. The second thermometer is inserted into Taiwan — citing this site's published ANK-2026-06-01-001 (the retail-investor phenomenon of Taiwan's AI stock boom) and ANK-2026-06-16-001 (the great retail-money migration), which present a nationwide rush into the market via official statistics. Yardstick declaration first: the Japan side is sample surveys (each with population limits) and issuer releases, while the Taiwan side rests on official whole-population statistics (the stock exchange, the central bank, the Financial Supervisory Commission) underlying this site's published cards. Populations, statistical natures and time points all differ — this card performs a juxtaposed observation of a "temperature gap," not a single-scale comparison, and passes no verdict of superiority.
The "defense" of wealthy Japan: 36.8% keep half or more in cash, "preservation" leads at 34.0%
Per the survey released on July 1, 2026 by LIS LLC (Minato-ku, Tokyo): among 250 men and women aged 20 or over with net financial assets of ¥100 million or more (online questionnaire with screening plus main survey, conducted June 22-23, 2026), respondents whose cash-and-deposit share of financial assets (including idle standby funds) is 50% or more account for 36.8% of the total; those at 80% or more account for 24.0% — which the source also phrases as "one in four" (PRTIMES #1283286). [F-001] Notation note: within the same release, the summary section says "exceeding 50%" while the detail section says "50% or more"; this card adopts the detail section's wording and flags the variance.
The source attributes the high cash ratios to a "preservation orientation": the top priority in asset management among the surveyed wealthy is "not losing assets (preservation)" at 34.0%, with "preparing for inflation and yen depreciation" second at 19.2% (PRTIMES #1283286). [F-002] As background, the release cites a Nomura Research Institute (野村総合研究所) estimate: as of 2023, Japan's wealthy and ultra-wealthy households totaled 1.653 million, up 11.3% from 2021 — a third-party estimate cited by the LIS release, not a figure from this survey. [F-002]
The baseline allocation: domestic and traditional assets, with crypto at 33.6% above overseas real estate
Shares answering that they hold "1% or more" of each asset class (excluding 0% answers): cash and deposits 98.0% (the source says "almost everyone"), domestic equities 85.6%, crypto assets 33.6%, alternative investments 30.0%, overseas real estate 26.0% (PRTIMES #1283286). [F-003] The source draws two structural points: first, a domestic-centered, traditional-asset-centered allocation is the baseline for the surveyed wealthy; second, the breadth of crypto ownership (33.6%) exceeding overseas real estate (26.0%) and alternatives (30.0%) is a distinguishing feature.
On physical assets: precious metals such as gold and platinum 44.8%, luxury watches and jewelry 42.8%, art and antiques 24.4%, wine and whisky and the like 23.6%, antique coins 16.0%; 31.6% hold none of these. Among owner-executives, luxury watch and jewelry ownership rises to 56.45%, far above the overall average (PRTIMES #1283286). [F-004]
Offshore assets are polarized: on the share of financial assets held overseas (overseas accounts, overseas private banks, foreign-currency products and the like), 0% (domestic only) is the most common answer at 43.2%, yet 21.2% keep 30% or more offshore. By age, the 0% answers run 63.16% for those in their 50s and 60.66% for 60 and over — the older the respondent, the more domestically concentrated. The source adds that among those in their 30s and 40s, a "certain number" report offshore ratios of 80% or more (not quantified in the source; this card does not fill in numbers) (PRTIMES #1283286). [F-005]
The generational and positional fault lines: crypto at 51.6% versus 6.6%, "borrow, don't sell" known but unused by 33.2%
Current ownership of crypto assets (Bitcoin, Ethereum and the like) stands at 21.6% of all respondents, rising to 26.4% including past ownership. The generational fault line is stark: 40.0% for the 20s (n=20, a reference value the source flags for its small count), 51.6% for the 30s (n=31), 30.8% for the 40s (n=39), 26.3% for the 50s (n=38), and 6.6% for 60 and over (n=122). By occupation: regular-employee company workers 37.4% versus owner-executives 6.5%. Owner-executives are conservative across other assets too — 80.7% hold 0% in alternatives and 83.9% hold 0% in overseas real estate (PRTIMES #1283286). [F-006]
The "borrow, don't sell" approach — securities-backed loans and crypto-backed loans that raise funds against holdings without selling — shows this awareness structure: "know the mechanism and have used it" 18.8%, "know the mechanism but have never used it" 33.2% (the most common), "did not know" 24.4%. The generational and positional gaps recur: usage experience is 48.39% for the 30s versus 6.56% for 60 and over; owner-executives' usage runs 11.29%, below the overall average (another passage of the same release writes "11.3%" against an overall average of "18.8%"; the notations differ slightly and this card records both) (PRTIMES #1283286). [F-007] The source also cites the National Tax Agency (国税庁) Tax Answer No. 1524 (crypto assets are in principle taxed at the point of sale or use) and links it to the interest signaled by the 33.2% who know of such loans but have not used them — this is the LIS release's own argument and citation; this card reproduces it as-is and makes no tax or legal judgment whatsoever.
