JIAHON Group · JIAHON Think Tank perspective · Road to NT$100 Million, Part Four. Published by JIAHON Think Tank | Written by Marcus (Senior Vice President) | Published 2 October 2026 | jiahon.com.tw
Part Two said that the 65% step requires customers to write you into their specifications, and that it takes ten years with no shortcut. This part takes those ten years apart: why customers are willing to pay more, how supplier qualification releases trust step by step, when “trusting this person” becomes “trusting this specification” — and why Taiwanese small and medium enterprises especially need to move trust from the owner into the company.
In Brief
Part two said the top rung of the margin ladder—65%, where a customer writes you into the specification—takes ten years and has no shortcut. This part opens up those ten years. Margin, it argues, is the price of trust. Virtually every commercial transaction contains an element of trust, certainly any transaction conducted over a period of time (Arrow, 1972); when buyers cannot verify quality, good sellers are driven out unless guarantees, brands or licensing step in (Akerlof, 1970). Across 344 supplier–automaker relationships in the US, Japan and Korea, the automaker its suppliers trusted least bore procurement costs five times those of the most-trusted (Dyer and Chu, 2003); the study measures suppliers’ trust in buyers, but the logic runs both ways. Customers therefore extend trust the way banks extend credit: a quote, then a sample and a trial order, then a vendor code, and finally a part number in the specification—each rung lowering their transaction costs and raising the supplier’s margin. The last rung changes what is trusted: not a person but a mechanism, which is why it survives buyers being reassigned and engineers retiring. For Taiwanese SMEs—in Fukuyama’s (1995) account, trust in Chinese societies has traditionally stayed within the family—the task is to move it from the owner into the company. On the series’ own regression, the same NT$100 million of revenue yields about NT$8.1 million of net profit at a 19% gross margin and NT$24.1 million at 65%; the NT$16 million between them is what customers pay to stop worrying.
1. Both listed as wholesale — why do margins differ threefold?
At the bottom of Part One‘s margin ladder is 19%, the typical value for wholesale in the Ministry of Finance’s industry profit standards; above it is the 57% a niche technical channel can reach, and at the top, 65%, which requires customers to write you into their specifications. Part Two matched each step to the question the customer asks: at 19%, the customer asks “how much?”; at 30% to 40%, “will it work?”; at 50% to 57%, “what do you think?”; at 65%, the customer stops asking.
The same item might be sold cheaper elsewhere. What the customer pays on top is not buying the thing.
It is buying not having to worry. Margin is the price of trust.
2. There is trust in every deal
The Nobel economics laureate Kenneth Arrow wrote a much-quoted passage in 1972: almost every commercial transaction contains an element of trust, above all one that stretches over time; and there is reason to think that much of the world’s economic backwardness comes from a lack of mutual trust.
In the same era, George Akerlof used the used-car market to show the reverse: when buyers cannot tell good from bad, they will pay only the average price, sellers of good cars leave, and only lemons remain. In the same paper he pointed out that warranties, brands, chains and licences are the institutions markets use against this — in the end, they are all selling trust.
Someone has measured what trust is worth in a supply chain. Jeffrey Dyer and Wujin Chu studied 344 automaker–supplier relationships in the United States, Japan and Korea: the automaker its suppliers did not trust spent clearly more of its meeting time with suppliers negotiating contracts and haggling, and its procurement transaction costs were five times those of the trusted automaker. This study measures suppliers’ trust in the automaker, the opposite direction from this article, but the logic carries over: the trusted party saves the whole transaction the cost of verifying, negotiating and guarding.
Figure 1: Procurement transaction costs (relative) — trusted automaker 1, untrusted automaker 5. 344 relationships in the U.S., Japan and Korea; Dyer, Chu (2003).
What the customer saves on verification, negotiation and the risk of mistakes, it is willing to share in part with the supplier; this article’s view is that this share is the margin. What JIAHON has always said — “one-stop sourcing does not sell items; it saves the customer the trouble of purchasing” — is exactly the slice of transaction cost Dyer and Chu measured.
3. The trust limit: how customers extend it step by step
A customer gives a supplier trust the way a bank gives a credit-card limit: a little at first, raised as the record builds, withdrawn when something goes wrong. A customer’s supplier qualification does exactly that.
Around 1980 the political scientist Robert Axelrod held two computer tournaments, written up in 1984 in The Evolution of Cooperation. Both were won by simple “tit for tat”: cooperate on the first move, then do whatever the other side did on its last move. He pointed out that the key to cooperation is that both sides expect to meet again — what he called “the shadow of the future.” A supply chain is a long repeated game: every small order is a test, and every on-time delivery raises the next limit a little.
