# Capacity Can Be Grown: A Former Personnel Commissioner on What a Manager Still Owns in the AI Era

> Notes on Investor's Sunday, 2026-09-13, part two of the conversation with Katsura Ito. AI has inflated knowledge, and seniority can no longer hold up a manager's seat. She uses adult development theory and the Japanese word utsuwa (capacity) to explain how managers grow, and a French boss's habit of saying 'But' to explain feedback. Educational content, not investment advice, no stock picks or price targets.

Published: 2026-09-13
Locale: en
Tags: leadership, adult development theory, AI era, feedback culture, podcast notes


![A long Tokyo office corridor at dusk. On the corner of a wooden desk in the foreground sits a handmade ceramic cup with an uneven glaze. Deep down the corridor, by an open door, an older female manager leans on the frame, listening to a young colleague holding a notebook. Warm light from the room and blue dusk light from the windows meet on the floor](/covers/invsunday-2026-09-13-ai-cover.png)

> Using bronze as a mirror, one can straighten one's clothes; using history as a mirror, one can understand the rise and fall of states; using people as a mirror, one can see one's own gains and losses.
>
> —— *Old Book of Tang*, Biography of Wei Zheng (Later Jin, 945; my translation)

## What this episode is about

The 2026-09-13 episode of Investor's Sunday is part two with Katsura Ito. She worked at IBM Japan and Adobe, rose to Chief Learning Officer at Microsoft Japan, and from 2022 served one four-year term as a Commissioner of Japan's National Personnel Authority. Last week she talked about organizational change. This week host Rie Murayama narrowed the topic to one group of people: Japanese managers today.

Murayama opened with a list of their complaints. You can't hire. The people you hire leave within three years. Try to communicate properly and you get accused of power harassment. Loosen standards until the company looks like a "white company," and people still leave. Her words came close to "what am I supposed to do?"

Ito's answer didn't spare anyone: these managers are still living in the fog of the Showa era. They copy the way they were hired, treated, and spoken to by their own bosses, and they get confused when it stops working. "Let's forget our past glory," she said. "It's Reiwa now."

The second half of the episode circles two words: the "knowledge inflation" brought by AI, and the Japanese word *utsuwa*, a vessel, meaning a person's capacity. Near the end she named three things she wants to do next: set up a foundation supporting women founders, use agile methods from software development to transform large companies, and take Japanese sake to the world. Along the way she gave one direction of travel: global alcohol consumption is falling, while premium sake (*tokutei meishō-shu*) and exports are rising.

## Key points

**Picking a successor who resembles you is, in her word, "the worst."** The Reiwa world moves fast and technology keeps changing; nobody stays right just because they've been around longest. The manager she has in mind first turns the skills a mission needs into a checklist, then finds people to fill the boxes they can't fill themselves: colleagues, subordinates, friends outside the company. Labels like seniority, gender, mid-career hire, insider or outsider come off first. The precondition is admitting one thing: part of how you got promoted was timing and luck. Once you admit that, you can leave behind "I need to manage my subordinates" and move to "I need this team." Murayama added a memory from her thirties: Masatoshi Ito, founder of Seven & i, would pick up the phone the moment he didn't understand something and call people inside or outside the company, regardless of their rank. Seeing how fast and how accurately information came back left her stunned.

![A skill checklist with two rows of eight boxes: the top row relies on yourself alone, with four boxes filled and four empty boxes marked "Gap"; the bottom row is the whole team, where colleagues, direct reports, outside friends, and AI fill every empty box, showing that a manager does not have to fill every box personally.](/figures/skill-checklist-team-fill-en.svg)

**Adult development theory: adults keep developing, and the highest stage rewrites its own frame.** The theory started with Robert Kegan at Harvard. Ito described three stages. Stage three does assigned work properly, common in one's twenties and thirties. Stage four judges by its own values and principles, roughly the manager in their forties. Stage five can question its own frame of judgment; when society changes or relations between countries shift, the frame shifts with them. She said only some executives reach stage five.