Who advises the wealthy? Respondents' current main counterparts: bank (main-bank) representatives lead at 32.0%, brokerage representatives follow at 30.4%, while "consulting no one (deciding alone)" also reaches 31.2%. Satisfaction with current proposals from institutions and professionals: "satisfied" 16.4% and "somewhat satisfied" 26.8%, against "none of these apply" 24.0% and "cannot say either way" 26.4%; outright dissatisfaction ("somewhat dissatisfied" 3.6% plus "dissatisfied" 2.8% = 6.4%) is comparatively rare — the source reads this as a largely passive state of "no settled adviser, no proactive proposals received" (PRTIMES #1283286). [F-008]
The mass layer's "entry": 28 million NISA accounts, motivated by "putting idle cash to work"
Turn the lens to Japan's mass-market layer. In its release of July 1, 2026, Rakuten Securities, Inc. (楽天証券株式会社) cites: at end-2025, Japan's domestic NISA accounts topped 28 million, with 23% of the total population holding a NISA account; of these, more than 7 million people (about one quarter) hold their NISA account at Rakuten Securities. The claims of having the most general brokerage accounts of any single domestic securities firm (disclosed basis) and the industry's largest NISA account count are the issuer's own asterisked wording (the footnote text is not included in the excerpt this card relies on) (PRTIMES #1282501). [F-009]
Motives and money levels come from usage data Oricon Inc. (オリコン株式会社) published on July 1, 2026, gathered separately from the 7,545 respondents of its 2026 satisfaction survey: the reason for starting NISA investing (multiple answers) is led by "putting idle cash to work" — 65.6% at brokerages and 64.9% at banks — followed by "saving for old age" in the 50-percent range and "saving toward a comfortable life" in the 30-percent range. Reasons for choosing the current NISA account: "low fees" leads at brokerages, "already using other services there" at banks. Growth-quota usage: "both regular contributions and lump sums" leads at brokerages, "regular contributions only" at banks. Average monthly amounts: "¥10,000 to under ¥30,000" leads for regular contributions and "¥100,000 or more" for lump sums (PRTIMES #1282493). [F-010]
The competitive map follows Oricon's 2026 satisfaction rankings (published July 1, 2026): overall No. 1 among NISA brokerages is Rakuten Securities — three straight years since the survey began in 2024; SBI Securities took "product lineup" for the third straight year. Overall No. 1 among NISA banks is PayPay Bank (PayPay銀行) — its first overall win, ranked first in every evaluated category including "account opening" and "ease of management"; Japan Post Bank took the "beginner investors" level for the second straight year; the regional rankings debuted with the 2026 edition (PRTIMES #1282495, #1282501). [F-011]
Discipline and experience: only 7.9% of crypto investors go all-in; retirement-money investing splits 25.1% versus 5.7%
Where crypto sits in Japanese retail portfolios comes from a survey Clabo Inc. released on July 2, 2026 covering 992 domestic crypto-experienced investors: the "all-in on crypto" camp is 7.9%; the source states that about 95% of investors practice diversification alongside other assets such as the new NISA and equities (the source presents "7.9%" and "about 95%" side by side without explaining the gap between the two yardsticks; this card reproduces both without reconciling them). The most common crypto share of the portfolio is "a sub-allocation of 30% or less" at 38.8%; the most common pairing is "investment trusts/NISA" at a 22.4% implementation rate (PRTIMES #1295068). [F-012]
The experience gap leaves its starkest numbers on retirement money. Monicle Financial (株式会社モニクルフィナンシャル), operator of the money-advice service Moneiro, published on July 2, 2026: analyzing 6,000 users randomly drawn from its "3-minute investment diagnosis" service — 3,000 each in their 40s (ages 40-49) and 50s (ages 50-59) — those choosing "invest it as seed money for asset management" as the use of their retirement money split 25.1% among experienced investors versus 5.7% among the inexperienced, a roughly 4.4x gap; across ages 40-59 overall, about 15%. Even among those expecting retirement payouts of ¥30 million or more, just 4.3% of the inexperienced chose to invest it (n=23, which the source flags as a small-sample guide value); the experienced chose to invest at 20.7% even when expecting under ¥5 million. Even awareness of the amount differs: "don't know / undecided" was answered by 8.3% of the experienced versus 14.8% of the inexperienced — about 1.8x more (PRTIMES #1295455). [F-013] The source's conclusion: what decides the use of retirement money is working-life investment experience rather than the amount itself — this is that survey's reading of its own sample; this card does not extrapolate it to the general population.
