Underwriting In A New World Order Of Higher-Longer Rates

Over the years, I’ve been fortunate to work alongside truly remarkable and visionary people who have created some amazing homes and communities. I love this business and respect those who do it so well. Still, we have a bad habit of over-promising and underdelivering when it comes to returns. I believe there are common causes of why that happens.

Intuition Vs. Analysis

Intuition is highly valuable – specifically, when it comes to identifying potential opportunities. But, it's not a substitute for analysis. Great ideas – and bad ones – come from intuition. That is the starting point for digging deep. Too often, when no comps are at hand and our intuition says “this will work,” we don’t do what we could to analyze it. The fact that there are not perfect comps should not be an excuse for not doing the best we can to test the idea and bracket – if not precisely -- a thorough analysis.  

You’re building townhomes in an infill location with no new housing. Okay, what is the relationship in other infill areas of new townhomes to old single-family homes? What is the relationship of projected monthly payments, to rental community rates, etc.?

Don’t leave the market analysis for last. So often, we find projects, think a bit about the high-level conceptual product we would do and then move on to all the “technical” due diligence. Then a market study is ordered and delivered right before an approval decision. By which time everyone involved in the process has gotten attached to doing the project and stopped asking, “should we?”

Inevitably, the “we can kill it selling homes here for $500K” comment early on no longer holds when all the costs are in. Now it needs to be $550K. That's then followed by “well, we were conservative at $500K, $550K is fine.” Except that $550K won’t truly support the absorption we assumed. So, we miss our projections again. Do a little detailed work on the market up front (revenue is the biggest number in the pro forma – just sayin') and have the discipline to not change your mind just because the costs change.  Consider re-designing the product and spec list, not just automatically raising the price to make it work.

What does the site “want to be?” This applies to both land plan and product. That's a sharply different question than "what would work with the land price?" That might be okay -- let’s face it, we respond to what is available. However, first ask, "what does it feel like and what does the data say it should be?" If there is a big gap between the answer to those questions and what would “work,” think twice about whether it will work. Do you just want it to?

Determine “what’s the bet?” What are we really banking on or believing in that leads us to do this deal? You can’t answer this question rigorously unless you are ruthlessly honest in your pro formas. Being [a little] aggressive on a bunch of aspects of the deal because in your heart you believe prices are going up leaves you unsure just how much prices need to go up for you to be right. Somehow, mildly aggressive pro formas are expected and tolerated, but an absolutely straight-up projection of today’s numbers with explicit inflation is considered risky or aggressive.  

It’s nonsense. If you are clear-eyed about the bet and the numbers associated with it, you'll know what you have to focus on in the underwriting. If you think prices are going up substantially, you’d spend less time on the details of current comps, and much more on the hard work of identifying the pipeline and timing of future projects compared with macro projections of demand.

Outdated Data

If you are going to evaluate projects with cash flows and IRRs, you are wasting time unless you actually update those projections periodically, and especially when the project is completed. Nine times out of 10, the business does not do this. The result? The business has no idea how good it is at projecting projects and learns less than it could about looking at projects and choosing risks. Were it not for private equity firms requiring updated cash flows, one wonders if anyone in the business would do it.

The bulk of the industry works off of updated annual business plans. And so, it pats itself on the back when a project exceeds the annual budget, even, if overall, it is behind and has not caught up. You can’t learn to be better at evaluating projects doing that. Better to skip the cash flow projections if you will never update them and face reality.

The biggest loss in this annual plan approach to projects, is failure to truly understand how long things take. If you were going to sum up the homebuilding business it would be, costs will be up, revenue will be up, absorption will be down and the time to first closing will be longer than in the projection.

I can’t tell you how many times I've heard, "we hope to do that. We need to do that. Our goal is to do that."  It doesn’t matter. How long will it actually take?

Having said that, IRR isn’t everything and is nothing more than another benchmark. Comparing a 3-year project that is a 20% IRR with a 12-year project that is a 12% IRR, and stating the first one is clearly superior is nonsense.

Be aware of typical industry margins, and if you’re saying you’re projecting better than that, articulate in detail why/how. "We bought it right” is no answer. Everyone says that. Was it actually "off-market" or "below-market?" Did we really get a great margin on a deal sold to the market by a broker? Is our land plan or product unique compared with other bidders? How was an unusually large up-front margin achieved?

Where does the profit in the project come from? Base house? Lot premiums?  Options? Make a chart and stop to think about it if there is little to no base house profit. At least if there is not a base-house profit, then do better homework confirming the lot premiums, options, etc.