![A staircase with three steps rising upward: Stage 3 is doing assigned work well, Stage 4 is judging by your own values, and Stage 5 sits at the top with a looping arrow that curves back on itself, showing that this stage returns to revise its own framework of judgment.](/figures/adult-development-three-stages-en.svg)

**Growth runs on two axes, ability and *utsuwa*, and *utsuwa* can be grown.** Ability is data analysis, presentations, the things that go on a résumé. *Utsuwa* corresponds to the English "development" or "capacity"; the person who rendered it as *utsuwa* in Japanese is Yohei Kato. Ito said English can't carry the word, while any Japanese listener gets it at once: smart, capable, and still missing something, and that missing thing is *utsuwa*. How does it grow? Two moves. One is reflection: after something goes wrong, being able to say "I used this approach, so it failed" or "I had this bias." The other is perspective-taking: how large a unit you can look at, from yourself and your friends, to the whole company, to Japan, to the world. Her negative example: "You can't pin that on me, blame that person." Her positive example: "I should have been more grown-up and gone to the next department to ask for help," or "I should have hired that person." Murayama added that the people she has seen promote a capable subordinate they personally disliked went on to rise very high, while everyone around them wondered, "Wait, don't you two clash?"

![A chart with skill on the horizontal axis and caliber on the vertical axis: one point sits where skill is very high but caliber is low, labeled brilliant yet lacking something, and an arrow runs straight up from it to a hollow point, showing that caliber can be grown separately.](/figures/ability-and-capacity-two-axes-en.svg)

**AI has inflated knowledge, and the premise of seniority pay has collapsed.** Knowledge used to require your own digging or a call to an expert. Older people had more experience and knew more, so they got better pay and became department heads. Now AI hands knowledge to everyone in plain language, and "knowing more" no longer sets anyone apart. She said two things remain. One is lived experience: the atmosphere of a real room, a customer complaint at a store counter, which AI doesn't have. The other is the ability to keep learning: when a thought comes up, type the question to AI yourself, ask younger people, and your questions get sharper along the way, with fewer "just deal with this." A skill matrix doesn't need every box filled; people around you and AI fill the gaps. Murayama chimed in that Ito reads a huge amount, and the books Ito recommends pile up half-read on Murayama's shelf. Ito's reply: "These days I have AI summarize them. Done in no time."

![Four concentric circles, from inside to outside: me and my friends, the whole company, Japan, and the world; "Not my fault" links to the innermost circle, while "Ask the next team" and "Should have hired him" link to the company circle, showing that as the unit you think in grows, your reactions change with it.](/figures/perspective-nested-circles-en.svg)

**Feedback takes daily practice; sudden effort tears a muscle.** The manager model Microsoft taught her is Model, Coach, Care: first show what learning and enjoying your work looks like, coach so that people discover things themselves, and Care, which she simply called love. When someone says harsh words now get labeled harassment, her reply was, "That's because you're not serious enough." She said a few of her one-on-ones were held with the resolve of going down together, with subordinates and with bosses. Murayama used a gym comparison: train every day and the muscle is there, so praise comes as easily as breathing; skip training and a sudden effort pulls a muscle. At Goldman Sachs her boss taught her that, beyond the smoking room, the chats in passing in the hallway or in the restroom matter most; once you've asked "how are the kids?" on ordinary days, the hard words can land when you need them. She also joked about the line foreign-firm managers love, "My door is always open for you": the door is always open, and nobody is ever in.

![Two groups of bars side by side: before, four people's knowledge rose steadily with age; now AI lifts everyone's knowledge to the same height, and the only difference left is a small cap on top labeled "lived experience plus continuous learning," whose height has nothing to do with age.](/figures/knowledge-inflation-before-after-en.svg)

**Ask for feedback, and don't reach a conclusion within two seconds of hearing it.** At IBM, after a foreign boss finished giving her feedback, he asked, "So, any feedback for me?" She was floored. Later she learned to speak up. She told a French boss, Patrice: you're logical, you treat me well, thank you; but you start every sentence with "But," and Japanese conversation starts with "Yes," so every time you say "But," my heart shrinks a little. He apologized, and the following week he kept saying "But." What mattered to her was something else: he learned that one word could hit another person that hard emotionally, a thing he had never considered. She went on: after hearing someone out, don't rush to a conclusion or cut them off at two seconds. "Conclusion first" is for emergencies only, like surgery or a fire.

## Further thoughts

### When I'm stuck in a losing position, who do I blame first?

A stock drops 30% after I buy it, and the first thoughts are usually "big players are dumping," "the market dragged it down," "the earnings were dressed up." When Ito quoted "You can't pin that on me, blame that person," I felt called out.