The Taiwan contrast: a nationwide rush into the market in official statistics (via this site's published cards)
The Taiwan side cites two of this site's published cards. ANK-2026-06-01-001 (the retail-investor phenomenon of Taiwan's AI stock boom) records, per Taiwan Stock Exchange statistics: Taiwan's cumulative brokerage accounts reached a record 14,332,896 in May 2026; single-month regular-savings investment hit NT$32.579 billion, breaking the NT$30 billion mark; and those aged 30 and under made up 51.8% of May's new accounts. The same card records the flip side per Financial Supervisory Commission data: the share of natural persons in centralized-market turnover fell over five years from 58% in 2022 to 53.8% in Q1 2026 — national frenzy and share recession coexist. ANK-2026-06-16-001 (the great retail-money migration) records: Taiwan's active ETFs blew past NT$902.2 billion within a year (with Uni-President assets running a 59% share), while on Japan's NISA side "NISA poverty" surfaced — per the BRITA survey (580 people) cited in that card, 52.5% of NISA users said belt-tightening had lowered their quality of life.
Where is the "temperature gap"? Japan's top wealthy layer puts preservation first (36.8% of respondents keep half or more in cash); the mass layer enters gradually via NISA (28 million accounts; the most common contribution band ¥10,000 to under ¥30,000 a month). Taiwan shows whole-population penetration in official statistics — 14,332,896 cumulative accounts, NT$32.579 billion of regular-savings investment in a single month, and the under-30s at more than half of new accounts. The direction is the same (Japan's "from savings to investment" and Taiwan's nationwide entry are both retail money moving into markets); the temperature and structure differ. Three honest boundary lines: first, Japan's wealthy survey (a 250-person sample) and Taiwan's whole-population account statistics are entirely different populations and statistical natures, and cannot be divided into each other; second, Japan's "¥10,000 to under ¥30,000" is a per-person modal band while Taiwan's "NT$32.579 billion" is a whole-market single-month total — different unit levels, not equivalent figures; third, each side has its own flip side — Taiwan's share recession and active-ETF concentration (recorded in this site's published cards), Japan's "NISA poverty" and household resilience (recorded in ANK-2026-06-16-001). A temperature gap does not mean one side is better.
Risk factors
- Sample and population limits (the LIS survey): an online questionnaire of 250 people, not a census; 122 of the 250 people are 60 or over (an age-skewed composition) and only 20 people are in their 20s (a reference value the source itself flags). Every ratio is a share "among surveyed wealthy respondents" and must not be scaled up to all wealthy Japanese (the 1.653 million households figure is a separate Nomura Research Institute estimate) (PRTIMES #1283286).
- Issuer positioning: LIS operates the ultra-wealthy asset-management media "AFFLUENT THEORY," and the release contains explainer links on securities-backed and crypto-backed loans, so mind the issuer's position when using the numbers; Rakuten Securities' award release is a first-party release; Oricon's satisfaction rankings are a commercial survey (PRTIMES #1283286, #1282501, #1282495).
- Notation variance and yardstick gaps: within the same LIS release, "exceeding 50%" coexists with "50% or more," and owner-executives' loan usage appears as both "11.3%" and "11.29%"; Clabo's "all-in 7.9%" and "about 95% diversified" carry an unexplained gap between yardsticks — this card records all as-is without reconciling (PRTIMES #1283286, #1295068).
- Tax citations not extended: the National Tax Agency Tax Answer No. 1524 taxation-timing explanation is a citation within the LIS release (paired there with the 33.2% know-but-unused share); this card makes no tax or legal judgment (PRTIMES #1283286).
