PulteGroup: How much land should the builder own?
Download case PDF (5 pages)Request teacher’s pack
Strategy case for MBA students and working professionals | Hypothetical capital-allocation review: July 2026 | Information cutoff: July 22, 2026, 4:07:44 p.m. ET
A flexible pipeline, with a cost to waiting
In July 2026, PulteGroup (PHM) had 228,241 lots under its control: 102,323 owned and 125,918 optioned. Management said options let the company defer land acquisition until it decides whether and when to exercise, reducing risks associated with holding land for a long time. At the same time, management described competition for well-positioned land as robust. A land option can preserve flexibility, but a site not secured or exercised at the right time may be unavailable later.
The filings do not identify a land shortage, a missed sale caused by delayed land acquisition, or a market where PHM should buy more. They also do not show that owned lots earn better returns than optioned lots. The board’s hypothetical capital-allocation review is therefore a policy question: how should PHM decide when to continue using options, when to purchase or exercise sooner, and when to walk away?
The decision has become more salient as the controlled-lot portfolio shifted. Between December 31, 2025 and June 30, 2026, owned lots increased modestly while optioned lots declined. Those changes may reflect acquisitions, deliveries, exercise, abandonment, timing, or other factors; the public totals do not isolate the causes. Nor do they establish that either side of the portfolio is at the wrong level.
The business and its operating context
PHM develops and sells homes through its brands and operating markets. At June 30, 2026, the company reported operations in 48 markets across 26 states and six reporting segments. A lot can require acquisition, entitlement, development and infrastructure before it is ready for home construction. The timing and terms differ by parcel and contract.
PHM reported that 37% of its controlled lots were developed at June 30. The share was 50% among owned lots and 26% among optioned lots. These portfolio averages describe development status; they do not establish comparative quality, readiness, cost, expected sales pace or return. A less-developed optioned lot may be at a different stage by design, while an owned lot may carry more capital for longer.
The latest quarter showed both demand and affordability pressures. Q2 2026 home-sale revenue was $3.8 billion, down 11% year over year. Closings fell 8% to 6,997, average selling price declined 3% to $544,000, and gross margin was 25.0%, compared with 27.0% a year earlier. Net new orders increased 6% to 7,536, while average community count rose 8% to 1,074. Management reported differences across segments and connected orders in part to a higher community count. These results do not identify land mix as a cause of revenue, margin or order changes. [C2, C3]
PHM reported $9.7 billion of remaining purchase price under land-option agreements and $717.4 million in deposits, of which $15.8 million was refundable. These amounts call attention to exposure, but they are not equivalent: the remaining purchase price is not necessarily debt or an unavoidable cash outflow, and contract terms, exercise schedules and abandonment costs vary. In 2025, PHM abandoned 24,596 previously controlled lots that no longer met its underwriting criteria. That is evidence that the company has exited land positions; it does not mean every option is freely cancellable or that every abandoned lot would have been unprofitable. [C1, C2]
A decision under uncertainty and competition
The trade-off is not simply flexibility versus commitment. Maintaining options can defer cash deployment and preserve an opportunity to reassess demand, feasibility and costs. But options require deposits and pre-acquisition spending, may include scheduled takedowns, and can leave PHM without a site if it waits too long or another buyer secures it. Buying sooner can improve control over a selected parcel, while increasing capital tied up before homes are sold. Under affordability pressure, committing earlier may also expose the company to slower sales or lower prices.
Academic work offers a mechanism, not a forecast for PHM. A fully read but earlier 2002 working-paper version of Bulan, Mayer and Somerville models irreversible real-estate investment and reports that uncertainty can delay development; its Vancouver condominium estimates also suggest that competition attenuates some of the estimated delay effect. The estimates are provisional, version-specific and non-load-bearing here. The setting differs from PHM’s U.S. homebuilding operations, and the model’s irreversible investment assumption does not describe PHM’s varied land contracts. No coefficient should be applied to PHM’s lot decisions. [A1]
Industry peers illustrate different structures, but not a performance ranking. D.R. Horton reported 147,000 owned and 444,900 contract-controlled lots at September 30, 2025. Its contract-controlled total included 40,400 lots owned by its 62%-owned public affiliate Forestar: 22,800 were under purchase contract for D.R. Horton homebuilding and 17,600 had a right of first offer. Forestar sold 11,751 lots to D.R. Horton during fiscal 2025. These categories and the affiliate relationship are not equivalent to PHM’s optioned-lot count. [C4]
Lennar’s land-light structure provides another example. In February 2025, Lennar spun off about 80% of Millrose after contributing land assets and cash. It temporarily retained about 20%, then exchanged or disposed of approximately that interest in November 2025, retaining only an immaterial amount of forfeited shares at year-end. Millrose was a separate public company. This is an external land-financing architecture described by Lennar, not a like-for-like lot-control ratio or evidence of superior economics. [C5]
Three policy directions
The review team has not prescribed a new target mix. It is considering three broad directions for discussion:
| Direction | Policy logic | Questions for PHM |
|---|---|---|
| Maintain option-first discipline | Keep using options as the default, with existing underwriting and exercise procedures. | Does the policy preserve adequate access to well-positioned sites? Which contract costs and lost opportunities should be tracked? |
| Selectively own or exercise sooner | Increase commitment only where a market-specific case supports securing control earlier. | What evidence would justify the additional cash and carrying exposure? How should affordability, competition, entitlement and expected sales pace enter the decision? |
| Raise hurdles and let the pipeline contract | Tighten return or demand requirements and allow more opportunities to expire or be abandoned. | Could a smaller pipeline constrain future community availability? How can PHM distinguish prudent discipline from avoidable loss of access? |
The choices need not imply one company-wide ratio. A policy could vary by market, parcel, stage of development and contract terms. But the information available publicly is aggregated. It does not disclose community-level option conversion rates, deposits at risk by contract, option expiration dates, competitive bids, lost sites, market-level land costs, expected returns, or the sales impact of insufficient community supply.