Map this back to the margin ladder and you have this part’s framework — the trust-limit ladder:
| Step | Limit the customer gives | What the customer is verifying | What the supplier must deliver | Matching margin step |
|---|---|---|---|---|
| 1 Quote | One request for quotation | Whether the price is reasonable | A price breakdown that can be explained | 19%: the customer asks how much |
| 2 Sample | One trial, one small order | Whether it works, whether each batch is the same | Verifiable samples, on-time delivery | 30–40%: the customer asks whether it will work |
| 3 Vendor registration | A vendor code, the approved supplier list | Whether a long-term relationship is possible | Documents, audit records, a non-disclosure agreement | 50–57%: the customer asks what you think |
| 4 Written into the specification | Part number designated | No further verification needed | Years of records without errors | 65%: the customer stops asking |
Figure 2: The trust-limit ladder (quote, sample, registration, specification) set against the margin ladder (19%, 30%, 40%, 50%, 57%, 65%, true scale). The 19% margin is the Ministry of Finance’s typical value for wholesale; 57% and 65% are industry estimates (Part One appendix).
What this table shows is the order: no step can be skipped. JIAHON’s rule: settle the price first; samples are only for confirming the physical item, not a tool for price comparison; complete the non-disclosure agreement and vendor registration before handing over specific dimensions and drawings. Skipping a step means giving away a limit you have not yet received.
4. The last step: from trusting a person to trusting a specification
At the fourth step, what the customer trusts changes. In the first three steps, the customer trusts the salesperson, the engineer, the owner — a person; at the fourth, the customer trusts a specification, a part number and the stack of documents behind it.
The automotive industry turned this into a system. The Automotive Industry Action Group (AIAG) Production Part Approval Process, PPAP (4th edition, 2006), lists 18 requirements, including design records, failure mode and effects analysis (FMEA), the control plan, measurement system analysis, dimensional results, and material and performance test records; depending on the submission level the customer specifies, some are submitted to the customer and some are retained at the supplier for review, after which the supplier signs a Part Submission Warrant (PSW). Quality is no longer a promise; it is evidence the customer can verify.
Why is this step stable? The Nobel economics laureate Oliver Williamson called it the “fundamental transformation”: before the contract, many suppliers compete; once the relationship builds up dedicated investment — dedicated equipment, tooling, knowledge worked out together — the original many-sided competition becomes a two-sided lock-in after the contract. Changing suppliers means the customer has to climb the whole trust-limit ladder again.
Trust in a mechanism runs both ways. The specification protects the customer and also the supplier: JIAHON’s rule is that when a customer claims poor quality, we first check whether the drawing has the criteria for judging it — datum surfaces, measurement direction, measurement conditions, sampling rules — so the judgement goes back to the specification, not to whoever is louder.
Part One said the ticket expires: buyers are reassigned, engineers retire. But a part number written into a specification does not leave with the person.
5. The problem for Taiwanese SMEs: trust lives in the owner
In Trust (1995), the political scientist Francis Fukuyama placed Chinese societies on the low-trust side: the family is the core, and there are few intermediate organisations between family and state. He argued that whether a society can build organisations beyond the family is key to whether its companies can become large modern organisations.
Part One had already pointed to the consequence: most of the value of a B2B distributor is its principal’s ticket, and the ticket does not transfer with the assets, so the recovery rate on failure is only 20% (industry estimate). In other words, in Taiwanese SMEs trust mostly lives in the owner — customers trust the boss, not the company. The day the boss retires, the trust leaves with them.
The solution is to move trust from the person into the company:
Write it down: turn every answer to “will it work?” into a knowledge base that engineering answers from (the amplifier in Part Two, Section 4).
Document it: specification comparisons, test data and audit records let customers verify for themselves, without relying on personal ties.
Face outward as a department, not an individual: JIAHON’s replies are signed “Engineering Department,” so that what customers remember is a department’s speed and answers, not one particular person.
Do these three things, and the ticket recovery rate Part Two talked about really starts to move up from zero.
6. The price of trust, worked out
Using Part One’s regression, net margin = 0.348 × gross margin + 1.51:
19% step: 0.348 × 19 + 1.51 = 8.1%
65% step: 0.348 × 65 + 1.51 = 24.1%
On the same NT$100 million of revenue:
19% step: net profit about NT$8.1 million
65% step: net profit about NT$24.1 million
The difference is about NT$16 million; net profit at the 65% step is about three times that at the 19% step (24.1 ÷ 8.1 ≈ 2.97).