- Population bias (Clabo, Monicle): Clabo's subjects are 992 people with crypto-asset investment experience (not a general-investor population); Monicle's subjects are users of its "3-minute investment diagnosis" (possible self-selection bias toward the finance-minded; the source states random sampling from among users) (PRTIMES #1295068, #1295455).
- Taiwan and Japan cannot be compared on one scale: the Japan side is sample surveys and issuer releases, the Taiwan side official whole-population statistics; "23% of the total population" is wording cited within the Rakuten Securities release — this card juxtaposes the temperature gap without division or ranking.
FAQ
Q: What is the core finding of the Japanese wealthy survey?
Per LIS LLC's July 1, 2026 release (250 people with net financial assets of ¥100 million or more, online survey of June 22-23, 2026): 36.8% of respondents keep 50% or more in cash and deposits and 24.0% keep 80% or more; the top asset-management priority is "not losing assets (preservation)" at 34.0%, with "preparing for inflation and yen depreciation" second at 19.2% — a "defensive" temperature is the survey's main axis.
The baseline allocation is domestic and traditional: cash and deposits 98.0% and domestic equities 85.6% held (at 1% or more, excluding 0% answers), against overseas real estate 26.0% and alternatives 30.0% (PRTIMES #1283286).
Q: Can this 250-person survey represent all wealthy Japanese?
No. It is an online sample survey (250 people, including screening), and every ratio is a share "among respondents" only; the sample skews older (122 aged 60 or over) with just 20 in their 20s (a reference value the source itself flags). For the population size of Japan's wealthy, the LIS release cites a Nomura Research Institute estimate: 1.653 million wealthy and ultra-wealthy households as of 2023, up 11.3% from 2021.
Issuer positioning also warrants care: LIS operates the wealthy-focused asset-management media "AFFLUENT THEORY," and the release contains explainer links on collateral loans (PRTIMES #1283286).
Q: How widely do wealthy Japanese hold crypto assets? How large is the generational gap?
Among surveyed wealthy respondents, current crypto ownership is 21.6%, or 26.4% including past ownership; the generational fault line is stark — 51.6% for the 30s, 40.0% for the 20s (n=20, reference value), 30.8% for the 40s and 26.3% for the 50s, versus just 6.6% for 60 and over. Occupation splits too: regular-employee company workers 37.4% versus owner-executives 6.5%.
On breadth of ownership, crypto at 33.6% (held at 1% or more, excluding 0% answers) exceeds overseas real estate at 26.0% and alternatives at 30.0% — the feature the source highlights (PRTIMES #1283286).
Q: What is "borrow, don't sell," and how well known is it?
It refers to securities-backed loans and crypto-backed loans that raise funds against held securities or crypto assets without selling them. The awareness structure: "know it and have used it" 18.8%, "know it but never used it" 33.2% (most common), "did not know" 24.4%; usage experience runs 48.39% for the 30s versus 6.56% for 60 and over, and owner-executives at 11.29% sit below the overall average.
The source cites National Tax Agency Tax Answer No. 1524 (crypto assets are in principle taxed at the point of sale or use), pairing it with the 33.2% know-but-unused share — a citation within the release; this card makes no tax or legal judgment (PRTIMES #1283286).
Q: How far has NISA penetrated Japan's mass layer, and what motivates entry?
A Rakuten Securities release of July 1, 2026 cites: at end-2025, Japan's domestic NISA accounts topped 28 million, equivalent to 23% of the total population; more than 7 million people (about one quarter) hold theirs at Rakuten Securities. Per Oricon's 7,545-user usage survey (2026): the top reason for starting is "putting idle cash to work" (65.6% brokerage, 64.9% bank), and the most common monthly amounts are "¥10,000 to under ¥30,000" for regular contributions and "¥100,000 or more" for lump sums.
"23% of the total population" and "industry No. 1" are the wording of the Rakuten Securities release (asterisked; footnote text not in the excerpt) (PRTIMES #1282501, #1282493).
Q: What does the retirement-money survey show?
Per Monicle Financial's July 2, 2026 release (6,000 users of its "3-minute investment diagnosis," 3,000 each in their 40s and 50s): choosing to "invest retirement money as seed capital" splits 25.1% among experienced investors versus 5.7% among the inexperienced — a roughly 4.4x gap; about 15% across ages 40-59 overall. Even inexperienced respondents expecting ¥30 million or more chose investing at just 4.3% (n=23, guide value), while the experienced chose it at 20.7% even when expecting under ¥5 million — the source's reading: working-life investment experience, not the amount, decides the use.
