Seventh in a series breaking down real prompts our members run on Ora, and why the wording matters as much as the data underneath it.
Every secondaries manager raising capital in Asia gets the same advice about the region’s insurers: big balance sheets, long-dated liabilities, and private markets books that have grown for a decade. All true, and all about the LP side. What gets less attention is that the same balance sheets are now a source of secondary supply, as new capital rules change what it costs an insurer to hold a portfolio of fund stakes.
The prompt below tries to catch both at once: insurers that might commit to the fund, and insurers the fund might buy from. It is a sharp instinct. It also produces a list that needs careful handling, because the same institution can sit on both sides of your business, and the people who answer each question are rarely the same.
The prompt
I’m raising capital for a secondaries fund and want to explore Asian insurance companies as both LPs and potential secondary sellers. Identify insurance companies in Japan, South Korea, and Taiwan with $10B+ in invested assets that have participated in a secondary transaction in the last 3 years, either as buyer or seller. Show key investment contact and any details of their most recent secondary transaction.
What nobody says out loud: the capital rules decide both sides
In three years, all three markets in this prompt moved to economic, market-value solvency rules. That shift, more than appetite, is what decides whether an insurer buys fund stakes, sells them, or does neither.
| Market | Regime | In force | What changed |
|---|---|---|---|
| South Korea | K-ICS | January 2023 | 100% minimum ratio, 130% recommended; a 50% basic-capital requirement starts in 2027 with a nine-year transition (Insurance Business) |
| Taiwan | TW-ICS with IFRS 17 | January 2026 | Fair-value balance sheets and higher capital charges, with some factors phased in over 15 years; foreign-exchange exposure becomes more visible (AllianceBernstein) |
| Japan | J-ICS (ESR) | Fiscal year ended March 31, 2026 | Economic solvency ratio with a 100% minimum, replacing the old 200% solvency margin floor (Insurance Business) |
Under these rules, a fund stake costs an insurer its capital charge as well as its expected return. When that charge rises, selling the stake becomes a capital decision, not just a portfolio one. Taiwan showed this early: its life insurers offloaded fund stakes and halved annual commitments, partly because of current and coming capital requirements (Private Equity International, March 2024). Nan Shan Life then shopped a portfolio reported at around $2 billion (Secondaries Investor, February 2025).
The same rules shape the LP side. A secondaries fund can look attractive to a capital-constrained insurer, with cash coming back sooner than from a new primary fund. Whether that actually lowers its capital charge depends on how each regime treats the vehicle, and that is a question to ask, not assume. Supervisors are watching too: the global insurance standard-setter has flagged growing investment in alternative assets as a concern (Insurance Business).
So the useful question is not who did a secondary in the last three years. It is whose capital position makes a secondary decision likely next, and which way it will go.
The anatomy
“I’m raising capital for a secondaries fund.” This sets the purpose, but not the strategy. LP-led portfolio purchases, GP-led continuation vehicles, credit secondaries and real estate secondaries attract different insurers on both sides. Name the strategy, the fund size and your typical deal size, so Ora can match sellers whose portfolios you would actually buy and LPs whose tickets fit.
“Explore Asian insurance companies as both LPs and potential secondary sellers.” The sharpest idea in the prompt, and the one that needs the most care. The team that commits to new funds is often not the team that decides to sell old ones, which tends to sit closer to finance, risk or asset-liability management. Ask Ora to return the two roles as separate lists, and keep the two conversations separate. Pitching an insurer for a commitment while bidding on its portfolio mixes a relationship with a negotiation.
“In Japan, South Korea, and Taiwan.” Three markets at three different points in the same transition. Korea has lived under K-ICS since 2023, Taiwan moved to TW-ICS this year, and Japan published its first ESR results in May 2026. Ask for each insurer’s latest solvency ratio and its direction, because that is the leading indicator. Much of the best disclosure is in Japanese, Korean or Chinese, so tell Ora to include local-language sources.
“With $10B+ in invested assets.” A low bar for these markets, and that is not a flaw. The largest groups are the likeliest LPs. Smaller and capital-stretched carriers are the likelier sellers. In Korea, several insurers sat below the coming 50% basic-capital line in the first half of 2026 (Insurance Business). Keep the floor, but ask Ora to tier results by size and capital ratio instead of treating everyone above $10 billion alike.
