Agent-to-Broker Migration Economics: What It Costs and What It Returns

A buyer choosing between a broker and an agent is asking which channel to purchase through, a different question answered in insurance broker vs agent for commercial cover. An individual weighing three rungs against each other has yet another question, which the advisor career ladder takes apart from the person's side.

This post is the practitioner's version, at firm level, of one move: you sell insurance today under somebody else's authorisation, and you are considering standing up a licensed broking entity of your own. It is a balance-sheet decision, and it most often goes wrong when it gets made as a status upgrade with the economics reverse-engineered afterwards.

Two things changed on 5 February 2026, when the Sabka Bima Sabki Raksha (Amendment of Insurance Laws) Act, 2025 came into force. Intermediary licences became perpetual, turning the registration from a renewable permission into a durable asset with a resale value. And 100 percent FDI in intermediaries opened, putting capital in the market that is actively looking for licensed entities to buy. Both make the far end of the migration worth more than it was in January 2026. Neither makes the near end cheaper.

The IRDAI (Insurance Brokers) Regulations, 2018 set the requirements, and they are requirements of an entity rather than a person. Nothing about this move is a qualification you sit for.

Capital that stays in. The 2018 Regulations set minimum net worth and paid-up capital by broker category. For a direct broker those thresholds stand at INR 50 lakh net worth with a related minimum paid-up capital floor of INR 75 lakh, and higher figures for reinsurance and composite categories. Read the current instrument before budgeting against them, since a recalibration has been widely anticipated, as analysed in the broker net worth question. The operative words are net worth: not a fee, but capital that has to be there and stay there, a test the firm keeps passing on an ordinary Tuesday, because there is no renewal any more.

A Principal Officer. The firm must have one, meeting the qualification and experience requirements in the regulations. Most discover that this, not the money, is the binding constraint. Either you are that person, and your time splits between producing and running a regulated entity, or you hire one, and a senior fixed salary lands on a P&L that previously had almost no fixed cost.

Professional indemnity. The 2018 Regulations require the firm to hold and maintain professional indemnity cover, sized by reference to its remuneration with a floor beneath it. An agency does not carry this cost because it does not carry the exposure; the principal does. Once you advise in your own right, you are the one being sued for the advice.

The entity and everything trailing it. A company, audited accounts, returns, an IT and data posture, an AML process, and a compliance function that exists in months with no revenue to pay for it. The draft IRDAI (Insurance Intermediaries) (Amendment) Regulations floated in June 2026 would add a separate financial-statement schedule for intermediation revenue, audited filings to IRDAI by 30 September, and website publication. That draft is not final, but it indicates the direction of the fixed cost, which is up. The application itself then costs time: the stretch between committing the capital and placing the first policy in your own name.

Everything above is knowable in advance. The book question decides whether the model works: when you stand up a broking entity, does your book come with you?

Start with your current arrangement, and read it rather than remembering it. What it says about client data, solicitation after termination, renewal rights on business written during the tie, and notice, is the actual answer. Firms discover clauses at the point of resignation that they could have read four years earlier.

Then separate three things that get conflated:

1. The relationship. Who the client calls. This travels with the person, and no clause changes it. It is the only asset here that is genuinely yours.
2. The record. Who bought what, when it renews, what was claimed, what was promised. This travels only if you have it. If your book lives in the principal's system and you leave with a memory and a phone, you have the relationship without the information to service it.
3. The policy. The in-force contract sits with the insurer and names an intermediary. Moving it is not your decision. It is the policyholder's, at renewal, with paperwork.

So migration is a renewal-cycle exercise, not an event. Your book arrives over twelve months as each policy comes up, client by client, each of whom has to choose you again in a slightly different form and, at a corporate account, get that choice through their own procurement. Some fraction will not. Model the fraction honestly, because the difference between assuming 90 percent and 65 percent is usually the difference between a viable plan and a fantasy.

Assume the book travels at whatever rate you modelled. What does the same business earn on the other side? The temptation is to compare your commission as an agent against gross brokerage as a broker and treat the difference as the prize. That is not the bridge. The bridge has four segments and two are negative.

Segment one: what the same policy pays. Channel economics differ, and levelling those differences is among the stated aims of the reform direction currently under consultation, as set out in the channel arbitrage analysis. Read it as a risk: a migration case premised on the broker channel paying better for the same product rests on a differential the regulator has said it is looking at. Nothing has been decided, and a consultation paper on commission reform was expected by end-July 2026 but had not been published as at the time of writing. Build a model that survives the differential narrowing.