The population is that service's users (possible self-selection bias); "don't know / undecided" on the amount was answered by 14.8% of the inexperienced, about 1.8x the experienced at 8.3% (PRTIMES #1295455).
Q: What are the Taiwan contrast figures, and where do they come from?
Citing this site's published cards: ANK-2026-06-01-001 records, per Taiwan Stock Exchange statistics, a record 14,332,896 cumulative brokerage accounts in May 2026, single-month regular-savings investment of NT$32.579 billion, and those 30 and under at 51.8% of new accounts; the same card records, per Financial Supervisory Commission data, the natural-person turnover share falling from 58% in 2022 to 53.8% in Q1 2026. ANK-2026-06-16-001 records Taiwan's active ETFs blowing past NT$902.2 billion within a year (a 59% share for the leading issuer) and Japan's "NISA poverty" (52.5% of NISA users in the 580-person BRITA survey reporting a lower quality of life from belt-tightening).
The Taiwan side is official whole-population statistics (underlying this site's published cards); the Japan side is sample surveys and issuer releases — different populations and statistical natures (ANK-2026-06-01-001, ANK-2026-06-16-001).
Q: Can the Taiwan-Japan "temperature gap" be compared directly?
Not on a single scale. First, Japan's wealthy survey (a sample of 250 people) and Taiwan's whole-population account statistics differ in population and statistical nature. Second, Japan's "¥10,000 to under ¥30,000" is a per-person modal band while Taiwan's "NT$32.579 billion" is a whole-market single-month total — different unit levels. Third, each side has its own flip side (Taiwan's share recession and concentration; Japan's "NISA poverty"). What this card performs is juxtaposed observation: same direction (retail money moving into markets), different temperature and structure — no division, no ranking, no verdict of superiority.
The content of the "temperature gap": Japan's top layer puts preservation first (36.8% of respondents with half or more in cash; preservation leading at 34.0%) and its mass layer enters gradually via NISA; Taiwan shows whole-population penetration in official statistics (PRTIMES #1283286, #1282501; ANK-2026-06-01-001).
F-Units
F-001: LIS LLC survey (250 men and women aged 20 or over with net financial assets of ¥100 million or more; online questionnaire with screening plus main survey; conducted June 22-23, 2026): respondents with a cash-and-deposit ratio of 50% or more account for 36.8% of the total, and 80% or more for 24.0% (also phrased "one in four" in the source) - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: high - basis: official_statement - period: Survey 2026-06-22 to 06-23; released 2026-07-01 - caveat: A sample survey, not a census; within the same release the summary section says "exceeding 50%" while the detail section says "50% or more"; the referent of "of these, 80% or more" follows the release's summary line "one in four holds 80% or more in cash" as a whole-sample yardstick; cash and deposits include idle standby funds
F-002: The surveyed wealthy's top asset-management priority is "not losing assets (preservation)" at 34.0%, with "preparing for inflation and yen depreciation" second at 19.2%; as background the release cites a Nomura Research Institute estimate: 1.653 million wealthy and ultra-wealthy Japanese households as of 2023, up 11.3% from 2021 - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: high - basis: official_statement - period: Survey 2026-06-22 to 06-23; the Nomura Research Institute estimate is as of 2023 - caveat: 34.0% and 19.2% are shares among the 250 respondents; 1.653 million households is a third-party (Nomura Research Institute) estimate cited by the LIS release, not a figure from this survey
F-003: Shares holding "1% or more" of each asset class (excluding 0% answers): cash and deposits 98.0% (the source: "almost everyone"), domestic equities 85.6%, crypto assets 33.6%, alternative investments 30.0%, overseas real estate 26.0% - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: high - basis: official_statement - period: Survey 2026-06-22 to 06-23 - caveat: A "1% or more held" yardstick (share excluding 0% answers); breadth of ownership, not allocation weight by amount
