Distributor Margin: You Are Not Setting a Percentage, You Are Buying a List of Things

Benchmark tables answer how much. The only part of a distributor margin you can actually decide is what it buys.

Every brand that appoints a partner asks the same question early, usually in the first or second meeting, and almost always in the same form: what margin is normal?

It is a reasonable question and it has no answer. Not a contested answer, or a regionally variable answer. No answer, in the way "what is a normal salary" has no answer until you have said for what work.

The benchmark tables are the proof. Search the question and you will be given ranges — a band for FMCG, a wider band for electronics, another for cosmetics — wide enough that both twenty per cent and forty per cent sit comfortably inside them. A range that contains both of your candidate numbers has told you nothing. It cannot, because the number is not the variable. It is the output of variables nobody listed.

What follows is the version of the question that can be answered: margin is a price you pay for a specific list of things, and if you cannot write the list, you are not negotiating a margin — you are funding an unnamed cost that both sides will resent within two years.

Why the benchmark is structurally useless

The tables are not wrong. They are describing something other than what you are deciding.

A published margin band aggregates partners who do completely different work. Inside any single band you will find a partner who imports, clears, warehouses, holds receivables for ninety days, employs a retail-facing sales team, funds in-store display, and registers your product dossiers — and another who takes a container, invoices two wholesalers, and touches nothing else. Both are "distributors". Both show up in the same industry average.

So the number you copy from the table carries an assumption about which of those two you are talking to, and you did not make that assumption. Whoever wrote the table did.

There is a second problem with using the band as a target. Once a margin percentage becomes the thing being managed, it stops describing the thing it was supposed to describe — the familiar failure Goodhart identified and Strathern later put plainly: a measure used as a target ceases to be a good measure (Goodhart's law). A partner held to a margin number will defend the number, including by removing the activities the margin was supposed to pay for. The margin survives; the work does not.

What the partner actually earns, which is never the margin

The partner's return is not the margin. It is closer to margin × stock turns − the costs they absorb.

All three terms move, and the brand usually only sees the first one.

An illustration with invented numbers, because our own figures do not belong in a public article. Two partners, both on a 25% gross margin on the same product:

• Partner A turns stock four times a year and absorbs import duty, clearance, warehousing, and a two-person retail team. The 25% is being spread across real operating expense and four cycles of working capital.
• Partner B turns stock once a year, warehouses nothing because they drop-ship to two wholesalers, and absorbs no selling cost at all.

Partner B is in a far better business at the same percentage, and will still ask for more, because their alternative use of the same working capital is another brand's container. Partner A may be close to break-even and never say so.

The consequence for us is not fairness. It is prediction. These two partners fail in opposite directions when volume disappoints. Partner A, carrying fixed selling cost, sees profit collapse faster than sales fall — high operating leverage, the sensitivity of profit to volume when costs are fixed. That partner comes to us asking for a lower supply price, which is the conversation in What to Do When Your Distributor Asks for a Lower Price. Partner B has almost no fixed cost to defend; when volume disappoints they simply stop reordering and go quiet, which is the conversation in The Month-Seven Test.

Same margin. Same disappointing quarter. Two completely different symptoms, and two different correct responses. You cannot tell which you are holding by looking at the percentage.

Same percentage, different businesses. What fills the bar decides how the partner behaves when volume disappoints. Operating leverage is a standard management-accounting concept.

So write the list: what is this margin buying

Here is the move that changes the negotiation. Stop quoting a percentage and start quoting a percentage against a list.

The list has four kinds of item, and they are worth separating because they behave differently:

• Costs the partner absorbs that you would otherwise pay. Import duty, clearance, local registration and dossier maintenance, warehousing, last-mile delivery, local returns handling.
• Working capital the partner provides. Stock held, and receivables carried for retailers — often the single largest and least discussed item. A partner financing ninety-day retail terms is lending money on your behalf.
• Selling capacity the partner employs. Named headcount facing retailers, merchandising, order-taking, category conversations. This is the item that disappears first when margin is squeezed, and the only item whose absence you will not notice for two quarters.
• Market-specific investment the partner makes that has no value outside this relationship. Display furniture, a trained team on your category, listings negotiated in your name.