Figure 3: Net profit on the same NT$100 million of revenue — about NT$8.1 million at the 19% step, about NT$24.1 million at the 65% step, a difference of about NT$16 million. Net margin = 0.348 × gross margin + 1.51 (regression in Part One of this series).
That NT$16 million is what customers pay so they do not have to worry. Another of Part One’s rulers measures the same thing: climbing from 19% to 65%, the ten-year multiple for reaching NT$100 million goes from 21× to 77×, a 3.7× difference.
Two things must be said honestly: the 57% and 65% steps are industry estimates, not official statistics; and the industry profit standards are a lower bound set for tax assessment, so the whole ladder should be read as conservative.
7. Trust breaks: one broken promise, and the limit goes to zero
The psychologist Paul Slovic put the nature of trust plainly in 1993: trust is usually built slowly, but it can be destroyed in an instant by a single mishap. He asked 103 university students to rate, on a 1-to-7 scale, the effect on trust of a series of news items about a nuclear power plant. For the same item — “the county medical examiner reports on the health of residents near the plant” — 50% rated the version saying residents’ health was worse than average as having a very strong effect (6 or 7); for the version saying it was better than average, only 18.3%.
Figure 4: Share rating the same news item as having a very strong effect on trust (6 or 7) — negative version (residents’ health worse than average) 50%, positive version (better than average) 18.3%. 7-point scale; Slovic (1993), 103 university students.
In a supply chain, this means one late delivery or one inconsistent batch can drop you from the third step back to the first, or off the list altogether. So trust also has a floor: do not promise what you cannot do. JIAHON’s rule is that a technical report does not promise quality specifications on behalf of upstream suppliers; we first confirm the conditions they can actually meet, then report to the customer.
This connects to the supply-chain-break window in The Remedy: when large manufacturers start looking for unfamiliar suppliers, every one of them has to climb again from the first step. The ones who climb faster are not those who promise more, but those who have never broken a promise.
8. What this means for mid-stream supply chain companies
Write trust into something the customer can verify — customers extend a limit only when they can verify for themselves.
Specification comparisons, test data and audit documents are all tools that let customers judge without relying on personal ties. PPAP’s 18 requirements are this taken to its fullest.
Don’t skip steps, and don’t let the customer skip them — each step should earn its matching limit.
Price settled first, samples not used as a price-comparison tool, drawings after registration. Skipping a step means giving away trust you have not yet received.
Don’t let trust live in one person — when the owner retires, trust should stay in the company.
Sign outward as a department, write knowledge into documents, and put the company’s name on the approved supplier list. This is the step at which a Taiwanese SME goes from “the boss’s business” to “the company’s asset.”
Conclusion
Part One said the 3.7× difference on the margin ladder comes not from effort or luck but from which product line you chose. This part adds one line: choosing the right product line is only the entry; climbing comes from trust — trust bought step by step. The last step is making it no longer depend on any one person.
Margin is the price of trust; the specification is the receipt.
Frequently asked questions
Why do margins differ so much between suppliers of the same product?
Customers pay not only for the product but for “not having to worry”: the verification and negotiation costs saved, and the risk of mistakes. A study of 344 automaker–supplier relationships in the U.S., Japan and Korea found that the automaker its suppliers did not trust had procurement transaction costs five times those of the trusted one — what trust saves is the cost of the whole transaction.
How does a supplier get written into a customer’s specification?
Usually by passing the four steps of the customer’s supplier qualification: quote, sample and small order, registration on the approved supplier list, and part-number designation. Each step verifies something different, and skipping a step often sends you back to the start.
What is PPAP?
The automotive industry’s Production Part Approval Process (AIAG 4th edition, 2006). It lists 18 requirements, including design records, failure mode and effects analysis (FMEA), the control plan, measurement system analysis and dimensional results; depending on the submission level, some are submitted to the customer and some retained at the supplier, after which the supplier signs a Part Submission Warrant (PSW), turning the promise of quality into evidence the customer can verify.
Why do Taiwanese SMEs find it hard to grow large?
One common reason is that trust is tied to the owner. Fukuyama argued that trust in Chinese societies has traditionally stayed within the family, and that the ability to build organisations beyond the family decides whether companies can grow large; institutionalising knowledge and customer relationships is how trust moves into the company.
Can trust be repaired once it breaks?
With difficulty. The psychologist Slovic pointed out that trust is usually built slowly but can be destroyed by a single mishap; negative news about the same event also affects trust far more than positive news. The best approach is not to promise what you cannot do.
