Case Study 2 — The Book That Was One Relationship

A composite failure, set against a real market shock

How to read this case. The market events described in Part 1 are real and public: the 2024 changes to real estate broker practices that followed the settlement of antitrust litigation involving the National Association of REALTORS®, and the affiliated-business structures that RESPA expressly permits under defined conditions. The loan officer, the team, and every figure in Parts 2 through 5 are a clearly labeled composite — assembled from documented industry patterns that recur in every rate cycle and in every market, not from any identifiable individual's production. All figures in the composite are constructed. Nothing here should be read as a benchmark; the arithmetic is the point, not the values.

Part 1 — The real backdrop: 2024 and the reorganization of the buyer side

For most of the modern history of American residential real estate, buyer-broker compensation was commonly communicated through the Multiple Listing Service as an offer from the listing side, and the buyer's agent's fee was, from the buyer's perspective, largely invisible — folded into a price and a process the buyer did not negotiate directly.

That changed. Following the settlement of antitrust litigation involving the National Association of REALTORS® — announced in March 2024, with a settlement payment widely reported at approximately \$418 million payable over roughly four years, and which you should verify at the source rather than take from this book — practice changes took effect in August 2024. The two that reorganized the buyer side:

1. Offers of compensation to buyer brokers were removed from the MLS. Compensation could still be negotiated, but not communicated through that channel.
2. MLS-participating buyer brokers were required to enter into written agreements with buyers before touring a home.

The second-order effects landed squarely on loan officers, and they are the reason this case study opens here.

Buyer-broker compensation became an explicit, negotiated, visible number. Which meant it became a financing question. Whether and how it could be paid by a seller, credited at closing, or treated as an interested-party contribution subject to program limits had direct consequences for cash to close. Each agency addressed the treatment through its own guidance, and the VA — which had historically restricted veterans from paying buyer-broker fees — issued guidance in 2024 addressing the question for VA buyers. Do not memorize any of these treatments from a textbook. Verify the current treatment with your investor, your compliance department, and the applicable agency on every file, because this is precisely the kind of rule that is revised on a schedule.

Buyer agents came under pressure to demonstrate value in writing, to a client who now sees what they cost. An agent who must justify their fee in a signed agreement before showing a house needs tools: a lender who can produce a credible cash-to-close conversation early, who can explain program limits on concessions, and who does not surprise the buyer in the last week.

And agents moved. Periods of structural change in brokerage economics reliably produce brokerage switching, team formation, team dissolution, and consolidation. Agents change firms. Teams break apart and reassemble. That is the ordinary, documented mechanism by which a loan officer's largest referral relationship disappears — not through betrayal, and usually not through anything the loan officer did.

Which brings us to the composite.

Part 2 — The book (composite begins here; all figures constructed)

A loan officer with nine years in the business, well regarded, technically strong, closing steadily. Last full year:

CLOSINGS BY SOURCE — year 1                      [constructed composite]

  A four-agent buyer's team at a single brokerage        29
  Past-client and personal referrals                      8
  Two other individual agents (4 + 2)                     6
  A builder's on-site relationship                        4
  ─────────────────────────────────────────────────────────
  TOTAL                                                  47

  Average loan amount (constructed)                $340,000
  Total volume                                  $15,980,000

Concentration in the single largest source:

$$\frac{29}{47} = 61.7\%$$

Nearly two out of every three files came from one team, and — this is the detail that matters — from one person on that team. The team's lead agent had brought the loan officer in four years earlier. The other three agents on the team used the loan officer because the lead agent told them to. The loan officer had never had a substantive business conversation with any of the three.

The relationship was genuinely good. The loan officer closed the team's files on time, updated without being chased, and called early when something broke. All four of §38.3's currencies were being paid, competently, for four years. This is important: the failure that follows is not a failure of service.

Part 3 — The event

In the spring, the lead agent moved her team to a different brokerage. Her reasons were ordinary and had nothing to do with the loan officer: better split, better support, a brand she preferred in a market that was reorganizing.

The new brokerage had an affiliated mortgage company.

Understand precisely what that does and does not mean, because loan officers get this wrong in both directions. An affiliated business arrangement is expressly permitted under RESPA when its conditions are met: the arrangement is disclosed to the consumer, the consumer is not required to use the affiliated provider, and the only thing of value received is a return on ownership interest. Chapter 24 covers the conditions. So:

• The affiliated lender could not require the team to send it business, and could not require consumers to use it.
• The affiliated lender could be the default — introduced at the first meeting, integrated into the brokerage's systems, present at every sales meeting, and named on every piece of the team's marketing.

"Not required" and "not the default" are different things, and the second one is where the business actually goes. The team did not fire the loan officer. Nobody made a decision. Files simply started arriving somewhere else, because the path of least resistance had moved.

Part 4 — The arithmetic of the year that followed

CLOSINGS BY SOURCE — year 2                      [constructed composite]

  The team (clients who specifically asked for the LO)     6
  Past-client and personal referrals                       9
  Two other individual agents                              7
  Builder's on-site relationship                           3
  ─────────────────────────────────────────────────────────
  TOTAL                                                   25

$$47 - 25 = 22 \text{ closings lost}$$

$$\frac{22}{47} = 46.8\% \text{ decline in units}$$

$$25 \times \$340{,}000 = \$8{,}500{,}000 \text{ in volume, against } \$15{,}980{,}000$$

$$\$15{,}980{,}000 - \$8{,}500{,}000 = \$7{,}480{,}000 \text{ of volume, gone in four quarters}$$

Because loan officer compensation is a function of volume rather than of price — Chapter 26 explains why it cannot legally be a function of price — a 46.8% decline in volume is, at any plausible compensation plan, roughly a 46.8% decline in income. Not a bad year. Half a career's income, removed by an event the loan officer had no part in and no vote on.

