Someone asked me Friday on Twitter where he can learn about starting a mortgage company. Here's how we narrowed it down to retail mortgage, which is what I do:
@appflyer just register for the 10 year term at any nearby school of hard knocks. joking. retail or banking/trading?
— Julian Hebron (@TheBasisPoint) August 18, 2012
— Yoenis Cespedes (@AppFlyer) August 18, 2012
This question rattled around in my head all weekend, in fact it's been rattling me for the past 9.5 years since I made a career change into retail mortgage banking.
The reason is because there are two distinct models for building a retail lending operation.
(1) The obvious model is to build a branch. The brainstorm most people have about this model is that they just hire a bunch of loan agents and make a lot of money. Easy breezy. Except for... your heavy fixed cost for things like nice space and top-tier technical, operations and sales talent. Or when you hire loan agents who are good at selling but bad at doing loans. Or they put together one or two good months and think they're worth more than they are. Or their complaining-to-production ratios are off the charts. Or they're so afraid of clients and Realtors that they make promises they know they can't keep then expect the branch to bail them out on a daily basis. Or any number of similar things that make building and running a profitable branch anything but easy breezy.
And this is to say nothing of whether the person building the branch is an actual loan agent themselves. If the branch operator is a loan agent (like me), they have their own time constraints and any time building/running the branch is time away from building/running your own client base. But they also contribute to profitability and they know exactly how to guide their agents through all of the daily deal chaos to keep clients and business partners happy. If they're not a loan agent, they can face credibility issues with their agents because it's very difficult to navigate the minefield of ever-changing (and deal-killing) rules if you're not in the trenches. But a non-producing manager can also be there for their agents at the drop of a hat because they're not bogged down in their own production.
It's also worth noting that most big banks are skewed toward running their retail lending operations with non-producing branch managers, and most midsize firms have a pretty fair mix of producing and non-producing branch managers. And as midsize firms grow, their number of non-producing branch managers tends to grow accordingly.
(2) The less obvious model is to build a production team. In an industry so rooted in tradition, this model is usually overlooked. It's where you are a loan agent that does so much production, you're the same size or larger than most traditional multi-agent branches. In addition to the fixed costs noted in the traditional branch model above, this model can carry even more fixed cost because it accomplishes it's mission by having senior sales and technical team members that are often more weighted on base salaries rather than compensation based on total loan volume (I word it that way instead of saying "commissions" because it more accurately describes Dodd Frank's loan agent compensation rules).
What you get in exchange for this higher fixed cost is direct control over the client experience.
The interesting thing about this model is that, done right, the senior production team members make more than most loan agents anyway. They're just not directly responsible for bringing in business and as such, they don't have the costs associated with generating that business. They just come into work, close loans and keep clients and business partners happy. Obviously that's an oversimplification, but the point is these are career roles if the model is done right.
So how do you do it right? You have to run large volume, period. Minimum of $100m funded loans per year. And how do you get to large volume without compromising the client service? Well, you have to incur more fixed cost, sometimes for years, to build the team until there's enough volume to where it works. Which leads to scenario three.
(3) Hybrid of 1 & 2. This is where a producing branch operator is slowly building model 1 which enables them to build and develop technical and operations team (and of course cover all other fixed cost), and if it works, it also allows for some investment into building model 2.
I'll leave it at that for definition of the models. And I realize model 3 is light on detail. Just writing this brings up years of the same old question---that I ask myself, that my industry mentors ask me, that loan agents I interview ask me:
Which model are you doing? Are you a producer or a branch manager?
The question is always posed just like that. Always. Even by me, until the last 2 years.
The reason people ask it that way is because it's viewed (justifiably) by most as futile to do both. From a technical skills standpoint, and from time commitment standpoint.
Nevertheless, my answer is both.
I've definitely seen the best of both and have made it a point to befriend company colleagues and industry competitors alike who do each extremely well so I can glean the best methods for both.
The reason I believe in both is because I've seen too many give up their own production in favor of model 1, only to see their company sold (or fail, as has been the case since 2006), then they're left with nothing ... except for a multi-year rebuilding process.
I think model 3 is possible with the right team and a firm commitment to ensuring each underlying model (1 and 2) could stand alone if the other went away. But that's a post for another day.
As for you @Appflyer, I hope this helps a bit. This is all from the context of me working for the same midsize mortgage bank for the last 9.5 years---so I've got a great deal of banking infrastructure that I'm working from, and this post doesn't even get into that, this is just some color on the retail part of it. Like I said to you on Twitter: If you're serious, you should try to partner with a local firm. Trying to build a retail lending firm from scratch has too much financial and regulatory risk for the uninitiated---this is intentional by regulators in the wake of the bust and I agree with it. The barriers to entry were too low before and that's what led to most of the trouble we're all still trying to recover from.