The Broker Times · Market Data

A record share, and a record amount of book on the move

Two data releases landed a day apart. One says the broker channel has never been bigger. The other says the loans inside it have never changed hands faster.

81.6%

Of all new residential home loans written through brokers in the June 2026 quarter — a record.

MFAA / Cotality, 3 Sep 2026

$139.08bn

Value of broker-originated residential loans in the quarter, up $17.49bn year on year.

MFAA / Cotality, 3 Sep 2026

$371bn

Home loan balances that moved between lenders in the year to 30 June 2026.

Elula analysis of APRA data, via The Adviser

664,000

Customers who switched lender over the year — about 1,800 a day, or roughly $1bn a day by balance.

Elula analysis of APRA data, via The Adviser

+13%

Increase in churn on FY25 — $41bn more balance and 48,000 more customers leaving their lender.

Elula analysis of APRA data, via The Adviser

−14%

Year-on-year fall in overall mortgage demand in June 2026, with external refinancing enquiries down 15.1%.

Equifax Consumer Market Pulse, June 2026

Where the churn actually comes from

Refinancing — 63%Rate-driven. This is the slice every pricing alert, fixed-expiry list and rate-review campaign is built to catch.
Property sales — 37%Life-event driven. Almost no retention system has a trigger for it, because there is no rate movement to fire on.

The gap in one line: on those figures, roughly 246,000 of the 664,000 customers who changed lender did so off the back of a property transaction — the one form of churn where you usually get months of warning and no system alert.

The back-book triage

1

Segment by life event

Sort the book by likelihood of a move, not by rate: fixed expiries, loans 3+ years old on likely-outgrown properties, and older investment holdings.

2

Ask the sale question

Put a direct question about selling or moving in the next 18 months into every review, and record the answer on file.

3

Build the perimeter

Agents, conveyancers and accountants hear about a sale at listing. Your CRM hears about it at discharge.

4

Measure your own churn

Discharges over the last 12 months divided by average book. If you cannot split yours 63/37, that is the finding.

The takeaway

At 81.6% channel share, the loan that replaces the one you lose is, on the arithmetic, most likely written by another broker. Retention is no longer a defensive line against the branch network — it is a competitive one, and the biggest untended door is the sale, not the rate.

Sources: MFAA quarterly market share report (research by Cotality), published 3 September 2026; Elula analysis of APRA banking data for the year to 30 June 2026, as reported by The Adviser, 4 September 2026; Equifax Consumer Market Pulse, June 2026, as reported by The Adviser, 20 July 2026. Figures are as published by those sources.

Growth · Market Data

Brokers Hit a Record 81.6% Share. The Same Year, $371bn Changed Lender — and 37% of It Left Through a Sale, Not a Refinance

The MFAA’s record share number and Elula’s churn number describe the same market from opposite ends. Read together, they point at a gap sitting in almost every brokerage’s retention system.

  • Australian market
  • Retention & back book
  • 8 min read

Two data releases landed within about 24 hours of each other this week. Most brokers will have seen the flattering one. The useful one is the other.

Two numbers, one market

On 3 September the MFAA published its quarterly market share report — the 55th consecutive edition, dating back to 2013, with research by Cotality. Mortgage brokers facilitated 81.6% of all new residential home loans in the June 2026 quarter. That is a record. It is up from 81.0% in the March 2026 quarter and 77.6% in the June 2025 quarter. In dollars, broker-originated residential lending totalled $139.08 billion for the quarter, $17.49 billion more than the same quarter a year earlier.

On 4 September, The Adviser reported an analysis by the Australian financial services technology firm Elula of APRA banking data for the year to 30 June 2026. Across that year, $371 billion of home loan balances moved between lenders. 664,000 customers switched — about 1,800 a day, or roughly $1 billion a day measured by balance. Churn rose by $41 billion, or 13%, on FY25, with 48,000 more customers leaving their lender than the year before.