“Participated in a secondary transaction in the last 3 years, either as buyer or seller.” “Participated” is too loose. An insurer that simply rolled its stake when a GP ran a continuation vehicle technically participated, and tells you nothing. Split this into three signals: sold a portfolio of fund stakes, bought secondaries directly, and committed to a secondaries fund. Each points to a different conversation.
“Show key investment contact.” Singular, when you need two: the head of alternatives or private equity for the LP side, and the CIO or portfolio management lead for the seller side. Many groups also invest through an asset management affiliate or an overseas office, so the right person may not sit at the insurer itself.
“Any details of their most recent secondary transaction.” Expect gaps. Insurer portfolio sales are rarely announced, and pricing almost never is. Ask for what can be found (date, approximate size, buyer, adviser and source), and have Ora label each detail as reported or confirmed.
What you do with the output
- Split the list in two before anyone makes a call. One list for LP prospects, one for potential sellers. Assign different owners, and agree internally how you will handle an insurer that appears on both.
- Rank sellers by capital pressure, not by past deals. A falling or thin solvency ratio, a capital raise, or a public pullback from private equity is a stronger signal than a sale three years ago. Revisit the ranking every time new ratios are published.
- Lead LP conversations with capital treatment. Ask early how each regime treats a commitment to your fund, and come with a view on how your structure affects it. For many of these insurers, that question decides the commitment.
- Approach sellers through the channels they use. Insurer portfolio sales usually run through advisers. Make sure your fund is known to the advisers active in Asia, not just to the insurers.
- Plan for currency. Taiwan’s new rules make foreign-exchange exposure more visible, and many of these balance sheets hold U.S. dollar assets. Be ready to discuss share classes, hedging or local feeder options.
Variants
Here is the prompt rewritten with the changes above:
I’m raising a $[size] LP-led secondaries fund focused on [strategy], with typical deals of $[range]. Identify insurance companies in Japan, South Korea and Taiwan with $10B+ in invested assets, including local-language sources. Return two lists. List 1, potential sellers: insurers that sold a portfolio of fund stakes in the last 3 years or show capital pressure (latest solvency ratio and trend, recent capital raises, reported cuts to private equity). List 2, potential LPs: insurers that bought secondaries directly or committed to a secondaries fund in the last 3 years. For each, give the head of alternatives and the CIO or portfolio management lead, any asset management affiliate that invests on their behalf, and details of the most recent transaction (date, size, counterparty, adviser, source), labeled as reported or confirmed.
Three swaps that each open a different channel:
- Swap insurers for pensions and sovereign funds in the same three markets. Korea’s sovereign fund, KIC, was reported in September 2026 to be exploring a sale of more than $1 billion in private equity interests (Private Equity Wire). Pensions and sovereigns face different pressures, but the same secondary market.
- Swap the seller list for insurers below a capital threshold. For example, Korean insurers below the 50% basic-capital line due in 2027. This finds tomorrow’s sellers rather than yesterday’s.
- Keep the LP list and swap the asset class to credit secondaries. Insurers already understand credit, and the capital conversation may be easier.
Why this is a prompt and not a database query
No field in a conventional investor database is called “Taiwanese life insurer whose solvency position makes a portfolio sale likely this year.” That picture is spread across solvency disclosures, regulator statistics, annual reports in three languages, secondaries trade press and adviser announcements. Assembling it by hand takes days per market, and the ratios it depends on change every reporting cycle.
Ora reasons across those sources and returns the synthesis with its working shown, so you can check a solvency figure or a reported sale against the source before you act on it. The prompt is the specification. The more precisely you write the constraints, the less time you spend cleaning the output.
One practical note: marketing a fund to institutions in Japan, Korea and Taiwan is regulated in each market, and insurer portfolio sales are usually confidential. Confirm your private placement route with counsel in each jurisdiction, treat reported transactions as leads rather than facts, and keep seller and LP conversations separate. The list gets you to the right institutions. How you approach each one is still your call.
This is the seventh post in Prompt Anatomy, where we take a real prompt run on Ora and break down why it is written the way it is. Have one you want dissected? Send it to info@octum.ai.
New to Octum? Ora is our research engine for institutional investing, built on GAR, our own cognitive architecture. Join in about three minutes.