F-004: Physical-asset ownership: precious metals (gold, platinum, etc.) 44.8%, luxury watches and jewelry 42.8%, art and antiques 24.4%, wine and whisky and the like 23.6%, antique coins 16.0%, none of these 31.6%; among owner-executives, luxury watch and jewelry ownership is 56.45% - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: high - basis: official_statement - period: Survey 2026-06-22 to 06-23 - caveat: Shares among the 250 respondents; the framing around inheritance planning and clear valuation standards is the source's own reading
F-005: Offshore-asset ratios: 0% (domestic only) most common at 43.2%; 30% or more at 21.2%; 0% answers by age: 50s at 63.16%, 60 and over at 60.66%; a "certain number" of those in their 30s and 40s report 80% or more (not quantified in the source) - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: high - basis: official_statement - period: Survey 2026-06-22 to 06-23 - caveat: "Polarization" is the source's wording; "a certain number" is not quantified in the source and this card does not fill in a figure
F-006: Crypto assets currently held by 21.6%, or 26.4% including past ownership; current ownership by generation: 20s 40.0% (n=20, a reference value flagged by the source), 30s 51.6% (n=31), 40s 30.8% (n=39), 50s 26.3% (n=38), 60 and over 6.6% (n=122); by occupation: regular-employee company workers 37.4%, owner-executives 6.5%; among owner-executives, 80.7% hold 0% in alternatives and 83.9% hold 0% in overseas real estate - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: medium - basis: official_statement - period: Survey 2026-06-22 to 06-23 - caveat: Generational sub-samples are small (20s n=20 flagged by the source as a reference value); those 60 and over at n=122 make up nearly half the sample, so overall ratios reflect the composition
F-007: Securities-backed and crypto-backed loans: "know it and have used it" 18.8%, "know it but never used it" 33.2% (most common), "did not know" 24.4%; usage experience 48.39% for the 30s, 6.56% for 60 and over; owner-executives' usage 11.29% (another passage of the same release writes "11.3%"), below the overall average (18.8%); the source cites National Tax Agency Tax Answer No. 1524 (crypto assets are in principle taxed at the point of sale or use) - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: medium - basis: official_statement - period: Survey 2026-06-22 to 06-23 - caveat: "11.3%" and "11.29%" coexist in the same release and this card records both; the tax explanation is a citation within the release and this card makes no tax or legal judgment; LIS operates a wealthy-focused asset-management media outlet and the release includes explainer links on the related theme
F-008: Main asset-management counterpart: bank representatives lead at 32.0%, brokerage representatives 30.4%, "consulting no one (deciding alone)" 31.2%; satisfaction with proposals: "satisfied" 16.4%, "somewhat satisfied" 26.8%, "none of these apply" 24.0%, "cannot say either way" 26.4%, outright dissatisfaction ("somewhat dissatisfied" 3.6% plus "dissatisfied" 2.8%) = 6.4% - source: PRTIMES #1283286 - source_url: https://prtimes.jp/main/html/rd/p/000000002.000181252.html - confidence: high - basis: official_statement - period: Survey 2026-06-22 to 06-23 - caveat: The "no settled adviser, no proactive proposals — a passive state" framing is the source's reading; shares among the 250 respondents
F-009: A Rakuten Securities release of July 1, 2026 cites: Japan's domestic NISA accounts topped 28 million at end-2025, with 23% of the total population holding a NISA account; more than 7 million people (about one quarter) hold theirs at Rakuten Securities; Rakuten Securities won overall No. 1 in the "2026 Oricon Customer Satisfaction Survey, NISA Brokerage" for the third straight year since the survey began in 2024 - source: PRTIMES #1282501 - source_url: https://prtimes.jp/main/html/rd/p/000000728.000011088.html - confidence: medium - basis: official_statement - period: NISA account count as of end-2025; release 2026-07-01 - caveat: The 28 million and 23% figures are citations within the Rakuten Securities release (asterisked; footnote text not included in the excerpt this card relies on); "most general brokerage accounts of any single firm" and "industry No. 1 in NISA accounts" are the issuer's own wording; the award release is a first-party release