Read on their own, the first is a good-news story about the channel and the second is a lender problem. Read together, they describe one market. And the second number is not only about the banks’ books. It is about yours.

Key takeaway

At 81.6% channel share, the loan that replaces one you lose is, on the arithmetic, most likely written by another broker. And with roughly 37% of FY26 switching tied to property sales rather than refinancing, the biggest gap in most retention systems is that they only watch for rate triggers.

What a record 81.6% actually changes

The temptation with a share record is to treat it as a scoreboard. It is more useful as a statement about who you now compete with.

For most of the last decade, the broker channel’s growth story was share taken from proprietary channels. Branch networks shrank, borrowers got more complicated, and the value of someone who knows forty credit policies compounded. That story is close to finished as a source of growth. When 81.6 cents in every dollar of new residential lending already comes through a broker, there are 18.4 cents left. Even capturing every remaining cent would only add about a fifth again to the channel — and no single brokerage captures a channel-level shift anyway.

The practical consequence is this. When one of your clients refinances away, the loan they move into is, on the same arithmetic, most likely a broker-written loan. Not a branch. Not a call centre. Another broker, with an aggregator panel that looks a lot like yours, running the same servicing calculators.

That reframes retention. It is no longer a defensive line held against the bank’s own retention team offering a rate hold. It is a competitive contest against a peer who is doing to your book precisely what you have been trained to do to theirs.

The 37% nobody has a trigger for

The single most useful figure in Elula’s analysis is not the headline. It is the split. According to the analysis as reported, 63% of churning borrowers were refinancing and 37% of the churn was linked to property sales.

Almost every retention tool this industry has built points at the 63%. Lender pricing alerts. Fixed-rate expiry lists. Annual review calendars. Aggregator flags when a client’s credit file is pinged. Rate-review email campaigns. Every one of them fires on a rate event or a date.

The 37% does not move for a rate. It moves because the client sold something. An upsize after a second child. A downsize once the last one leaves. A separation. A job that relocated to another city. An investment property cashed out after a capital gains conversation with an accountant. An estate being wound up.

On those figures, roughly 246,000 of the 664,000 customers who changed lender last financial year did so off the back of a property transaction. That is not a rounding error, and it has a particular quality: it is the one form of churn where you usually get months of warning and no system alert at all. The intent forms at a kitchen table, sometimes a year out. The signal your CRM finally receives is a discharge authority, by which point the new loan is conditionally approved somewhere else and the conversation is over.

The asymmetry worth sitting with: the 63% is contested with pricing, which you do not control. The 37% is contested with relationship and timing, which you do. Most brokerages have automated the half they cannot win on and left the half they can to memory.

Fewer enquiries, same volume of book on the move

The backdrop matters. Equifax’s Consumer Market Pulse for June 2026, as reported by The Adviser in July, put overall mortgage demand down 14% year on year. First home buyer enquiries were down 17.2%. Demand from 26- to 35-year-olds was down 18.2%. Refinancing enquiries to a different lender were down 15.1%; refinancing with the same lender was down 10.4%.

Kevin James, Equifax’s Chief Solutions Officer, described the shift in that release as consumers moving from proactive risk management to “a far more conservative, defensive approach”.

These are different measures and they do not net against each other — one counts applications in a single month, the other counts settled balance movement across a full year. But the direction is worth holding in the same frame. The pool of fresh enquiry is contracting while an enormous volume of existing balance keeps changing hands.

In a market shaped like that, the cheapest loan you will write next year is one you already own. Replacing a lost $700,000 loan costs you the acquisition, the fact-find, the assessment, the submission and the settlement — and, depending on how soon after settlement it discharges, may expose you to commission clawback. Clawback terms vary by lender and by aggregator agreement; if you cannot state yours from memory, that is worth an hour with your aggregator before you plan next year’s growth targets.

Where the balances are landing

Elula’s analysis also mapped who won and lost. As reported, Macquarie Bank grew at roughly 3.9 times the system rate, adding about $39 billion; CBA grew at around system; NAB, Westpac and ANZ each grew below system; and BOQ and Bendigo and Adelaide Bank contracted.