Appendix: the data grade of this article
| Parameter | Grade | Source |
|---|---|---|
| Margin ladder 19% | Hard | Ministry of Finance industry profit standards (Part One) |
| Margin ladder 57%, 65% | Industry estimate | Part One appendix |
| Net margin regression 0.348 × gross margin + 1.51 | Hard (regression) | Part One, 1,283 categories |
| Net profit NT$8.1 million and NT$24.1 million on NT$100 million of revenue | Derived | Calculated from the regression |
| Ten-year multiples of 21× and 77× for reaching NT$100 million | Calibrated | Part One, Table 5 |
| Recovery rate 20% | Industry estimate | Part One appendix |
| 344 relationships, fivefold transaction costs | Hard (research) | Dyer, Chu (2003) |
| Two computer tournaments, won by tit for tat | Hard (research) | Axelrod (1980, 1984) |
| PPAP’s 18 requirements, submission levels, PSW | Hard (industry standard) | AIAG PPAP 4th edition (2006) |
| 50% vs 18.3% (6 or 7 on a 7-point scale) | Hard (research) | Slovic (1993), 103 university students |
| Mapping between the trust-limit ladder and the margin ladder | Practical observation | JIAHON shop-floor experience; Part Two, Figure 2 |
| JIAHON’s rules (price first, registration before drawings, judging criteria on drawings, no promises on behalf of upstream suppliers, department signatures) | Practice | JIAHON internal working rules |
Road to NT$100 Million series: Part One, “The Odds and Costs of Three Roads” | Part Two, “You Write the Sign” | The Remedy, “The Window Opens Only for Those with a Floor” | Part Four, “Margin Is the Price of Trust”
References
[1] JIAHON Think Tank (2026), “Choosing a Road to NT$100 Million — Lottery, Salary, Startup: The Odds and Costs of Three Roads” (Part One). https://jiahon.com.tw/three-roads-to-100-million-part-one/
[2] JIAHON Think Tank (2026), “You Write the Sign — The Same Numbers, Read Upward” (Part Two). https://jiahon.com.tw/you-write-the-sign-part-two/
[3] JIAHON Think Tank (2026), “The Window Opens Only for Those with a Floor — Fuel, Floor, Window” (The Remedy). https://jiahon.com.tw/fuel-floor-window-part-three/
[4] Kenneth J. Arrow (1972), “Gifts and Exchanges,” Philosophy & Public Affairs 1(4): 343–362.
[5] George A. Akerlof (1970), “The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism,” Quarterly Journal of Economics 84(3): 488–500. https://doi.org/10.2307/1879431
[6] Jeffrey H. Dyer, Wujin Chu (2003), “The Role of Trustworthiness in Reducing Transaction Costs and Improving Performance: Empirical Evidence from the United States, Japan, and Korea,” Organization Science 14(1): 57–68. https://doi.org/10.1287/orsc.14.1.57.12806
[7] Robert Axelrod (1980), “Effective Choice in the Prisoner’s Dilemma,” Journal of Conflict Resolution 24(1): 3–25; “More Effective Choice in the Prisoner’s Dilemma,” Journal of Conflict Resolution 24(3): 379–403. Robert Axelrod (1984), The Evolution of Cooperation, Basic Books; Taiwanese edition translated by Hu Wei-shan, Locus Publishing, 2012.
[8] Oliver E. Williamson (1985), The Economic Institutions of Capitalism, Free Press.
[9] Automotive Industry Action Group (AIAG) (2006), Production Part Approval Process (PPAP), 4th edition.
[10] Francis Fukuyama (1995), Trust: The Social Virtues and the Creation of Prosperity, Free Press.
[11] Paul Slovic (1993), “Perceived Risk, Trust, and Democracy,” Risk Analysis 13(6): 675–682. https://doi.org/10.1111/j.1539-6924.1993.tb01329.x
[12] Ministry of Finance, 2025 (ROC 114) Industry Income and Profit Standards for Profit-Seeking Enterprises (regression in Part One).
Note
This is an opinion article. All figures are drawn from the public statistics and research listed, with their data grade stated; conversion between gross and net margin follows the regression in Part One of this series and involves no financial data of any specific company. This article does not constitute investment, entrepreneurship, tax or legal advice.
License
This work is published by JIAHON Group. All rights reserved. Quotation and redistribution are permitted with proper attribution and a link to jiahon.com.tw. Commercial use—including but not limited to paid courses, reports, or consulting services incorporating this content or its frameworks (including the “trust-limit ladder” and the “margin is the price of trust” framework)—requires prior written authorization from JIAHON. For commercial licensing enquiries, please contact us through the website.