F-010: Usage data Oricon published on July 1, 2026 from the 7,545 respondents of its 2026 NISA satisfaction survey: the reason for starting (multiple answers) is led by "putting idle cash to work" — brokerages 65.6%, banks 64.9% — followed by "saving for old age" in the 50-percent range and "saving toward a comfortable life" in the 30-percent range; account-choice reasons: "low fees" leads at brokerages, "already using other services there" at banks; growth-quota usage: "both regular and lump-sum" leads at brokerages, "regular contributions only" at banks; average monthly amounts: "¥10,000 to under ¥30,000" leads for regular contributions, "¥100,000 or more" for lump sums - source: PRTIMES #1282493 - source_url: https://prtimes.jp/main/html/rd/p/000000780.000034467.html - confidence: high - basis: official_statement - period: Published 2026-07-01 (2026 survey respondents) - caveat: Subjects are actual NISA users (a user population, not the general public); the "50-percent range" and "30-percent range" carry no precise values in the source; the source notes composition ratios are rounded to one decimal place and may not sum to 100
F-011: Oricon 2026 satisfaction rankings (published July 1, 2026): overall No. 1 NISA brokerage is Rakuten Securities (third straight year); SBI Securities first in "product lineup" for the third straight year; overall No. 1 NISA bank is PayPay Bank (first time; first in every evaluated category including "account opening" and "ease of management"); Japan Post Bank first in the "beginner investors" level for the second straight year; regional rankings debuted - source: PRTIMES #1282495 - source_url: https://prtimes.jp/main/html/rd/p/000000774.000034467.html - confidence: high - basis: official_statement - period: Published 2026-07-01 (2026 edition) - caveat: A commercial satisfaction survey (conducted by oricon ME); rankings are results within that survey's framework, not market shares
F-012: Clabo Inc. survey (992 domestic crypto-asset-experienced investors; released July 2, 2026): the "all-in on crypto" camp is 7.9%; the source states about 95% practice diversification with other assets; the most common crypto share is "a sub-allocation of 30% or less" at 38.8%; the most common pairing is "investment trusts/NISA" at a 22.4% implementation rate - source: PRTIMES #1295068 - source_url: https://prtimes.jp/main/html/rd/p/000000054.000178703.html - confidence: medium - basis: official_statement - period: Released 2026-07-02 (survey period not stated in the source) - caveat: Subjects are crypto-experienced investors (not a general-investor population); the source presents "7.9%" and "about 95%" side by side without explaining the yardstick gap, and this card reproduces both without reconciling
F-013: Monicle Financial survey (6,000 users randomly drawn from its "3-minute investment diagnosis" — 3,000 each in their 40s and 50s; released July 2, 2026): "invest retirement money" — experienced investors 25.1%, inexperienced 5.7%, a roughly 4.4x gap; about 15% across ages 40-59 overall; among those expecting ¥30 million or more, the inexperienced chose investing at 4.3% (n=23, flagged by the source as a small-sample guide value) while the experienced chose it at 20.7% even when expecting under ¥5 million; "don't know / undecided" on the amount: experienced 8.3%, inexperienced 14.8% (about 1.8x) - source: PRTIMES #1295455 - source_url: https://prtimes.jp/main/html/rd/p/000000041.000049079.html - confidence: medium - basis: official_statement - period: Released 2026-07-02 (survey period not stated in the source) - caveat: The population is that service's users (possible self-selection bias); "experience decides the use" is that survey's reading of its own sample and must not be extrapolated to the general population; the n=23 sub-sample is flagged by the source as a guide value
J-Units
J-001: Two temperatures coexist within the same Japan — the top wealthy layer plays "defense" (36.8% of respondents with half or more in cash; "preservation" first at 34.0%; owner-executives especially conservative on volatile assets and leverage), while the mass layer "enters" (28 million NISA accounts; 23% of the population per the Rakuten Securities citation; "putting idle cash to work" above 60% as the trigger; ¥10,000-to-under-¥30,000 monthly contributions the modal band) — the "temperature gap" exists first inside Japan's own asset strata; the two rest on different data natures (a sample survey versus issuer-cited statistics), so this is juxtaposed observation - confidence: medium - basis: official_statement