Josh Shipman, identified by The Adviser as Elula’s chief executive, said: “At this point in time, we are experiencing an almost perfect storm that is driving customer churn in the banking sector. Churn is accelerating due to cost-of-living challenges; government policy changes to negative gearing; rising interest rates; and falling property prices.”

For brokers the read is less about league tables than concentration. When growth clusters in a small number of lenders, panel behaviour follows: turnaround times stretch at the winners, credit appetite loosens at the ones losing balance, and the spread between the fastest and slowest lender on your panel widens. That is a servicing and expectation-setting issue on live files, not just a market observation.

The 90-minute back-book triage

This is a defined piece of work, not a mindset shift. Block out ninety minutes and run the following.

Step 1 — Segment by life event, not by rate

Pull your trail book and build three lists: fixed rates expiring in the next twelve months; loans settled three or more years ago on properties the client may well have outgrown; and investment holdings bought before the current cycle. The first list is your rate risk. The second and third are your sale risk, and they are the ones that have no automated trigger behind them.

Step 2 — Put the sale question into every review

Four questions surface most of the 37% before it becomes a discharge:

  1. Is this still the right home for the next three years?
  2. Has anything changed at work — role, employer, or location?
  3. Is there any property you are thinking about selling in the next eighteen months?
  4. Has anyone in the family started talking to you about buying?

Ask them, and write the answers into the file. An unrecorded answer is not an early-warning system; it is a memory.

Step 3 — Build the perimeter

Agents, conveyancers, buyer’s agents and accountants learn about a sale at listing or earlier. Your CRM learns about it at discharge. The referral relationships worth deepening this quarter are the ones that close that gap — not because they send you new leads, but because they tell you about your existing clients first.

Step 4 — Measure your own churn rate

Most brokerages track settlements closely and their own attrition barely at all. Calculate discharges over the last twelve months as a share of your average book, then try to split that number the way Elula splits the market: how much left on a rate, and how much left on a sale? If your data cannot answer that question, that is itself the finding, and fixing the tagging is the first job.

Step 5 — Decide what a “save” is worth

Before you spend on retention, price it. Work out the trail value of a typical loan over its remaining expected life, then compare that with the cost of acquiring a replacement at current lead prices. That comparison usually settles the argument about whether the review calls are worth the diary space.

Retention is still regulated activity

A word of caution, because retention campaigns can drift. A conversation with an existing client that ends in a recommendation is a recommendation, whether the recommendation is to move or to stay put.

ASIC’s Regulatory Guide 273 Mortgage brokers: Best interests duty explains, in ASIC’s own words, “what ASIC looks for when we assess compliance with the best interests obligations in Pt 3-5A of the National Consumer Credit Protection Act 2009”. The reasoning behind a recommendation to stay with an incumbent lender belongs in the file with the same care as the reasoning behind a recommendation to move.

If you are planning a campaign that nudges a segment of your existing clients toward a particular lender or product, the framing and the record-keeping should be signed off by your licensee’s compliance team rather than by a marketing calendar. This article is general information only — confirm how these obligations apply to your business with your licensee or compliance adviser.

What to review this week

  • Your own twelve-month churn rate, split between rate-driven and sale-driven departures — or the tagging change needed to produce it.
  • The list of loans three or more years old where the client’s circumstances have most likely changed.
  • Your review script, and whether it asks a direct question about selling or moving.
  • Your clawback exposure by lender, confirmed with your aggregator rather than from memory.
  • Whether your file notes would show a reviewer why a client was advised to stay put.

What to watch next

Three things will tell you whether this week’s pairing was a moment or a trend. The September 2026 quarter MFAA report, due later in the year, will show whether 81.6% is a ceiling or a step. The next Equifax quarterly pulse will show whether external refinancing enquiries keep falling while balances keep moving — the tension at the centre of this story. And the monthly APRA ADI statistics will show whether the concentration Elula identified is holding, which is what determines how your panel behaves on live files.

