Good morning. Selling more because you opened more places to sell is not the same thing as selling more. Toll Brothers — which calls itself "the nation's leading builder of luxury homes" — published both numbers in the same release, and they point in opposite directions.
Top story — Growth, and the number underneath it
Toll reported its quarter ended 31 July. The chief executive's line was this: "We also grew net signed contracts by 5% year over year." That is true — 2,508 contracts against 2,388.
Now the other number in the same release. Toll ended the quarter with 471 selling communities, against 420 at the same point a year earlier. Fifty-one more places to sell at quarter end, and a hundred and twenty more orders in the quarter.
Toll publishes the ratio itself, and does not bury it. Its printed metric "Net Signed Contracts per Community" fell from 5.6 to 5.4. More locations, fewer orders at each one.
Meanwhile the work going out the door fell: 2,662 homes delivered against 2,959, down ten percent, on home sales revenue of $2.65 billion against $2.88 billion. And the cost of running the business went the other way. Toll prints selling, general and administrative expense at $266.1 million, or 10.0% of home sales revenues, against $253.7 million, or 8.8% — its own overhead-intensity number, up 1.2 points.
Do the division on those two pairs of printed figures — $266.1 million over 2,662 homes, $253.7 million over 2,959 — and it comes out at roughly $100,000 of overhead per home delivered, against about $85,700. Up nearly seventeen percent, per house, in a year. It is a rough ratio, and worth saying so: that overhead also supports the contracts Toll signed and the communities it opened, not only the homes it closed. The direction is the point, not the decimal.
None of which says Toll is in trouble. It made $280.1 million in the quarter, against $369.6 million a year earlier. What it says is more useful than that. This is what it looks like, in public, when the capacity line and the output line stop moving together — more places to sell, fewer homes delivered, and an overhead line that went up anyway. It is the same shape as a shop that adds two vans and a dispatcher and finds the revenue line no better and the overhead line permanent.
Quick hits
The margin fell, and write-downs are not the reason. Home sales gross margin — the GAAP figure, not the adjusted one Toll leads with — came in at 23.9%, against 25.6% a year earlier. The easy explanation would be impairments, except Toll prints them and they went down: pre-tax inventory impairments inside home sales cost of revenues were $17.7 million this quarter against $23.3 million. Interest inside that same cost line was 1.1% of home sales revenues in both years. Toll's own adjusted margin — a non-GAAP measure that strips both of those out — also fell, 27.5% to 25.6%. The release does not say what drove it, and neither will we.
One flat national number, five regions going different ways. Backlog finished at 5,312 homes against 5,492 — down about three percent, which reads like nothing happened. Underneath, by units: North 1,073 against 907. South 1,742 against 1,659. Mountain 1,265 against 1,317. Mid-Atlantic 708 against 856. Pacific 524 against 753 — down roughly thirty percent. If you read a national headline about new-home backlog and adjusted your hiring, you read the wrong number. (Toll's regions are named groups of states, listed in the release. It publishes no margin, community count or cancellation figure by region.)
Three average prices, and only one of them is about today. Homes delivered in the quarter averaged $996,400, up from $973,600. Homes in backlog averaged $1,174,400, up from $1,161,000. But the contracts signed in the quarter averaged $1,002,900, down from $1,010,100. Deliveries are last year's pricing arriving; backlog is a stock, and its average moves with whatever mix is left in the book. Only the third number is today's market. Two of the three are up, and the one that is down is the one that counts.
The Tip — Divide it, don't total it
Take everything you spend that is not on a job — office, dispatch, non-producing trucks, software, advertising, your own time — and divide it by the number of jobs you actually completed last quarter. Do it again for the same quarter last year. Then divide the same total by vans, or by locations if you run more than one. Totals always go up and always look like growth; the per-unit number tells you whether the extra capacity paid for itself. Toll publishes half of this about itself every ninety days — orders per community — and lets anyone watch it fall. The overhead-per-job half it does not publish, which is why we had to do the division above. Almost no shop does either one.
Tool watch — [Not sponsored]
The release is free on SEC EDGAR, and the useful part is at the back, not the front. Skip the headline bullets and go to the five-segment tables — revenues, contracts and backlog, each split by region with units, dollars and average price. Three tables, one page, and they will tell you more about where residential work is going than any summary of them will. The permanent link is in Sources at the bottom of this email; it is an EDGAR archive address, which does not move when a company redesigns its investor page. One note if you go looking: the release marks its own figures "(Unaudited)" and says they are "subject to finalization of the Company's regulatory filings, related financial and accounting reporting procedures and external auditor procedures."
That's the callout. See you next time.
The Callout — the business brief for the trades. We label every commercial placement: [Sponsor], [Affiliate], or [Not sponsored]. Today's Tool watch is our own honest read — nobody paid for it, and the only thing named is a free federal filing system. This is not investment advice and nothing here is a view on the share. 231J Ramparts Road, Te Anau 9600, New Zealand · Unsubscribe anytime.