The fourth category needs the most care, because that is the investment a rational partner will decline to make unless the margin funds it and the relationship protects it. Klein, Crawford and Alchian's hold-up problem — once an investment is specific to a relationship, its value can be appropriated by the other side after the fact — predicts under-investment in exactly this category. Williamson's asset specificity is the property that creates the exposure.

Under-investment is not visible as under-investment. It is visible as a partner who orders on time, pays on time, and sells nothing — the failure mode described in Distributor Performance Evaluation.

Practical form: a one-page annex to the agreement, listing which side carries each of the four categories. Not amounts. Just which side. Most agreements never state it, and both parties then assume in their own favour.

Unspecified margin is the margin that gets cut every year

If the margin is attached to a list, a request to reduce it becomes a request to remove something, and the conversation is about which item. That conversation is survivable and often productive.

If the margin is attached to nothing, it reads — to a procurement department, to a new finance lead, to a successor on either side — as pure cost with no stated function. It will therefore be challenged every budget cycle, from both directions. We will be asked to justify why they get 25%. They will be asked by their own management why they accept 25%. Neither side can answer with a document, so the answer becomes history: it has always been this.

History is a weak defence, and it loses to a spreadsheet. The partner's new commercial director will arrive with a benchmark table, which, as established, contains any number they want it to contain.

The asymmetry worth noticing: the list protects the partner at least as much as it protects us. That makes it an unusually easy thing to propose, and it is the reason to propose it at appointment rather than during the first squeeze. Introduced early, it is process. Introduced mid-conflict, it is leverage, and it will be read that way.

A margin has to be protected or it will not be spent

Even a correctly sized margin funding a correctly specified list will fail if someone else can sell the same product without carrying any of the list.

This is the oldest result in the subject. Telser's argument for resale price maintenance was not about prices at all: a dealer margin exists to pay for dealer services — demonstration, stock availability, after-sales, category education — and those services are vulnerable to free-riding, because a competitor who provides none of them can undercut the dealer who does and still capture the sale the services created. If free-riding is unchecked, the rational dealer stops providing the service while keeping the margin, or stops carrying the product.

For a brand appointing distribution partners across markets, free-riding has two everyday faces:

• A second partner or reseller inside the same market who sells without carrying the selling cost. This is what territory and exclusivity clauses are actually for — not to reward loyalty, which is the usual framing, but to make a funded margin spendable. Exclusive or Not covers the decision of whether to grant it.
• Our own product arriving from outside the market at a price the funded partner cannot match. That is Parallel Imports Are a Price Gap, Not a Piracy Problem, the version where the free-rider is created by our own pricing rather than by a competitor.

A legal note that is not legal advice: resale price maintenance — a supplier dictating the price at which a buyer resells — is restricted or prohibited in many jurisdictions, including on maximum prices in some of them, and the rules differ by market. Telser's point is useful here as an explanation of why a margin needs protecting, not as a recommendation to set resale prices. The protections available to you are generally structural: territory and channel definition, product differentiation, and governance of your own promotional calendar.

Protected and unprotected margin. A reseller who carries none of the list can still reach the same shelf. Concept: Telser (1960).

The margin a partner accepts tells you which business they think they are in

Ask two candidates for the same market to price the same four categories, and you will get two structurally different answers.

One asks for a high margin and declines to name any item in the third or fourth category. They are pricing a trading business: buy, resell, carry no fixed cost, remain free to switch brands. That is a legitimate business and sometimes the right partner — for a market you are testing, or a product that sells itself.

The other accepts a thinner margin and names headcount, display investment, and registration work they will carry. They are pricing a market-building business, and the thinner margin is only acceptable to them because they expect the volume curve to justify it later.

Spence's point about signalling applies: a signal is informative only when it costs something to send. A number in a proposal costs nothing. A commitment to carry two named salespeople costs something, and so it carries information the proposal document cannot. This is the same test applied to track record in How to Evaluate a Distributor Before You Sign — promises are free, investments are not.

The failure mode to watch for is the third pattern: a partner who asks for the market-building margin and names nothing. That combination is not a negotiating position. It is an answer.

Margin asked against commitments named. Where a candidate lands tells you more than the number they quote. Concept: Spence (1973).