Three details in that table deserve attention, because they are where the recovery came from.

The 6. Six clients of the team followed the loan officer rather than the team's new default. Every one of them was a borrower who had had a difficult file — an appraisal problem, a self-employment analysis, a condition that nearly killed the deal — and who therefore had personal evidence of §38.1's certainty rather than a general impression. Easy files produced no loyalty. Hard ones did.

The 9. Past-client referrals went up, from 8 to 9, in the year volume collapsed — a lagging indicator of the prior year's high closing count. This was the only source in the book that did not depend on anyone else's business decisions, and it was the smallest one the loan officer had ever invested in.

The 7. The two other individual agents went from 6 to 7 as the loan officer, suddenly available, gave them the attention the team had been absorbing. That is the uncomfortable finding: capacity had been the constraint on those relationships all along, and nobody had noticed because the team was filling the calendar.

Part 5 — What was visible in advance

Every warning sign was available a year before the event, in data the loan officer already had.

Now the counterfactual, and it is uncomfortable in a specific way.

Suppose the loan officer had adopted a concentration rule — no single source above 35% — and enforced it. At 47 closings:

$$47 \times 0.35 = 16.45 \rightarrow 16 \text{ files from the largest source}$$

$$29 - 16 = 13 \text{ files that would have had to come from somewhere else}$$

Thirteen files from new sources means developing perhaps two or three additional productive relationships, during the busiest years of the loan officer's career, at the cost of hours that were already fully committed. That is the real reason nobody does this. Redundancy has to be built when you are busiest, which is exactly when it feels least necessary and costs the most. The loan officer in this composite was not lazy or foolish. They were fully occupied, and they were being rewarded, every month, for the thing that was going to hurt them.

Part 6 — The rebuild, and how long it took

Seven quarters to get back to 47. Not two. Seven — and only because the past-client base and the public reviews were portable in a way the agent relationship was not.

What the rebuild actually consisted of, in order:

1. Computing the numbers that had never been computed (§38.10's dashboard) — because the first conversation with a new partner needs evidence, and there was none.
2. Building the post-close sequence (§38.7) on every file going forward, which is why past-client referrals kept growing while everything else was flat.
3. Choosing a niche (§38.9) — in this composite, self-employed borrowers, because two of the six clients who followed had been self-employed and the loan officer had done that analysis well. The niche produced nothing for three quarters and then produced steadily.
4. Twelve target partners, not fifty, and the Tuesday block held even in the months when it felt pointless.
5. A different relationship shape with every new partner: an explicit relationship with each agent on a team, not with the team's leader.

Note what is not on that list: any attempt to win the team back, and any attempt to compete with an affiliated lender on price. Both are losing positions, and the hours spent on them are the hours the rebuild needs.

What this case teaches that Case Study 1 does not

Case Study 1 is about the arrangements you must refuse. This one is about the exposure you create by succeeding at the arrangements you are permitted.

There is no rule broken anywhere in Parts 2 through 6. The loan officer did nothing wrong. The lead agent did nothing wrong. The affiliated lender operated inside a structure RESPA expressly permits. And a nine-year business lost nearly half its income in four quarters, because it had one point of failure and its owner was too busy being successful to notice.

The honest caveat, because this chapter does not deal in slogans: concentration is not always a mistake. A deliberate bet on a single builder in a growth market, or on one dominant team, can be the correct strategy — the returns to depth are real, and diversifying a book costs hours that produce nothing for quarters. What separates a strategy from an accident is whether it was decided, and whether a contingency was written down. A loan officer who says "I am concentrated in this relationship on purpose, and if it ends my plan is X, and I will know it is ending when I see Y" has made a bet. A loan officer who discovers their concentration by computing it for the first time after the source has left has not made a bet. They have had one made for them.

Discussion questions

1. Compute your own source concentration for the last twelve months. State the percentage. Then answer: was that number a decision, or a discovery?

2. The six clients who followed the loan officer had all had difficult files. Explain the mechanism in terms of §38.1's argument about what is actually being sold. Then say what that implies about which of your current borrowers you should be investing the most post-close effort in.

3. The counterfactual in Part 5 requires developing thirteen files' worth of new relationships during the busiest year of a career. Design the smallest realistic version of that — what would you actually cut to make room, and what would you accept getting worse?

4. An affiliated business arrangement may not require consumers to use the affiliated provider, yet in this composite the business moved anyway. Explain the difference between a requirement and a default, and say what — if anything — a competing loan officer can legitimately do about a default.

5. Past-client referrals were the only source that grew during the collapse, and they were the source the loan officer had invested in least. Why is this pattern so common? Give a structural answer, not a moral one.

6. The rebuild list deliberately excludes competing on price with the affiliated lender. Argue against that exclusion as strongly as you can, and then say where your argument runs into Chapter 26's compensation rule and Chapter 1's first theme.

7. Write the contingency plan you would attach to your own largest referral relationship: the bet you are making, the plan if it ends, and the specific early signal you would watch for. One page.