The record share is genuinely worth celebrating; the channel earned it. But a record share of new lending and a record amount of existing lending on the move are the same market seen from two ends. Growing from here means winning a larger portion of a book that is changing hands faster than ever — and, first, holding onto your own.

Questions brokers are asking

Does a record 81.6% market share mean there is less room to grow?

At a channel level, yes — the share available from proprietary channels is now about 18.4%, so channel-level growth is close to its natural limit. At an individual brokerage level it means something different: the growth available to you increasingly comes from books already written by other brokers, and from holding your own, rather than from converting bank-direct borrowers.

Why does the split between refinancing and property sales matter so much?

Because the two are lost in completely different ways. Refinancing churn is triggered by rates and dates, which most retention systems already monitor. Sale-driven churn — about 37% of FY26 switching on the figures reported — is triggered by life events that no pricing alert can detect. It is the larger blind spot precisely because it is quieter.

Isn’t falling mortgage demand a reason to focus on new leads rather than the back book?

It is a reason to do the arithmetic. With overall mortgage demand reported down 14% year on year in June 2026, acquisition is getting harder and more expensive at the same time as an enormous volume of balance is changing hands. Comparing the cost of a replacement client against the remaining trail value of an existing one usually settles where the next hour of effort should go.

Is there a compliance risk in running a retention campaign?

Retention conversations that end in a recommendation are recommendations. ASIC’s RG 273 sets out what ASIC looks for when assessing compliance with the best interests obligations in Part 3-5A of the National Consumer Credit Protection Act 2009. The reasoning for advising a client to stay should be documented as carefully as the reasoning for advising them to move. Confirm the specifics with your licensee or compliance adviser before launching a campaign.

Breaking news for modern brokers

Market data, lender policy and compliance developments, read through what they actually change on your files.

More at The Broker Times →

Sources: MFAA quarterly market share report for the June 2026 quarter, research by Cotality, published 3 September 2026. Elula analysis of APRA banking data for the year ending 30 June 2026, as reported by The Adviser, 4 September 2026 — the churn figures, the 63/37 split, the lender growth comparisons and the Josh Shipman quote are as reported by that outlet. Equifax Consumer Market Pulse for June 2026, including the Kevin James quote, as reported by The Adviser, 20 July 2026. ASIC Regulatory Guide 273 Mortgage brokers: Best interests duty. Figures are as published by those sources at the time of writing.

Interactive · Self-assessment

How exposed is your back book to sale-driven churn?

Six questions about how your brokerage actually runs its book. Answer honestly — the point is to find the gap, not to score well. Nothing is sent anywhere; this runs entirely in your browser.


0 of 6 answered
1. Can you produce your own twelve-month churn rate today?

Discharges over the last twelve months as a share of your average book.



2. Can you split that churn into rate-driven and sale-driven departures?

The market split reported for FY26 was about 63% refinancing, 37% property sales.



3. Does your review script ask directly about selling or moving?

A specific question about plans over the next eighteen months, with the answer recorded on file.



4. Do you have a segment for loans three or more years old on likely-outgrown properties?

The clients most likely to sell, and the ones no rate alert will flag.



5. Would a referral partner tell you a client had listed, before the discharge arrived?

Agents, conveyancers and accountants usually know months earlier than your CRM does.



6. Can you state your clawback exposure by lender without checking?

Terms vary by lender and by aggregator agreement — confirm yours rather than assuming.




A note on scope: this is a prompt for your own planning, not advice. Any retention activity that ends in a recommendation is still subject to your obligations as a credit representative — check how they apply to your business with your licensee or compliance adviser before you run a campaign.

Disclaimer: This article is for general information and professional development purposes only. It does not constitute legal, compliance, or financial advice. Brokers should consult their aggregator's compliance team and, where required, seek independent legal advice regarding their obligations under the National Consumer Credit Protection Act 2009 and ASIC’s responsible lending guidelines.