What to do instead of looking up a benchmark

Build the partner's income statement before you quote a number. Roughly is enough.

Take the retail price in that market, work back through retailer margin, local taxes, logistics and duty to a landed cost. Then put the four categories above against whichever side is carrying them, and ask what is left for the partner after the costs they absorb, multiplied by the turns they realistically achieve.

Two things usually fall out of that exercise.

First, you discover that the question was never "what margin" but "who carries which cost", and that several of those costs are cheaper on your side than theirs, or the reverse. Moving one item across the line is often worth more to both parties than moving the percentage.

Second, you find the floor below which the partner cannot fund selling capacity at all. That floor is the number worth knowing, because a margin below it buys you an order-taker regardless of what the contract says about marketing obligations. Arnold's long-standing argument about international distribution is relevant here: treating partners as a low-cost way to enter a market, rather than as something requiring investment and direction from the brand, is the recurring error — and a margin set from a benchmark table is precisely that error expressed as a number.

The percentage you arrive at this way will probably sit inside the published band. That is fine. The difference is that you will know what it is paying for, which means you will know what to remove if it ever has to come down.

The five-line version

IF YOU READ NOTHING ELSE

01  "What margin is normal" has no answer — any benchmark band that contains both of your candidate numbers has told you nothing.

02  The partner's return is margin × turns minus the costs they absorb, so one percentage describes two businesses that fail in opposite directions.

03  Quote the margin against a written list: costs absorbed, working capital provided, selling capacity employed, market-specific investment made.

04  A margin attached to nothing gets challenged every budget cycle; a margin attached to a list turns that into a conversation about which item to remove.

05  A funded margin still has to be protected from free-riding, and what a partner accepts against the list is the cheapest signal of which business they are in.

Your turn

For your largest market: can you write, in four lines, what the partner's margin is buying — and would they write the same four lines?

Do you know how many times a year that partner turns your stock? If not, you do not know what they earn.

Which of your partners carries retail receivables, and have you ever priced that as the loan it is?

If your partner's margin had to fall two points next year, which item on the list would you remove — and have you ever said so out loud?

Of your last three appointments, which named a specific commitment in exchange for a thinner margin, and did that correlate with how they performed?

Sources

• Lester G. Telser, "Why Should Manufacturers Want Fair Trade?", Journal of Law and Economics 3, 1960, pp. 86–105.
• Benjamin Klein, Robert G. Crawford & Armen A. Alchian, "Vertical Integration, Appropriable Rents, and the Competitive Contracting Process", Journal of Law and Economics 21(2), 1978, pp. 297–326.
• Oliver E. Williamson, "Transaction-Cost Economics: The Governance of Contractual Relations", Journal of Law and Economics 22(2), 1979, pp. 233–261.
• Michael Spence, "Job Market Signaling", Quarterly Journal of Economics 87(3), 1973, pp. 355–374.
• Goodhart's law — Charles Goodhart, 1975; the widely quoted formulation is Marilyn Strathern, "Improving Ratings: Audit in the British University System", European Review 5(3), 1997, pp. 305–321.
• David Arnold, "Seven Rules of International Distribution", Harvard Business Review, November–December 2000 — cited for its general argument only.
• Operating leverage is a standard management-accounting concept and is not attributed to a single author.

Nothing in this article is legal advice; resale pricing rules differ by market and change. All figures are illustrative. Countries, companies, brands, individuals and proprietary figures have been removed; only the structure remains.

Written from the brand side of this table — twenty-two years appointing and managing distribution partners across Asian markets, across categories and price tiers.

Previously in this series: How to Evaluate a Distributor Before You Sign · Exclusive or Not · When to Stop Selling Cross-Border and Set Up Locally · What to Do When Your Distributor Asks for a Lower Price · When Is a Market Ready for a Local Launch? · Marketing Money in a Distribution Agreement · When Your Distributor Stops Paying · The Month-Seven Test · How to Manage the Distributors You Never Visit · Parallel Imports Are a Price Gap, Not a Piracy Problem · Stop Pricing One Export Market at a Time · Your Marketplace Store Is Four Rights, Not One Decision · Distributor Performance Evaluation · Distributor Onboarding