Her two moves carry over to investing as one way of breaking the problem down.

The first layer is reflection, and it needs something to check against. If I never wrote down why I bought, all I have afterward is emotion, and I can't even find which assumption failed. I only recently started recording a reason for every trade; for a long stretch before that I logged only price and share count, and when I look back at the losses from that period, the causes I can name mostly point outward.

The second layer is widening the unit. Looking at this one stock alone, a 30% drop is a disaster. Put it back in the whole portfolio and the questions become how much of it this position is, and whether it's falling together with the other holdings. One layer up, if a whole class of assets is being repriced in the same window, say a rate move hitting every richly valued stock, that's structure, and the news on this one name becomes noise.

External causes are sometimes real, so I need a test that separates them from my own mistakes: if an outside cause is real, it should hit similar companies too. If peers are flat and only my stock fell, I go back and check my own assumptions first.

![Two groups of downward bars: on the left, my stock and four peers fall by about the same depth, so the reading is to look at the broader structure; on the right, only my bar drops deeply while the peers barely move, so the reading is to go back and recheck my own assumptions.](/figures/peer-drop-check-en.svg)

### If AI can look up anything, and whatever I find others can find, what's left?

I once spent a night pulling together the key points of a company's earnings, then found the next day that AI could produce roughly the same thing in three minutes. I know that deflated feeling.

Ito was talking about managers when she described knowledge inflation, and I heard it as an investor's situation: everyone gets the same neatly organized public information, and that information reaches prices faster too (this is my inference; the episode didn't discuss markets). The two things she said remain apply to investing as well.

One is firsthand experience: visiting stores, using the product, listening to customer complaints. The sample is small. A line outside one store can't tell you company-wide revenue. Its use is to form a hypothesis, then test it against filings and numbers; drawing a conclusion from it directly treats a small sample as a large one.

The other is how fast you can change your frame, which maps to stage five. When new technology rewrites how an industry makes money, measuring it with the old valuation ruler makes expensive look cheap, or the reverse. The habit I've set for myself is to ask once a quarter: the criteria I'm using now, what year and what environment were they set in?

### When I look at a company's management, what am I looking at?

A headline says the new CEO was promoted from within after twenty years at the predecessor's side, and the market shrugs. I used to not get it: with a résumé that complete, what's the worry?

Ito called picking people like yourself the worst, because the environment moves fast and one person's experience can't carry every judgment. At a company, a successor from the same mold inherits the predecessor's blind spots along with the job.

The boundary matters here. In slow-moving industries with stable rules, a like-for-like succession has real benefits; continuity has value in itself. The worry only holds when the environment speeds up: a technology shift, a change in the customer base, a regulatory upheaval.

![Two sets of field-of-view bars: a predecessor and a successor cast from the same mold see exactly the same range, and the same empty patch on the right is a blind spot they share; a successor with a different background shifts to the right and fills in that empty patch.](/figures/succession-shared-blind-spot-en.svg)

I watch two signals. One is the makeup of the senior team: whether anyone was brought in from outside, or whether every background traces the same promotion path. The other is how the executive talks about misses. Asked on an earnings call about last quarter's shortfall, do they talk about "the external environment," or about "which of our calls was wrong"? The second maps to the reflection side of *utsuwa*, and you can hear it in the Q&A.

## References

- Investor's Sunday (InterFM), 2026-09-13, part two with Katsura Ito; part one on 2026-09-06 covered organizational change, and there is a separate post on it on this site.
- Robert Kegan and Lisa Lahey, *Immunity to Change* (2009): one of the entry points into adult development theory.
- Yohei Kato, *Naze buka to umaku ikanai no ka: "Jita henkaku" no hattatsu shinrigaku* (2016): adult development theory and *utsuwa*, in Japanese.
- Microsoft has spoken publicly about its three manager expectations, Model, Coach, Care; searching "Microsoft model coach care" brings up the original description.

## One thing to take with you

*Utsuwa* is trained, and the training is letting other people show you the side of yourself you can't see.

One thing I've tried: this week, pick someone you spend a lot of time with, a colleague, a partner, a family member, and ask, "Is there one thing you'd like me to do a little less of, or a little more of?" When they finish, say thank you, then go two seconds without explaining or defending. That day, write down what they said word for word, and a week later take it out and see how much you've changed.