Segment two: what widens. The real gain, and not about rate. As an agent you place what your principal offers. As a broker you place across insurers in your own right, so you can quote risks you currently decline, keep clients whose needs outgrew your principal's range, and compete for accounts needing a market exercise rather than a product. That is new revenue, not repriced revenue, and the segment most migration models understate.

Segment three: the fee line. The 2018 Regulations permit a broker to charge clients fees for risk management services and claims consultancy under written agreements, for work distinct from what placement brokerage remunerates. An agency has no such line. A firm reaching 8 to 15 percent of revenue from fees within two years has income that does not move when commission rules move, which given segment one is the most defensible part of the bridge.

Segment four: what it costs to earn any of it. Principal Officer, professional indemnity, audit, compliance, technology. Subtract it before you compare.

Build the model on the cash, over three years, monthly for the first twelve.

The outflow is front-loaded and certain. Capital committed at the start, and not recoverable while the firm operates, because net worth is a continuing test. Then the annual fixed base from the day the entity exists: Principal Officer, professional indemnity, audit, compliance, technology, office. That base does not scale down in a bad quarter, which is the line most often skipped, because an agency's cost base is almost entirely variable and the practitioner has no muscle memory for a fixed one.

The inflow is back-loaded and uncertain. Nothing during the application period. Then a partial book arriving across a renewal cycle at your modelled retention. Then, later than you think, the widened placement and the fee line, both needing a capability that does not exist on day one.

The shape is a trough, so the payback question is two questions: how deep does it get, and can you fund the bottom. A firm that breaks even in month twenty-two but runs out of cash in month fourteen does not reach month twenty-two.

Four tests to run before you believe your own model:

• The retention test. Rerun at 65 percent first-year book retention. If it only works above 85 percent, you have modelled a hope, not a business.
• The differential test. Rerun with the channel rate advantage removed. What is left is the part that does not depend on a differential under consultation.
• The Principal Officer test. If you are the Principal Officer, rerun with your production halved, because running a regulated entity does not happen in the evenings. If you are hiring one, rerun with that salary starting three months before your first placement, because it will.
• The bad-year test. Apply the new fixed base to the worst twelve months your practice has had. That is the question fixed cost asks, and it only asks at the worst time.

One threshold is worth reading correctly. Broker viability at scale sits at roughly INR 25 crore to INR 50 crore of annual revenue for a firm competing on general strength, with smaller firms surviving on niche depth and shared cost bases instead. That is not an entry requirement. It describes the small-generalist position, which is exactly the position a migrating practitioner should not be planning to occupy.

The two changes from the Sabka Bima Act matter mainly at exit, and exit is a legitimate reason to make this move.

Perpetual licences. A broking registration no longer expires, which is the precondition for it having a market value at all: nobody pays a real multiple for a permission that has to be re-earned on a cycle. For a practitioner who expects to sell in seven years rather than operate for twenty-five, the terminal value in the model stopped being speculative.

The corollary is not comfortable. With no renewal cycle, enforcement is the only route by which a licence is lost. The regulatory downside of the entity you are creating is concentrated entirely into conduct, a different risk profile from anything an agency has carried.

100 percent FDI in intermediaries. Foreign capital can now own an Indian intermediary outright, and the effect is straightforward: more buyers with more capital are looking at broking firms, and mid-market firms are the ones being looked at, as analysed in 100% FDI for insurance intermediaries.

Both facts improve the terminal value of what you are building, and terminal value is a real part of the return. Neither does anything about the trough. A practitioner who migrates because licences are worth more now, without funding the eighteen months in between, has correctly identified an opportunity and incorrectly assumed they will be there for it.

The migration is wrong more often than the enthusiasm around it suggests, and the signals are legible in advance.

It is probably a no if: your book sits in products your principal happens to be strong in, so the widened placement segment is thin. Your clients are relationships with you but records with your principal. You cannot name the Principal Officer, or the answer is you and you cannot say what stops being done. You cannot fund eighteen months of a fixed base. Or the case only works with the differential intact.

It is probably a yes if: you are consistently declining or brokering out business your principal's range cannot hold, so there is quantified revenue waiting on the other side of the licence. You hold the record as well as the relationship. You have a niche where insurer relationships are earned by expertise rather than volume. You can name the Principal Officer and fund them. And your model survives all four tests.

And there is a middle that gets ignored. Joining an existing broking firm as a producer with a book, merging into a small firm that holds the licence but needs distribution, or affiliating with a network all put you on the broker side of the line without asking you to personally fund a trough.

So this is not a question about what you deserve to be called. It is whether there is quantified revenue on the other side of the licence, whether your book will walk across, and whether you can pay a fixed cost through the winter while it does. Three questions, all answerable with numbers before any capital moves.