J-002: The dividing line is generation and investment experience, not asset size — current crypto ownership 51.6% for the 30s versus 6.6% for 60 and over (within the wealthy sample); collateral-loan usage 48.39% for the 30s versus 6.56% for 60 and over; retirement-money investing intent: experienced 25.1% versus inexperienced 5.7% (about 4.4x), and even the inexperienced expecting ¥30 million or more chose it at just 4.3% (n=23 guide value) — "a bigger amount" does not mean "will invest"; all are juxtaposed observations within each survey's sample, not causal claims - confidence: medium - basis: official_statement
J-003: The Taiwan-Japan contrast — same direction (retail money moving into markets: Japan's "from savings to investment," Taiwan's nationwide entry), different temperature and structure: Taiwan shows whole-population penetration in official statistics (14,332,896 cumulative accounts in May 2026; NT$32.579 billion single-month regular-savings investment; 51.8% of new accounts aged 30 and under — recorded in this site's published cards), while Japan shows layered temperature differences via samples and issuer statistics; and each side has a flip side (Taiwan's natural-person share down from 58% to 53.8% in five years plus active-ETF concentration; Japan's "NISA poverty" at 52.5%) — different populations and yardsticks, juxtaposed without division - confidence: medium - basis: official_statement
P-Units
P-001: Whether wealthy Japan's cash ratios loosen — the LIS survey is a single cross-section (conducted June 2026); changes in the "preservation orientation" under inflation and yen depreciation need verification via same-yardstick follow-up surveys ### P-002: The pace of growth and age composition beyond 28 million NISA accounts (end-2025, per the Rakuten Securities citation) — pending follow-up figures on an official-statistics basis ### P-003: The bear-market test of Taiwan-Japan contribution discipline — Taiwan's regular-savings investing (NT$32.579 billion in a single month) and Japan's NISA contributions (the ¥10,000-to-under-¥30,000 modal band) largely formed in bull markets; their persistence through the next full bear market is unverified
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 document: - ANK-2026-06-01-001 (the retail-investor phenomenon of Taiwan's AI stock boom: record 14.33 million cumulative accounts in May 2026, regular-savings investment past NT$32.5 billion in a single month, 51.8% of new accounts aged 30 and under) → cited as the Taiwan side's official-statistics contrast of a "nationwide rush into the market": 14,332,896 cumulative accounts, NT$32.579 billion of single-month regular-savings investment, 51.8% of new accounts under 30, plus the flip side of the natural-person turnover share falling from 58% to 53.8% over five years. - ANK-2026-06-16-001 (the great retail-money migration: Taiwan's active ETFs surging to NT$902.2 billion versus the structural cracks of Japan's NISA "from savings to investment" in its third year) → cited as the prior observation of the Taiwan-Japan "temperature gap": Taiwan's explosive active-ETF concentration at NT$902.2 billion against Japan's "NISA poverty" (52.5% of surveyed users reporting a lower quality of life from belt-tightening) — this card adds the layer of wealthy Japan's "defense" on top of it.
Sources
1. [PRTIMES #1283286] 合同会社LIS, "【2026年版】金融資産1億円以上の富裕層250人に聞いた資産運用実態調査", 2026-07-01. https://prtimes.jp/main/html/rd/p/000000002.000181252.html 2. [PRTIMES #1282501] 楽天証券株式会社, "楽天証券、「2026年 オリコン顧客満足度®調査 NISA 証券会社」ランキング3年連続総合1位を受賞！", 2026-07-01. https://prtimes.jp/main/html/rd/p/000000728.000011088.html 3. [PRTIMES #1282493] オリコン株式会社, "＜計7,545人が回答＞『NISA 証券会社／銀行』利用実態データ┃NISAで投資を始めたきっかけは【証券会社】【銀行】ともに「手元資金の有効活用」が6割以上で最多（オリコン顧客満足度調査）", 2026-07-01. https://prtimes.jp/main/html/rd/p/000000780.000034467.html 4. [PRTIMES #1282495] オリコン株式会社, "満足度の高い『NISA 証券会社／銀行』ランキング証券会社は【楽天証券】が3年連続総合1位 銀行は【PayPay銀行】が初の総合1位（オリコン顧客満足度®調査）", 2026-07-01. https://prtimes.jp/main/html/rd/p/000000774.000034467.html 5. [PRTIMES #1295068] 株式会社Clabo, "暗号資産はNISA併用が最多。投資家992人が実践する「NISA・株式」併用の分散戦略", 2026-07-02. https://prtimes.jp/main/html/rd/p/000000054.000178703.html 6. [PRTIMES #1295455] 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 7. [ANK-2026-06-01-001] Rin Takenouchi, "AI台股熱的散戶現象：2026年5月開戶累計1433萬創高、定期定額單月衝破325億、證券劃撥存款4兆5039億創高，30歲以下占新增開戶51.8%全民入市，但自然人成交占比近5年從58%降到53.8%，熱絡與結構位移是同一枚硬幣", 2026-06-28. https://ainews.washinmura.jp/ainews/zh/ank/ANK-2026-06-01-001 8. [ANK-2026-06-16-001] Rin Takenouchi, "散戶資金大遷徙：台主動式ETF爆衝9022億（統一投信59%市占）vs 日NISA「貯蓄から投資」滿3年的結構裂縫", 2026-06-16. https://ainews.washinmura.jp/ainews/zh/ank/ANK-2026-06-16-001