Always “reference class" forecast. This means do not get lost in the current pricing graph by itself. If you project sales of 40 $500,000 homes each of the next three years, develop a graph showing all projects in your submarket (or relevant area) that achieved 40 sales per year for three years running. Do another graph that filters for $500K homes and above, and look at their absorption rates over time.

This is particularly meaningful with large-scale projects, where absorption projections for multiple segments can look reasonable, but taken as a whole, you may find out no large-scale projects have ever maintained this pace in your market. Or maybe they have, but with a much broader price band (more real segments). Just because something has never been done before does not mean it can’t be done, but it’s disturbing how often we are ignorant when we’re saying we can do something unprecedented.

If you are, can you explain why this time is different?

Scrap the sensitivity analysis.  

Five percent, 10%, and 15% down did not come down from the mountain top on stone tablets. Those rules of thumb are way too arbitrary and do not reflect real-world risk. A much more useful analysis is to imagine the project has been completed, you’re talking it over with your business partner and deciding whether on balance you are glad you did it. You might think, "not thrilled," but just past the cut-off for “wish we hadn’t bought it.” Now, what assumptions would create that “just barely good enough” result? If you’re highly confident in at least those assumptions happening, you’ve got a good deal. If you think they're at a material risk of not happening, walk away.

Do not use “the brand” to underwrite.

Builders put way too much stock in the idea of a brand. The holy grail is to be able to charge more for who you are, not what you are producing. It can work, but where it does, it’s typically with widely available consumer products or services, where paying at little extra for reputation is an easy choice to make.

Housing projects are not available anywhere, they are available where we found dirt, and more specifically where we were allowed to build it.

People buy houses using – in concept, if not actually -- a Venn diagram. They look at the intersection of schools, work commutes, family, friends, churches, access to outdoors, amenities, etc. with their pocket book. And for all but the most discretionary buyers, they usually have to compromise on something. So, by the time the consumer considers us, they’ve usually made some compromises and have certain house requirements. Beds/Baths/Garages/Yards, etc. The number of times we have exactly the same home in the same place as our competition is pretty limited.  So, with rare exceptions, the brand (reputation) needs to not be a reason not to buy our home. By the same token, it will not make someone buy our home if the value proposition is not right.

Excessive belief in our brand diminishes the quality of our comparable analysis. It’s a cheat. We look at other builders and come to a conclusion where we think we slot in above them in an area and expect a premium because “we build a better house than Builder XYZ.” But the consumer does the analysis we skimp on. They are looking at 10’ vs 9’ plates, fixtures, more windows vs less, bedroom, bath and garage counts, etc. That’s how they come up with a premium.

But our underwriting is so caught up in technical due diligence that we don’t create a serious feature list based on a competitive analysis that drives a good estimate of product cost. And then we’re surprised when our costs creep up, and so do our asking prices, and absorption ends up less than expected. We never firmly established our true value proposition for the buyer when we underwrote the deal. And you better really know your buyer if you are assuming they are going to give up room count for design.

Metrics That Matter

At the company level, make commitments to measure metrics that are comparable across builders:

  • ROA, not ROE
  • EBITDA, not net profit
  • Unlevered IRR, not Leveraged IRR
  • Revenue/Employee
  • EBITDA/Employee
  • Employees/Community

No two builders calculate Gross Margin exactly the same way, and the type of capital used impacts net margins, return on capital, etc. Unleveraged metrics measure our ability to choose projects and markets, and to execute. Leverage metrics cloud that with the type of capital financing and risk of the capital stack.

Further, 99% of your employees have nothing to do with the amount of risk you take. They should be measured on those things they do contribute to – namely execution. Not whether we choose to use our own equity, others' equity, mezzanine debt, etc. Only the CEO, CFO and a capital provider need be concerned with leveraged metrics.

A few final random thoughts:

  • We’ve all heard overhead is the enemy of profit. But it’s also the enemy of good decisions. The need to cover overhead causes deals to be pursued that would not otherwise be.
  • The first step to profitability is survival.
  • If it sounds too good to be true, it probably is not true.
  • Spend the money on the best consultants and if it’s important enough, consider two. We tend to be remarkably cheap about $25,000 studies with $50 million at stake.
  • We all suffer from confirmation bias. Find someone smart and tough to play devil’s advocate, and take the exercise seriously.
  • Never forget to analyze the project with a simple yellow pad analysis and update it periodically when new info is available during due diligence and planning. Cash flows are very useful, especially if period costs are handled correctly. But they can confuse and sometimes take too long.

Keep updating that yellow pad.