Your assignment
Assume you advise PHM’s capital-allocation committee in July 2026. Recommend a policy for deciding which sites to keep optioned, which to purchase or exercise sooner, and which to abandon or decline.
Your recommendation should:
- Explain how competition and affordability affect the value of waiting and the cost of committing.
- Set out a decision rule that can vary across markets or parcels without claiming a PHM-specific return advantage that the evidence does not show.
- Identify the operational and financial information needed to apply the rule, including contract terms, site readiness, demand, timing and alternatives.
- Propose a small set of indicators for monitoring access to land, cash exposure, community availability and realized outcomes, and explain what each indicator cannot establish by itself.
- Describe how PHM could learn from future decisions while keeping any pilot or comparison interpretable.
You may retain the current posture, change it selectively, or propose another approach. Treat reported results as context, not proof that lot mix caused performance. Do not use the Bulan estimates as PHM inputs or infer that either ownership or options are generally superior.
Exhibit 1. PHM controlled-lot position
| Date | Owned lots | Optioned lots | Total controlled | Owned / optioned mix | Developed share: owned / optioned |
|---|---|---|---|---|---|
| December 31, 2025 | 101,104 | 133,528 | 234,632 | 43% / 57% | 50% / 25% |
| June 30, 2026 | 102,323 | 125,918 | 228,241 | 45% / 55% | 50% / 26% |
| Six-month change | +1,219 | −7,610 | −6,391 | — | — |
Source: C2, “Controlled lots,” p. 27. Calculated changes are simple differences between reported counts. “Controlled” combines legally and economically different positions. The figures do not identify why lots entered or left the portfolio.
Exhibit 2. Selected operating and cash-exposure context
| Measure | Q2 or June 2026 | Comparison | Interpretation limit |
|---|---|---|---|
| Home-sale revenue | $3.8bn | −11% year over year | No causal link to lot mix shown. |
| Closings | 6,997 | −8% | Volume across the company, not lot-specific sales. |
| Average selling price | $544,000 | −3% | Aggregate price; product and market mix matter. |
| Gross margin | 25.0% | 27.0% in Q2 2025 | Does not isolate land acquisition structure. |
| Net new orders | 7,536 | +6% | Management cited higher community count among contributing context. |
| Average community count | 1,074 | +8% | Does not show whether land access was too high or too low. |
| Remaining option purchase price | $9.7bn | As of June 30, 2026 | Not debt or necessarily unavoidable future cash outflow. |
| Option deposits | $717.4m | $15.8m refundable | Aggregate; contract-level risk and timing vary. |
| H1 2026 write-offs of deposits and pre-acquisition costs | $11.5m | $15.7m in H1 2025 | Recognized costs, not a measure of total option value or future loss. |
Sources: C2, C3. Results and balances use the reporting dates stated. H1 amounts are six-month figures and should not be annualized without an explicit assumption.
Exhibit 3. Alternative industry structures
| Company / date | Reported structure | Comparison caveat |
|---|---|---|
| PHM, June 30, 2026 | 102,323 owned; 125,918 optioned | Company-defined position; option contracts vary. |
| D.R. Horton, Sept. 30, 2025 | 147,000 owned; 444,900 under land and lot purchase contracts | Includes 40,400 Forestar-owned lots: 22,800 under contract for DHI homebuilding and 17,600 with a right of first offer. Forestar sold DHI 11,751 lots during FY2025. Not equivalent to PHM options. |
| Lennar / Millrose, 2025 | Lennar contributed land and cash to Millrose, distributed about 80% in February, and disposed of approximately its remaining 20% in November | Millrose was a separate public company by year-end; a different financing and control architecture, not evidence of superior economics. |
Sources: C1, C4, C5. Dates, definitions and corporate structures differ. Do not rank these companies by raw owned/controlled percentages.
Source key
- C1. PulteGroup, 2025 Form 10-K, filed February 4, 2026. SEC filing. Item 1, land-position table, p. 7; Item 7, “Land and lot position,” pp. 29–30; Note 8, pp. 68–71.
- C2. PulteGroup, Q2 2026 Form 10-Q, filed July 22, 2026. SEC filing. “Controlled lots,” pp. 27–28; option purchase price and deposits, p. 37; write-offs, p. 26; cash flows, p. 37.
- C3. PulteGroup, Q2 2026 earnings release, July 22, 2026. SEC exhibit. Operating results and management commentary, paragraphs 1–15 and tables.
- C4. D.R. Horton, 2025 Form 10-K, fiscal year ended September 30, 2025. SEC filing. Lot-position table and Forestar footnotes, p. 41; Forestar sales table, p. 45.
- C5. Lennar, 2025 Form 10-K, fiscal year ended November 30, 2025. SEC filing. Millrose spin-off and exchange offer, pp. 4, 36, 62–63.
- A1. Bulan, L., C. J. Mayer, and C. T. Somerville, “Irreversible Investment, Real Options, and Competition: Evidence from Real Estate Development,” Columbia-hosted manuscript dated November 26, 2002. PDF. Sections III–IV and Tables 2, 4–5. The version-specific results are provisional and non-load-bearing; the accessible manuscript was not verified against the final 2009 article.
This case uses evidence public by July 22, 2026, 4:07:44 p.m. ET. It contains a hypothetical committee decision and no later company outcomes.